diff options
Diffstat (limited to 'contrib/libucl')
153 files changed, 26247 insertions, 0 deletions
diff --git a/contrib/libucl/CMakeLists.txt b/contrib/libucl/CMakeLists.txt new file mode 100644 index 0000000..ae28a89 --- /dev/null +++ b/contrib/libucl/CMakeLists.txt @@ -0,0 +1,260 @@ +PROJECT(libucl C) +CMAKE_MINIMUM_REQUIRED(VERSION 2.6.0 FATAL_ERROR) + +SET(LIBUCL_VERSION_MAJOR 0) +SET(LIBUCL_VERSION_MINOR 5) +SET(LIBUCL_VERSION_PATCH 0) + +SET(LIBUCL_VERSION + "${LIBUCL_VERSION_MAJOR}.${LIBUCL_VERSION_MINOR}.${LIBUCL_VERSION_PATCH}") + +INCLUDE(CheckCCompilerFlag) +INCLUDE(FindOpenSSL) + +OPTION(ENABLE_URL_INCLUDE "Enable urls in ucl includes (requires libcurl or libfetch) [default: OFF]" OFF) +OPTION(ENABLE_URL_SIGN "Enable signatures check in ucl includes (requires openssl) [default: OFF]" OFF) +OPTION(BUILD_SHARED_LIBS "Build Shared Libraries [default: OFF]" OFF) +OPTION(ENABLE_LUA "Enable lua support [default: OFF]" OFF) +OPTION(ENABLE_LUAJIT "Enable luajit support [default: OFF]" OFF) + +# Find lua installation +MACRO(FindLua) + # Find lua libraries + UNSET(LUA_INCLUDE_DIR CACHE) + UNSET(LUA_LIBRARY CACHE) + CMAKE_PARSE_ARGUMENTS(LUA "" "VERSION_MAJOR;VERSION_MINOR;ROOT" "" ${ARGN}) + + IF(NOT LUA_VERSION_MAJOR OR NOT LUA_VERSION_MINOR) + MESSAGE(FATAL_ERROR "Invalid FindLua invocation: ${ARGN}") + ENDIF() + + IF(ENABLE_LUAJIT MATCHES "ON") + MESSAGE(STATUS "Check for luajit ${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}") + FIND_PATH(LUA_INCLUDE_DIR luajit.h + HINTS + "${RSPAMD_SEARCH_PATH}" "${LUA_ROOT}" + $ENV{LUA_DIR} + PATH_SUFFIXES "include/luajit-2.0" + "include/luajit${LUA_VERSION_MAJOR}${LUA_VERSION_MINOR}" + "include/luajit${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + "include/luajit-${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + "include/luajit" + "include/lua${LUA_VERSION_MAJOR}${LUA_VERSION_MINOR}" + "include/lua${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + "include/lua-${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + include/lua include + PATHS ${RSPAMD_DEFAULT_INCLUDE_PATHS} + ) + FIND_LIBRARY(LUA_LIBRARY + NAMES luajit + "luajit-2.0" + "luajit2.0" + "luajit${LUA_VERSION_MAJOR}${LUA_VERSION_MINOR}" + "luajit${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + "luajit-${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + HINTS + "${RSPAMD_SEARCH_PATH}" "${LUA_ROOT}" + $ENV{LUA_DIR} + PATH_SUFFIXES lib64 lib + PATHS ${RSPAMD_DEFAULT_LIBRARY_PATHS} + DOC "Lua library" + ) + + IF(NOT LUA_LIBRARY OR NOT LUA_INCLUDE_DIR) + MESSAGE(STATUS "Fallback from luajit to plain lua") + SET(ENABLE_LUAJIT "OFF") + MESSAGE(STATUS "Check for lua ${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}") + FIND_PATH(LUA_INCLUDE_DIR lua.h + HINTS + "${RSPAMD_SEARCH_PATH}" "${LUA_ROOT}" + $ENV{LUA_DIR} + PATH_SUFFIXES "include/lua${LUA_VERSION_MAJOR}${LUA_VERSION_MINOR}" + "include/lua${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + "include/lua-${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + include/lua include + PATHS ${RSPAMD_DEFAULT_INCLUDE_PATHS} + ) + FIND_LIBRARY(LUA_LIBRARY + NAMES lua + "lua${LUA_VERSION_MAJOR}${LUA_VERSION_MINOR}" + "lua${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + "lua-${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + HINTS + "${RSPAMD_SEARCH_PATH}" "${LUA_ROOT}" + $ENV{LUA_DIR} + PATH_SUFFIXES lib64 lib + PATHS ${RSPAMD_DEFAULT_LIBRARY_PATHS} + DOC "Lua library" + ) + ENDIF() + ELSE(ENABLE_LUAJIT MATCHES "ON") + MESSAGE(STATUS "Check for lua ${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}") + FIND_PATH(LUA_INCLUDE_DIR lua.h + HINTS + "${RSPAMD_SEARCH_PATH}" "${LUA_ROOT}" + $ENV{LUA_DIR} + PATH_SUFFIXES "include/lua${LUA_VERSION_MAJOR}${LUA_VERSION_MINOR}" + "include/lua${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + "include/lua-${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + include/lua include + PATHS ${RSPAMD_DEFAULT_INCLUDE_PATHS} + ) + FIND_LIBRARY(LUA_LIBRARY + NAMES lua + "lua${LUA_VERSION_MAJOR}${LUA_VERSION_MINOR}" + "lua${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + "lua-${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}" + HINTS + "${RSPAMD_SEARCH_PATH}" "${LUA_ROOT}" + $ENV{LUA_DIR} + PATH_SUFFIXES lib64 lib + PATHS ${RSPAMD_DEFAULT_LIBRARY_PATHS} + DOC "Lua library" + ) + ENDIF(ENABLE_LUAJIT MATCHES "ON") + + IF(LUA_LIBRARY AND LUA_INCLUDE_DIR) + SET(LUA_FOUND 1) + IF(NOT LUA_VERSION_MAJOR OR NOT LUA_VERSION_MINOR) + SET(LUA_VERSION_MAJOR ${LUA_VERSION_MAJOR}) + SET(LUA_VERSION_MINOR ${LUA_VERSION_MINOR}) + ENDIF(NOT LUA_VERSION_MAJOR OR NOT LUA_VERSION_MINOR) + IF(ENABLE_LUAJIT MATCHES "ON") + MESSAGE(STATUS "Found luajit ${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}") + ELSE(ENABLE_LUAJIT MATCHES "ON") + MESSAGE(STATUS "Found lua ${LUA_VERSION_MAJOR}.${LUA_VERSION_MINOR}") + ENDIF(ENABLE_LUAJIT MATCHES "ON") + ENDIF(LUA_LIBRARY AND LUA_INCLUDE_DIR) +ENDMACRO() + +IF(CMAKE_SYSTEM_NAME STREQUAL "Linux") + LIST(APPEND CMAKE_REQUIRED_LIBRARIES rt) +ENDIF(CMAKE_SYSTEM_NAME STREQUAL "Linux") + +IF(ENABLE_URL_INCLUDE MATCHES "ON") + FIND_LIBRARY(LIBFETCH_LIBRARY NAMES fetch PATHS PATH_SUFFIXES lib64 lib + PATHS + ~/Library/Frameworks + /Library/Frameworks + /usr/local + /usr + /sw + /opt/local + /opt/csw + /opt + DOC "Path where the libfetch library can be found") + IF(LIBFETCH_LIBRARY) + FIND_FILE(HAVE_FETCH_H NAMES fetch.h PATHS /usr/include + /opt/include + /usr/local/include + DOC "Path to libfetch header") + ELSE(LIBFETCH_LIBRARY) + # Try to find libcurl + ProcessPackage(CURL libcurl) + IF(NOT CURL_FOUND) + MESSAGE(WARNING "Neither libcurl nor libfetch were found, no support of URL includes in configuration") + ENDIF(NOT CURL_FOUND) + ENDIF(LIBFETCH_LIBRARY) +ENDIF(ENABLE_URL_INCLUDE MATCHES "ON") + +SET(CMAKE_C_WARN_FLAGS "") +CHECK_C_COMPILER_FLAG(-Wall SUPPORT_WALL) +CHECK_C_COMPILER_FLAG(-W SUPPORT_W) +CHECK_C_COMPILER_FLAG(-Wno-unused-parameter SUPPORT_WPARAM) +CHECK_C_COMPILER_FLAG(-Wno-pointer-sign SUPPORT_WPOINTER_SIGN) +CHECK_C_COMPILER_FLAG(-Wstrict-prototypes SUPPORT_WSTRICT_PROTOTYPES) +IF(NOT "${CMAKE_C_COMPILER_ID}" MATCHES SunPro) + CHECK_C_COMPILER_FLAG("-std=c99" SUPPORT_STD_FLAG) +ENDIF(NOT "${CMAKE_C_COMPILER_ID}" MATCHES SunPro) +IF(SUPPORT_W) + SET(CMAKE_C_WARN_FLAGS "${CMAKE_C_WARN_FLAGS} -W") +ENDIF(SUPPORT_W) +IF(SUPPORT_WALL) + SET(CMAKE_C_WARN_FLAGS "${CMAKE_C_WARN_FLAGS} -Wall") +ENDIF(SUPPORT_WALL) +IF(SUPPORT_WPARAM) + SET(CMAKE_C_WARN_FLAGS "${CMAKE_C_WARN_FLAGS} -Wno-unused-parameter") +ENDIF(SUPPORT_WPARAM) +IF(SUPPORT_WPOINTER_SIGN) + SET(CMAKE_C_WARN_FLAGS "${CMAKE_C_WARN_FLAGS} -Wno-pointer-sign") +ENDIF(SUPPORT_WPOINTER_SIGN) +IF(SUPPORT_WSTRICT_PROTOTYPES) + SET(CMAKE_C_WARN_FLAGS "${CMAKE_C_WARN_FLAGS} -Wstrict-prototypes") +ENDIF(SUPPORT_WSTRICT_PROTOTYPES) +IF(SUPPORT_STD_FLAG) + SET(CMAKE_C_WARN_FLAGS "${CMAKE_C_WARN_FLAGS} -std=c99") +ENDIF(SUPPORT_STD_FLAG) + +IF(ENABLE_URL_SIGN MATCHES "ON") + IF(OPENSSL_FOUND) + SET(HAVE_OPENSSL 1) + INCLUDE_DIRECTORIES("${OPENSSL_INCLUDE_DIR}") + ENDIF(OPENSSL_FOUND) +ENDIF(ENABLE_URL_SIGN MATCHES "ON") + +INCLUDE_DIRECTORIES("src") +INCLUDE_DIRECTORIES("include") +INCLUDE_DIRECTORIES("uthash") +INCLUDE_DIRECTORIES("klib") + +SET(UCLSRC src/ucl_util.c + src/ucl_parser.c + src/ucl_emitter.c + src/ucl_emitter_streamline.c + src/ucl_emitter_utils.c + src/ucl_hash.c + src/ucl_schema.c + src/ucl_msgpack.c + src/ucl_sexp.c + src/xxhash.c) + + +SET (LIB_TYPE STATIC) +IF (BUILD_SHARED_LIBS) + SET (LIB_TYPE SHARED) +ENDIF (BUILD_SHARED_LIBS) +ADD_LIBRARY(ucl ${LIB_TYPE} ${UCLSRC}) +SET_TARGET_PROPERTIES(ucl PROPERTIES VERSION ${LIBUCL_VERSION} SOVERSION ${LIBUCL_VERSION_MAJOR}) + +IF(ENABLE_LUA MATCHES "ON") + IF(ENABLE_LUAJIT MATCHES "ON") + FindLua(VERSION_MAJOR "5" VERSION_MINOR "1" ROOT "${LUA_ROOT}") + IF(NOT LUA_FOUND) + MESSAGE(FATAL_ERROR "Lua not found, lua support is required") + ELSE(NOT LUA_FOUND) + INCLUDE_DIRECTORIES("${LUA_INCLUDE_DIR}") + ENDIF(NOT LUA_FOUND) + ELSE(ENABLE_LUAJIT MATCHES "ON") + FindLua(VERSION_MAJOR "5" VERSION_MINOR "2" ROOT "${LUA_ROOT}") + IF(NOT LUA_FOUND) + FindLua(VERSION_MAJOR "5" VERSION_MINOR "1" ROOT "${LUA_ROOT}") + ENDIF(NOT LUA_FOUND) + IF(NOT LUA_FOUND) + MESSAGE(FATAL_ERROR "Lua not found, lua support is required") + ELSE(NOT LUA_FOUND) + INCLUDE_DIRECTORIES("${LUA_INCLUDE_DIR}") + ENDIF(NOT LUA_FOUND) + ENDIF(ENABLE_LUAJIT MATCHES "ON") + SET(UCL_LUA_SRC lua/lua_ucl.c) + ADD_LIBRARY(lua-ucl ${LIB_TYPE} ${UCL_LUA_SRC}) + IF(ENABLE_LUAJIT MATCHES "ON") + TARGET_LINK_LIBRARIES(lua-ucl "${LUAJIT_LIBRARY}") + ELSE(ENABLE_LUAJIT MATCHES "ON") + TARGET_LINK_LIBRARIES(lua-ucl "${LUA_LIBRARY}") + ENDIF(ENABLE_LUAJIT MATCHES "ON") + TARGET_LINK_LIBRARIES(lua-ucl ucl) + SET_TARGET_PROPERTIES(lua-ucl PROPERTIES VERSION ${LIBUCL_VERSION} SOVERSION ${LIBUCL_VERSION_MAJOR}) +ENDIF() + +IF(HAVE_FETCH_H) + TARGET_LINK_LIBRARIES(ucl fetch) +ELSE(HAVE_FETCH_H) + IF(CURL_FOUND) + TARGET_LINK_LIBRARIES(ucl ${CURL_LIBRARIES}) + ENDIF(CURL_FOUND) +ENDIF(HAVE_FETCH_H) +IF(ENABLE_URL_SIGN MATCHES "ON") + IF(OPENSSL_FOUND) + TARGET_LINK_LIBRARIES(ucl ${OPENSSL_LIBRARIES}) + ENDIF(OPENSSL_FOUND) +ENDIF(ENABLE_URL_SIGN MATCHES "ON") diff --git a/contrib/libucl/COPYING b/contrib/libucl/COPYING new file mode 100644 index 0000000..33048e1 --- /dev/null +++ b/contrib/libucl/COPYING @@ -0,0 +1,23 @@ +Copyright (c) 2013-2014, Vsevolod Stakhov <vsevolod@highsecure.ru> +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. diff --git a/contrib/libucl/ChangeLog.md b/contrib/libucl/ChangeLog.md new file mode 100644 index 0000000..f714f40 --- /dev/null +++ b/contrib/libucl/ChangeLog.md @@ -0,0 +1,39 @@ +# Version history + +## Libucl 0.5 + +- Streamline emitter has been added, so it is now possible to output partial `ucl` objects +- Emitter now is more flexible due to emitter_context structure + +### 0.5.1 +- Fixed number of bugs and memory leaks + +### 0.5.2 + +- Allow userdata objects to be emitted and destructed +- Use userdata objects to store lua function references + +### Libucl 0.6 + +- Reworked macro interface + +### Libucl 0.6.1 + +- Various utilities fixes + +### Libucl 0.7.0 + +- Move to klib library from uthash to reduce memory overhead and increase performance + +### Libucl 0.7.1 + +- Added safe iterators API + +### Libucl 0.7.2 + +- Fixed serious bugs in schema and arrays iteration + +### Libucl 0.7.3 + +- Fixed a bug with macroes that come after an empty object +- Fixed a bug in include processing when an incorrect variable has been destroyed (use-after-free) diff --git a/contrib/libucl/Makefile.am b/contrib/libucl/Makefile.am new file mode 100644 index 0000000..957e3a2 --- /dev/null +++ b/contrib/libucl/Makefile.am @@ -0,0 +1,11 @@ +ACLOCAL_AMFLAGS = -I m4 +EXTRA_DIST = uthash klib README.md + +pkgconfigdir = $(libdir)/pkgconfig +pkgconfig_DATA = libucl.pc + +if LUA_SUB + LUA_SUBDIR = lua +endif + +SUBDIRS = src tests utils doc $(LUA_SUBDIR)
\ No newline at end of file diff --git a/contrib/libucl/Makefile.unix b/contrib/libucl/Makefile.unix new file mode 100644 index 0000000..121403d --- /dev/null +++ b/contrib/libucl/Makefile.unix @@ -0,0 +1,92 @@ +CC ?= gcc +DESTDIR ?= /usr/local +LD ?= gcc +C_COMMON_FLAGS ?= -fPIC -Wall -W -Wno-unused-parameter -Wno-pointer-sign -I./include -I./uthash -I./src +MAJOR_VERSION = 0 +MINOR_VERSION = 2 +PATCH_VERSION = 9 +VERSION = "$(MAJOR_VERSION).$(MINOR_VERSION).$(PATCH_VERSION)" +SONAME = libucl.so +SONAME_FULL = $(SONAME).$(MAJOR_VERSION) +OBJDIR ?= .obj +TESTDIR ?= tests +SRCDIR ?= src +INCLUDEDIR ?= include +MKDIR ?= mkdir +INSTALL ?= install +RM ?= rm +RMDIR ?= rmdir +LN ?= ln +LD_SHARED_FLAGS ?= -Wl,-soname,$(SONAME) -shared -lm +LD_UCL_FLAGS ?= -L$(OBJDIR) -Wl,-rpath,$(OBJDIR) -lucl +LD_ADD ?= -lrt +COPT_FLAGS ?= -O2 +HDEPS = $(SRCDIR)/ucl_hash.h \ + $(SRCDIR)/ucl_chartable.h \ + $(SRCDIR)/ucl_internal.h \ + $(INCLUDEDIR)/ucl.h \ + $(SRCDIR)/xxhash.h +OBJECTS = $(OBJDIR)/ucl_hash.o \ + $(OBJDIR)/ucl_util.o \ + $(OBJDIR)/ucl_parser.o \ + $(OBJDIR)/ucl_emitter.o \ + $(OBJDIR)/ucl_schema.o \ + $(OBJDIR)/xxhash.o + +all: $(OBJDIR) $(OBJDIR)/$(SONAME) + +$(OBJDIR)/$(SONAME): $(OBJDIR)/$(SONAME_FULL) + $(LN) -sf $(SONAME_FULL) $(OBJDIR)/$(SONAME) + +$(OBJDIR)/$(SONAME_FULL): $(OBJECTS) + $(CC) -o $(OBJDIR)/$(SONAME_FULL) $(OBJECTS) $(LD_SHARED_FLAGS) $(LDFLAGS) $(SSL_LIBS) $(FETCH_LIBS) + +$(OBJDIR): + @$(MKDIR) -p $(OBJDIR) + +# Compile rules +$(OBJDIR)/ucl_util.o: $(SRCDIR)/ucl_util.c $(HDEPS) + $(CC) -o $(OBJDIR)/ucl_util.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/ucl_util.c +$(OBJDIR)/ucl_parser.o: $(SRCDIR)/ucl_parser.c $(HDEPS) + $(CC) -o $(OBJDIR)/ucl_parser.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/ucl_parser.c +$(OBJDIR)/ucl_emitter.o: $(SRCDIR)/ucl_emitter.c $(HDEPS) + $(CC) -o $(OBJDIR)/ucl_emitter.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/ucl_emitter.c +$(OBJDIR)/ucl_hash.o: $(SRCDIR)/ucl_hash.c $(HDEPS) + $(CC) -o $(OBJDIR)/ucl_hash.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/ucl_hash.c +$(OBJDIR)/ucl_schema.o: $(SRCDIR)/ucl_schema.c $(HDEPS) + $(CC) -o $(OBJDIR)/ucl_schema.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/ucl_schema.c +$(OBJDIR)/xxhash.o: $(SRCDIR)/xxhash.c $(HDEPS) + $(CC) -o $(OBJDIR)/xxhash.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/xxhash.c + +clean: + $(RM) $(OBJDIR)/*.o $(OBJDIR)/$(SONAME_FULL) $(OBJDIR)/$(SONAME) $(OBJDIR)/chargen $(OBJDIR)/test_basic $(OBJDIR)/test_speed $(OBJDIR)/objdump $(OBJDIR)/test_generate $(OBJDIR)/test_schema || true + $(RMDIR) $(OBJDIR) + +# Utils + +chargen: utils/chargen.c $(OBJDIR)/$(SONAME) + $(CC) -o $(OBJDIR)/chargen $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) $(LDFLAGS) utils/chargen.c +objdump: utils/objdump.c $(OBJDIR)/$(SONAME) + $(CC) -o $(OBJDIR)/objdump $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) $(LDFLAGS) utils/objdump.c $(LD_UCL_FLAGS) + +# Tests + +test: $(OBJDIR) $(OBJDIR)/$(SONAME) $(OBJDIR)/test_basic $(OBJDIR)/test_speed $(OBJDIR)/test_generate $(OBJDIR)/test_schema + +run-test: test + TEST_DIR=$(TESTDIR) $(TESTDIR)/run_tests.sh $(OBJDIR)/test_basic $(OBJDIR)/test_speed $(OBJDIR)/test_generate $(OBJDIR)/test_schema + +$(OBJDIR)/test_basic: $(TESTDIR)/test_basic.c $(OBJDIR)/$(SONAME) + $(CC) -o $(OBJDIR)/test_basic $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) $(LDFLAGS) $(TESTDIR)/test_basic.c $(LD_UCL_FLAGS) +$(OBJDIR)/test_schema: $(TESTDIR)/test_schema.c $(OBJDIR)/$(SONAME) + $(CC) -o $(OBJDIR)/test_schema $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) $(LDFLAGS) $(TESTDIR)/test_schema.c $(LD_UCL_FLAGS) +$(OBJDIR)/test_speed: $(TESTDIR)/test_speed.c $(OBJDIR)/$(SONAME) + $(CC) -o $(OBJDIR)/test_speed $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) $(LDFLAGS) $(TESTDIR)/test_speed.c $(LD_UCL_FLAGS) $(LD_ADD) +$(OBJDIR)/test_generate: $(TESTDIR)/test_generate.c $(OBJDIR)/$(SONAME) + $(CC) -o $(OBJDIR)/test_generate $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) $(LDFLAGS) $(TESTDIR)/test_generate.c $(LD_UCL_FLAGS) $(LD_ADD) + +install: $(OBJDIR)/$(SONAME) + $(INSTALL) -m0755 $(SONAME) $(DESTDIR)/lib/$(SONAME) + $(INSTALL) -m0644 include/ucl.h $(DESTDIR)/include/ucl.h + +.PHONY: clean $(OBJDIR) diff --git a/contrib/libucl/Makefile.w32 b/contrib/libucl/Makefile.w32 new file mode 100644 index 0000000..0e61274 --- /dev/null +++ b/contrib/libucl/Makefile.w32 @@ -0,0 +1,93 @@ +CC ?= gcc +DESTDIR ?= /usr/local +LD ?= gcc +C_COMMON_FLAGS ?= -fPIC -Wall -W -Wno-unused-parameter -Wno-pointer-sign -I./include -I./uthash -I./src +MAJOR_VERSION = 0 +MINOR_VERSION = 2 +PATCH_VERSION = 9 +VERSION = "$(MAJOR_VERSION).$(MINOR_VERSION).$(PATCH_VERSION)" +SONAME = libucl.dll +OBJDIR ?= .obj +TESTDIR ?= tests +SRCDIR ?= src +INCLUDEDIR ?= include +MKDIR ?= mkdir +INSTALL ?= install +RM ?= rm +RMDIR ?= rmdir +ifeq (Windows_NT, $(OS)) +LN ?= ln +else +LN ?= rem ln +endif +LD_SHARED_FLAGS ?= -Wl,-soname,$(SONAME) -shared -lm +LD_UCL_FLAGS ?= -L$(OBJDIR) -Wl,-rpath,$(OBJDIR) -lucl +LD_ADD ?= -lrt +COPT_FLAGS ?= -O2 +HDEPS = $(SRCDIR)/ucl_hash.h \ + $(SRCDIR)/ucl_chartable.h \ + $(SRCDIR)/ucl_internal.h \ + $(INCLUDEDIR)/ucl.h \ + $(SRCDIR)/xxhash.h +OBJECTS = $(OBJDIR)/ucl_hash.o \ + $(OBJDIR)/ucl_util.o \ + $(OBJDIR)/ucl_parser.o \ + $(OBJDIR)/ucl_emitter.o \ + $(OBJDIR)/ucl_emitter_utils.o \ + $(OBJDIR)/ucl_schema.o \ + $(OBJDIR)/xxhash.o + +all: $(OBJDIR) $(OBJDIR)/$(SONAME) + +$(OBJDIR)/$(SONAME): $(OBJECTS) + $(CC) -o $(OBJDIR)/$(SONAME) $(OBJECTS) $(LD_SHARED_FLAGS) $(LDFLAGS) $(SSL_LIBS) $(FETCH_LIBS) + +$(OBJDIR): + @$(MKDIR) -p $(OBJDIR) + +# Compile rules +$(OBJDIR)/ucl_util.o: $(SRCDIR)/ucl_util.c $(HDEPS) + $(CC) -o $(OBJDIR)/ucl_util.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/ucl_util.c +$(OBJDIR)/ucl_parser.o: $(SRCDIR)/ucl_parser.c $(HDEPS) + $(CC) -o $(OBJDIR)/ucl_parser.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/ucl_parser.c +$(OBJDIR)/ucl_emitter.o: $(SRCDIR)/ucl_emitter.c $(HDEPS) + $(CC) -o $(OBJDIR)/ucl_emitter.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/ucl_emitter.c +$(OBJDIR)/ucl_emitter_utils.o: $(SRCDIR)/ucl_emitter_utils.c $(HDEPS) + $(CC) -o $(OBJDIR)/ucl_emitter_utils.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/ucl_emitter_utils.c +$(OBJDIR)/ucl_hash.o: $(SRCDIR)/ucl_hash.c $(HDEPS) + $(CC) -o $(OBJDIR)/ucl_hash.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/ucl_hash.c +$(OBJDIR)/ucl_schema.o: $(SRCDIR)/ucl_schema.c $(HDEPS) + $(CC) -o $(OBJDIR)/ucl_schema.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/ucl_schema.c +$(OBJDIR)/xxhash.o: $(SRCDIR)/xxhash.c $(HDEPS) + $(CC) -o $(OBJDIR)/xxhash.o $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) -c $(SRCDIR)/xxhash.c + +clean: + $(RM) $(OBJDIR)/*.o $(OBJDIR)/$(SONAME) $(OBJDIR)/$(SONAME) $(OBJDIR)/chargen $(OBJDIR)/test_basic $(OBJDIR)/test_speed $(OBJDIR)/objdump $(OBJDIR)/test_generate + $(RMDIR) $(OBJDIR) + +# Utils + +chargen: utils/chargen.c $(OBJDIR)/$(SONAME) + $(CC) -o $(OBJDIR)/chargen $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) $(LDFLAGS) utils/chargen.c +objdump: utils/objdump.c $(OBJDIR)/$(SONAME) + $(CC) -o $(OBJDIR)/objdump $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) $(LDFLAGS) utils/objdump.c $(LD_UCL_FLAGS) + +# Tests + +test: $(OBJDIR) $(OBJDIR)/$(SONAME) $(OBJDIR)/test_basic $(OBJDIR)/test_speed $(OBJDIR)/test_generate + +run-test: test + TEST_DIR=$(TESTDIR) $(TESTDIR)/run_tests.sh $(OBJDIR)/test_basic $(OBJDIR)/test_speed $(OBJDIR)/test_generate + +$(OBJDIR)/test_basic: $(TESTDIR)/test_basic.c $(OBJDIR)/$(SONAME) + $(CC) -o $(OBJDIR)/test_basic $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) $(LDFLAGS) $(TESTDIR)/test_basic.c $(LD_UCL_FLAGS) +$(OBJDIR)/test_speed: $(TESTDIR)/test_speed.c $(OBJDIR)/$(SONAME) + $(CC) -o $(OBJDIR)/test_speed $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) $(LDFLAGS) $(TESTDIR)/test_speed.c $(LD_UCL_FLAGS) $(LD_ADD) +$(OBJDIR)/test_generate: $(TESTDIR)/test_generate.c $(OBJDIR)/$(SONAME) + $(CC) -o $(OBJDIR)/test_generate $(CPPFLAGS) $(COPT_FLAGS) $(CFLAGS) $(C_COMMON_FLAGS) $(SSL_CFLAGS) $(FETCH_FLAGS) $(LDFLAGS) $(TESTDIR)/test_generate.c $(LD_UCL_FLAGS) $(LD_ADD) + +install: $(OBJDIR)/$(SONAME) + $(INSTALL) -m0755 $(SONAME) $(DESTDIR)/lib/$(SONAME) + $(INSTALL) -m0644 include/ucl.h $(DESTDIR)/include/ucl.h + +.PHONY: clean $(OBJDIR) diff --git a/contrib/libucl/README.md b/contrib/libucl/README.md new file mode 100644 index 0000000..a742db1 --- /dev/null +++ b/contrib/libucl/README.md @@ -0,0 +1,382 @@ +# LIBUCL + +[](https://travis-ci.org/vstakhov/libucl)[](https://scan.coverity.com/projects/4138) + +**Table of Contents** *generated with [DocToc](http://doctoc.herokuapp.com/)* + +- [Introduction](#introduction) +- [Basic structure](#basic-structure) +- [Improvements to the json notation](#improvements-to-the-json-notation) + - [General syntax sugar](#general-syntax-sugar) + - [Automatic arrays creation](#automatic-arrays-creation) + - [Named keys hierarchy](#named-keys-hierarchy) + - [Convenient numbers and booleans](#convenient-numbers-and-booleans) +- [General improvements](#general-improvements) + - [Commments](#commments) + - [Macros support](#macros-support) + - [Variables support](#variables-support) + - [Multiline strings](#multiline-strings) +- [Emitter](#emitter) +- [Validation](#validation) +- [Performance](#performance) +- [Conclusion](#conclusion) + +## Introduction + +This document describes the main features and principles of the configuration +language called `UCL` - universal configuration language. + +If you are looking for the libucl API documentation you can find it at [this page](doc/api.md). + +## Basic structure + +UCL is heavily infused by `nginx` configuration as the example of a convenient configuration +system. However, UCL is fully compatible with `JSON` format and is able to parse json files. +For example, you can write the same configuration in the following ways: + +* in nginx like: + +```nginx +param = value; +section { + param = value; + param1 = value1; + flag = true; + number = 10k; + time = 0.2s; + string = "something"; + subsection { + host = { + host = "hostname"; + port = 900; + } + host = { + host = "hostname"; + port = 901; + } + } +} +``` + +* or in JSON: + +```json +{ + "param": "value", + "param1": "value1", + "flag": true, + "subsection": { + "host": [ + { + "host": "hostname", + "port": 900 + }, + { + "host": "hostname", + "port": 901 + } + ] + } +} +``` + +## Improvements to the json notation. + +There are various things that make ucl configuration more convenient for editing than strict json: + +### General syntax sugar + +* Braces are not necessary to enclose a top object: it is automatically treated as an object: + +```json +"key": "value" +``` +is equal to: +```json +{"key": "value"} +``` + +* There is no requirement of quotes for strings and keys, moreover, `:` may be replaced `=` or even be skipped for objects: + +```nginx +key = value; +section { + key = value; +} +``` +is equal to: +```json +{ + "key": "value", + "section": { + "key": "value" + } +} +``` + +* No commas mess: you can safely place a comma or semicolon for the last element in an array or an object: + +```json +{ + "key1": "value", + "key2": "value", +} +``` +### Automatic arrays creation + +* Non-unique keys in an object are allowed and are automatically converted to the arrays internally: + +```json +{ + "key": "value1", + "key": "value2" +} +``` +is converted to: +```json +{ + "key": ["value1", "value2"] +} +``` + +### Named keys hierarchy + +UCL accepts named keys and organize them into objects hierarchy internally. Here is an example of this process: +```nginx +section "blah" { + key = value; +} +section foo { + key = value; +} +``` + +is converted to the following object: + +```nginx +section { + blah { + key = value; + } + foo { + key = value; + } +} +``` + +Plain definitions may be more complex and contain more than a single level of nested objects: + +```nginx +section "blah" "foo" { + key = value; +} +``` + +is presented as: + +```nginx +section { + blah { + foo { + key = value; + } + } +} +``` + +### Convenient numbers and booleans + +* Numbers can have suffixes to specify standard multipliers: + + `[kKmMgG]` - standard 10 base multipliers (so `1k` is translated to 1000) + + `[kKmMgG]b` - 2 power multipliers (so `1kb` is translated to 1024) + + `[s|min|d|w|y]` - time multipliers, all time values are translated to float number of seconds, for example `10min` is translated to 600.0 and `10ms` is translated to 0.01 +* Hexadecimal integers can be used by `0x` prefix, for example `key = 0xff`. However, floating point values can use decimal base only. +* Booleans can be specified as `true` or `yes` or `on` and `false` or `no` or `off`. +* It is still possible to treat numbers and booleans as strings by enclosing them in double quotes. + +## General improvements + +### Commments + +UCL supports different style of comments: + +* single line: `#` +* multiline: `/* ... */` + +Multiline comments may be nested: +```c +# Sample single line comment +/* + some comment + /* nested comment */ + end of comment +*/ +``` + +### Macros support + +UCL supports external macros both multiline and single line ones: +```nginx +.macro "sometext"; +.macro { + Some long text + .... +}; +``` + +Moreover, each macro can accept an optional list of arguments in braces. These +arguments themselves are the UCL object that is parsed and passed to a macro as +options: + +```nginx +.macro(param=value) "something"; +.macro(param={key=value}) "something"; +.macro(.include "params.conf") "something"; +.macro(#this is multiline macro +param = [value1, value2]) "something"; +.macro(key="()") "something"; +``` + +UCL also provide a convenient `include` macro to load content from another files +to the current UCL object. This macro accepts either path to file: + +```nginx +.include "/full/path.conf" +.include "./relative/path.conf" +.include "${CURDIR}/path.conf" +``` + +or URL (if ucl is built with url support provided by either `libcurl` or `libfetch`): + + .include "http://example.com/file.conf" + +`.include` macro supports a set of options: + +* `try` (default: **false**) - if this option is `true` than UCL treats errors on loading of +this file as non-fatal. For example, such a file can be absent but it won't stop the parsing +of the top-level document. +* `sign` (default: **false**) - if this option is `true` UCL loads and checks the signature for +a file from path named `<FILEPATH>.sig`. Trusted public keys should be provided for UCL API after +parser is created but before any configurations are parsed. +* `glob` (default: **false**) - if this option is `true` UCL treats the filename as GLOB pattern and load +all files that matches the specified pattern (normally the format of patterns is defined in `glob` manual page +for your operating system). This option is meaningless for URL includes. +* `url` (default: **true**) - allow URL includes. +* `path` (default: empty) - A UCL_ARRAY of directories to search for the include file. +Search ends after the first patch, unless `glob` is true, then all matches are included. +* `prefix` (default false) - Put included contents inside an object, instead +of loading them into the root. If no `key` is provided, one is automatically generated based on each files basename() +* `key` (default: <empty string>) - Key to load contents of include into. If +the key already exists, it must be the correct type +* `target` (default: object) - Specify if the `prefix` `key` should be an +object or an array. +* `priority` (default: 0) - specify priority for the include (see below). +* `duplicate` (default: 'append') - specify policy of duplicates resolving: + - `append` - default strategy, if we have new object of higher priority then it replaces old one, if we have new object with less priority it is ignored completely, and if we have two duplicate objects with the same priority then we have a multi-value key (implicit array) + - `merge` - if we have object or array, then new keys are merged inside, if we have a plain object then an implicit array is formed (regardeless of priorities) + - `error` - create error on duplicate keys and stop parsing + - `rewrite` - always rewrite an old value with new one (ignoring priorities) + +Priorities are used by UCL parser to manage the policy of objects rewriting during including other files +as following: + +* If we have two objects with the same priority then we form an implicit array +* If a new object has bigger priority then we overwrite an old one +* If a new object has lower priority then we ignore it + +By default, the priority of top-level object is set to zero (lowest priority). Currently, +you can define up to 16 priorities (from 0 to 15). Includes with bigger priorities will +rewrite keys from the objects with lower priorities as specified by the policy. + +### Variables support + +UCL supports variables in input. Variables are registered by a user of the UCL parser and can be presented in the following forms: + +* `${VARIABLE}` +* `$VARIABLE` + +UCL currently does not support nested variables. To escape variables one could use double dollar signs: + +* `$${VARIABLE}` is converted to `${VARIABLE}` +* `$$VARIABLE` is converted to `$VARIABLE` + +However, if no valid variables are found in a string, no expansion will be performed (and `$$` thus remains unchanged). This may be a subject +to change in future libucl releases. + +### Multiline strings + +UCL can handle multiline strings as well as single line ones. It uses shell/perl like notation for such objects: +``` +key = <<EOD +some text +splitted to +lines +EOD +``` + +In this example `key` will be interpreted as the following string: `some text\nsplitted to\nlines`. +Here are some rules for this syntax: + +* Multiline terminator must start just after `<<` symbols and it must consist of capital letters only (e.g. `<<eof` or `<< EOF` won't work); +* Terminator must end with a single newline character (and no spaces are allowed between terminator and newline character); +* To finish multiline string you need to include a terminator string just after newline and followed by a newline (no spaces or other characters are allowed as well); +* The initial and the final newlines are not inserted to the resulting string, but you can still specify newlines at the begin and at the end of a value, for example: + +``` +key <<EOD + +some +text + +EOD +``` + +## Emitter + +Each UCL object can be serialized to one of the three supported formats: + +* `JSON` - canonic json notation (with spaces indented structure); +* `Compacted JSON` - compact json notation (without spaces or newlines); +* `Configuration` - nginx like notation; +* `YAML` - yaml inlined notation. + +## Validation + +UCL allows validation of objects. It uses the same schema that is used for json: [json schema v4](http://json-schema.org). UCL supports the full set of json schema with the exception of remote references. This feature is unlikely useful for configuration objects. Of course, a schema definition can be in UCL format instead of JSON that simplifies schemas writing. Moreover, since UCL supports multiple values for keys in an object it is possible to specify generic integer constraints `maxValues` and `minValues` to define the limits of values count in a single key. UCL currently is not absolutely strict about validation schemas themselves, therefore UCL users should supply valid schemas (as it is defined in json-schema draft v4) to ensure that the input objects are validated properly. + +## Performance + +Are UCL parser and emitter fast enough? Well, there are some numbers. +I got a 19Mb file that consist of ~700 thousands lines of json (obtained via +http://www.json-generator.com/). Then I checked jansson library that performs json +parsing and emitting and compared it with UCL. Here are results: + +``` +jansson: parsed json in 1.3899 seconds +jansson: emitted object in 0.2609 seconds + +ucl: parsed input in 0.6649 seconds +ucl: emitted config in 0.2423 seconds +ucl: emitted json in 0.2329 seconds +ucl: emitted compact json in 0.1811 seconds +ucl: emitted yaml in 0.2489 seconds +``` + +So far, UCL seems to be significantly faster than jansson on parsing and slightly faster on emitting. Moreover, +UCL compiled with optimizations (-O3) performs significantly faster: +``` +ucl: parsed input in 0.3002 seconds +ucl: emitted config in 0.1174 seconds +ucl: emitted json in 0.1174 seconds +ucl: emitted compact json in 0.0991 seconds +ucl: emitted yaml in 0.1354 seconds +``` + +You can do your own benchmarks by running `make check` in libucl top directory. + +## Conclusion + +UCL has clear design that should be very convenient for reading and writing. At the same time it is compatible with +JSON language and therefore can be used as a simple JSON parser. Macroes logic provides an ability to extend configuration +language (for example by including some lua code) and comments allows to disable or enable the parts of a configuration +quickly. diff --git a/contrib/libucl/autogen.sh b/contrib/libucl/autogen.sh new file mode 100755 index 0000000..68f4a17 --- /dev/null +++ b/contrib/libucl/autogen.sh @@ -0,0 +1,2 @@ +#!/bin/sh +autoreconf -i diff --git a/contrib/libucl/configure.ac b/contrib/libucl/configure.ac new file mode 100644 index 0000000..73d3227 --- /dev/null +++ b/contrib/libucl/configure.ac @@ -0,0 +1,185 @@ +m4_define([maj_ver], [0]) +m4_define([med_ver], [7]) +m4_define([min_ver], [3]) +m4_define([so_version], [5:0:2]) +m4_define([ucl_version], [maj_ver.med_ver.min_ver]) + +AC_INIT([libucl],[ucl_version],[https://github.com/vstakhov/libucl],[libucl]) +AC_CONFIG_SRCDIR([configure.ac]) +AM_INIT_AUTOMAKE([1.11 foreign -Wall -Wportability no-dist-gzip dist-xz]) +m4_ifdef([AM_SILENT_RULES], [AM_SILENT_RULES([yes])]) + +UCL_VERSION=ucl_version +SO_VERSION=so_version + +AC_SUBST(UCL_VERSION) +AC_SUBST(SO_VERSION) + +AC_PROG_CC_C99 +AM_PROG_CC_C_O +AM_PROG_AR +LT_INIT +AC_CONFIG_MACRO_DIR([m4]) +AC_CONFIG_HEADERS([config.h]) + +AC_C_CONST +AC_TYPE_SIZE_T + +AC_CHECK_HEADERS_ONCE([fcntl.h unistd.h]) +AC_TYPE_OFF_T +AC_FUNC_MMAP +AC_CHECK_HEADERS_ONCE([fcntl.h]) +AC_CHECK_HEADERS_ONCE([sys/types.h]) +AC_CHECK_HEADERS_ONCE([sys/stat.h]) +AC_CHECK_HEADERS_ONCE([sys/param.h]) +AC_CHECK_HEADERS_ONCE([sys/mman.h]) +AC_CHECK_HEADERS_ONCE([stdlib.h]) +AC_CHECK_HEADERS_ONCE([stddef.h]) +AC_CHECK_HEADERS_ONCE([stdarg.h]) +AC_CHECK_HEADERS_ONCE([stdbool.h]) +AC_CHECK_HEADERS_ONCE([stdint.h]) +AC_CHECK_HEADERS_ONCE([string.h]) +AC_CHECK_HEADERS_ONCE([unistd.h]) +AC_CHECK_HEADERS_ONCE([ctype.h]) +AC_CHECK_HEADERS_ONCE([errno.h]) +AC_CHECK_HEADERS_ONCE([limits.h]) +AC_CHECK_HEADERS_ONCE([libgen.h]) +AC_CHECK_HEADERS_ONCE([stdio.h]) +AC_CHECK_HEADERS_ONCE([float.h]) +AC_CHECK_HEADERS_ONCE([math.h]) +AC_CHECK_HEADERS_ONCE([endian.h sys/endian.h machine/endian.h]) + +dnl Example of default-disabled feature +AC_ARG_ENABLE([urls], AS_HELP_STRING([--enable-urls], + [Enable URLs fetch (requires libfetch or libcurl) @<:@default=no@:>@]), [], + [enable_urls=no]) +AC_ARG_ENABLE([regex], AS_HELP_STRING([--enable-regex], + [Enable regex checking for schema @<:@default=yes@:>@]), [], + [enable_regex=yes]) +AC_ARG_ENABLE([signatures], AS_HELP_STRING([--enable-signatures], + [Enable signatures check (requires openssl) @<:@default=no@:>@]), [], + [enable_signatures=no]) +AC_ARG_ENABLE([lua], AS_HELP_STRING([--enable-lua], + [Enable lua API build (requires lua libraries and headers) @<:@default=no@:>@]), [], + [enable_lua=no]) +AC_ARG_ENABLE([utils], + AS_HELP_STRING([--enable-utils], [Build and install utils @<:@default=no@:>@]), + [case "${enableval}" in + yes) utils=true ;; + no) utils=false ;; + *) AC_MSG_ERROR([bad value ${enableval} for --enable-utils]) ;; + esac],[utils=false]) +AM_CONDITIONAL([UTILS], [test x$utils = xtrue]) + +AS_IF([test "x$enable_signatures" = "xyes"], [ + AC_SEARCH_LIBS([EVP_MD_CTX_create], [crypto], [ + AC_DEFINE(HAVE_OPENSSL, 1, [Define to 1 if you have the 'crypto' library (-lcrypto).]) + LIBCRYPTO_LIB="-lcrypto" + LIBS_EXTRA="${LIBS_EXTRA} -lcrypto" + ], [AC_MSG_ERROR([unable to find the EVP_MD_CTX_create() function])]) +]) +AC_SUBST(LIBCRYPTO_LIB) +AC_PATH_PROG(PANDOC, pandoc, [/non/existent]) + +AC_SEARCH_LIBS([clock_gettime], [rt], [], [ + AC_CHECK_HEADER([mach/mach_time.h], [ + AC_DEFINE(HAVE_MACH_MACH_TIME_H, 1, [Define to 1 on Darwin]) + ], [AC_MSG_ERROR([unable to find clock_gettime or mach_absolute_time])]) +]) +AC_SEARCH_LIBS([remainder], [m], [], [AC_MSG_ERROR([unable to find remainder() function])]) + +AS_IF([test "x$enable_regex" = "xyes"], [ + AC_CHECK_HEADER([regex.h], [ + AC_DEFINE(HAVE_REGEX_H, 1, [Define to 1 if you have the <regex.h> header file.]) + AC_SEARCH_LIBS([regexec], [regex], [ + AS_IF([test "x$ac_cv_search_regexec" = "x-lregex"], [ + LIBREGEX_LIB="-lregex" + LIBS_EXTRA="${LIBS_EXTRA} -lregex" + ] + )], + [AC_MSG_ERROR([unable to find the regexec() function])])], + [AC_MSG_ERROR([unable to find the regex.h header]) + ], + [#include <sys/types.h>]) +]) +AC_SUBST(LIBREGEX_LIB) + +AS_IF([test "x$enable_lua" = "xyes"], [ + AX_PROG_LUA([5.1], [], [ + AX_LUA_HEADERS([ + AX_LUA_LIBS([ + AC_DEFINE(HAVE_LUA, 1, [Define to 1 for lua support.]) + with_lua="yes" + ], [AC_MSG_ERROR([unable to find the lua libraries]) + ]) + ], [AC_MSG_ERROR([unable to find the lua header files]) + ]) + ], [AC_MSG_ERROR([unable to find the lua interpreter])]) +], [with_lua="no"]) + +AM_CONDITIONAL([LUA_SUB], [test "$with_lua" = "yes"]) + +AS_IF([test "x$enable_urls" = "xyes"], [ + AC_CHECK_HEADER([fetch.h], [ + AC_DEFINE(HAVE_FETCH_H, 1, [Define to 1 if you have the <fetch.h> header file.]) + AC_CHECK_LIB(fetch, fetchXGet, [ + AC_DEFINE(HAVE_LIBFETCH, 1, [Define to 1 if you have the 'fetch' library (-lfetch).]) + LIBFETCH_LIBS="-lfetch" + have_libfetch="yes" + LIBS_EXTRA="${LIBS_EXTRA} -lfetch" + ]) + ], [],[ + #include <stdio.h> + #ifdef HAVE_SYS_PARAM_H + #include <sys/param.h> + #endif + ]) + AC_SUBST(LIBFETCH_LIBS) + + AS_IF([ test "x$have_libfetch" != "xyes"], [ + dnl Fallback to libcurl + PKG_CHECK_MODULES([CURL], [libcurl], [ + AC_DEFINE(CURL_FOUND, 1, [Use libcurl]) + LIBS_EXTRA="${LIBS_EXTRA} -lcurl"], + [AC_MSG_ERROR([unable to find neither libfetch nor libcurl])]) + ]) + AC_SUBST(CURL_FOUND) + AC_SUBST(CURL_LIBS) + AC_SUBST(CURL_CFLAGS) +]) + +AC_SUBST(LIBS_EXTRA) + +AC_MSG_CHECKING(for GCC atomic builtins) +AC_LINK_IFELSE([ + AC_LANG_SOURCE([[ + int main() { + volatile unsigned long val = 1; + __sync_synchronize(); + __sync_val_compare_and_swap(&val, 1, 0); + __sync_add_and_fetch(&val, 1); + __sync_sub_and_fetch(&val, 1); + return 0; + } + ]]) +], +[ + AC_MSG_RESULT([yes]) + AC_DEFINE([HAVE_ATOMIC_BUILTINS], [1], [Has gcc/MSVC atomic intrinsics]) +], +[ + AC_MSG_RESULT([no]) + AC_DEFINE([HAVE_ATOMIC_BUILTINS], [0], [Has gcc/MSVC atomic intrinsics]) + AC_MSG_WARN([Libucl references could be thread-unsafe because atomic builtins are missing]) +]) + +AC_CONFIG_FILES(Makefile \ + src/Makefile \ + lua/Makefile + tests/Makefile \ + utils/Makefile \ + doc/Makefile \ + lua/libucl.rockspec \ + libucl.pc) +AC_CONFIG_FILES([stamp-h], [echo timestamp > stamp-h]) +AC_OUTPUT diff --git a/contrib/libucl/doc/Makefile.am b/contrib/libucl/doc/Makefile.am new file mode 100644 index 0000000..dcfacf6 --- /dev/null +++ b/contrib/libucl/doc/Makefile.am @@ -0,0 +1,9 @@ +EXTRA_DIST = api.md + +dist_man_MANS = libucl.3 + +gen-man: @PANDOC@ + tail -n +$$(grep -n '# Synopsis' api.md | cut -d':' -f1) api.md | \ + cat pandoc.template - | sed -e 's/^# \(.*\)/# \U\1/' \ + -e "s/%%date%%/$$(LANG=C date +'%d %B, %Y')/" | \ + @PANDOC@ -s -f markdown -t man -o libucl.3 diff --git a/contrib/libucl/doc/api.md b/contrib/libucl/doc/api.md new file mode 100644 index 0000000..75b954b --- /dev/null +++ b/contrib/libucl/doc/api.md @@ -0,0 +1,495 @@ +# API documentation + +**Table of Contents** *generated with [DocToc](http://doctoc.herokuapp.com/)* + +- [Synopsis](#synopsis) +- [Description](#description) + - [Parser functions](#parser-functions) + - [Emitting functions](#emitting-functions) + - [Conversion functions](#conversion-functions) + - [Generation functions](#generation-functions) + - [Iteration functions](#iteration-functions) + - [Validation functions](#validation-functions) + - [Utility functions](#utility-functions) +- [Parser functions](#parser-functions-1) + - [ucl_parser_new](#ucl_parser_new) + - [ucl_parser_register_macro](#ucl_parser_register_macro) + - [ucl_parser_register_variable](#ucl_parser_register_variable) + - [ucl_parser_add_chunk](#ucl_parser_add_chunk) + - [ucl_parser_add_string](#ucl_parser_add_string) + - [ucl_parser_add_file](#ucl_parser_add_file) + - [ucl_parser_get_object](#ucl_parser_get_object) + - [ucl_parser_get_error](#ucl_parser_get_error) + - [ucl_parser_free](#ucl_parser_free) + - [ucl_pubkey_add](#ucl_pubkey_add) + - [ucl_parser_set_filevars](#ucl_parser_set_filevars) + - [Parser usage example](#parser-usage-example) +- [Emitting functions](#emitting-functions-1) + - [ucl_object_emit](#ucl_object_emit) + - [ucl_object_emit_full](#ucl_object_emit_full) +- [Conversion functions](#conversion-functions-1) +- [Generation functions](#generation-functions-1) + - [ucl_object_new](#ucl_object_new) + - [ucl_object_typed_new](#ucl_object_typed_new) + - [Primitive objects generation](#primitive-objects-generation) + - [ucl_object_fromstring_common](#ucl_object_fromstring_common) +- [Iteration functions](#iteration-functions-1) + - [ucl_iterate_object](#ucl_iterate_object) +- [Validation functions](#validation-functions-1) + - [ucl_object_validate](#ucl_object_validate) + +# Synopsis + +`#include <ucl.h>` + +# Description + +Libucl is a parser and `C` API to parse and generate `ucl` objects. Libucl consist of several groups of functions: + +### Parser functions +Used to parse `ucl` files and provide interface to extract `ucl` object. Currently, `libucl` can parse only full `ucl` documents, for instance, it is impossible to parse a part of document and therefore it is impossible to use `libucl` as a streaming parser. In future, this limitation can be removed. + +### Emitting functions +Convert `ucl` objects to some textual or binary representation. Currently, libucl supports the following exports: + +- `JSON` - valid json format (can possibly lose some original data, such as implicit arrays) +- `Config` - human-readable configuration format (lossless) +- `YAML` - embedded yaml format (has the same limitations as `json` output) + +### Conversion functions +Help to convert `ucl` objects to C types. These functions are used to convert `ucl_object_t` to C primitive types, such as numbers, strings or boolean values. + +### Generation functions +Allow creation of `ucl` objects from C types and creating of complex `ucl` objects, such as hashes or arrays from primitive `ucl` objects, such as numbers or strings. + +### Iteration functions +Iterate over `ucl` complex objects or over a chain of values, for example when a key in an object has multiple values (that can be treated as implicit array or implicit consolidation). + +### Validation functions +Validation functions are used to validate some object `obj` using json-schema compatible object `schema`. Both input and schema must be UCL objects to perform validation. + +### Utility functions +Provide basic utilities to manage `ucl` objects: creating, removing, retaining and releasing reference count and so on. + +# Parser functions + +Parser functions operates with `struct ucl_parser`. + +### ucl_parser_new + +~~~C +struct ucl_parser* ucl_parser_new (int flags); +~~~ + +Creates new parser with the specified flags: + +- `UCL_PARSER_KEY_LOWERCASE` - lowercase keys parsed +- `UCL_PARSER_ZEROCOPY` - try to use zero-copy mode when reading files (in zero-copy mode text chunk being parsed without copying strings so it should exist till any object parsed is used) +- `UCL_PARSER_NO_TIME` - treat time values as strings without parsing them as floats + +### ucl_parser_register_macro + +~~~C +void ucl_parser_register_macro (struct ucl_parser *parser, + const char *macro, ucl_macro_handler handler, void* ud); +~~~ + +Register new macro with name .`macro` parsed by handler `handler` that accepts opaque data pointer `ud`. Macro handler should be of the following type: + +~~~C +bool (*ucl_macro_handler) (const unsigned char *data, + size_t len, void* ud);` +~~~ + +Handler function accepts macro text `data` of length `len` and the opaque pointer `ud`. If macro is parsed successfully the handler should return `true`. `false` indicates parsing failure and the parser can be terminated. + +### ucl_parser_register_variable + +~~~C +void ucl_parser_register_variable (struct ucl_parser *parser, + const char *var, const char *value); +~~~ + +Register new variable $`var` that should be replaced by the parser to the `value` string. + +### ucl_parser_add_chunk + +~~~C +bool ucl_parser_add_chunk (struct ucl_parser *parser, + const unsigned char *data, size_t len); +~~~ + +Add new text chunk with `data` of length `len` to the parser. At the moment, `libucl` parser is not a streamlined parser and chunk *must* contain the *valid* ucl object. For example, this object should be valid: + +~~~json +{ "var": "value" } +~~~ + +while this one won't be parsed correctly: + +~~~json +{ "var": +~~~ + +This limitation may possible be removed in future. + +### ucl_parser_add_string +~~~C +bool ucl_parser_add_string (struct ucl_parser *parser, + const char *data, size_t len); +~~~ + +This function acts exactly like `ucl_parser_add_chunk` does but if `len` argument is zero, then the string `data` must be zero-terminated and the actual length is calculated up to `\0` character. + +### ucl_parser_add_file + +~~~C +bool ucl_parser_add_file (struct ucl_parser *parser, + const char *filename); +~~~ + +Load file `filename` and parse it with the specified `parser`. This function uses `mmap` call to load file, therefore, it should not be `shrunk` during parsing. Otherwise, `libucl` can cause memory corruption and terminate the calling application. This function is also used by the internal handler of `include` macro, hence, this macro has the same limitation. + +### ucl_parser_get_object + +~~~C +ucl_object_t* ucl_parser_get_object (struct ucl_parser *parser); +~~~ + +If the `ucl` data has been parsed correctly this function returns the top object for the parser. Otherwise, this function returns the `NULL` pointer. The reference count for `ucl` object returned is increased by one, therefore, a caller should decrease reference by using `ucl_object_unref` to free object after usage. + +### ucl_parser_get_error + +~~~C +const char *ucl_parser_get_error(struct ucl_parser *parser); +~~~ + +Returns the constant error string for the parser object. If no error occurred during parsing a `NULL` object is returned. A caller should not try to free or modify this string. + +### ucl_parser_free + +~~~C +void ucl_parser_free (struct ucl_parser *parser); +~~~ + +Frees memory occupied by the parser object. The reference count for top object is decreased as well, however if the function `ucl_parser_get_object` was called previously then the top object won't be freed. + +### ucl_pubkey_add + +~~~C +bool ucl_pubkey_add (struct ucl_parser *parser, + const unsigned char *key, size_t len); +~~~ + +This function adds a public key from text blob `key` of length `len` to the `parser` object. This public key should be in the `PEM` format and can be used by `.includes` macro for checking signatures of files included. `Openssl` support should be enabled to make this function working. If a key cannot be added (e.g. due to format error) or `openssl` was not linked to `libucl` then this function returns `false`. + +### ucl_parser_set_filevars + +~~~C +bool ucl_parser_set_filevars (struct ucl_parser *parser, + const char *filename, bool need_expand); +~~~ + +Add the standard file variables to the `parser` based on the `filename` specified: + +- `$FILENAME` - a filename of `ucl` input +- `$CURDIR` - a current directory of the input + +For example, if a `filename` param is `../something.conf` then the variables will have the following values: + +- `$FILENAME` - "../something.conf" +- `$CURDIR` - ".." + +if `need_expand` parameter is `true` then all relative paths are expanded using `realpath` call. In this example if `..` is `/etc/dir` then variables will have these values: + +- `$FILENAME` - "/etc/something.conf" +- `$CURDIR` - "/etc" + +## Parser usage example + +The following example loads, parses and extracts `ucl` object from stdin using `libucl` parser functions (the length of input is limited to 8K): + +~~~C +char inbuf[8192]; +struct ucl_parser *parser = NULL; +int ret = 0, r = 0; +ucl_object_t *obj = NULL; +FILE *in; + +in = stdin; +parser = ucl_parser_new (0); +while (!feof (in) && r < (int)sizeof (inbuf)) { + r += fread (inbuf + r, 1, sizeof (inbuf) - r, in); +} +ucl_parser_add_chunk (parser, inbuf, r); +fclose (in); + +if (ucl_parser_get_error (parser)) { + printf ("Error occurred: %s\n", ucl_parser_get_error (parser)); + ret = 1; +} +else { + obj = ucl_parser_get_object (parser); +} + +if (parser != NULL) { + ucl_parser_free (parser); +} +if (obj != NULL) { + ucl_object_unref (obj); +} +return ret; +~~~ + +# Emitting functions + +Libucl can transform UCL objects to a number of tectual formats: + +- configuration (`UCL_EMIT_CONFIG`) - nginx like human readable configuration file where implicit arrays are transformed to the duplicate keys +- compact json: `UCL_EMIT_JSON_COMPACT` - single line valid json without spaces +- formatted json: `UCL_EMIT_JSON` - pretty formatted JSON with newlines and spaces +- compact yaml: `UCL_EMIT_YAML` - compact YAML output + +Moreover, libucl API allows to select a custom set of emitting functions allowing +efficient and zero-copy output of libucl objects. Libucl uses the following structure to support this feature: + +~~~C +struct ucl_emitter_functions { + /** Append a single character */ + int (*ucl_emitter_append_character) (unsigned char c, size_t nchars, void *ud); + /** Append a string of a specified length */ + int (*ucl_emitter_append_len) (unsigned const char *str, size_t len, void *ud); + /** Append a 64 bit integer */ + int (*ucl_emitter_append_int) (int64_t elt, void *ud); + /** Append floating point element */ + int (*ucl_emitter_append_double) (double elt, void *ud); + /** Opaque userdata pointer */ + void *ud; +}; +~~~ + +This structure defines the following callbacks: + +- `ucl_emitter_append_character` - a function that is called to append `nchars` characters equal to `c` +- `ucl_emitter_append_len` - used to append a string of length `len` starting from pointer `str` +- `ucl_emitter_append_int` - this function applies to integer numbers +- `ucl_emitter_append_double` - this function is intended to output floating point variable + +The set of these functions could be used to output text formats of `UCL` objects to different structures or streams. + +Libucl provides the following functions for emitting UCL objects: + +### ucl_object_emit + +~~~C +unsigned char *ucl_object_emit (const ucl_object_t *obj, enum ucl_emitter emit_type); +~~~ + +Allocate a string that is suitable to fit the underlying UCL object `obj` and fill it with the textual representation of the object `obj` according to style `emit_type`. The caller should free the returned string after using. + +### ucl_object_emit_full + +~~~C +bool ucl_object_emit_full (const ucl_object_t *obj, enum ucl_emitter emit_type, + struct ucl_emitter_functions *emitter); +~~~ + +This function is similar to the previous with the exception that it accepts the additional argument `emitter` that defines the concrete set of output functions. This emit function could be useful for custom structures or streams emitters (including C++ ones, for example). + +# Conversion functions + +Conversion functions are used to convert UCL objects to primitive types, such as strings, numbers, or boolean values. There are two types of conversion functions: + +- safe: try to convert an ucl object to a primitive type and fail if such a conversion is not possible +- unsafe: return primitive type without additional checks, if the object cannot be converted then some reasonable default is returned (NULL for strings and 0 for numbers) + +Also there is a single `ucl_object_tostring_forced` function that converts any UCL object (including compound types - arrays and objects) to a string representation. For objects, arrays, booleans and numeric types this function performs emitting to a compact json format actually. + +Here is a list of all conversion functions: + +- `ucl_object_toint` - returns `int64_t` of UCL object +- `ucl_object_todouble` - returns `double` of UCL object +- `ucl_object_toboolean` - returns `bool` of UCL object +- `ucl_object_tostring` - returns `const char *` of UCL object (this string is NULL terminated) +- `ucl_object_tolstring` - returns `const char *` and `size_t` len of UCL object (string does not need to be NULL terminated) +- `ucl_object_tostring_forced` - returns string representation of any UCL object + +Strings returned by these pointers are associated with the UCL object and exist over its lifetime. A caller should not free this memory. + +# Generation functions + +It is possible to generate UCL objects from C primitive types. Moreover, libucl allows creation and modifying complex UCL objects, such as arrays or associative objects. + +## ucl_object_new +~~~C +ucl_object_t * ucl_object_new (void) +~~~ + +Creates new object of type `UCL_NULL`. This object should be released by caller. + +## ucl_object_typed_new +~~~C +ucl_object_t * ucl_object_typed_new (unsigned int type) +~~~ + +Create an object of a specified type: +- `UCL_OBJECT` - UCL object - key/value pairs +- `UCL_ARRAY` - UCL array +- `UCL_INT` - integer number +- `UCL_FLOAT` - floating point number +- `UCL_STRING` - NULL terminated string +- `UCL_BOOLEAN` - boolean value +- `UCL_TIME` - time value (floating point number of seconds) +- `UCL_USERDATA` - opaque userdata pointer (may be used in macros) +- `UCL_NULL` - null value + +This object should be released by caller. + +## Primitive objects generation +Libucl provides the functions similar to inverse conversion functions called with the specific C type: +- `ucl_object_fromint` - converts `int64_t` to UCL object +- `ucl_object_fromdouble` - converts `double` to UCL object +- `ucl_object_fromboolean` - converts `bool` to UCL object +- `ucl_object_fromstring` - converts `const char *` to UCL object (this string should be NULL terminated) +- `ucl_object_fromlstring` - converts `const char *` and `size_t` len to UCL object (string does not need to be NULL terminated) + +Also there is a function to generate UCL object from a string performing various parsing or conversion operations called `ucl_object_fromstring_common`. + +## ucl_object_fromstring_common +~~~C +ucl_object_t * ucl_object_fromstring_common (const char *str, + size_t len, enum ucl_string_flags flags) +~~~ + +This function is used to convert a string `str` of size `len` to a UCL object applying `flags` conversions. If `len` is equal to zero then a `str` is assumed as NULL-terminated. This function supports the following flags (a set of flags can be specified using logical `OR` operation): + +- `UCL_STRING_ESCAPE` - perform JSON escape +- `UCL_STRING_TRIM` - trim leading and trailing whitespaces +- `UCL_STRING_PARSE_BOOLEAN` - parse passed string and detect boolean +- `UCL_STRING_PARSE_INT` - parse passed string and detect integer number +- `UCL_STRING_PARSE_DOUBLE` - parse passed string and detect integer or float number +- `UCL_STRING_PARSE_TIME` - parse time values as floating point numbers +- `UCL_STRING_PARSE_NUMBER` - parse passed string and detect number (both float, integer and time types) +- `UCL_STRING_PARSE` - parse passed string (and detect booleans, numbers and time values) +- `UCL_STRING_PARSE_BYTES` - assume that numeric multipliers are in bytes notation, for example `10k` means `10*1024` and not `10*1000` as assumed without this flag + +If parsing operations fail then the resulting UCL object will be a `UCL_STRING`. A caller should always check the type of the returned object and release it after using. + +# Iteration functions + +Iteration are used to iterate over UCL compound types: arrays and objects. Moreover, iterations could be performed over the keys with multiple values (implicit arrays). +There are two types of iterators API: old and unsafe one via `ucl_iterate_object` and the proposed interface of safe iterators. + + +## ucl_iterate_object +~~~C +const ucl_object_t* ucl_iterate_object (const ucl_object_t *obj, + ucl_object_iter_t *iter, bool expand_values); +~~~ + +This function accepts opaque iterator pointer `iter`. In the first call this iterator *must* be initialized to `NULL`. Iterator is changed by this function call. `ucl_iterate_object` returns the next UCL object in the compound object `obj` or `NULL` if all objects have been iterated. The reference count of the object returned is not increased, so a caller should not unref the object or modify its content (e.g. by inserting to another compound object). The object `obj` should not be changed during the iteration process as well. `expand_values` flag speicifies whether `ucl_iterate_object` should expand keys with multiple values. The general rule is that if you need to iterate through the *object* or *explicit array*, then you always need to set this flag to `true`. However, if you get some key in the object and want to extract all its values then you should set `expand_values` to `false`. Mixing of iteration types is not permitted since the iterator is set according to the iteration type and cannot be reused. Here is an example of iteration over the objects using libucl API (assuming that `top` is `UCL_OBJECT` in this example): + +~~~C +ucl_object_iter_t it = NULL, it_obj = NULL; +const ucl_object_t *cur, *tmp; + +/* Iterate over the object */ +while ((obj = ucl_iterate_object (top, &it, true))) { + printf ("key: \"%s\"\n", ucl_object_key (obj)); + /* Iterate over the values of a key */ + while ((cur = ucl_iterate_object (obj, &it_obj, false))) { + printf ("value: \"%s\"\n", + ucl_object_tostring_forced (cur)); + } +} +~~~ + +## Safe iterators API + +Safe iterators are defined to clarify iterating over UCL objects and simplify flattening of UCL objects in non-trivial cases. +For example, if there is an implicit array that contains another array and a boolean value it is extremely unclear how to iterate over +such an object. Safe iterators are desinged to define two sorts of iteration: + +1. Iteration over complex objects with expanding all values +2. Iteration over complex objects without expanding of values + +The following example demonstrates the difference between these two types of iteration: + +~~~ +key = 1; +key = [2, 3, 4]; + +Iteration with expansion: + +1, 2, 3, 4 + +Iteration without expansion: + +1, [2, 3, 4] +~~~ + +UCL defines the following functions to manage safe iterators: + +- `ucl_object_iterate_new` - creates new safe iterator +- `ucl_object_iterate_reset` - resets iterator to a new object +- `ucl_object_iterate_safe` - safely iterate the object inside iterator +- `ucl_object_iterate_free` - free memory associated with the safe iterator + +Please note that unlike unsafe iterators, safe iterators *must* be explicitly initialized and freed. +An assert is likely generated if you use uninitialized or `NULL` iterator in all safe iterators functions. + +~~~C +ucl_object_iter_t it; +const ucl_object_t *cur; + +it = ucl_object_iterate_new (obj); + +while ((cur = ucl_object_iterate_safe (it, true)) != NULL) { + /* Do something */ +} + +/* Switch to another object */ +it = ucl_object_iterate_reset (it, another_obj); + +while ((cur = ucl_object_iterate_safe (it, true)) != NULL) { + /* Do something else */ +} + +ucl_object_iterate_free (it); +~~~ + +# Validation functions + +Currently, there is only one validation function called `ucl_object_validate`. It performs validation of object using the specified schema. This function is defined as following: + +## ucl_object_validate +~~~C +bool ucl_object_validate (const ucl_object_t *schema, + const ucl_object_t *obj, struct ucl_schema_error *err); +~~~ + +This function uses ucl object `schema`, that must be valid in terms of `json-schema` draft v4, to validate input object `obj`. If this function returns `true` then validation procedure has been succeed. Otherwise, `false` is returned and `err` is set to a specific value. If a caller sets `err` to NULL then this function does not set any error just returning `false`. Error is the structure defined as following: + +~~~C +struct ucl_schema_error { + enum ucl_schema_error_code code; /* error code */ + char msg[128]; /* error message */ + ucl_object_t *obj; /* object where error occurred */ +}; +~~~ + +Caller may use `code` field to get a numeric error code: + +~~~C +enum ucl_schema_error_code { + UCL_SCHEMA_OK = 0, /* no error */ + UCL_SCHEMA_TYPE_MISMATCH, /* type of object is incorrect */ + UCL_SCHEMA_INVALID_SCHEMA, /* schema is invalid */ + UCL_SCHEMA_MISSING_PROPERTY,/* missing properties */ + UCL_SCHEMA_CONSTRAINT, /* constraint found */ + UCL_SCHEMA_MISSING_DEPENDENCY, /* missing dependency */ + UCL_SCHEMA_UNKNOWN /* generic error */ +}; +~~~ + +`msg` is a string description of an error and `obj` is an object where error has occurred. Error object is not allocated by libucl, so there is no need to free it after validation (a static object should thus be used). diff --git a/contrib/libucl/doc/libucl.3 b/contrib/libucl/doc/libucl.3 new file mode 100644 index 0000000..ec50463 --- /dev/null +++ b/contrib/libucl/doc/libucl.3 @@ -0,0 +1,708 @@ +.TH "LIBUCL" "3" "27 December, 2014" "Libucl manual" "" +.SH NAME +.PP +\f[B]ucl_parser_new\f[], \f[B]ucl_parser_register_macro\f[], +\f[B]ucl_parser_register_variable\f[], \f[B]ucl_parser_add_chunk\f[], +\f[B]ucl_parser_add_string\f[], \f[B]ucl_parser_add_file\f[], +\f[B]ucl_parser_get_object\f[], \f[B]ucl_parser_get_error\f[], +\f[B]ucl_parser_free\f[], \f[B]ucl_pubkey_add\f[], +\f[B]ucl_parser_set_filevars\f[] \- universal configuration library +parser and utility functions +.SH LIBRARY +.PP +UCL library (libucl, \-lucl) +.SH SYNOPSIS +.PP +\f[C]#include\ <ucl.h>\f[] +.SH DESCRIPTION +.PP +Libucl is a parser and \f[C]C\f[] API to parse and generate \f[C]ucl\f[] +objects. +Libucl consist of several groups of functions: +.SS Parser functions +.PP +Used to parse \f[C]ucl\f[] files and provide interface to extract +\f[C]ucl\f[] object. +Currently, \f[C]libucl\f[] can parse only full \f[C]ucl\f[] documents, +for instance, it is impossible to parse a part of document and therefore +it is impossible to use \f[C]libucl\f[] as a streaming parser. +In future, this limitation can be removed. +.SS Emitting functions +.PP +Convert \f[C]ucl\f[] objects to some textual or binary representation. +Currently, libucl supports the following exports: +.IP \[bu] 2 +\f[C]JSON\f[] \- valid json format (can possibly lose some original +data, such as implicit arrays) +.IP \[bu] 2 +\f[C]Config\f[] \- human\-readable configuration format (lossless) +.IP \[bu] 2 +\f[C]YAML\f[] \- embedded yaml format (has the same limitations as +\f[C]json\f[] output) +.SS Conversion functions +.PP +Help to convert \f[C]ucl\f[] objects to C types. +These functions are used to convert \f[C]ucl_object_t\f[] to C primitive +types, such as numbers, strings or boolean values. +.SS Generation functions +.PP +Allow creation of \f[C]ucl\f[] objects from C types and creating of +complex \f[C]ucl\f[] objects, such as hashes or arrays from primitive +\f[C]ucl\f[] objects, such as numbers or strings. +.SS Iteration functions +.PP +Iterate over \f[C]ucl\f[] complex objects or over a chain of values, for +example when a key in an object has multiple values (that can be treated +as implicit array or implicit consolidation). +.SS Validation functions +.PP +Validation functions are used to validate some object \f[C]obj\f[] using +json\-schema compatible object \f[C]schema\f[]. +Both input and schema must be UCL objects to perform validation. +.SS Utility functions +.PP +Provide basic utilities to manage \f[C]ucl\f[] objects: creating, +removing, retaining and releasing reference count and so on. +.SH PARSER FUNCTIONS +.PP +Parser functions operates with \f[C]struct\ ucl_parser\f[]. +.SS ucl_parser_new +.IP +.nf +\f[C] +struct\ ucl_parser*\ ucl_parser_new\ (int\ flags); +\f[] +.fi +.PP +Creates new parser with the specified flags: +.IP \[bu] 2 +\f[C]UCL_PARSER_KEY_LOWERCASE\f[] \- lowercase keys parsed +.IP \[bu] 2 +\f[C]UCL_PARSER_ZEROCOPY\f[] \- try to use zero\-copy mode when reading +files (in zero\-copy mode text chunk being parsed without copying +strings so it should exist till any object parsed is used) +.IP \[bu] 2 +\f[C]UCL_PARSER_NO_TIME\f[] \- treat time values as strings without +parsing them as floats +.SS ucl_parser_register_macro +.IP +.nf +\f[C] +void\ ucl_parser_register_macro\ (struct\ ucl_parser\ *parser, +\ \ \ \ const\ char\ *macro,\ ucl_macro_handler\ handler,\ void*\ ud); +\f[] +.fi +.PP +Register new macro with name .\f[C]macro\f[] parsed by handler +\f[C]handler\f[] that accepts opaque data pointer \f[C]ud\f[]. +Macro handler should be of the following type: +.IP +.nf +\f[C] +bool\ (*ucl_macro_handler)\ (const\ unsigned\ char\ *data, +\ \ \ \ size_t\ len,\ void*\ ud);` +\f[] +.fi +.PP +Handler function accepts macro text \f[C]data\f[] of length \f[C]len\f[] +and the opaque pointer \f[C]ud\f[]. +If macro is parsed successfully the handler should return \f[C]true\f[]. +\f[C]false\f[] indicates parsing failure and the parser can be +terminated. +.SS ucl_parser_register_variable +.IP +.nf +\f[C] +void\ ucl_parser_register_variable\ (struct\ ucl_parser\ *parser, +\ \ \ \ const\ char\ *var,\ const\ char\ *value); +\f[] +.fi +.PP +Register new variable $\f[C]var\f[] that should be replaced by the +parser to the \f[C]value\f[] string. +.SS ucl_parser_add_chunk +.IP +.nf +\f[C] +bool\ ucl_parser_add_chunk\ (struct\ ucl_parser\ *parser,\ +\ \ \ \ const\ unsigned\ char\ *data,\ size_t\ len); +\f[] +.fi +.PP +Add new text chunk with \f[C]data\f[] of length \f[C]len\f[] to the +parser. +At the moment, \f[C]libucl\f[] parser is not a streamlined parser and +chunk \f[I]must\f[] contain the \f[I]valid\f[] ucl object. +For example, this object should be valid: +.IP +.nf +\f[C] +{\ "var":\ "value"\ } +\f[] +.fi +.PP +while this one won\[aq]t be parsed correctly: +.IP +.nf +\f[C] +{\ "var":\ +\f[] +.fi +.PP +This limitation may possible be removed in future. +.SS ucl_parser_add_string +.IP +.nf +\f[C] +bool\ ucl_parser_add_string\ (struct\ ucl_parser\ *parser,\ +\ \ \ \ const\ char\ *data,\ size_t\ len); +\f[] +.fi +.PP +This function acts exactly like \f[C]ucl_parser_add_chunk\f[] does but +if \f[C]len\f[] argument is zero, then the string \f[C]data\f[] must be +zero\-terminated and the actual length is calculated up to \f[C]\\0\f[] +character. +.SS ucl_parser_add_file +.IP +.nf +\f[C] +bool\ ucl_parser_add_file\ (struct\ ucl_parser\ *parser,\ +\ \ \ \ const\ char\ *filename); +\f[] +.fi +.PP +Load file \f[C]filename\f[] and parse it with the specified +\f[C]parser\f[]. +This function uses \f[C]mmap\f[] call to load file, therefore, it should +not be \f[C]shrunk\f[] during parsing. +Otherwise, \f[C]libucl\f[] can cause memory corruption and terminate the +calling application. +This function is also used by the internal handler of \f[C]include\f[] +macro, hence, this macro has the same limitation. +.SS ucl_parser_get_object +.IP +.nf +\f[C] +ucl_object_t*\ ucl_parser_get_object\ (struct\ ucl_parser\ *parser); +\f[] +.fi +.PP +If the \f[C]ucl\f[] data has been parsed correctly this function returns +the top object for the parser. +Otherwise, this function returns the \f[C]NULL\f[] pointer. +The reference count for \f[C]ucl\f[] object returned is increased by +one, therefore, a caller should decrease reference by using +\f[C]ucl_object_unref\f[] to free object after usage. +.SS ucl_parser_get_error +.IP +.nf +\f[C] +const\ char\ *ucl_parser_get_error(struct\ ucl_parser\ *parser); +\f[] +.fi +.PP +Returns the constant error string for the parser object. +If no error occurred during parsing a \f[C]NULL\f[] object is returned. +A caller should not try to free or modify this string. +.SS ucl_parser_free +.IP +.nf +\f[C] +void\ ucl_parser_free\ (struct\ ucl_parser\ *parser); +\f[] +.fi +.PP +Frees memory occupied by the parser object. +The reference count for top object is decreased as well, however if the +function \f[C]ucl_parser_get_object\f[] was called previously then the +top object won\[aq]t be freed. +.SS ucl_pubkey_add +.IP +.nf +\f[C] +bool\ ucl_pubkey_add\ (struct\ ucl_parser\ *parser,\ +\ \ \ \ const\ unsigned\ char\ *key,\ size_t\ len); +\f[] +.fi +.PP +This function adds a public key from text blob \f[C]key\f[] of length +\f[C]len\f[] to the \f[C]parser\f[] object. +This public key should be in the \f[C]PEM\f[] format and can be used by +\f[C]\&.includes\f[] macro for checking signatures of files included. +\f[C]Openssl\f[] support should be enabled to make this function +working. +If a key cannot be added (e.g. +due to format error) or \f[C]openssl\f[] was not linked to +\f[C]libucl\f[] then this function returns \f[C]false\f[]. +.SS ucl_parser_set_filevars +.IP +.nf +\f[C] +bool\ ucl_parser_set_filevars\ (struct\ ucl_parser\ *parser,\ +\ \ \ \ const\ char\ *filename,\ bool\ need_expand); +\f[] +.fi +.PP +Add the standard file variables to the \f[C]parser\f[] based on the +\f[C]filename\f[] specified: +.IP \[bu] 2 +\f[C]$FILENAME\f[] \- a filename of \f[C]ucl\f[] input +.IP \[bu] 2 +\f[C]$CURDIR\f[] \- a current directory of the input +.PP +For example, if a \f[C]filename\f[] param is +\f[C]\&../something.conf\f[] then the variables will have the following +values: +.IP \[bu] 2 +\f[C]$FILENAME\f[] \- "../something.conf" +.IP \[bu] 2 +\f[C]$CURDIR\f[] \- ".." +.PP +if \f[C]need_expand\f[] parameter is \f[C]true\f[] then all relative +paths are expanded using \f[C]realpath\f[] call. +In this example if \f[C]\&..\f[] is \f[C]/etc/dir\f[] then variables +will have these values: +.IP \[bu] 2 +\f[C]$FILENAME\f[] \- "/etc/something.conf" +.IP \[bu] 2 +\f[C]$CURDIR\f[] \- "/etc" +.SS Parser usage example +.PP +The following example loads, parses and extracts \f[C]ucl\f[] object +from stdin using \f[C]libucl\f[] parser functions (the length of input +is limited to 8K): +.IP +.nf +\f[C] +char\ inbuf[8192]; +struct\ ucl_parser\ *parser\ =\ NULL; +int\ ret\ =\ 0,\ r\ =\ 0; +ucl_object_t\ *obj\ =\ NULL; +FILE\ *in; + +in\ =\ stdin; +parser\ =\ ucl_parser_new\ (0); +while\ (!feof\ (in)\ &&\ r\ <\ (int)sizeof\ (inbuf))\ { +\ \ \ \ r\ +=\ fread\ (inbuf\ +\ r,\ 1,\ sizeof\ (inbuf)\ \-\ r,\ in); +} +ucl_parser_add_chunk\ (parser,\ inbuf,\ r); +fclose\ (in); + +if\ (ucl_parser_get_error\ (parser))\ { +\ \ \ \ printf\ ("Error\ occurred:\ %s\\n",\ ucl_parser_get_error\ (parser)); +\ \ \ \ ret\ =\ 1; +} +else\ { +\ \ \ \ obj\ =\ ucl_parser_get_object\ (parser); +} + +if\ (parser\ !=\ NULL)\ { +\ \ \ \ ucl_parser_free\ (parser); +} +if\ (obj\ !=\ NULL)\ { +\ \ \ \ ucl_object_unref\ (obj); +} +return\ ret; +\f[] +.fi +.SH EMITTING FUNCTIONS +.PP +Libucl can transform UCL objects to a number of tectual formats: +.IP \[bu] 2 +configuration (\f[C]UCL_EMIT_CONFIG\f[]) \- nginx like human readable +configuration file where implicit arrays are transformed to the +duplicate keys +.IP \[bu] 2 +compact json: \f[C]UCL_EMIT_JSON_COMPACT\f[] \- single line valid json +without spaces +.IP \[bu] 2 +formatted json: \f[C]UCL_EMIT_JSON\f[] \- pretty formatted JSON with +newlines and spaces +.IP \[bu] 2 +compact yaml: \f[C]UCL_EMIT_YAML\f[] \- compact YAML output +.PP +Moreover, libucl API allows to select a custom set of emitting functions +allowing efficient and zero\-copy output of libucl objects. +Libucl uses the following structure to support this feature: +.IP +.nf +\f[C] +struct\ ucl_emitter_functions\ { +\ \ \ \ /**\ Append\ a\ single\ character\ */ +\ \ \ \ int\ (*ucl_emitter_append_character)\ (unsigned\ char\ c,\ size_t\ nchars,\ void\ *ud); +\ \ \ \ /**\ Append\ a\ string\ of\ a\ specified\ length\ */ +\ \ \ \ int\ (*ucl_emitter_append_len)\ (unsigned\ const\ char\ *str,\ size_t\ len,\ void\ *ud); +\ \ \ \ /**\ Append\ a\ 64\ bit\ integer\ */ +\ \ \ \ int\ (*ucl_emitter_append_int)\ (int64_t\ elt,\ void\ *ud); +\ \ \ \ /**\ Append\ floating\ point\ element\ */ +\ \ \ \ int\ (*ucl_emitter_append_double)\ (double\ elt,\ void\ *ud); +\ \ \ \ /**\ Opaque\ userdata\ pointer\ */ +\ \ \ \ void\ *ud; +}; +\f[] +.fi +.PP +This structure defines the following callbacks: +.IP \[bu] 2 +\f[C]ucl_emitter_append_character\f[] \- a function that is called to +append \f[C]nchars\f[] characters equal to \f[C]c\f[] +.IP \[bu] 2 +\f[C]ucl_emitter_append_len\f[] \- used to append a string of length +\f[C]len\f[] starting from pointer \f[C]str\f[] +.IP \[bu] 2 +\f[C]ucl_emitter_append_int\f[] \- this function applies to integer +numbers +.IP \[bu] 2 +\f[C]ucl_emitter_append_double\f[] \- this function is intended to +output floating point variable +.PP +The set of these functions could be used to output text formats of +\f[C]UCL\f[] objects to different structures or streams. +.PP +Libucl provides the following functions for emitting UCL objects: +.SS ucl_object_emit +.IP +.nf +\f[C] +unsigned\ char\ *ucl_object_emit\ (const\ ucl_object_t\ *obj,\ enum\ ucl_emitter\ emit_type); +\f[] +.fi +.PP +Allocate a string that is suitable to fit the underlying UCL object +\f[C]obj\f[] and fill it with the textual representation of the object +\f[C]obj\f[] according to style \f[C]emit_type\f[]. +The caller should free the returned string after using. +.SS ucl_object_emit_full +.IP +.nf +\f[C] +bool\ ucl_object_emit_full\ (const\ ucl_object_t\ *obj,\ enum\ ucl_emitter\ emit_type, +\ \ \ \ \ \ \ \ struct\ ucl_emitter_functions\ *emitter); +\f[] +.fi +.PP +This function is similar to the previous with the exception that it +accepts the additional argument \f[C]emitter\f[] that defines the +concrete set of output functions. +This emit function could be useful for custom structures or streams +emitters (including C++ ones, for example). +.SH CONVERSION FUNCTIONS +.PP +Conversion functions are used to convert UCL objects to primitive types, +such as strings, numbers, or boolean values. +There are two types of conversion functions: +.IP \[bu] 2 +safe: try to convert an ucl object to a primitive type and fail if such +a conversion is not possible +.IP \[bu] 2 +unsafe: return primitive type without additional checks, if the object +cannot be converted then some reasonable default is returned (NULL for +strings and 0 for numbers) +.PP +Also there is a single \f[C]ucl_object_tostring_forced\f[] function that +converts any UCL object (including compound types \- arrays and objects) +to a string representation. +For objects, arrays, booleans and numeric types this function performs +emitting to a compact json format actually. +.PP +Here is a list of all conversion functions: +.IP \[bu] 2 +\f[C]ucl_object_toint\f[] \- returns \f[C]int64_t\f[] of UCL object +.IP \[bu] 2 +\f[C]ucl_object_todouble\f[] \- returns \f[C]double\f[] of UCL object +.IP \[bu] 2 +\f[C]ucl_object_toboolean\f[] \- returns \f[C]bool\f[] of UCL object +.IP \[bu] 2 +\f[C]ucl_object_tostring\f[] \- returns \f[C]const\ char\ *\f[] of UCL +object (this string is NULL terminated) +.IP \[bu] 2 +\f[C]ucl_object_tolstring\f[] \- returns \f[C]const\ char\ *\f[] and +\f[C]size_t\f[] len of UCL object (string does not need to be NULL +terminated) +.IP \[bu] 2 +\f[C]ucl_object_tostring_forced\f[] \- returns string representation of +any UCL object +.PP +Strings returned by these pointers are associated with the UCL object +and exist over its lifetime. +A caller should not free this memory. +.SH GENERATION FUNCTIONS +.PP +It is possible to generate UCL objects from C primitive types. +Moreover, libucl allows creation and modifying complex UCL objects, such +as arrays or associative objects. +.SS ucl_object_new +.IP +.nf +\f[C] +ucl_object_t\ *\ ucl_object_new\ (void) +\f[] +.fi +.PP +Creates new object of type \f[C]UCL_NULL\f[]. +This object should be released by caller. +.SS ucl_object_typed_new +.IP +.nf +\f[C] +ucl_object_t\ *\ ucl_object_typed_new\ (unsigned\ int\ type) +\f[] +.fi +.PP +Create an object of a specified type: \- \f[C]UCL_OBJECT\f[] \- UCL +object \- key/value pairs \- \f[C]UCL_ARRAY\f[] \- UCL array \- +\f[C]UCL_INT\f[] \- integer number \- \f[C]UCL_FLOAT\f[] \- floating +point number \- \f[C]UCL_STRING\f[] \- NULL terminated string \- +\f[C]UCL_BOOLEAN\f[] \- boolean value \- \f[C]UCL_TIME\f[] \- time value +(floating point number of seconds) \- \f[C]UCL_USERDATA\f[] \- opaque +userdata pointer (may be used in macros) \- \f[C]UCL_NULL\f[] \- null +value +.PP +This object should be released by caller. +.SS Primitive objects generation +.PP +Libucl provides the functions similar to inverse conversion functions +called with the specific C type: \- \f[C]ucl_object_fromint\f[] \- +converts \f[C]int64_t\f[] to UCL object \- +\f[C]ucl_object_fromdouble\f[] \- converts \f[C]double\f[] to UCL object +\- \f[C]ucl_object_fromboolean\f[] \- converts \f[C]bool\f[] to UCL +object \- \f[C]ucl_object_fromstring\f[] \- converts +\f[C]const\ char\ *\f[] to UCL object (this string should be NULL +terminated) \- \f[C]ucl_object_fromlstring\f[] \- converts +\f[C]const\ char\ *\f[] and \f[C]size_t\f[] len to UCL object (string +does not need to be NULL terminated) +.PP +Also there is a function to generate UCL object from a string performing +various parsing or conversion operations called +\f[C]ucl_object_fromstring_common\f[]. +.SS ucl_object_fromstring_common +.IP +.nf +\f[C] +ucl_object_t\ *\ ucl_object_fromstring_common\ (const\ char\ *str,\ +\ \ \ \ size_t\ len,\ enum\ ucl_string_flags\ flags) +\f[] +.fi +.PP +This function is used to convert a string \f[C]str\f[] of size +\f[C]len\f[] to a UCL object applying \f[C]flags\f[] conversions. +If \f[C]len\f[] is equal to zero then a \f[C]str\f[] is assumed as +NULL\-terminated. +This function supports the following flags (a set of flags can be +specified using logical \f[C]OR\f[] operation): +.IP \[bu] 2 +\f[C]UCL_STRING_ESCAPE\f[] \- perform JSON escape +.IP \[bu] 2 +\f[C]UCL_STRING_TRIM\f[] \- trim leading and trailing whitespaces +.IP \[bu] 2 +\f[C]UCL_STRING_PARSE_BOOLEAN\f[] \- parse passed string and detect +boolean +.IP \[bu] 2 +\f[C]UCL_STRING_PARSE_INT\f[] \- parse passed string and detect integer +number +.IP \[bu] 2 +\f[C]UCL_STRING_PARSE_DOUBLE\f[] \- parse passed string and detect +integer or float number +.IP \[bu] 2 +\f[C]UCL_STRING_PARSE_TIME\f[] \- parse time values as floating point +numbers +.IP \[bu] 2 +\f[C]UCL_STRING_PARSE_NUMBER\f[] \- parse passed string and detect +number (both float, integer and time types) +.IP \[bu] 2 +\f[C]UCL_STRING_PARSE\f[] \- parse passed string (and detect booleans, +numbers and time values) +.IP \[bu] 2 +\f[C]UCL_STRING_PARSE_BYTES\f[] \- assume that numeric multipliers are +in bytes notation, for example \f[C]10k\f[] means \f[C]10*1024\f[] and +not \f[C]10*1000\f[] as assumed without this flag +.PP +If parsing operations fail then the resulting UCL object will be a +\f[C]UCL_STRING\f[]. +A caller should always check the type of the returned object and release +it after using. +.SH ITERATION FUNCTIONS +.PP +Iteration are used to iterate over UCL compound types: arrays and +objects. +Moreover, iterations could be performed over the keys with multiple +values (implicit arrays). +There are two types of iterators API: old and unsafe one via +\f[C]ucl_iterate_object\f[] and the proposed interface of safe +iterators. +.SS ucl_iterate_object +.IP +.nf +\f[C] +const\ ucl_object_t*\ ucl_iterate_object\ (const\ ucl_object_t\ *obj,\ +\ \ \ \ ucl_object_iter_t\ *iter,\ bool\ expand_values); +\f[] +.fi +.PP +This function accepts opaque iterator pointer \f[C]iter\f[]. +In the first call this iterator \f[I]must\f[] be initialized to +\f[C]NULL\f[]. +Iterator is changed by this function call. +\f[C]ucl_iterate_object\f[] returns the next UCL object in the compound +object \f[C]obj\f[] or \f[C]NULL\f[] if all objects have been iterated. +The reference count of the object returned is not increased, so a caller +should not unref the object or modify its content (e.g. +by inserting to another compound object). +The object \f[C]obj\f[] should not be changed during the iteration +process as well. +\f[C]expand_values\f[] flag speicifies whether +\f[C]ucl_iterate_object\f[] should expand keys with multiple values. +The general rule is that if you need to iterate through the +\f[I]object\f[] or \f[I]explicit array\f[], then you always need to set +this flag to \f[C]true\f[]. +However, if you get some key in the object and want to extract all its +values then you should set \f[C]expand_values\f[] to \f[C]false\f[]. +Mixing of iteration types is not permitted since the iterator is set +according to the iteration type and cannot be reused. +Here is an example of iteration over the objects using libucl API +(assuming that \f[C]top\f[] is \f[C]UCL_OBJECT\f[] in this example): +.IP +.nf +\f[C] +ucl_object_iter_t\ it\ =\ NULL,\ it_obj\ =\ NULL; +const\ ucl_object_t\ *cur,\ *tmp; + +/*\ Iterate\ over\ the\ object\ */ +while\ ((obj\ =\ ucl_iterate_object\ (top,\ &it,\ true)))\ { +\ \ \ \ printf\ ("key:\ \\"%s\\"\\n",\ ucl_object_key\ (obj)); +\ \ \ \ /*\ Iterate\ over\ the\ values\ of\ a\ key\ */ +\ \ \ \ while\ ((cur\ =\ ucl_iterate_object\ (obj,\ &it_obj,\ false)))\ { +\ \ \ \ \ \ \ \ printf\ ("value:\ \\"%s\\"\\n",\ +\ \ \ \ \ \ \ \ \ \ \ \ ucl_object_tostring_forced\ (cur)); +\ \ \ \ } +} +\f[] +.fi +.SS Safe iterators API +.PP +Safe iterators are defined to clarify iterating over UCL objects and +simplify flattening of UCL objects in non\-trivial cases. +For example, if there is an implicit array that contains another array +and a boolean value it is extremely unclear how to iterate over such an +object. +Safe iterators are desinged to define two sorts of iteration: +.IP "1." 3 +Iteration over complex objects with expanding all values +.IP "2." 3 +Iteration over complex objects without expanding of values +.PP +The following example demonstrates the difference between these two +types of iteration: +.IP +.nf +\f[C] +key\ =\ 1; +key\ =\ [2,\ 3,\ 4]; + +Iteration\ with\ expansion: + +1,\ 2,\ 3,\ 4 + +Iteration\ without\ expansion: + +1,\ [2,\ 3,\ 4] +\f[] +.fi +.PP +UCL defines the following functions to manage safe iterators: +.IP \[bu] 2 +\f[C]ucl_object_iterate_new\f[] \- creates new safe iterator +.IP \[bu] 2 +\f[C]ucl_object_iterate_reset\f[] \- resets iterator to a new object +.IP \[bu] 2 +\f[C]ucl_object_iterate_safe\f[] \- safely iterate the object inside +iterator +.IP \[bu] 2 +\f[C]ucl_object_iterate_free\f[] \- free memory associated with the safe +iterator +.PP +Please note that unlike unsafe iterators, safe iterators \f[I]must\f[] +be explicitly initialized and freed. +An assert is likely generated if you use uninitialized or \f[C]NULL\f[] +iterator in all safe iterators functions. +.IP +.nf +\f[C] +ucl_object_iter_t\ it; +const\ ucl_object_t\ *cur; + +it\ =\ ucl_object_iterate_new\ (obj); + +while\ ((cur\ =\ ucl_object_iterate_safe\ (it,\ true))\ !=\ NULL)\ { +\ \ \ \ /*\ Do\ something\ */ +} + +/*\ Switch\ to\ another\ object\ */ +it\ =\ ucl_object_iterate_reset\ (it,\ another_obj); + +while\ ((cur\ =\ ucl_object_iterate_safe\ (it,\ true))\ !=\ NULL)\ { +\ \ \ \ /*\ Do\ something\ else\ */ +} + +ucl_object_iterate_free\ (it); +\f[] +.fi +.SH VALIDATION FUNCTIONS +.PP +Currently, there is only one validation function called +\f[C]ucl_object_validate\f[]. +It performs validation of object using the specified schema. +This function is defined as following: +.SS ucl_object_validate +.IP +.nf +\f[C] +bool\ ucl_object_validate\ (const\ ucl_object_t\ *schema, +\ \ \ \ const\ ucl_object_t\ *obj,\ struct\ ucl_schema_error\ *err); +\f[] +.fi +.PP +This function uses ucl object \f[C]schema\f[], that must be valid in +terms of \f[C]json\-schema\f[] draft v4, to validate input object +\f[C]obj\f[]. +If this function returns \f[C]true\f[] then validation procedure has +been succeed. +Otherwise, \f[C]false\f[] is returned and \f[C]err\f[] is set to a +specific value. +If a caller sets \f[C]err\f[] to NULL then this function does not set +any error just returning \f[C]false\f[]. +Error is the structure defined as following: +.IP +.nf +\f[C] +struct\ ucl_schema_error\ { +\ \ \ \ enum\ ucl_schema_error_code\ code;\ \ \ \ /*\ error\ code\ */ +\ \ \ \ char\ msg[128];\ \ \ \ \ \ \ \ \ \ \ \ \ \ /*\ error\ message\ */ +\ \ \ \ ucl_object_t\ *obj;\ \ \ \ \ \ \ \ \ \ /*\ object\ where\ error\ occurred\ */ +}; +\f[] +.fi +.PP +Caller may use \f[C]code\f[] field to get a numeric error code: +.IP +.nf +\f[C] +enum\ ucl_schema_error_code\ { +\ \ \ \ UCL_SCHEMA_OK\ =\ 0,\ \ \ \ \ \ \ \ \ \ /*\ no\ error\ */ +\ \ \ \ UCL_SCHEMA_TYPE_MISMATCH,\ \ \ /*\ type\ of\ object\ is\ incorrect\ */ +\ \ \ \ UCL_SCHEMA_INVALID_SCHEMA,\ \ /*\ schema\ is\ invalid\ */ +\ \ \ \ UCL_SCHEMA_MISSING_PROPERTY,/*\ missing\ properties\ */ +\ \ \ \ UCL_SCHEMA_CONSTRAINT,\ \ \ \ \ \ /*\ constraint\ found\ */ +\ \ \ \ UCL_SCHEMA_MISSING_DEPENDENCY,\ /*\ missing\ dependency\ */ +\ \ \ \ UCL_SCHEMA_UNKNOWN\ \ \ \ \ \ \ \ \ \ /*\ generic\ error\ */ +}; +\f[] +.fi +.PP +\f[C]msg\f[] is a string description of an error and \f[C]obj\f[] is an +object where error has occurred. +Error object is not allocated by libucl, so there is no need to free it +after validation (a static object should thus be used). +.SH AUTHORS +Vsevolod Stakhov <vsevolod@highsecure.ru>. diff --git a/contrib/libucl/doc/lua_api.md b/contrib/libucl/doc/lua_api.md new file mode 100644 index 0000000..f7af3ca --- /dev/null +++ b/contrib/libucl/doc/lua_api.md @@ -0,0 +1,196 @@ +## Module `ucl` + +This lua module allows to parse objects from strings and to store data into +ucl objects. It uses `libucl` C library to parse and manipulate with ucl objects. + +Example: + +~~~lua +local ucl = require("ucl") + +local parser = ucl.parser() +local res,err = parser:parse_string('{key=value}') + +if not res then + print('parser error: ' .. err) +else + local obj = parser:get_object() + local got = ucl.to_format(obj, 'json') +end + +local table = { + str = 'value', + num = 100500, + null = ucl.null, + func = function () + return 'huh' + end +} + + +print(ucl.to_format(table, 'ucl')) +-- Output: +--[[ +num = 100500; +str = "value"; +null = null; +func = "huh"; +--]] +~~~ + +###Brief content: + +**Functions**: + +> [`ucl_object_push_lua(L, obj, allow_array)`](#function-ucl_object_push_lual-obj-allow_array) + +> [`ucl.to_format(var, format)`](#function-uclto_formatvar-format) + + + +**Methods**: + +> [`parser:parse_file(name)`](#method-parserparse_filename) + +> [`parser:parse_string(input)`](#method-parserparse_stringinput) + +> [`parser:get_object()`](#method-parserget_object) + + +## Functions + +The module `ucl` defines the following functions. + +### Function `ucl_object_push_lua(L, obj, allow_array)` + +This is a `C` function to push `UCL` object as lua variable. This function +converts `obj` to lua representation using the following conversions: + +- *scalar* values are directly presented by lua objects +- *userdata* values are converted to lua function objects using `LUA_REGISTRYINDEX`, +this can be used to pass functions from lua to c and vice-versa +- *arrays* are converted to lua tables with numeric indicies suitable for `ipairs` iterations +- *objects* are converted to lua tables with string indicies + +**Parameters:** + +- `L {lua_State}`: lua state pointer +- `obj {ucl_object_t}`: object to push +- `allow_array {bool}`: expand implicit arrays (should be true for all but partial arrays) + +**Returns:** + +- `{int}`: `1` if an object is pushed to lua + +Back to [module description](#module-ucl). + +### Function `ucl.to_format(var, format)` + +Converts lua variable `var` to the specified `format`. Formats supported are: + +- `json` - fine printed json +- `json-compact` - compacted json +- `config` - fine printed configuration +- `ucl` - same as `config` +- `yaml` - embedded yaml + +If `var` contains function, they are called during output formatting and if +they return string value, then this value is used for ouptut. + +**Parameters:** + +- `var {variant}`: any sort of lua variable (if userdata then metafield `__to_ucl` is searched for output) +- `format {string}`: any available format + +**Returns:** + +- `{string}`: string representation of `var` in the specific `format`. + +Example: + +~~~lua +local table = { + str = 'value', + num = 100500, + null = ucl.null, + func = function () + return 'huh' + end +} + + +print(ucl.to_format(table, 'ucl')) +-- Output: +--[[ +num = 100500; +str = "value"; +null = null; +func = "huh"; +--]] +~~~ + +Back to [module description](#module-ucl). + + +## Methods + +The module `ucl` defines the following methods. + +### Method `parser:parse_file(name)` + +Parse UCL object from file. + +**Parameters:** + +- `name {string}`: filename to parse + +**Returns:** + +- `{bool[, string]}`: if res is `true` then file has been parsed successfully, otherwise an error string is also returned + +Example: + +~~~lua +local parser = ucl.parser() +local res,err = parser:parse_file('/some/file.conf') + +if not res then + print('parser error: ' .. err) +else + -- Do something with object +end +~~~ + +Back to [module description](#module-ucl). + +### Method `parser:parse_string(input)` + +Parse UCL object from file. + +**Parameters:** + +- `input {string}`: string to parse + +**Returns:** + +- `{bool[, string]}`: if res is `true` then file has been parsed successfully, otherwise an error string is also returned + +Back to [module description](#module-ucl). + +### Method `parser:get_object()` + +Get top object from parser and export it to lua representation. + +**Parameters:** + + nothing + +**Returns:** + +- `{variant or nil}`: ucl object as lua native variable + +Back to [module description](#module-ucl). + + +Back to [top](#). + diff --git a/contrib/libucl/doc/pandoc.template b/contrib/libucl/doc/pandoc.template new file mode 100644 index 0000000..2effe1a --- /dev/null +++ b/contrib/libucl/doc/pandoc.template @@ -0,0 +1,12 @@ +% LIBUCL(3) Libucl manual +% Vsevolod Stakhov <vsevolod@highsecure.ru> +% %%date%% + +# Name + +**ucl_parser_new**, **ucl_parser_register_macro**, **ucl_parser_register_variable**, **ucl_parser_add_chunk**, **ucl_parser_add_string**, **ucl_parser_add_file**, **ucl_parser_get_object**, **ucl_parser_get_error**, **ucl_parser_free**, **ucl_pubkey_add**, **ucl_parser_set_filevars** - universal configuration library parser and utility functions + +# Library + +UCL library (libucl, -lucl) + diff --git a/contrib/libucl/examples/ucl_cpp.cc b/contrib/libucl/examples/ucl_cpp.cc new file mode 100644 index 0000000..2d15d84 --- /dev/null +++ b/contrib/libucl/examples/ucl_cpp.cc @@ -0,0 +1,26 @@ +#include <iostream> +#include <string> +#include "ucl++.h" + +int main(int argc, char **argv) +{ + std::string input, err; + + input.assign((std::istreambuf_iterator<char>(std::cin)), + std::istreambuf_iterator<char>()); + + auto obj = ucl::Ucl::parse(input, err); + + if (obj) { + std::cout << obj.dump(UCL_EMIT_CONFIG) << std::endl; + + for (const auto &o : obj) { + std::cout << o.dump(UCL_EMIT_CONFIG) << std::endl; + } + } + else { + std::cerr << "Error: " << err << std::endl; + + return 1; + } +} diff --git a/contrib/libucl/include/lua_ucl.h b/contrib/libucl/include/lua_ucl.h new file mode 100644 index 0000000..38e74d3 --- /dev/null +++ b/contrib/libucl/include/lua_ucl.h @@ -0,0 +1,69 @@ +/* Copyright (c) 2014, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ +#ifndef LUA_UCL_H_ +#define LUA_UCL_H_ + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include <lua.h> +#include <lauxlib.h> +#include <lualib.h> +#include "ucl.h" + +/** + * Closure structure for lua function storing inside UCL + */ +struct ucl_lua_funcdata { + lua_State *L; + int idx; + char *ret; +}; + +/** + * Initialize lua UCL API + */ +UCL_EXTERN int luaopen_ucl (lua_State *L); + +/** + * Import UCL object from lua state + * @param L lua state + * @param idx index of object at the lua stack to convert to UCL + * @return new UCL object or NULL, the caller should unref object after using + */ +UCL_EXTERN ucl_object_t* ucl_object_lua_import (lua_State *L, int idx); + +/** + * Push an object to lua + * @param L lua state + * @param obj object to push + * @param allow_array traverse over implicit arrays + */ +UCL_EXTERN int ucl_object_push_lua (lua_State *L, + const ucl_object_t *obj, bool allow_array); + +UCL_EXTERN struct ucl_lua_funcdata* ucl_object_toclosure ( + const ucl_object_t *obj); + +#endif /* LUA_UCL_H_ */ diff --git a/contrib/libucl/include/ucl++.h b/contrib/libucl/include/ucl++.h new file mode 100644 index 0000000..87d2041 --- /dev/null +++ b/contrib/libucl/include/ucl++.h @@ -0,0 +1,422 @@ +/* + * Copyright (c) 2015, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY AUTHOR ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#pragma once +#include <string> +#include <memory> +#include <iostream> +#include <strstream> + +#include "ucl.h" + +// C++11 API inspired by json11: https://github.com/dropbox/json11/ + +namespace ucl { + +struct ucl_map_construct_t { }; +constexpr ucl_map_construct_t ucl_map_construct = ucl_map_construct_t(); +struct ucl_array_construct_t { }; +constexpr ucl_array_construct_t ucl_array_construct = ucl_array_construct_t(); + +class Ucl final { +private: + + struct ucl_deleter { + void operator() (ucl_object_t *obj) { + ucl_object_unref (obj); + } + }; + + static int + append_char (unsigned char c, size_t nchars, void *ud) + { + std::string *out = reinterpret_cast<std::string *>(ud); + + out->append (nchars, (char)c); + + return nchars; + } + static int + append_len (unsigned const char *str, size_t len, void *ud) + { + std::string *out = reinterpret_cast<std::string *>(ud); + + out->append ((const char *)str, len); + + return len; + } + static int + append_int (int64_t elt, void *ud) + { + std::string *out = reinterpret_cast<std::string *>(ud); + auto nstr = std::to_string (elt); + + out->append (nstr); + + return nstr.size (); + } + static int + append_double (double elt, void *ud) + { + std::string *out = reinterpret_cast<std::string *>(ud); + auto nstr = std::to_string (elt); + + out->append (nstr); + + return nstr.size (); + } + + static struct ucl_emitter_functions default_emit_funcs() + { + struct ucl_emitter_functions func = { + Ucl::append_char, + Ucl::append_len, + Ucl::append_int, + Ucl::append_double, + nullptr, + nullptr + }; + + return func; + }; + + std::unique_ptr<ucl_object_t, ucl_deleter> obj; + +public: + class const_iterator { + private: + struct ucl_iter_deleter { + void operator() (ucl_object_iter_t it) { + ucl_object_iterate_free (it); + } + }; + std::shared_ptr<void> it; + std::unique_ptr<Ucl> cur; + public: + typedef std::forward_iterator_tag iterator_category; + + const_iterator(const Ucl &obj) { + it = std::shared_ptr<void>(ucl_object_iterate_new (obj.obj.get()), + ucl_iter_deleter()); + cur.reset (new Ucl(ucl_object_iterate_safe (it.get(), true))); + } + + const_iterator() {} + const_iterator(const const_iterator &other) { + it = other.it; + } + ~const_iterator() {} + + const_iterator& operator=(const const_iterator &other) { + it = other.it; + return *this; + } + + bool operator==(const const_iterator &other) const + { + if (cur && other.cur) { + return cur->obj.get() == other.cur->obj.get(); + } + + return !cur && !other.cur; + } + + bool operator!=(const const_iterator &other) const + { + return !(*this == other); + } + + const_iterator& operator++() + { + if (it) { + cur.reset (new Ucl(ucl_object_iterate_safe (it.get(), true))); + } + + if (!*cur) { + it.reset (); + cur.reset (); + } + + return *this; + } + + const Ucl& operator*() const + { + return *cur; + } + const Ucl* operator->() const + { + return cur.get(); + } + }; + + // We grab ownership if get non-const ucl_object_t + Ucl(ucl_object_t *other) { + obj.reset (other); + } + + // Shared ownership + Ucl(const ucl_object_t *other) { + obj.reset (ucl_object_ref (other)); + } + + Ucl(const Ucl &other) { + obj.reset (ucl_object_ref (other.obj.get())); + } + + Ucl(Ucl &&other) { + obj.swap (other.obj); + } + + Ucl() noexcept { + obj.reset (ucl_object_typed_new (UCL_NULL)); + } + Ucl(std::nullptr_t) noexcept { + obj.reset (ucl_object_typed_new (UCL_NULL)); + } + Ucl(double value) { + obj.reset (ucl_object_typed_new (UCL_FLOAT)); + obj->value.dv = value; + } + Ucl(int64_t value) { + obj.reset (ucl_object_typed_new (UCL_INT)); + obj->value.iv = value; + } + Ucl(bool value) { + obj.reset (ucl_object_typed_new (UCL_BOOLEAN)); + obj->value.iv = static_cast<int64_t>(value); + } + Ucl(const std::string &value) { + obj.reset (ucl_object_fromstring_common (value.data (), value.size (), + UCL_STRING_RAW)); + } + Ucl(const char * value) { + obj.reset (ucl_object_fromstring_common (value, 0, UCL_STRING_RAW)); + } + + // Implicit constructor: anything with a to_json() function. + template <class T, class = decltype(&T::to_ucl)> + Ucl(const T & t) : Ucl(t.to_ucl()) {} + + // Implicit constructor: map-like objects (std::map, std::unordered_map, etc) + template <class M, typename std::enable_if< + std::is_constructible<std::string, typename M::key_type>::value + && std::is_constructible<Ucl, typename M::mapped_type>::value, + int>::type = 0> + Ucl(const M & m) { + obj.reset (ucl_object_typed_new (UCL_OBJECT)); + auto cobj = obj.get (); + + for (const auto &e : m) { + ucl_object_insert_key (cobj, ucl_object_ref (e.second.obj.get()), + e.first.data (), e.first.size (), true); + } + } + + // Implicit constructor: vector-like objects (std::list, std::vector, std::set, etc) + template <class V, typename std::enable_if< + std::is_constructible<Ucl, typename V::value_type>::value, + int>::type = 0> + Ucl(const V & v) { + obj.reset (ucl_object_typed_new (UCL_ARRAY)); + auto cobj = obj.get (); + + for (const auto &e : v) { + ucl_array_append (cobj, ucl_object_ref (e.obj.get())); + } + } + + ucl_type_t type () const { + if (obj) { + return ucl_object_type (obj.get ()); + } + return UCL_NULL; + } + + const std::string key () const { + std::string res; + + if (obj->key) { + res.assign (obj->key, obj->keylen); + } + + return res; + } + + double number_value () const + { + if (obj) { + return ucl_object_todouble (obj.get()); + } + + return 0.0; + } + + int64_t int_value () const + { + if (obj) { + return ucl_object_toint (obj.get()); + } + + return 0; + } + + bool bool_value () const + { + if (obj) { + return ucl_object_toboolean (obj.get()); + } + + return false; + } + + const std::string string_value () const + { + std::string res; + + if (obj) { + res.assign (ucl_object_tostring (obj.get())); + } + + return res; + } + + const Ucl operator[] (size_t i) const + { + if (type () == UCL_ARRAY) { + return Ucl (ucl_array_find_index (obj.get(), i)); + } + + return Ucl (nullptr); + } + + const Ucl operator[](const std::string &key) const + { + if (type () == UCL_OBJECT) { + return Ucl (ucl_object_find_keyl (obj.get(), + key.data (), key.size ())); + } + + return Ucl (nullptr); + } + // Serialize. + void dump (std::string &out, ucl_emitter_t type = UCL_EMIT_JSON) const + { + struct ucl_emitter_functions cbdata; + + cbdata = Ucl::default_emit_funcs(); + cbdata.ud = reinterpret_cast<void *>(&out); + + ucl_object_emit_full (obj.get(), type, &cbdata); + } + + std::string dump (ucl_emitter_t type = UCL_EMIT_JSON) const + { + std::string out; + + dump (out, type); + + return out; + } + + static Ucl parse (const std::string & in, std::string & err) + { + auto parser = ucl_parser_new (UCL_PARSER_DEFAULT); + + if (!ucl_parser_add_chunk (parser, (const unsigned char *)in.data (), + in.size ())) { + err.assign (ucl_parser_get_error (parser)); + ucl_parser_free (parser); + + return nullptr; + } + + auto obj = ucl_parser_get_object (parser); + ucl_parser_free (parser); + + // Obj will handle ownership + return Ucl (obj); + } + + static Ucl parse (const char * in, std::string & err) + { + if (in) { + return parse (std::string(in), err); + } else { + err = "null input"; + return nullptr; + } + } + + static Ucl parse (std::istream &ifs, std::string &err) + { + return Ucl::parse (std::string(std::istreambuf_iterator<char>(ifs), + std::istreambuf_iterator<char>()), err); + } + + bool operator== (const Ucl &rhs) const + { + return ucl_object_compare (obj.get(), rhs.obj.get ()) == 0; + } + bool operator< (const Ucl &rhs) const + { + return ucl_object_compare (obj.get(), rhs.obj.get ()) < 0; + } + bool operator!= (const Ucl &rhs) const { return !(*this == rhs); } + bool operator<= (const Ucl &rhs) const { return !(rhs < *this); } + bool operator> (const Ucl &rhs) const { return (rhs < *this); } + bool operator>= (const Ucl &rhs) const { return !(*this < rhs); } + + operator bool () const + { + if (!obj || type() == UCL_NULL) { + return false; + } + + if (type () == UCL_BOOLEAN) { + return bool_value (); + } + + return true; + } + + const_iterator begin() const + { + return const_iterator(*this); + } + const_iterator cbegin() const + { + return const_iterator(*this); + } + const_iterator end() const + { + return const_iterator(); + } + const_iterator cend() const + { + return const_iterator(); + } +}; + +}; diff --git a/contrib/libucl/include/ucl.h b/contrib/libucl/include/ucl.h new file mode 100644 index 0000000..82a1fd1 --- /dev/null +++ b/contrib/libucl/include/ucl.h @@ -0,0 +1,1332 @@ +/* Copyright (c) 2013-2015, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef UCL_H_ +#define UCL_H_ + +#include <string.h> +#include <stddef.h> +#include <stdlib.h> +#include <stdint.h> +#include <stdbool.h> +#include <stdarg.h> +#include <stdio.h> + +#ifdef _WIN32 +# define UCL_EXTERN __declspec(dllexport) +#else +# define UCL_EXTERN +#endif + +/** + * @mainpage + * This is a reference manual for UCL API. You may find the description of UCL format by following this + * [github repository](https://github.com/vstakhov/libucl). + * + * This manual has several main sections: + * - @ref structures + * - @ref utils + * - @ref parser + * - @ref emitter + */ + +/** + * @file ucl.h + * @brief UCL parsing and emitting functions + * + * UCL is universal configuration language, which is a form of + * JSON with less strict rules that make it more comfortable for + * using as a configuration language + */ +#ifdef __cplusplus +extern "C" { +#endif +/* + * Memory allocation utilities + * UCL_ALLOC(size) - allocate memory for UCL + * UCL_FREE(size, ptr) - free memory of specified size at ptr + * Default: malloc and free + */ +#ifndef UCL_ALLOC +#define UCL_ALLOC(size) malloc(size) +#endif +#ifndef UCL_FREE +#define UCL_FREE(size, ptr) free(ptr) +#endif + +#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 4) +#define UCL_WARN_UNUSED_RESULT \ + __attribute__((warn_unused_result)) +#else +#define UCL_WARN_UNUSED_RESULT +#endif + +#ifdef __GNUC__ +#define UCL_DEPRECATED(func) func __attribute__ ((deprecated)) +#elif defined(_MSC_VER) +#define UCL_DEPRECATED(func) __declspec(deprecated) func +#else +#define UCL_DEPRECATED(func) func +#endif + +/** + * @defgroup structures Structures and types + * UCL defines several enumeration types used for error reporting or specifying flags and attributes. + * + * @{ + */ + +/** + * The common error codes returned by ucl parser + */ +typedef enum ucl_error { + UCL_EOK = 0, /**< No error */ + UCL_ESYNTAX, /**< Syntax error occurred during parsing */ + UCL_EIO, /**< IO error occurred during parsing */ + UCL_ESTATE, /**< Invalid state machine state */ + UCL_ENESTED, /**< Input has too many recursion levels */ + UCL_EMACRO, /**< Error processing a macro */ + UCL_EINTERNAL, /**< Internal unclassified error */ + UCL_ESSL /**< SSL error */ +} ucl_error_t; + +/** + * #ucl_object_t may have one of specified types, some types are compatible with each other and some are not. + * For example, you can always convert #UCL_TIME to #UCL_FLOAT. Also you can convert #UCL_FLOAT to #UCL_INTEGER + * by loosing floating point. Every object may be converted to a string by #ucl_object_tostring_forced() function. + * + */ +typedef enum ucl_type { + UCL_OBJECT = 0, /**< UCL object - key/value pairs */ + UCL_ARRAY, /**< UCL array */ + UCL_INT, /**< Integer number */ + UCL_FLOAT, /**< Floating point number */ + UCL_STRING, /**< Null terminated string */ + UCL_BOOLEAN, /**< Boolean value */ + UCL_TIME, /**< Time value (floating point number of seconds) */ + UCL_USERDATA, /**< Opaque userdata pointer (may be used in macros) */ + UCL_NULL /**< Null value */ +} ucl_type_t; + +/** + * You can use one of these types to serialise #ucl_object_t by using ucl_object_emit(). + */ +typedef enum ucl_emitter { + UCL_EMIT_JSON = 0, /**< Emit fine formatted JSON */ + UCL_EMIT_JSON_COMPACT, /**< Emit compacted JSON */ + UCL_EMIT_CONFIG, /**< Emit human readable config format */ + UCL_EMIT_YAML, /**< Emit embedded YAML format */ + UCL_EMIT_MSGPACK, /**< Emit msgpack output */ + UCL_EMIT_MAX /**< Unsupported emitter type */ +} ucl_emitter_t; + +/** + * These flags defines parser behaviour. If you specify #UCL_PARSER_ZEROCOPY you must ensure + * that the input memory is not freed if an object is in use. Moreover, if you want to use + * zero-terminated keys and string values then you should not use zero-copy mode, as in this case + * UCL still has to perform copying implicitly. + */ +typedef enum ucl_parser_flags { + UCL_PARSER_DEFAULT = 0x0, /**< No special flags */ + UCL_PARSER_KEY_LOWERCASE = 0x1, /**< Convert all keys to lower case */ + UCL_PARSER_ZEROCOPY = 0x2, /**< Parse input in zero-copy mode if possible */ + UCL_PARSER_NO_TIME = 0x4, /**< Do not parse time and treat time values as strings */ + UCL_PARSER_NO_IMPLICIT_ARRAYS = 0x8 /** Create explicit arrays instead of implicit ones */ +} ucl_parser_flags_t; + +/** + * String conversion flags, that are used in #ucl_object_fromstring_common function. + */ +typedef enum ucl_string_flags { + UCL_STRING_RAW = 0x0, /**< Treat string as is */ + UCL_STRING_ESCAPE = 0x1, /**< Perform JSON escape */ + UCL_STRING_TRIM = 0x2, /**< Trim leading and trailing whitespaces */ + UCL_STRING_PARSE_BOOLEAN = 0x4, /**< Parse passed string and detect boolean */ + UCL_STRING_PARSE_INT = 0x8, /**< Parse passed string and detect integer number */ + UCL_STRING_PARSE_DOUBLE = 0x10, /**< Parse passed string and detect integer or float number */ + UCL_STRING_PARSE_TIME = 0x20, /**< Parse time strings */ + UCL_STRING_PARSE_NUMBER = UCL_STRING_PARSE_INT|UCL_STRING_PARSE_DOUBLE|UCL_STRING_PARSE_TIME, /**< + Parse passed string and detect number */ + UCL_STRING_PARSE = UCL_STRING_PARSE_BOOLEAN|UCL_STRING_PARSE_NUMBER, /**< + Parse passed string (and detect booleans and numbers) */ + UCL_STRING_PARSE_BYTES = 0x40 /**< Treat numbers as bytes */ +} ucl_string_flags_t; + +/** + * Basic flags for an object + */ +typedef enum ucl_object_flags { + UCL_OBJECT_ALLOCATED_KEY = (1 << 0), /**< An object has key allocated internally */ + UCL_OBJECT_ALLOCATED_VALUE = (1 << 1), /**< An object has a string value allocated internally */ + UCL_OBJECT_NEED_KEY_ESCAPE = (1 << 2), /**< The key of an object need to be escaped on output */ + UCL_OBJECT_EPHEMERAL = (1 << 3), /**< Temporary object that does not need to be freed really */ + UCL_OBJECT_MULTILINE = (1 << 4), /**< String should be displayed as multiline string */ + UCL_OBJECT_MULTIVALUE = (1 << 5), /**< Object is a key with multiple values */ + UCL_OBJECT_INHERITED = (1 << 6), /**< Object has been inherited from another */ + UCL_OBJECT_BINARY = (1 << 7) /**< Object contains raw binary data */ +} ucl_object_flags_t; + +/** + * Duplicate policy types + */ +enum ucl_duplicate_strategy { + UCL_DUPLICATE_APPEND = 0, /**< Default policy to merge based on priorities */ + UCL_DUPLICATE_MERGE, /**< Merge new object with old one */ + UCL_DUPLICATE_REWRITE, /**< Rewrite old keys */ + UCL_DUPLICATE_ERROR /**< Stop parsing on duplicate found */ +}; + +/** + * Input format type + */ +enum ucl_parse_type { + UCL_PARSE_UCL = 0, /**< Default ucl format */ + UCL_PARSE_MSGPACK, /**< Message pack input format */ + UCL_PARSE_CSEXP /**< Canonical S-expressions */ +}; + +/** + * UCL object structure. Please mention that the most of fields should not be touched by + * UCL users. In future, this structure may be converted to private one. + */ +typedef struct ucl_object_s { + /** + * Variant value type + */ + union { + int64_t iv; /**< Int value of an object */ + const char *sv; /**< String value of an object */ + double dv; /**< Double value of an object */ + void *av; /**< Array */ + void *ov; /**< Object */ + void* ud; /**< Opaque user data */ + } value; + const char *key; /**< Key of an object */ + struct ucl_object_s *next; /**< Array handle */ + struct ucl_object_s *prev; /**< Array handle */ + uint32_t keylen; /**< Lenght of a key */ + uint32_t len; /**< Size of an object */ + uint32_t ref; /**< Reference count */ + uint16_t flags; /**< Object flags */ + uint16_t type; /**< Real type */ + unsigned char* trash_stack[2]; /**< Pointer to allocated chunks */ +} ucl_object_t; + +/** + * Destructor type for userdata objects + * @param ud user specified data pointer + */ +typedef void (*ucl_userdata_dtor)(void *ud); +typedef const char* (*ucl_userdata_emitter)(void *ud); + +/** @} */ + +/** + * @defgroup utils Utility functions + * A number of utility functions simplify handling of UCL objects + * + * @{ + */ +/** + * Copy and return a key of an object, returned key is zero-terminated + * @param obj CL object + * @return zero terminated key + */ +UCL_EXTERN char* ucl_copy_key_trash (const ucl_object_t *obj); + +/** + * Copy and return a string value of an object, returned key is zero-terminated + * @param obj CL object + * @return zero terminated string representation of object value + */ +UCL_EXTERN char* ucl_copy_value_trash (const ucl_object_t *obj); + +/** + * Creates a new object + * @return new object + */ +UCL_EXTERN ucl_object_t* ucl_object_new (void) UCL_WARN_UNUSED_RESULT; + +/** + * Create new object with type specified + * @param type type of a new object + * @return new object + */ +UCL_EXTERN ucl_object_t* ucl_object_typed_new (ucl_type_t type) UCL_WARN_UNUSED_RESULT; + +/** + * Create new object with type and priority specified + * @param type type of a new object + * @param priority priority of an object + * @return new object + */ +UCL_EXTERN ucl_object_t* ucl_object_new_full (ucl_type_t type, unsigned priority) + UCL_WARN_UNUSED_RESULT; + +/** + * Create new object with userdata dtor + * @param dtor destructor function + * @return new object + */ +UCL_EXTERN ucl_object_t* ucl_object_new_userdata (ucl_userdata_dtor dtor, + ucl_userdata_emitter emitter) UCL_WARN_UNUSED_RESULT; + +/** + * Perform deep copy of an object copying everything + * @param other object to copy + * @return new object with refcount equal to 1 + */ +UCL_EXTERN ucl_object_t * ucl_object_copy (const ucl_object_t *other) + UCL_WARN_UNUSED_RESULT; + +/** + * Return the type of an object + * @return the object type + */ +UCL_EXTERN ucl_type_t ucl_object_type (const ucl_object_t *obj); + +/** + * Convert any string to an ucl object making the specified transformations + * @param str fixed size or NULL terminated string + * @param len length (if len is zero, than str is treated as NULL terminated) + * @param flags conversion flags + * @return new object + */ +UCL_EXTERN ucl_object_t * ucl_object_fromstring_common (const char *str, size_t len, + enum ucl_string_flags flags) UCL_WARN_UNUSED_RESULT; + +/** + * Create a UCL object from the specified string + * @param str NULL terminated string, will be json escaped + * @return new object + */ +UCL_EXTERN ucl_object_t *ucl_object_fromstring (const char *str) UCL_WARN_UNUSED_RESULT; + +/** + * Create a UCL object from the specified string + * @param str fixed size string, will be json escaped + * @param len length of a string + * @return new object + */ +UCL_EXTERN ucl_object_t *ucl_object_fromlstring (const char *str, + size_t len) UCL_WARN_UNUSED_RESULT; + +/** + * Create an object from an integer number + * @param iv number + * @return new object + */ +UCL_EXTERN ucl_object_t* ucl_object_fromint (int64_t iv) UCL_WARN_UNUSED_RESULT; + +/** + * Create an object from a float number + * @param dv number + * @return new object + */ +UCL_EXTERN ucl_object_t* ucl_object_fromdouble (double dv) UCL_WARN_UNUSED_RESULT; + +/** + * Create an object from a boolean + * @param bv bool value + * @return new object + */ +UCL_EXTERN ucl_object_t* ucl_object_frombool (bool bv) UCL_WARN_UNUSED_RESULT; + +/** + * Insert a object 'elt' to the hash 'top' and associate it with key 'key' + * @param top destination object (must be of type UCL_OBJECT) + * @param elt element to insert (must NOT be NULL) + * @param key key to associate with this object (either const or preallocated) + * @param keylen length of the key (or 0 for NULL terminated keys) + * @param copy_key make an internal copy of key + * @return true if key has been inserted + */ +UCL_EXTERN bool ucl_object_insert_key (ucl_object_t *top, ucl_object_t *elt, + const char *key, size_t keylen, bool copy_key); + +/** + * Replace a object 'elt' to the hash 'top' and associate it with key 'key', old object will be unrefed, + * if no object has been found this function works like ucl_object_insert_key() + * @param top destination object (must be of type UCL_OBJECT) + * @param elt element to insert (must NOT be NULL) + * @param key key to associate with this object (either const or preallocated) + * @param keylen length of the key (or 0 for NULL terminated keys) + * @param copy_key make an internal copy of key + * @return true if key has been inserted + */ +UCL_EXTERN bool ucl_object_replace_key (ucl_object_t *top, ucl_object_t *elt, + const char *key, size_t keylen, bool copy_key); + +/** + * Merge the keys from one object to another object. Overwrite on conflict + * @param top destination object (must be of type UCL_OBJECT) + * @param elt element to insert (must be of type UCL_OBJECT) + * @param copy copy rather than reference the elements + * @return true if all keys have been merged + */ +UCL_EXTERN bool ucl_object_merge (ucl_object_t *top, ucl_object_t *elt, bool copy); + +/** + * Delete a object associated with key 'key', old object will be unrefered, + * @param top object + * @param key key associated to the object to remove + * @param keylen length of the key (or 0 for NULL terminated keys) + */ +UCL_EXTERN bool ucl_object_delete_keyl (ucl_object_t *top, + const char *key, size_t keylen); + +/** + * Delete a object associated with key 'key', old object will be unrefered, + * @param top object + * @param key key associated to the object to remove + */ +UCL_EXTERN bool ucl_object_delete_key (ucl_object_t *top, + const char *key); + + +/** + * Removes `key` from `top` object, returning the object that was removed. This + * object is not released, caller must unref the returned object when it is no + * longer needed. + * @param top object + * @param key key to remove + * @param keylen length of the key (or 0 for NULL terminated keys) + * @return removed object or NULL if object has not been found + */ +UCL_EXTERN ucl_object_t* ucl_object_pop_keyl (ucl_object_t *top, const char *key, + size_t keylen) UCL_WARN_UNUSED_RESULT; + +/** + * Removes `key` from `top` object returning the object that was removed. This + * object is not released, caller must unref the returned object when it is no + * longer needed. + * @param top object + * @param key key to remove + * @return removed object or NULL if object has not been found + */ +UCL_EXTERN ucl_object_t* ucl_object_pop_key (ucl_object_t *top, const char *key) + UCL_WARN_UNUSED_RESULT; + +/** + * Insert a object 'elt' to the hash 'top' and associate it with key 'key', if + * the specified key exist, try to merge its content + * @param top destination object (must be of type UCL_OBJECT) + * @param elt element to insert (must NOT be NULL) + * @param key key to associate with this object (either const or preallocated) + * @param keylen length of the key (or 0 for NULL terminated keys) + * @param copy_key make an internal copy of key + * @return true if key has been inserted + */ +UCL_EXTERN bool ucl_object_insert_key_merged (ucl_object_t *top, ucl_object_t *elt, + const char *key, size_t keylen, bool copy_key); + +/** + * Append an element to the end of array object + * @param top destination object (must NOT be NULL) + * @param elt element to append (must NOT be NULL) + * @return true if value has been inserted + */ +UCL_EXTERN bool ucl_array_append (ucl_object_t *top, + ucl_object_t *elt); + +/** + * Append an element to the start of array object + * @param top destination object (must NOT be NULL) + * @param elt element to append (must NOT be NULL) + * @return true if value has been inserted + */ +UCL_EXTERN bool ucl_array_prepend (ucl_object_t *top, + ucl_object_t *elt); + +/** + * Merge all elements of second array into the first array + * @param top destination array (must be of type UCL_ARRAY) + * @param elt array to copy elements from (must be of type UCL_ARRAY) + * @param copy copy elements instead of referencing them + * @return true if arrays were merged + */ +UCL_EXTERN bool ucl_array_merge (ucl_object_t *top, ucl_object_t *elt, + bool copy); + +/** + * Removes an element `elt` from the array `top`, returning the object that was + * removed. This object is not released, caller must unref the returned object + * when it is no longer needed. + * @param top array ucl object + * @param elt element to remove + * @return removed element or NULL if `top` is NULL or not an array + */ +UCL_EXTERN ucl_object_t* ucl_array_delete (ucl_object_t *top, + ucl_object_t *elt); + +/** + * Returns the first element of the array `top` + * @param top array ucl object + * @return element or NULL if `top` is NULL or not an array + */ +UCL_EXTERN const ucl_object_t* ucl_array_head (const ucl_object_t *top); + +/** + * Returns the last element of the array `top` + * @param top array ucl object + * @return element or NULL if `top` is NULL or not an array + */ +UCL_EXTERN const ucl_object_t* ucl_array_tail (const ucl_object_t *top); + +/** + * Removes the last element from the array `top`, returning the object that was + * removed. This object is not released, caller must unref the returned object + * when it is no longer needed. + * @param top array ucl object + * @return removed element or NULL if `top` is NULL or not an array + */ +UCL_EXTERN ucl_object_t* ucl_array_pop_last (ucl_object_t *top); + +/** + * Removes the first element from the array `top`, returning the object that was + * removed. This object is not released, caller must unref the returned object + * when it is no longer needed. + * @param top array ucl object + * @return removed element or NULL if `top` is NULL or not an array + */ +UCL_EXTERN ucl_object_t* ucl_array_pop_first (ucl_object_t *top); + +/** + * Return object identified by index of the array `top` + * @param top object to get a key from (must be of type UCL_ARRAY) + * @param index array index to return + * @return object at the specified index or NULL if index is not found + */ +UCL_EXTERN const ucl_object_t* ucl_array_find_index (const ucl_object_t *top, + unsigned int index); + +/** + * Return the index of `elt` in the array `top` + * @param top object to get a key from (must be of type UCL_ARRAY) + * @param elt element to find index of (must NOT be NULL) + * @return index of `elt` in the array `top or (unsigned int)-1 if `elt` is not found + */ +UCL_EXTERN unsigned int ucl_array_index_of (ucl_object_t *top, + ucl_object_t *elt); + +/** + * Replace an element in an array with a different element, returning the object + * that was replaced. This object is not released, caller must unref the + * returned object when it is no longer needed. + * @param top destination object (must be of type UCL_ARRAY) + * @param elt element to append (must NOT be NULL) + * @param index array index in destination to overwrite with elt + * @return object that was replaced or NULL if index is not found + */ +ucl_object_t * +ucl_array_replace_index (ucl_object_t *top, ucl_object_t *elt, + unsigned int index); + +/** + * Append a element to another element forming an implicit array + * @param head head to append (may be NULL) + * @param elt new element + * @return the new implicit array + */ +UCL_EXTERN ucl_object_t * ucl_elt_append (ucl_object_t *head, + ucl_object_t *elt); + +/** + * Converts an object to double value + * @param obj CL object + * @param target target double variable + * @return true if conversion was successful + */ +UCL_EXTERN bool ucl_object_todouble_safe (const ucl_object_t *obj, double *target); + +/** + * Unsafe version of \ref ucl_obj_todouble_safe + * @param obj CL object + * @return double value + */ +UCL_EXTERN double ucl_object_todouble (const ucl_object_t *obj); + +/** + * Converts an object to integer value + * @param obj CL object + * @param target target integer variable + * @return true if conversion was successful + */ +UCL_EXTERN bool ucl_object_toint_safe (const ucl_object_t *obj, int64_t *target); + +/** + * Unsafe version of \ref ucl_obj_toint_safe + * @param obj CL object + * @return int value + */ +UCL_EXTERN int64_t ucl_object_toint (const ucl_object_t *obj); + +/** + * Converts an object to boolean value + * @param obj CL object + * @param target target boolean variable + * @return true if conversion was successful + */ +UCL_EXTERN bool ucl_object_toboolean_safe (const ucl_object_t *obj, bool *target); + +/** + * Unsafe version of \ref ucl_obj_toboolean_safe + * @param obj CL object + * @return boolean value + */ +UCL_EXTERN bool ucl_object_toboolean (const ucl_object_t *obj); + +/** + * Converts an object to string value + * @param obj CL object + * @param target target string variable, no need to free value + * @return true if conversion was successful + */ +UCL_EXTERN bool ucl_object_tostring_safe (const ucl_object_t *obj, const char **target); + +/** + * Unsafe version of \ref ucl_obj_tostring_safe + * @param obj CL object + * @return string value + */ +UCL_EXTERN const char* ucl_object_tostring (const ucl_object_t *obj); + +/** + * Convert any object to a string in JSON notation if needed + * @param obj CL object + * @return string value + */ +UCL_EXTERN const char* ucl_object_tostring_forced (const ucl_object_t *obj); + +/** + * Return string as char * and len, string may be not zero terminated, more efficient that \ref ucl_obj_tostring as it + * allows zero-copy (if #UCL_PARSER_ZEROCOPY has been used during parsing) + * @param obj CL object + * @param target target string variable, no need to free value + * @param tlen target length + * @return true if conversion was successful + */ +UCL_EXTERN bool ucl_object_tolstring_safe (const ucl_object_t *obj, + const char **target, size_t *tlen); + +/** + * Unsafe version of \ref ucl_obj_tolstring_safe + * @param obj CL object + * @return string value + */ +UCL_EXTERN const char* ucl_object_tolstring (const ucl_object_t *obj, size_t *tlen); + +/** + * Return object identified by a key in the specified object + * @param obj object to get a key from (must be of type UCL_OBJECT) + * @param key key to search + * @return object matching the specified key or NULL if key was not found + */ +UCL_EXTERN const ucl_object_t* ucl_object_find_key (const ucl_object_t *obj, + const char *key); + +/** + * Return object identified by a key in the specified object, if the first key is + * not found then look for the next one. This process is repeated unless + * the next argument in the list is not NULL. So, `ucl_object_find_any_key(obj, key, NULL)` + * is equal to `ucl_object_find_key(obj, key)` + * @param obj object to get a key from (must be of type UCL_OBJECT) + * @param key key to search + * @param ... list of alternative keys to search (NULL terminated) + * @return object matching the specified key or NULL if key was not found + */ +UCL_EXTERN const ucl_object_t* ucl_object_find_any_key (const ucl_object_t *obj, + const char *key, ...); + +/** + * Return object identified by a fixed size key in the specified object + * @param obj object to get a key from (must be of type UCL_OBJECT) + * @param key key to search + * @param klen length of a key + * @return object matching the specified key or NULL if key was not found + */ +UCL_EXTERN const ucl_object_t* ucl_object_find_keyl (const ucl_object_t *obj, + const char *key, size_t klen); + +/** + * Return object identified by dot notation string + * @param obj object to search in + * @param path dot.notation.path to the path to lookup. May use numeric .index on arrays + * @return object matched the specified path or NULL if path is not found + */ +UCL_EXTERN const ucl_object_t *ucl_lookup_path (const ucl_object_t *obj, + const char *path); + +/** + * Return object identified by object notation string using arbitrary delimiter + * @param obj object to search in + * @param path dot.notation.path to the path to lookup. May use numeric .index on arrays + * @param sep the sepatorator to use in place of . (incase keys have . in them) + * @return object matched the specified path or NULL if path is not found + */ +UCL_EXTERN const ucl_object_t *ucl_lookup_path_char (const ucl_object_t *obj, + const char *path, char sep); + +/** + * Returns a key of an object as a NULL terminated string + * @param obj CL object + * @return key or NULL if there is no key + */ +UCL_EXTERN const char* ucl_object_key (const ucl_object_t *obj); + +/** + * Returns a key of an object as a fixed size string (may be more efficient) + * @param obj CL object + * @param len target key length + * @return key pointer + */ +UCL_EXTERN const char* ucl_object_keyl (const ucl_object_t *obj, size_t *len); + +/** + * Increase reference count for an object + * @param obj object to ref + * @return the referenced object + */ +UCL_EXTERN ucl_object_t* ucl_object_ref (const ucl_object_t *obj); + +/** + * Free ucl object + * @param obj ucl object to free + */ +UCL_DEPRECATED(UCL_EXTERN void ucl_object_free (ucl_object_t *obj)); + +/** + * Decrease reference count for an object + * @param obj object to unref + */ +UCL_EXTERN void ucl_object_unref (ucl_object_t *obj); + +/** + * Compare objects `o1` and `o2` + * @param o1 the first object + * @param o2 the second object + * @return values >0, 0 and <0 if `o1` is more than, equal and less than `o2`. + * The order of comparison: + * 1) Type of objects + * 2) Size of objects + * 3) Content of objects + */ +UCL_EXTERN int ucl_object_compare (const ucl_object_t *o1, + const ucl_object_t *o2); + +/** + * Sort UCL array using `cmp` compare function + * @param ar + * @param cmp + */ +UCL_EXTERN void ucl_object_array_sort (ucl_object_t *ar, + int (*cmp)(const ucl_object_t **o1, const ucl_object_t **o2)); + +/** + * Get the priority for specific UCL object + * @param obj any ucl object + * @return priority of an object + */ +UCL_EXTERN unsigned int ucl_object_get_priority (const ucl_object_t *obj); + +/** + * Set explicit priority of an object. + * @param obj any ucl object + * @param priority new priroity value (only 4 least significant bits are considred) + */ +UCL_EXTERN void ucl_object_set_priority (ucl_object_t *obj, + unsigned int priority); + +/** + * Opaque iterator object + */ +typedef void* ucl_object_iter_t; + +/** + * Get next key from an object + * @param obj object to iterate + * @param iter opaque iterator, must be set to NULL on the first call: + * ucl_object_iter_t it = NULL; + * while ((cur = ucl_iterate_object (obj, &it)) != NULL) ... + * @return the next object or NULL + */ +UCL_EXTERN const ucl_object_t* ucl_iterate_object (const ucl_object_t *obj, + ucl_object_iter_t *iter, bool expand_values); + +/** + * Create new safe iterator for the specified object + * @param obj object to iterate + * @return new iterator object that should be used with safe iterators API only + */ +UCL_EXTERN ucl_object_iter_t ucl_object_iterate_new (const ucl_object_t *obj) + UCL_WARN_UNUSED_RESULT; +/** + * Reset initialized iterator to a new object + * @param obj new object to iterate + * @return modified iterator object + */ +UCL_EXTERN ucl_object_iter_t ucl_object_iterate_reset (ucl_object_iter_t it, + const ucl_object_t *obj); + +/** + * Get the next object from the `obj`. This fucntion iterates over arrays, objects + * and implicit arrays + * @param iter safe iterator + * @return the next object in sequence + */ +UCL_EXTERN const ucl_object_t* ucl_object_iterate_safe (ucl_object_iter_t iter, + bool expand_values); + +/** + * Free memory associated with the safe iterator + * @param it safe iterator object + */ +UCL_EXTERN void ucl_object_iterate_free (ucl_object_iter_t it); + +/** @} */ + + +/** + * @defgroup parser Parsing functions + * These functions are used to parse UCL objects + * + * @{ + */ + +/** + * Macro handler for a parser + * @param data the content of macro + * @param len the length of content + * @param arguments arguments object + * @param ud opaque user data + * @param err error pointer + * @return true if macro has been parsed + */ +typedef bool (*ucl_macro_handler) (const unsigned char *data, size_t len, + const ucl_object_t *arguments, + void* ud); + +/** + * Context dependent macro handler for a parser + * @param data the content of macro + * @param len the length of content + * @param arguments arguments object + * @param context previously parsed context + * @param ud opaque user data + * @param err error pointer + * @return true if macro has been parsed + */ +typedef bool (*ucl_context_macro_handler) (const unsigned char *data, size_t len, + const ucl_object_t *arguments, + const ucl_object_t *context, + void* ud); + +/* Opaque parser */ +struct ucl_parser; + +/** + * Creates new parser object + * @param pool pool to allocate memory from + * @return new parser object + */ +UCL_EXTERN struct ucl_parser* ucl_parser_new (int flags); + +/** + * Sets the default priority for the parser applied to chunks that does not + * specify priority explicitly + * @param parser parser object + * @param prio default priority (0 .. 16) + * @return true if parser's default priority was set + */ +UCL_EXTERN bool ucl_parser_set_default_priority (struct ucl_parser *parser, + unsigned prio); +/** + * Register new handler for a macro + * @param parser parser object + * @param macro macro name (without leading dot) + * @param handler handler (it is called immediately after macro is parsed) + * @param ud opaque user data for a handler + */ +UCL_EXTERN void ucl_parser_register_macro (struct ucl_parser *parser, + const char *macro, + ucl_macro_handler handler, void* ud); + +/** + * Register new context dependent handler for a macro + * @param parser parser object + * @param macro macro name (without leading dot) + * @param handler handler (it is called immediately after macro is parsed) + * @param ud opaque user data for a handler + */ +UCL_EXTERN void ucl_parser_register_context_macro (struct ucl_parser *parser, + const char *macro, + ucl_context_macro_handler handler, + void* ud); + +/** + * Handler to detect unregistered variables + * @param data variable data + * @param len length of variable + * @param replace (out) replace value for variable + * @param replace_len (out) replace length for variable + * @param need_free (out) UCL will free `dest` after usage + * @param ud opaque userdata + * @return true if variable + */ +typedef bool (*ucl_variable_handler) (const unsigned char *data, size_t len, + unsigned char **replace, size_t *replace_len, bool *need_free, void* ud); + +/** + * Register new parser variable + * @param parser parser object + * @param var variable name + * @param value variable value + */ +UCL_EXTERN void ucl_parser_register_variable (struct ucl_parser *parser, const char *var, + const char *value); + +/** + * Set handler for unknown variables + * @param parser parser structure + * @param handler desired handler + * @param ud opaque data for the handler + */ +UCL_EXTERN void ucl_parser_set_variables_handler (struct ucl_parser *parser, + ucl_variable_handler handler, void *ud); + +/** + * Load new chunk to a parser + * @param parser parser structure + * @param data the pointer to the beginning of a chunk + * @param len the length of a chunk + * @return true if chunk has been added and false in case of error + */ +UCL_EXTERN bool ucl_parser_add_chunk (struct ucl_parser *parser, + const unsigned char *data, size_t len); + +/** + * Load new chunk to a parser with the specified priority + * @param parser parser structure + * @param data the pointer to the beginning of a chunk + * @param len the length of a chunk + * @param priority the desired priority of a chunk (only 4 least significant bits + * are considered for this parameter) + * @return true if chunk has been added and false in case of error + */ +UCL_EXTERN bool ucl_parser_add_chunk_priority (struct ucl_parser *parser, + const unsigned char *data, size_t len, unsigned priority); + +/** + * Full version of ucl_add_chunk with priority and duplicate strategy + * @param parser parser structure + * @param data the pointer to the beginning of a chunk + * @param len the length of a chunk + * @param priority the desired priority of a chunk (only 4 least significant bits + * are considered for this parameter) + * @param strat duplicates merging strategy + * @param parse_type input format + * @return true if chunk has been added and false in case of error + */ +UCL_EXTERN bool ucl_parser_add_chunk_full (struct ucl_parser *parser, + const unsigned char *data, size_t len, unsigned priority, + enum ucl_duplicate_strategy strat, enum ucl_parse_type parse_type); + +/** + * Load ucl object from a string + * @param parser parser structure + * @param data the pointer to the string + * @param len the length of the string, if `len` is 0 then `data` must be zero-terminated string + * @return true if string has been added and false in case of error + */ +UCL_EXTERN bool ucl_parser_add_string (struct ucl_parser *parser, + const char *data,size_t len); + +/** + * Load ucl object from a string + * @param parser parser structure + * @param data the pointer to the string + * @param len the length of the string, if `len` is 0 then `data` must be zero-terminated string + * @param priority the desired priority of a chunk (only 4 least significant bits + * are considered for this parameter) + * @return true if string has been added and false in case of error + */ +UCL_EXTERN bool ucl_parser_add_string_priority (struct ucl_parser *parser, + const char *data, size_t len, unsigned priority); + +/** + * Load and add data from a file + * @param parser parser structure + * @param filename the name of file + * @param err if *err is NULL it is set to parser error + * @return true if chunk has been added and false in case of error + */ +UCL_EXTERN bool ucl_parser_add_file (struct ucl_parser *parser, + const char *filename); + +/** + * Load and add data from a file + * @param parser parser structure + * @param filename the name of file + * @param err if *err is NULL it is set to parser error + * @param priority the desired priority of a chunk (only 4 least significant bits + * are considered for this parameter) + * @return true if chunk has been added and false in case of error + */ +UCL_EXTERN bool ucl_parser_add_file_priority (struct ucl_parser *parser, + const char *filename, unsigned priority); + +/** + * Load and add data from a file descriptor + * @param parser parser structure + * @param filename the name of file + * @param err if *err is NULL it is set to parser error + * @return true if chunk has been added and false in case of error + */ +UCL_EXTERN bool ucl_parser_add_fd (struct ucl_parser *parser, + int fd); + +/** + * Load and add data from a file descriptor + * @param parser parser structure + * @param filename the name of file + * @param err if *err is NULL it is set to parser error + * @param priority the desired priority of a chunk (only 4 least significant bits + * are considered for this parameter) + * @return true if chunk has been added and false in case of error + */ +UCL_EXTERN bool ucl_parser_add_fd_priority (struct ucl_parser *parser, + int fd, unsigned priority); + +/** + * Provide a UCL_ARRAY of paths to search for include files. The object is + * copied so caller must unref the object. + * @param parser parser structure + * @param paths UCL_ARRAY of paths to search + * @return true if the path search array was replaced in the parser + */ +UCL_EXTERN bool ucl_set_include_path (struct ucl_parser *parser, + ucl_object_t *paths); + +/** + * Get a top object for a parser (refcount is increased) + * @param parser parser structure + * @param err if *err is NULL it is set to parser error + * @return top parser object or NULL + */ +UCL_EXTERN ucl_object_t* ucl_parser_get_object (struct ucl_parser *parser); + +/** + * Get the error string if parsing has been failed + * @param parser parser object + * @return error description + */ +UCL_EXTERN const char *ucl_parser_get_error(struct ucl_parser *parser); + +/** + * Get the code of the last error + * @param parser parser object + * @return error code + */ +UCL_EXTERN int ucl_parser_get_error_code(struct ucl_parser *parser); + +/** + * Get the current column number within parser + * @param parser parser object + * @return current column number + */ +UCL_EXTERN unsigned ucl_parser_get_column(struct ucl_parser *parser); + +/** + * Get the current line number within parser + * @param parser parser object + * @return current line number + */ +UCL_EXTERN unsigned ucl_parser_get_linenum(struct ucl_parser *parser); + +/** + * Clear the error in the parser + * @param parser parser object + */ +UCL_EXTERN void ucl_parser_clear_error(struct ucl_parser *parser); + +/** + * Free ucl parser object + * @param parser parser object + */ +UCL_EXTERN void ucl_parser_free (struct ucl_parser *parser); + +/** + * Add new public key to parser for signatures check + * @param parser parser object + * @param key PEM representation of a key + * @param len length of the key + * @param err if *err is NULL it is set to parser error + * @return true if a key has been successfully added + */ +UCL_EXTERN bool ucl_pubkey_add (struct ucl_parser *parser, const unsigned char *key, size_t len); + +/** + * Set FILENAME and CURDIR variables in parser + * @param parser parser object + * @param filename filename to set or NULL to set FILENAME to "undef" and CURDIR to getcwd() + * @param need_expand perform realpath() if this variable is true and filename is not NULL + * @return true if variables has been set + */ +UCL_EXTERN bool ucl_parser_set_filevars (struct ucl_parser *parser, const char *filename, + bool need_expand); + +/** @} */ + +/** + * @defgroup emitter Emitting functions + * These functions are used to serialise UCL objects to some string representation. + * + * @{ + */ + +struct ucl_emitter_context; +/** + * Structure using for emitter callbacks + */ +struct ucl_emitter_functions { + /** Append a single character */ + int (*ucl_emitter_append_character) (unsigned char c, size_t nchars, void *ud); + /** Append a string of a specified length */ + int (*ucl_emitter_append_len) (unsigned const char *str, size_t len, void *ud); + /** Append a 64 bit integer */ + int (*ucl_emitter_append_int) (int64_t elt, void *ud); + /** Append floating point element */ + int (*ucl_emitter_append_double) (double elt, void *ud); + /** Free userdata */ + void (*ucl_emitter_free_func)(void *ud); + /** Opaque userdata pointer */ + void *ud; +}; + +struct ucl_emitter_operations { + /** Write a primitive element */ + void (*ucl_emitter_write_elt) (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool first, bool print_key); + /** Start ucl object */ + void (*ucl_emitter_start_object) (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool print_key); + /** End ucl object */ + void (*ucl_emitter_end_object) (struct ucl_emitter_context *ctx, + const ucl_object_t *obj); + /** Start ucl array */ + void (*ucl_emitter_start_array) (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool print_key); + void (*ucl_emitter_end_array) (struct ucl_emitter_context *ctx, + const ucl_object_t *obj); +}; + +/** + * Structure that defines emitter functions + */ +struct ucl_emitter_context { + /** Name of emitter (e.g. json, compact_json) */ + const char *name; + /** Unique id (e.g. UCL_EMIT_JSON for standard emitters */ + int id; + /** A set of output functions */ + const struct ucl_emitter_functions *func; + /** A set of output operations */ + const struct ucl_emitter_operations *ops; + /** Current amount of indent tabs */ + unsigned int indent; + /** Top level object */ + const ucl_object_t *top; + /** The rest of context */ + unsigned char data[1]; +}; + +/** + * Emit object to a string + * @param obj object + * @param emit_type if type is #UCL_EMIT_JSON then emit json, if type is + * #UCL_EMIT_CONFIG then emit config like object + * @return dump of an object (must be freed after using) or NULL in case of error + */ +UCL_EXTERN unsigned char *ucl_object_emit (const ucl_object_t *obj, + enum ucl_emitter emit_type); + +/** + * Emit object to a string that can contain `\0` inside + * @param obj object + * @param emit_type if type is #UCL_EMIT_JSON then emit json, if type is + * #UCL_EMIT_CONFIG then emit config like object + * @param len the resulting length + * @return dump of an object (must be freed after using) or NULL in case of error + */ +UCL_EXTERN unsigned char *ucl_object_emit_len (const ucl_object_t *obj, + enum ucl_emitter emit_type, size_t *len); + +/** + * Emit object to a string + * @param obj object + * @param emit_type if type is #UCL_EMIT_JSON then emit json, if type is + * #UCL_EMIT_CONFIG then emit config like object + * @param emitter a set of emitter functions + * @return dump of an object (must be freed after using) or NULL in case of error + */ +UCL_EXTERN bool ucl_object_emit_full (const ucl_object_t *obj, + enum ucl_emitter emit_type, + struct ucl_emitter_functions *emitter); + +/** + * Start streamlined UCL object emitter + * @param obj top UCL object + * @param emit_type emit type + * @param emitter a set of emitter functions + * @return new streamlined context that should be freed by + * `ucl_object_emit_streamline_finish` + */ +UCL_EXTERN struct ucl_emitter_context* ucl_object_emit_streamline_new ( + const ucl_object_t *obj, enum ucl_emitter emit_type, + struct ucl_emitter_functions *emitter); + +/** + * Start object or array container for the streamlined output + * @param ctx streamlined context + * @param obj container object + */ +UCL_EXTERN void ucl_object_emit_streamline_start_container ( + struct ucl_emitter_context *ctx, const ucl_object_t *obj); +/** + * Add a complete UCL object to streamlined output + * @param ctx streamlined context + * @param obj object to output + */ +UCL_EXTERN void ucl_object_emit_streamline_add_object ( + struct ucl_emitter_context *ctx, const ucl_object_t *obj); +/** + * End previously added container + * @param ctx streamlined context + */ +UCL_EXTERN void ucl_object_emit_streamline_end_container ( + struct ucl_emitter_context *ctx); +/** + * Terminate streamlined container finishing all containers in it + * @param ctx streamlined context + */ +UCL_EXTERN void ucl_object_emit_streamline_finish ( + struct ucl_emitter_context *ctx); + +/** + * Returns functions to emit object to memory + * @param pmem target pointer (should be freed by caller) + * @return emitter functions structure + */ +UCL_EXTERN struct ucl_emitter_functions* ucl_object_emit_memory_funcs ( + void **pmem); + +/** + * Returns functions to emit object to FILE * + * @param fp FILE * object + * @return emitter functions structure + */ +UCL_EXTERN struct ucl_emitter_functions* ucl_object_emit_file_funcs ( + FILE *fp); +/** + * Returns functions to emit object to a file descriptor + * @param fd file descriptor + * @return emitter functions structure + */ +UCL_EXTERN struct ucl_emitter_functions* ucl_object_emit_fd_funcs ( + int fd); + +/** + * Free emitter functions + * @param f pointer to functions + */ +UCL_EXTERN void ucl_object_emit_funcs_free (struct ucl_emitter_functions *f); + +/** @} */ + +/** + * @defgroup schema Schema functions + * These functions are used to validate UCL objects using json schema format + * + * @{ + */ + +/** + * Used to define UCL schema error + */ +enum ucl_schema_error_code { + UCL_SCHEMA_OK = 0, /**< no error */ + UCL_SCHEMA_TYPE_MISMATCH, /**< type of object is incorrect */ + UCL_SCHEMA_INVALID_SCHEMA, /**< schema is invalid */ + UCL_SCHEMA_MISSING_PROPERTY,/**< one or more missing properties */ + UCL_SCHEMA_CONSTRAINT, /**< constraint found */ + UCL_SCHEMA_MISSING_DEPENDENCY, /**< missing dependency */ + UCL_SCHEMA_UNKNOWN /**< generic error */ +}; + +/** + * Generic ucl schema error + */ +struct ucl_schema_error { + enum ucl_schema_error_code code; /**< error code */ + char msg[128]; /**< error message */ + const ucl_object_t *obj; /**< object where error occured */ +}; + +/** + * Validate object `obj` using schema object `schema`. + * @param schema schema object + * @param obj object to validate + * @param err error pointer, if this parameter is not NULL and error has been + * occured, then `err` is filled with the exact error definition. + * @return true if `obj` is valid using `schema` + */ +UCL_EXTERN bool ucl_object_validate (const ucl_object_t *schema, + const ucl_object_t *obj, struct ucl_schema_error *err); + +/** @} */ + +#ifdef __cplusplus +} +#endif +/* + * XXX: Poorly named API functions, need to replace them with the appropriate + * named function. All API functions *must* use naming ucl_object_*. Usage of + * ucl_obj* should be avoided. + */ +#define ucl_obj_todouble_safe ucl_object_todouble_safe +#define ucl_obj_todouble ucl_object_todouble +#define ucl_obj_tostring ucl_object_tostring +#define ucl_obj_tostring_safe ucl_object_tostring_safe +#define ucl_obj_tolstring ucl_object_tolstring +#define ucl_obj_tolstring_safe ucl_object_tolstring_safe +#define ucl_obj_toint ucl_object_toint +#define ucl_obj_toint_safe ucl_object_toint_safe +#define ucl_obj_toboolean ucl_object_toboolean +#define ucl_obj_toboolean_safe ucl_object_toboolean_safe +#define ucl_obj_get_key ucl_object_find_key +#define ucl_obj_get_keyl ucl_object_find_keyl +#define ucl_obj_unref ucl_object_unref +#define ucl_obj_ref ucl_object_ref +#define ucl_obj_free ucl_object_free + +#endif /* UCL_H_ */ diff --git a/contrib/libucl/klib/khash.h b/contrib/libucl/klib/khash.h new file mode 100644 index 0000000..afc3ce3 --- /dev/null +++ b/contrib/libucl/klib/khash.h @@ -0,0 +1,627 @@ +/* The MIT License + + Copyright (c) 2008, 2009, 2011 by Attractive Chaos <attractor@live.co.uk> + + Permission is hereby granted, free of charge, to any person obtaining + a copy of this software and associated documentation files (the + "Software"), to deal in the Software without restriction, including + without limitation the rights to use, copy, modify, merge, publish, + distribute, sublicense, and/or sell copies of the Software, and to + permit persons to whom the Software is furnished to do so, subject to + the following conditions: + + The above copyright notice and this permission notice shall be + included in all copies or substantial portions of the Software. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + SOFTWARE. +*/ + +/* + An example: + +#include "khash.h" +KHASH_MAP_INIT_INT(32, char) +int main() { + int ret, is_missing; + khiter_t k; + khash_t(32) *h = kh_init(32); + k = kh_put(32, h, 5, &ret); + kh_value(h, k) = 10; + k = kh_get(32, h, 10); + is_missing = (k == kh_end(h)); + k = kh_get(32, h, 5); + kh_del(32, h, k); + for (k = kh_begin(h); k != kh_end(h); ++k) + if (kh_exist(h, k)) kh_value(h, k) = 1; + kh_destroy(32, h); + return 0; +} +*/ + +/* + 2013-05-02 (0.2.8): + + * Use quadratic probing. When the capacity is power of 2, stepping function + i*(i+1)/2 guarantees to traverse each bucket. It is better than double + hashing on cache performance and is more robust than linear probing. + + In theory, double hashing should be more robust than quadratic probing. + However, my implementation is probably not for large hash tables, because + the second hash function is closely tied to the first hash function, + which reduce the effectiveness of double hashing. + + Reference: http://research.cs.vt.edu/AVresearch/hashing/quadratic.php + + 2011-12-29 (0.2.7): + + * Minor code clean up; no actual effect. + + 2011-09-16 (0.2.6): + + * The capacity is a power of 2. This seems to dramatically improve the + speed for simple keys. Thank Zilong Tan for the suggestion. Reference: + + - http://code.google.com/p/ulib/ + - http://nothings.org/computer/judy/ + + * Allow to optionally use linear probing which usually has better + performance for random input. Double hashing is still the default as it + is more robust to certain non-random input. + + * Added Wang's integer hash function (not used by default). This hash + function is more robust to certain non-random input. + + 2011-02-14 (0.2.5): + + * Allow to declare global functions. + + 2009-09-26 (0.2.4): + + * Improve portability + + 2008-09-19 (0.2.3): + + * Corrected the example + * Improved interfaces + + 2008-09-11 (0.2.2): + + * Improved speed a little in kh_put() + + 2008-09-10 (0.2.1): + + * Added kh_clear() + * Fixed a compiling error + + 2008-09-02 (0.2.0): + + * Changed to token concatenation which increases flexibility. + + 2008-08-31 (0.1.2): + + * Fixed a bug in kh_get(), which has not been tested previously. + + 2008-08-31 (0.1.1): + + * Added destructor +*/ + + +#ifndef __AC_KHASH_H +#define __AC_KHASH_H + +/*! + @header + + Generic hash table library. + */ + +#define AC_VERSION_KHASH_H "0.2.8" + +#include <stdlib.h> +#include <string.h> +#include <limits.h> + +/* compiler specific configuration */ + +#if UINT_MAX == 0xffffffffu +typedef unsigned int khint32_t; +#elif ULONG_MAX == 0xffffffffu +typedef unsigned long khint32_t; +#endif + +#if ULONG_MAX == ULLONG_MAX +typedef unsigned long khint64_t; +#else +typedef unsigned long long khint64_t; +#endif + +#ifndef kh_inline +#ifdef _MSC_VER +#define kh_inline __inline +#else +#define kh_inline inline +#endif +#endif /* kh_inline */ + +#ifndef kh_unused +# ifdef __GNUC__ +# define kh_unused(x) __attribute__((__unused__)) x +# else +# define kh_unused(x) x +# endif +#endif + +typedef khint32_t khint_t; +typedef khint_t khiter_t; + +#define __ac_isempty(flag, i) ((flag[i>>4]>>((i&0xfU)<<1))&2) +#define __ac_isdel(flag, i) ((flag[i>>4]>>((i&0xfU)<<1))&1) +#define __ac_iseither(flag, i) ((flag[i>>4]>>((i&0xfU)<<1))&3) +#define __ac_set_isdel_false(flag, i) (flag[i>>4]&=~(1ul<<((i&0xfU)<<1))) +#define __ac_set_isempty_false(flag, i) (flag[i>>4]&=~(2ul<<((i&0xfU)<<1))) +#define __ac_set_isboth_false(flag, i) (flag[i>>4]&=~(3ul<<((i&0xfU)<<1))) +#define __ac_set_isdel_true(flag, i) (flag[i>>4]|=1ul<<((i&0xfU)<<1)) + +#define __ac_fsize(m) ((m) < 16? 1 : (m)>>4) + +#ifndef kroundup32 +#define kroundup32(x) (--(x), (x)|=(x)>>1, (x)|=(x)>>2, (x)|=(x)>>4, (x)|=(x)>>8, (x)|=(x)>>16, ++(x)) +#endif + +#ifndef kcalloc +#define kcalloc(N,Z) calloc(N,Z) +#endif +#ifndef kmalloc +#define kmalloc(Z) malloc(Z) +#endif +#ifndef krealloc +#define krealloc(P,Z) realloc(P,Z) +#endif +#ifndef kfree +#define kfree(P) free(P) +#endif + +static const double __ac_HASH_UPPER = 0.77; + +#define __KHASH_TYPE(name, khkey_t, khval_t) \ + typedef struct kh_##name##_s { \ + khint_t n_buckets, size, n_occupied, upper_bound; \ + khint32_t *flags; \ + khkey_t *keys; \ + khval_t *vals; \ + } kh_##name##_t; + +#define __KHASH_PROTOTYPES(name, khkey_t, khval_t) \ + extern kh_##name##_t * kh_init_##name(void); \ + extern void kh_destroy_##name(kh_##name##_t *h); \ + extern void kh_clear_##name(kh_##name##_t *h); \ + extern khint_t kh_get_##name(const kh_##name##_t *h, khkey_t key); \ + extern int kh_resize_##name(kh_##name##_t *h, khint_t new_n_buckets); \ + extern khint_t kh_put_##name(kh_##name##_t *h, khkey_t key, int *ret); \ + extern void kh_del_##name(kh_##name##_t *h, khint_t x); + +#define __KHASH_IMPL(name, SCOPE, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) \ + SCOPE kh_##name##_t *kh_init_##name(void) { \ + return (kh_##name##_t*)kcalloc(1, sizeof(kh_##name##_t)); \ + } \ + SCOPE void kh_destroy_##name(kh_##name##_t *h) \ + { \ + if (h) { \ + kfree((void *)h->keys); kfree(h->flags); \ + kfree((void *)h->vals); \ + kfree(h); \ + } \ + } \ + SCOPE void kh_unused(kh_clear_##name)(kh_##name##_t *h) \ + { \ + if (h && h->flags) { \ + memset(h->flags, 0xaa, __ac_fsize(h->n_buckets) * sizeof(khint32_t)); \ + h->size = h->n_occupied = 0; \ + } \ + } \ + SCOPE khint_t kh_get_##name(const kh_##name##_t *h, khkey_t key) \ + { \ + if (h->n_buckets) { \ + khint_t k, i, last, mask, step = 0; \ + mask = h->n_buckets - 1; \ + k = __hash_func(key); i = k & mask; \ + last = i; \ + while (!__ac_isempty(h->flags, i) && (__ac_isdel(h->flags, i) || !__hash_equal(h->keys[i], key))) { \ + i = (i + (++step)) & mask; \ + if (i == last) return h->n_buckets; \ + } \ + return __ac_iseither(h->flags, i)? h->n_buckets : i; \ + } else return 0; \ + } \ + SCOPE int kh_resize_##name(kh_##name##_t *h, khint_t new_n_buckets) \ + { /* This function uses 0.25*n_buckets bytes of working space instead of [sizeof(key_t+val_t)+.25]*n_buckets. */ \ + khint32_t *new_flags = 0; \ + khint_t j = 1; \ + { \ + kroundup32(new_n_buckets); \ + if (new_n_buckets < 4) new_n_buckets = 4; \ + if (h->size >= (khint_t)(new_n_buckets * __ac_HASH_UPPER + 0.5)) j = 0; /* requested size is too small */ \ + else { /* hash table size to be changed (shrink or expand); rehash */ \ + new_flags = (khint32_t*)kmalloc(__ac_fsize(new_n_buckets) * sizeof(khint32_t)); \ + if (!new_flags) return -1; \ + memset(new_flags, 0xaa, __ac_fsize(new_n_buckets) * sizeof(khint32_t)); \ + if (h->n_buckets < new_n_buckets) { /* expand */ \ + khkey_t *new_keys = (khkey_t*)krealloc((void *)h->keys, new_n_buckets * sizeof(khkey_t)); \ + if (!new_keys) { kfree(new_flags); return -1; } \ + h->keys = new_keys; \ + if (kh_is_map) { \ + khval_t *new_vals = (khval_t*)krealloc((void *)h->vals, new_n_buckets * sizeof(khval_t)); \ + if (!new_vals) { kfree(new_flags); return -1; } \ + h->vals = new_vals; \ + } \ + } /* otherwise shrink */ \ + } \ + } \ + if (j) { /* rehashing is needed */ \ + for (j = 0; j != h->n_buckets; ++j) { \ + if (__ac_iseither(h->flags, j) == 0) { \ + khkey_t key = h->keys[j]; \ + khval_t val; \ + khint_t new_mask; \ + new_mask = new_n_buckets - 1; \ + if (kh_is_map) val = h->vals[j]; \ + __ac_set_isdel_true(h->flags, j); \ + while (1) { /* kick-out process; sort of like in Cuckoo hashing */ \ + khint_t k, i, step = 0; \ + k = __hash_func(key); \ + i = k & new_mask; \ + while (!__ac_isempty(new_flags, i)) i = (i + (++step)) & new_mask; \ + __ac_set_isempty_false(new_flags, i); \ + if (i < h->n_buckets && __ac_iseither(h->flags, i) == 0) { /* kick out the existing element */ \ + { khkey_t tmp = h->keys[i]; h->keys[i] = key; key = tmp; } \ + if (kh_is_map) { khval_t tmp = h->vals[i]; h->vals[i] = val; val = tmp; } \ + __ac_set_isdel_true(h->flags, i); /* mark it as deleted in the old hash table */ \ + } else { /* write the element and jump out of the loop */ \ + h->keys[i] = key; \ + if (kh_is_map) h->vals[i] = val; \ + break; \ + } \ + } \ + } \ + } \ + if (h->n_buckets > new_n_buckets) { /* shrink the hash table */ \ + h->keys = (khkey_t*)krealloc((void *)h->keys, new_n_buckets * sizeof(khkey_t)); \ + if (kh_is_map) h->vals = (khval_t*)krealloc((void *)h->vals, new_n_buckets * sizeof(khval_t)); \ + } \ + kfree(h->flags); /* free the working space */ \ + h->flags = new_flags; \ + h->n_buckets = new_n_buckets; \ + h->n_occupied = h->size; \ + h->upper_bound = (khint_t)(h->n_buckets * __ac_HASH_UPPER + 0.5); \ + } \ + return 0; \ + } \ + SCOPE khint_t kh_put_##name(kh_##name##_t *h, khkey_t key, int *ret) \ + { \ + khint_t x; \ + if (h->n_occupied >= h->upper_bound) { /* update the hash table */ \ + if (h->n_buckets > (h->size<<1)) { \ + if (kh_resize_##name(h, h->n_buckets - 1) < 0) { /* clear "deleted" elements */ \ + *ret = -1; return h->n_buckets; \ + } \ + } else if (kh_resize_##name(h, h->n_buckets + 1) < 0) { /* expand the hash table */ \ + *ret = -1; return h->n_buckets; \ + } \ + } /* TODO: to implement automatically shrinking; resize() already support shrinking */ \ + { \ + khint_t k, i, site, last, mask = h->n_buckets - 1, step = 0; \ + x = site = h->n_buckets; k = __hash_func(key); i = k & mask; \ + if (__ac_isempty(h->flags, i)) x = i; /* for speed up */ \ + else { \ + last = i; \ + while (!__ac_isempty(h->flags, i) && (__ac_isdel(h->flags, i) || !__hash_equal(h->keys[i], key))) { \ + if (__ac_isdel(h->flags, i)) site = i; \ + i = (i + (++step)) & mask; \ + if (i == last) { x = site; break; } \ + } \ + if (x == h->n_buckets) { \ + if (__ac_isempty(h->flags, i) && site != h->n_buckets) x = site; \ + else x = i; \ + } \ + } \ + } \ + if (__ac_isempty(h->flags, x)) { /* not present at all */ \ + h->keys[x] = key; \ + __ac_set_isboth_false(h->flags, x); \ + ++h->size; ++h->n_occupied; \ + *ret = 1; \ + } else if (__ac_isdel(h->flags, x)) { /* deleted */ \ + h->keys[x] = key; \ + __ac_set_isboth_false(h->flags, x); \ + ++h->size; \ + *ret = 2; \ + } else *ret = 0; /* Don't touch h->keys[x] if present and not deleted */ \ + return x; \ + } \ + SCOPE void kh_del_##name(kh_##name##_t *h, khint_t x) \ + { \ + if (x != h->n_buckets && !__ac_iseither(h->flags, x)) { \ + __ac_set_isdel_true(h->flags, x); \ + --h->size; \ + } \ + } + +#define KHASH_DECLARE(name, khkey_t, khval_t) \ + __KHASH_TYPE(name, khkey_t, khval_t) \ + __KHASH_PROTOTYPES(name, khkey_t, khval_t) + +#define KHASH_INIT2(name, SCOPE, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) \ + __KHASH_TYPE(name, khkey_t, khval_t) \ + __KHASH_IMPL(name, SCOPE, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) + +#define KHASH_INIT(name, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) \ + KHASH_INIT2(name, static kh_inline, khkey_t, khval_t, kh_is_map, __hash_func, __hash_equal) + +/* --- BEGIN OF HASH FUNCTIONS --- */ + +/*! @function + @abstract Integer hash function + @param key The integer [khint32_t] + @return The hash value [khint_t] + */ +#define kh_int_hash_func(key) (khint32_t)(key) +/*! @function + @abstract Integer comparison function + */ +#define kh_int_hash_equal(a, b) ((a) == (b)) +/*! @function + @abstract 64-bit integer hash function + @param key The integer [khint64_t] + @return The hash value [khint_t] + */ +#define kh_int64_hash_func(key) (khint32_t)((key)>>33^(key)^(key)<<11) +/*! @function + @abstract 64-bit integer comparison function + */ +#define kh_int64_hash_equal(a, b) ((a) == (b)) +/*! @function + @abstract const char* hash function + @param s Pointer to a null terminated string + @return The hash value + */ +static kh_inline khint_t __ac_X31_hash_string(const char *s) +{ + khint_t h = (khint_t)*s; + if (h) for (++s ; *s; ++s) h = (h << 5) - h + (khint_t)*s; + return h; +} +/*! @function + @abstract Another interface to const char* hash function + @param key Pointer to a null terminated string [const char*] + @return The hash value [khint_t] + */ +#define kh_str_hash_func(key) __ac_X31_hash_string(key) +/*! @function + @abstract Const char* comparison function + */ +#define kh_str_hash_equal(a, b) (strcmp(a, b) == 0) + +static kh_inline khint_t __ac_Wang_hash(khint_t key) +{ + key += ~(key << 15); + key ^= (key >> 10); + key += (key << 3); + key ^= (key >> 6); + key += ~(key << 11); + key ^= (key >> 16); + return key; +} +#define kh_int_hash_func2(k) __ac_Wang_hash((khint_t)key) + +/* --- END OF HASH FUNCTIONS --- */ + +/* Other convenient macros... */ + +/*! + @abstract Type of the hash table. + @param name Name of the hash table [symbol] + */ +#define khash_t(name) kh_##name##_t + +/*! @function + @abstract Initiate a hash table. + @param name Name of the hash table [symbol] + @return Pointer to the hash table [khash_t(name)*] + */ +#define kh_init(name) kh_init_##name() + +/*! @function + @abstract Destroy a hash table. + @param name Name of the hash table [symbol] + @param h Pointer to the hash table [khash_t(name)*] + */ +#define kh_destroy(name, h) kh_destroy_##name(h) + +/*! @function + @abstract Reset a hash table without deallocating memory. + @param name Name of the hash table [symbol] + @param h Pointer to the hash table [khash_t(name)*] + */ +#define kh_clear(name, h) kh_clear_##name(h) + +/*! @function + @abstract Resize a hash table. + @param name Name of the hash table [symbol] + @param h Pointer to the hash table [khash_t(name)*] + @param s New size [khint_t] + */ +#define kh_resize(name, h, s) kh_resize_##name(h, s) + +/*! @function + @abstract Insert a key to the hash table. + @param name Name of the hash table [symbol] + @param h Pointer to the hash table [khash_t(name)*] + @param k Key [type of keys] + @param r Extra return code: -1 if the operation failed; + 0 if the key is present in the hash table; + 1 if the bucket is empty (never used); 2 if the element in + the bucket has been deleted [int*] + @return Iterator to the inserted element [khint_t] + */ +#define kh_put(name, h, k, r) kh_put_##name(h, k, r) + +/*! @function + @abstract Retrieve a key from the hash table. + @param name Name of the hash table [symbol] + @param h Pointer to the hash table [khash_t(name)*] + @param k Key [type of keys] + @return Iterator to the found element, or kh_end(h) if the element is absent [khint_t] + */ +#define kh_get(name, h, k) kh_get_##name(h, k) + +/*! @function + @abstract Remove a key from the hash table. + @param name Name of the hash table [symbol] + @param h Pointer to the hash table [khash_t(name)*] + @param k Iterator to the element to be deleted [khint_t] + */ +#define kh_del(name, h, k) kh_del_##name(h, k) + +/*! @function + @abstract Test whether a bucket contains data. + @param h Pointer to the hash table [khash_t(name)*] + @param x Iterator to the bucket [khint_t] + @return 1 if containing data; 0 otherwise [int] + */ +#define kh_exist(h, x) (!__ac_iseither((h)->flags, (x))) + +/*! @function + @abstract Get key given an iterator + @param h Pointer to the hash table [khash_t(name)*] + @param x Iterator to the bucket [khint_t] + @return Key [type of keys] + */ +#define kh_key(h, x) ((h)->keys[x]) + +/*! @function + @abstract Get value given an iterator + @param h Pointer to the hash table [khash_t(name)*] + @param x Iterator to the bucket [khint_t] + @return Value [type of values] + @discussion For hash sets, calling this results in segfault. + */ +#define kh_val(h, x) ((h)->vals[x]) + +/*! @function + @abstract Alias of kh_val() + */ +#define kh_value(h, x) ((h)->vals[x]) + +/*! @function + @abstract Get the start iterator + @param h Pointer to the hash table [khash_t(name)*] + @return The start iterator [khint_t] + */ +#define kh_begin(h) (khint_t)(0) + +/*! @function + @abstract Get the end iterator + @param h Pointer to the hash table [khash_t(name)*] + @return The end iterator [khint_t] + */ +#define kh_end(h) ((h)->n_buckets) + +/*! @function + @abstract Get the number of elements in the hash table + @param h Pointer to the hash table [khash_t(name)*] + @return Number of elements in the hash table [khint_t] + */ +#define kh_size(h) ((h)->size) + +/*! @function + @abstract Get the number of buckets in the hash table + @param h Pointer to the hash table [khash_t(name)*] + @return Number of buckets in the hash table [khint_t] + */ +#define kh_n_buckets(h) ((h)->n_buckets) + +/*! @function + @abstract Iterate over the entries in the hash table + @param h Pointer to the hash table [khash_t(name)*] + @param kvar Variable to which key will be assigned + @param vvar Variable to which value will be assigned + @param code Block of code to execute + */ +#define kh_foreach(h, kvar, vvar, code) { khint_t __i; \ + for (__i = kh_begin(h); __i != kh_end(h); ++__i) { \ + if (!kh_exist(h,__i)) continue; \ + (kvar) = kh_key(h,__i); \ + (vvar) = kh_val(h,__i); \ + code; \ + } } + +/*! @function + @abstract Iterate over the values in the hash table + @param h Pointer to the hash table [khash_t(name)*] + @param vvar Variable to which value will be assigned + @param code Block of code to execute + */ +#define kh_foreach_value(h, vvar, code) { khint_t __i; \ + for (__i = kh_begin(h); __i != kh_end(h); ++__i) { \ + if (!kh_exist(h,__i)) continue; \ + (vvar) = kh_val(h,__i); \ + code; \ + } } + +/* More conenient interfaces */ + +/*! @function + @abstract Instantiate a hash set containing integer keys + @param name Name of the hash table [symbol] + */ +#define KHASH_SET_INIT_INT(name) \ + KHASH_INIT(name, khint32_t, char, 0, kh_int_hash_func, kh_int_hash_equal) + +/*! @function + @abstract Instantiate a hash map containing integer keys + @param name Name of the hash table [symbol] + @param khval_t Type of values [type] + */ +#define KHASH_MAP_INIT_INT(name, khval_t) \ + KHASH_INIT(name, khint32_t, khval_t, 1, kh_int_hash_func, kh_int_hash_equal) + +/*! @function + @abstract Instantiate a hash map containing 64-bit integer keys + @param name Name of the hash table [symbol] + */ +#define KHASH_SET_INIT_INT64(name) \ + KHASH_INIT(name, khint64_t, char, 0, kh_int64_hash_func, kh_int64_hash_equal) + +/*! @function + @abstract Instantiate a hash map containing 64-bit integer keys + @param name Name of the hash table [symbol] + @param khval_t Type of values [type] + */ +#define KHASH_MAP_INIT_INT64(name, khval_t) \ + KHASH_INIT(name, khint64_t, khval_t, 1, kh_int64_hash_func, kh_int64_hash_equal) + +typedef const char *kh_cstr_t; +/*! @function + @abstract Instantiate a hash map containing const char* keys + @param name Name of the hash table [symbol] + */ +#define KHASH_SET_INIT_STR(name) \ + KHASH_INIT(name, kh_cstr_t, char, 0, kh_str_hash_func, kh_str_hash_equal) + +/*! @function + @abstract Instantiate a hash map containing const char* keys + @param name Name of the hash table [symbol] + @param khval_t Type of values [type] + */ +#define KHASH_MAP_INIT_STR(name, khval_t) \ + KHASH_INIT(name, kh_cstr_t, khval_t, 1, kh_str_hash_func, kh_str_hash_equal) + +#endif /* __AC_KHASH_H */ diff --git a/contrib/libucl/klib/kvec.h b/contrib/libucl/klib/kvec.h new file mode 100644 index 0000000..b0a7504 --- /dev/null +++ b/contrib/libucl/klib/kvec.h @@ -0,0 +1,103 @@ +/* The MIT License + + Copyright (c) 2008, by Attractive Chaos <attractor@live.co.uk> + + Permission is hereby granted, free of charge, to any person obtaining + a copy of this software and associated documentation files (the + "Software"), to deal in the Software without restriction, including + without limitation the rights to use, copy, modify, merge, publish, + distribute, sublicense, and/or sell copies of the Software, and to + permit persons to whom the Software is furnished to do so, subject to + the following conditions: + + The above copyright notice and this permission notice shall be + included in all copies or substantial portions of the Software. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + SOFTWARE. +*/ + +/* + An example: + +#include "kvec.h" +int main() { + kvec_t(int) array; + kv_init(array); + kv_push(int, array, 10); // append + kv_a(int, array, 20) = 5; // dynamic + kv_A(array, 20) = 4; // static + kv_destroy(array); + return 0; +} +*/ + +/* + 2008-09-22 (0.1.0): + + * The initial version. + +*/ + +#ifndef AC_KVEC_H +#define AC_KVEC_H + +#include <stdlib.h> + +#define kv_roundup32(x) (--(x), (x)|=(x)>>1, (x)|=(x)>>2, (x)|=(x)>>4, (x)|=(x)>>8, (x)|=(x)>>16, ++(x)) + +#define kvec_t(type) struct { size_t n, m; type *a; } +#define kv_init(v) ((v).n = (v).m = 0, (v).a = 0) +#define kv_destroy(v) free((v).a) +#define kv_A(v, i) ((v).a[(i)]) +#define kv_pop(v) ((v).a[--(v).n]) +#define kv_size(v) ((v).n) +#define kv_max(v) ((v).m) + +#define kv_resize(type, v, s) ((v).m = (s), (v).a = (type*)realloc((v).a, sizeof(type) * (v).m)) +#define kv_grow_factor 1.5 +#define kv_grow(type, v) ((v).m = ((v).m > 1 ? (v).m * kv_grow_factor : 2), \ + (v).a = (type*)realloc((v).a, sizeof(type) * (v).m)) + +#define kv_copy(type, v1, v0) do { \ + if ((v1).m < (v0).n) kv_resize(type, v1, (v0).n); \ + (v1).n = (v0).n; \ + memcpy((v1).a, (v0).a, sizeof(type) * (v0).n); \ + } while (0) \ + +#define kv_push(type, v, x) do { \ + if ((v).n == (v).m) { \ + kv_grow(type, v); \ + } \ + (v).a[(v).n++] = (x); \ + } while (0) + +#define kv_prepend(type, v, x) do { \ + if ((v).n == (v).m) { \ + kv_grow(type, v); \ + } \ + memmove((v).a + 1, (v).a, sizeof(type) * (v).n); \ + (v).a[0] = (x); \ + (v).n ++; \ +} while (0) + +#define kv_concat(type, v1, v0) do { \ + if ((v1).m < (v0).n + (v1).n) kv_resize(type, v1, (v0).n + (v1).n); \ + memcpy((v1).a + (v1).n, (v0).a, sizeof(type) * (v0).n); \ + (v1).n = (v0).n + (v1).n; \ + } while (0) + +#define kv_del(type, v, i) do { \ + if ((i) < (v).n) { \ + memmove((v).a + (i), (v).a + ((i) + 1), sizeof(type) * ((v).n - (i) - 1)); \ + (v).n --; \ + } \ +} while (0) + +#endif diff --git a/contrib/libucl/libucl.pc.in b/contrib/libucl/libucl.pc.in new file mode 100644 index 0000000..3433fa9 --- /dev/null +++ b/contrib/libucl/libucl.pc.in @@ -0,0 +1,11 @@ +prefix=@prefix@ +exec_prefix=@exec_prefix@ +libdir=@libdir@ +includedir=@includedir@ + +Name: LibUCL +Description: Universal configuration library +Version: @UCL_VERSION@ +Libs: -L${libdir} -lucl +Libs.private: @LIBS_EXTRA@ @LUA_LIB@ +Cflags: -I${includedir}/ diff --git a/contrib/libucl/lua/Makefile.am b/contrib/libucl/lua/Makefile.am new file mode 100644 index 0000000..95beafb --- /dev/null +++ b/contrib/libucl/lua/Makefile.am @@ -0,0 +1,26 @@ +ucl_common_cflags= -I$(top_srcdir)/src \ + -I$(top_srcdir)/include \ + -I$(top_srcdir)/uthash \ + -Wall -W -Wno-unused-parameter -Wno-pointer-sign +luaexec_LTLIBRARIES= ucl.la +ucl_la_SOURCES= lua_ucl.c +ucl_la_CFLAGS= $(ucl_common_cflags) \ + @LUA_INCLUDE@ +ucl_la_LDFLAGS = -module -export-dynamic -avoid-version +ucl_la_LIBADD= $(top_srcdir)/src/libucl.la \ + @LIBFETCH_LIBS@ \ + @LIBCRYPTO_LIB@ \ + @LIBREGEX_LIB@ \ + @CURL_LIBS@ \ + @LUA_LIB@ + +include_HEADERS= $(top_srcdir)/include/lua_ucl.h + +ROCKSPEC = $(PACKAGE)-$(VERSION)-1.rockspec +EXTRA_DIST = $(PACKAGE).rockspec.in \ + test.lua +DISTCLEANFILES = $(PACKAGE).rockspec + +$(ROCKSPEC): $(PACKAGE).rockspec dist + sed -e 's/@MD5@/'`$(MD5SUM) $(distdir).tar.gz | \ + cut -d " " -f 1`'/g' < $(PACKAGE).rockspec > $@
\ No newline at end of file diff --git a/contrib/libucl/lua/libucl.rockspec.in b/contrib/libucl/lua/libucl.rockspec.in new file mode 100644 index 0000000..52f3917 --- /dev/null +++ b/contrib/libucl/lua/libucl.rockspec.in @@ -0,0 +1,26 @@ +package="@PACKAGE@" +version="@VERSION@-1" +source = { + url = "https://github.com/downloads/vstakhov/@PACKAGE@/@PACKAGE@-@VERSION@.tar.gz", + md5 = "@MD5@", + dir = "@PACKAGE@-@VERSION@" +} +description = { + summary = "UCL - json like configuration language", + detailed = [[ + UCL is heavily infused by nginx configuration as the example + of a convenient configuration system. + However, UCL is fully compatible with JSON format and is able + to parse json files. + ]], + homepage = "http://github.com/vstakhov/@PACKAGE@/", + license = "" +} +dependencies = { + "lua >= 5.1" +} +build = { + type = "command", + build_command = "LUA=$(LUA) CPPFLAGS=-I$(LUA_INCDIR) ./configure --prefix=$(PREFIX) --libdir=$(LIBDIR) --datadir=$(LUADIR) && make clean && make", + install_command = "make install" +} diff --git a/contrib/libucl/lua/lua_ucl.c b/contrib/libucl/lua/lua_ucl.c new file mode 100644 index 0000000..b6162b4 --- /dev/null +++ b/contrib/libucl/lua/lua_ucl.c @@ -0,0 +1,1138 @@ +/* Copyright (c) 2014, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/** + * @file lua ucl bindings + */ + +#include "ucl.h" +#include "ucl_internal.h" +#include "lua_ucl.h" +#include <strings.h> + +/*** + * @module ucl + * This lua module allows to parse objects from strings and to store data into + * ucl objects. It uses `libucl` C library to parse and manipulate with ucl objects. + * @example +local ucl = require("ucl") + +local parser = ucl.parser() +local res,err = parser:parse_string('{key=value}') + +if not res then + print('parser error: ' .. err) +else + local obj = parser:get_object() + local got = ucl.to_format(obj, 'json') +endif + +local table = { + str = 'value', + num = 100500, + null = ucl.null, + func = function () + return 'huh' + end +} + +print(ucl.to_format(table, 'ucl')) +-- Output: +--[[ +num = 100500; +str = "value"; +null = null; +func = "huh"; +--]] + */ + +#define PARSER_META "ucl.parser.meta" +#define EMITTER_META "ucl.emitter.meta" +#define NULL_META "null.emitter.meta" +#define OBJECT_META "ucl.object.meta" + +static int ucl_object_lua_push_array (lua_State *L, const ucl_object_t *obj); +static int ucl_object_lua_push_scalar (lua_State *L, const ucl_object_t *obj, bool allow_array); +static ucl_object_t* ucl_object_lua_fromtable (lua_State *L, int idx); +static ucl_object_t* ucl_object_lua_fromelt (lua_State *L, int idx); + +static void *ucl_null; + +/** + * Push a single element of an object to lua + * @param L + * @param key + * @param obj + */ +static void +ucl_object_lua_push_element (lua_State *L, const char *key, + const ucl_object_t *obj) +{ + lua_pushstring (L, key); + ucl_object_push_lua (L, obj, true); + lua_settable (L, -3); +} + +static void +lua_ucl_userdata_dtor (void *ud) +{ + struct ucl_lua_funcdata *fd = (struct ucl_lua_funcdata *)ud; + + luaL_unref (fd->L, LUA_REGISTRYINDEX, fd->idx); + if (fd->ret != NULL) { + free (fd->ret); + } + free (fd); +} + +static const char * +lua_ucl_userdata_emitter (void *ud) +{ + struct ucl_lua_funcdata *fd = (struct ucl_lua_funcdata *)ud; + const char *out = ""; + + lua_rawgeti (fd->L, LUA_REGISTRYINDEX, fd->idx); + + lua_pcall (fd->L, 0, 1, 0); + out = lua_tostring (fd->L, -1); + + if (out != NULL) { + /* We need to store temporary string in a more appropriate place */ + if (fd->ret) { + free (fd->ret); + } + fd->ret = strdup (out); + } + + lua_settop (fd->L, 0); + + return fd->ret; +} + +/** + * Push a single object to lua + * @param L + * @param obj + * @return + */ +static int +ucl_object_lua_push_object (lua_State *L, const ucl_object_t *obj, + bool allow_array) +{ + const ucl_object_t *cur; + ucl_object_iter_t it = NULL; + int nelt = 0; + + if (allow_array && obj->next != NULL) { + /* Actually we need to push this as an array */ + return ucl_object_lua_push_array (L, obj); + } + + /* Optimize allocation by preallocation of table */ + while (ucl_iterate_object (obj, &it, true) != NULL) { + nelt ++; + } + + lua_createtable (L, 0, nelt); + it = NULL; + + while ((cur = ucl_iterate_object (obj, &it, true)) != NULL) { + ucl_object_lua_push_element (L, ucl_object_key (cur), cur); + } + + return 1; +} + +/** + * Push an array to lua as table indexed by integers + * @param L + * @param obj + * @return + */ +static int +ucl_object_lua_push_array (lua_State *L, const ucl_object_t *obj) +{ + const ucl_object_t *cur; + ucl_object_iter_t it; + int i = 1, nelt = 0; + + if (obj->type == UCL_ARRAY) { + nelt = obj->len; + it = ucl_object_iterate_new (obj); + lua_createtable (L, nelt, 0); + + while ((cur = ucl_object_iterate_safe (it, true))) { + ucl_object_push_lua (L, cur, false); + lua_rawseti (L, -2, i); + i ++; + } + } + else { + /* Optimize allocation by preallocation of table */ + LL_FOREACH (obj, cur) { + nelt ++; + } + + lua_createtable (L, nelt, 0); + + LL_FOREACH (obj, cur) { + ucl_object_push_lua (L, cur, false); + lua_rawseti (L, -2, i); + i ++; + } + } + + return 1; +} + +/** + * Push a simple object to lua depending on its actual type + */ +static int +ucl_object_lua_push_scalar (lua_State *L, const ucl_object_t *obj, + bool allow_array) +{ + struct ucl_lua_funcdata *fd; + + if (allow_array && obj->next != NULL) { + /* Actually we need to push this as an array */ + return ucl_object_lua_push_array (L, obj); + } + + switch (obj->type) { + case UCL_BOOLEAN: + lua_pushboolean (L, ucl_obj_toboolean (obj)); + break; + case UCL_STRING: + lua_pushstring (L, ucl_obj_tostring (obj)); + break; + case UCL_INT: +#if LUA_VERSION_NUM >= 501 + lua_pushinteger (L, ucl_obj_toint (obj)); +#else + lua_pushnumber (L, ucl_obj_toint (obj)); +#endif + break; + case UCL_FLOAT: + case UCL_TIME: + lua_pushnumber (L, ucl_obj_todouble (obj)); + break; + case UCL_NULL: + lua_getfield (L, LUA_REGISTRYINDEX, "ucl.null"); + break; + case UCL_USERDATA: + fd = (struct ucl_lua_funcdata *)obj->value.ud; + lua_rawgeti (L, LUA_REGISTRYINDEX, fd->idx); + break; + default: + lua_pushnil (L); + break; + } + + return 1; +} + +/*** + * @function ucl_object_push_lua(L, obj, allow_array) + * This is a `C` function to push `UCL` object as lua variable. This function + * converts `obj` to lua representation using the following conversions: + * + * - *scalar* values are directly presented by lua objects + * - *userdata* values are converted to lua function objects using `LUA_REGISTRYINDEX`, + * this can be used to pass functions from lua to c and vice-versa + * - *arrays* are converted to lua tables with numeric indicies suitable for `ipairs` iterations + * - *objects* are converted to lua tables with string indicies + * @param {lua_State} L lua state pointer + * @param {ucl_object_t} obj object to push + * @param {bool} allow_array expand implicit arrays (should be true for all but partial arrays) + * @return {int} `1` if an object is pushed to lua + */ +int +ucl_object_push_lua (lua_State *L, const ucl_object_t *obj, bool allow_array) +{ + switch (obj->type) { + case UCL_OBJECT: + return ucl_object_lua_push_object (L, obj, allow_array); + case UCL_ARRAY: + return ucl_object_lua_push_array (L, obj); + default: + return ucl_object_lua_push_scalar (L, obj, allow_array); + } +} + +/** + * Parse lua table into object top + * @param L + * @param top + * @param idx + */ +static ucl_object_t * +ucl_object_lua_fromtable (lua_State *L, int idx) +{ + ucl_object_t *obj, *top = NULL; + size_t keylen; + const char *k; + bool is_array = true; + int max = INT_MIN; + + if (idx < 0) { + /* For negative indicies we want to invert them */ + idx = lua_gettop (L) + idx + 1; + } + /* Check for array */ + lua_pushnil (L); + while (lua_next (L, idx) != 0) { + if (lua_type (L, -2) == LUA_TNUMBER) { + double num = lua_tonumber (L, -2); + if (num == (int)num) { + if (num > max) { + max = num; + } + } + else { + /* Keys are not integer */ + lua_pop (L, 2); + is_array = false; + break; + } + } + else { + /* Keys are not numeric */ + lua_pop (L, 2); + is_array = false; + break; + } + lua_pop (L, 1); + } + + /* Table iterate */ + if (is_array) { + int i; + + top = ucl_object_typed_new (UCL_ARRAY); + for (i = 1; i <= max; i ++) { + lua_pushinteger (L, i); + lua_gettable (L, idx); + obj = ucl_object_lua_fromelt (L, lua_gettop (L)); + if (obj != NULL) { + ucl_array_append (top, obj); + } + lua_pop (L, 1); + } + } + else { + lua_pushnil (L); + top = ucl_object_typed_new (UCL_OBJECT); + while (lua_next (L, idx) != 0) { + /* copy key to avoid modifications */ + k = lua_tolstring (L, -2, &keylen); + obj = ucl_object_lua_fromelt (L, lua_gettop (L)); + + if (obj != NULL) { + ucl_object_insert_key (top, obj, k, keylen, true); + } + lua_pop (L, 1); + } + } + + return top; +} + +/** + * Get a single element from lua to object obj + * @param L + * @param obj + * @param idx + */ +static ucl_object_t * +ucl_object_lua_fromelt (lua_State *L, int idx) +{ + int type; + double num; + ucl_object_t *obj = NULL; + struct ucl_lua_funcdata *fd; + + type = lua_type (L, idx); + + switch (type) { + case LUA_TSTRING: + obj = ucl_object_fromstring_common (lua_tostring (L, idx), 0, 0); + break; + case LUA_TNUMBER: + num = lua_tonumber (L, idx); + if (num == (int64_t)num) { + obj = ucl_object_fromint (num); + } + else { + obj = ucl_object_fromdouble (num); + } + break; + case LUA_TBOOLEAN: + obj = ucl_object_frombool (lua_toboolean (L, idx)); + break; + case LUA_TUSERDATA: + if (lua_topointer (L, idx) == ucl_null) { + obj = ucl_object_typed_new (UCL_NULL); + } + break; + case LUA_TTABLE: + case LUA_TFUNCTION: + case LUA_TTHREAD: + if (luaL_getmetafield (L, idx, "__gen_ucl")) { + if (lua_isfunction (L, -1)) { + lua_settop (L, 3); /* gen, obj, func */ + lua_insert (L, 1); /* func, gen, obj */ + lua_insert (L, 2); /* func, obj, gen */ + lua_call(L, 2, 1); + obj = ucl_object_lua_fromelt (L, 1); + } + lua_pop (L, 2); + } + else { + if (type == LUA_TTABLE) { + obj = ucl_object_lua_fromtable (L, idx); + } + else if (type == LUA_TFUNCTION) { + fd = malloc (sizeof (*fd)); + if (fd != NULL) { + lua_pushvalue (L, idx); + fd->L = L; + fd->ret = NULL; + fd->idx = luaL_ref (L, LUA_REGISTRYINDEX); + + obj = ucl_object_new_userdata (lua_ucl_userdata_dtor, + lua_ucl_userdata_emitter); + obj->type = UCL_USERDATA; + obj->value.ud = (void *)fd; + } + } + } + break; + } + + return obj; +} + +/** + * @function ucl_object_lua_import(L, idx) + * Extracts ucl object from lua variable at `idx` position, + * @see ucl_object_push_lua for conversion definitions + * @param {lua_state} L lua state machine pointer + * @param {int} idx index where the source variable is placed + * @return {ucl_object_t} new ucl object extracted from lua variable. Reference count of this object is 1, + * this object thus needs to be unref'ed after usage. + */ +ucl_object_t * +ucl_object_lua_import (lua_State *L, int idx) +{ + ucl_object_t *obj; + int t; + + t = lua_type (L, idx); + switch (t) { + case LUA_TTABLE: + obj = ucl_object_lua_fromtable (L, idx); + break; + default: + obj = ucl_object_lua_fromelt (L, idx); + break; + } + + return obj; +} + +static int +lua_ucl_to_string (lua_State *L, const ucl_object_t *obj, enum ucl_emitter type) +{ + unsigned char *result; + + result = ucl_object_emit (obj, type); + + if (result != NULL) { + lua_pushstring (L, (const char *)result); + free (result); + } + else { + lua_pushnil (L); + } + + return 1; +} + +static int +lua_ucl_parser_init (lua_State *L) +{ + struct ucl_parser *parser, **pparser; + int flags = 0; + + if (lua_gettop (L) >= 1) { + flags = lua_tonumber (L, 1); + } + + parser = ucl_parser_new (flags); + if (parser == NULL) { + lua_pushnil (L); + } + + pparser = lua_newuserdata (L, sizeof (parser)); + *pparser = parser; + luaL_getmetatable (L, PARSER_META); + lua_setmetatable (L, -2); + + return 1; +} + +static struct ucl_parser * +lua_ucl_parser_get (lua_State *L, int index) +{ + return *((struct ucl_parser **) luaL_checkudata(L, index, PARSER_META)); +} + +static ucl_object_t * +lua_ucl_object_get (lua_State *L, int index) +{ + return *((ucl_object_t **) luaL_checkudata(L, index, OBJECT_META)); +} + +/*** + * @method parser:parse_file(name) + * Parse UCL object from file. + * @param {string} name filename to parse + * @return {bool[, string]} if res is `true` then file has been parsed successfully, otherwise an error string is also returned +@example +local parser = ucl.parser() +local res,err = parser:parse_file('/some/file.conf') + +if not res then + print('parser error: ' .. err) +else + -- Do something with object +end + */ +static int +lua_ucl_parser_parse_file (lua_State *L) +{ + struct ucl_parser *parser; + const char *file; + int ret = 2; + + parser = lua_ucl_parser_get (L, 1); + file = luaL_checkstring (L, 2); + + if (parser != NULL && file != NULL) { + if (ucl_parser_add_file (parser, file)) { + lua_pushboolean (L, true); + ret = 1; + } + else { + lua_pushboolean (L, false); + lua_pushstring (L, ucl_parser_get_error (parser)); + } + } + else { + lua_pushboolean (L, false); + lua_pushstring (L, "invalid arguments"); + } + + return ret; +} + +/*** + * @method parser:parse_string(input) + * Parse UCL object from file. + * @param {string} input string to parse + * @return {bool[, string]} if res is `true` then file has been parsed successfully, otherwise an error string is also returned + */ +static int +lua_ucl_parser_parse_string (lua_State *L) +{ + struct ucl_parser *parser; + const char *string; + size_t llen; + int ret = 2; + + parser = lua_ucl_parser_get (L, 1); + string = luaL_checklstring (L, 2, &llen); + + if (parser != NULL && string != NULL) { + if (ucl_parser_add_chunk (parser, (const unsigned char *)string, llen)) { + lua_pushboolean (L, true); + ret = 1; + } + else { + lua_pushboolean (L, false); + lua_pushstring (L, ucl_parser_get_error (parser)); + } + } + else { + lua_pushboolean (L, false); + lua_pushstring (L, "invalid arguments"); + } + + return ret; +} + +/*** + * @method parser:get_object() + * Get top object from parser and export it to lua representation. + * @return {variant or nil} ucl object as lua native variable + */ +static int +lua_ucl_parser_get_object (lua_State *L) +{ + struct ucl_parser *parser; + ucl_object_t *obj; + int ret = 1; + + parser = lua_ucl_parser_get (L, 1); + obj = ucl_parser_get_object (parser); + + if (obj != NULL) { + ret = ucl_object_push_lua (L, obj, false); + /* no need to keep reference */ + ucl_object_unref (obj); + } + else { + lua_pushnil (L); + } + + return ret; +} + +/*** + * @method parser:get_object_wrapped() + * Get top object from parser and export it to userdata object without + * unwrapping to lua. + * @return {ucl.object or nil} ucl object wrapped variable + */ +static int +lua_ucl_parser_get_object_wrapped (lua_State *L) +{ + struct ucl_parser *parser; + ucl_object_t *obj, **pobj; + int ret = 1; + + parser = lua_ucl_parser_get (L, 1); + obj = ucl_parser_get_object (parser); + + if (obj != NULL) { + pobj = lua_newuserdata (L, sizeof (*pobj)); + *pobj = obj; + luaL_getmetatable (L, OBJECT_META); + lua_setmetatable (L, -2); + } + else { + lua_pushnil (L); + } + + return ret; +} + +/*** + * @method parser:validate(schema) + * Validates the top object in the parser against schema. Schema might be + * another object or a string that represents file to load schema from. + * + * @param {string/table} schema input schema + * @return {result,err} two values: boolean result and the corresponding error + * + */ +static int +lua_ucl_parser_validate (lua_State *L) +{ + struct ucl_parser *parser, *schema_parser; + ucl_object_t *schema; + const char *schema_file; + struct ucl_schema_error err; + + parser = lua_ucl_parser_get (L, 1); + + if (parser && parser->top_obj) { + if (lua_type (L, 2) == LUA_TTABLE) { + schema = ucl_object_lua_import (L, 2); + + if (schema == NULL) { + lua_pushboolean (L, false); + lua_pushstring (L, "cannot load schema from lua table"); + + return 2; + } + } + else if (lua_type (L, 2) == LUA_TSTRING) { + schema_parser = ucl_parser_new (0); + schema_file = luaL_checkstring (L, 2); + + if (!ucl_parser_add_file (schema_parser, schema_file)) { + lua_pushboolean (L, false); + lua_pushfstring (L, "cannot parse schema file \"%s\": " + "%s", schema_file, ucl_parser_get_error (parser)); + ucl_parser_free (schema_parser); + + return 2; + } + + schema = ucl_parser_get_object (schema_parser); + ucl_parser_free (schema_parser); + } + else { + lua_pushboolean (L, false); + lua_pushstring (L, "invalid schema argument"); + + return 2; + } + + if (!ucl_object_validate (schema, parser->top_obj, &err)) { + lua_pushboolean (L, false); + lua_pushfstring (L, "validation error: " + "%s", err.msg); + } + else { + lua_pushboolean (L, true); + lua_pushnil (L); + } + + ucl_object_unref (schema); + } + else { + lua_pushboolean (L, false); + lua_pushstring (L, "invalid parser or empty top object"); + } + + return 2; +} + +static int +lua_ucl_parser_gc (lua_State *L) +{ + struct ucl_parser *parser; + + parser = lua_ucl_parser_get (L, 1); + ucl_parser_free (parser); + + return 0; +} + +/*** + * @method object:unwrap() + * Unwraps opaque ucl object to the native lua object (performing copying) + * @return {variant} any lua object + */ +static int +lua_ucl_object_unwrap (lua_State *L) +{ + ucl_object_t *obj; + + obj = lua_ucl_object_get (L, 1); + + if (obj) { + ucl_object_push_lua (L, obj, true); + } + else { + lua_pushnil (L); + } + + return 1; +} + +/*** + * @method object:tostring(type) + * Unwraps opaque ucl object to string (json by default). Optionally you can + * specify output format: + * + * - `json` - fine printed json + * - `json-compact` - compacted json + * - `config` - fine printed configuration + * - `ucl` - same as `config` + * - `yaml` - embedded yaml + * @param {string} type optional + * @return {string} string representation of the opaque ucl object + */ +static int +lua_ucl_object_tostring (lua_State *L) +{ + ucl_object_t *obj; + enum ucl_emitter format = UCL_EMIT_JSON_COMPACT; + + obj = lua_ucl_object_get (L, 1); + + if (obj) { + if (lua_gettop (L) > 1) { + if (lua_type (L, 2) == LUA_TSTRING) { + const char *strtype = lua_tostring (L, 2); + + if (strcasecmp (strtype, "json") == 0) { + format = UCL_EMIT_JSON; + } + else if (strcasecmp (strtype, "json-compact") == 0) { + format = UCL_EMIT_JSON_COMPACT; + } + else if (strcasecmp (strtype, "yaml") == 0) { + format = UCL_EMIT_YAML; + } + else if (strcasecmp (strtype, "config") == 0 || + strcasecmp (strtype, "ucl") == 0) { + format = UCL_EMIT_CONFIG; + } + } + } + + return lua_ucl_to_string (L, obj, format); + } + else { + lua_pushnil (L); + } + + return 1; +} + +/*** + * @method object:validate(schema, path) + * Validates the given ucl object using schema object represented as another + * opaque ucl object. You can also specify path in the form `#/path/def` to + * specify the specific schema element to perform validation. + * + * @param {ucl.object} schema schema object + * @param {string} path optional path for validation procedure + * @return {result,err} two values: boolean result and the corresponding error + */ +static int +lua_ucl_object_validate (lua_State *L) +{ + ucl_object_t *obj, *schema; + const ucl_object_t *schema_elt; + bool res = false; + struct ucl_schema_error err; + const char *path = NULL; + + obj = lua_ucl_object_get (L, 1); + schema = lua_ucl_object_get (L, 2); + + if (schema && obj && ucl_object_type (schema) == UCL_OBJECT) { + if (lua_gettop (L) > 2 && lua_type (L, 3) == LUA_TSTRING) { + path = lua_tostring (L, 3); + if (path[0] == '#') { + path ++; + } + } + + if (path) { + schema_elt = ucl_lookup_path_char (schema, path, '/'); + } + else { + /* Use the top object */ + schema_elt = schema; + } + + if (schema_elt) { + res = ucl_object_validate (schema_elt, obj, &err); + + if (res) { + lua_pushboolean (L, res); + lua_pushnil (L); + } + else { + lua_pushboolean (L, res); + lua_pushfstring (L, "validation error: %s", err.msg); + } + } + else { + lua_pushboolean (L, res); + + if (path) { + lua_pushfstring (L, "cannot find the requested path: %s", path); + } + else { + /* Should not be reached */ + lua_pushstring (L, "unknown error"); + } + } + } + else { + lua_pushboolean (L, res); + lua_pushstring (L, "invalid object or schema"); + } + + return 2; +} + +static int +lua_ucl_object_gc (lua_State *L) +{ + ucl_object_t *obj; + + obj = lua_ucl_object_get (L, 1); + + ucl_object_unref (obj); + + return 0; +} + +static void +lua_ucl_parser_mt (lua_State *L) +{ + luaL_newmetatable (L, PARSER_META); + + lua_pushvalue(L, -1); + lua_setfield(L, -2, "__index"); + + lua_pushcfunction (L, lua_ucl_parser_gc); + lua_setfield (L, -2, "__gc"); + + lua_pushcfunction (L, lua_ucl_parser_parse_file); + lua_setfield (L, -2, "parse_file"); + + lua_pushcfunction (L, lua_ucl_parser_parse_string); + lua_setfield (L, -2, "parse_string"); + + lua_pushcfunction (L, lua_ucl_parser_get_object); + lua_setfield (L, -2, "get_object"); + + lua_pushcfunction (L, lua_ucl_parser_get_object_wrapped); + lua_setfield (L, -2, "get_object_wrapped"); + + lua_pushcfunction (L, lua_ucl_parser_validate); + lua_setfield (L, -2, "validate"); + + lua_pop (L, 1); +} + +static void +lua_ucl_object_mt (lua_State *L) +{ + luaL_newmetatable (L, OBJECT_META); + + lua_pushvalue(L, -1); + lua_setfield(L, -2, "__index"); + + lua_pushcfunction (L, lua_ucl_object_gc); + lua_setfield (L, -2, "__gc"); + + lua_pushcfunction (L, lua_ucl_object_tostring); + lua_setfield (L, -2, "__tostring"); + + lua_pushcfunction (L, lua_ucl_object_tostring); + lua_setfield (L, -2, "tostring"); + + lua_pushcfunction (L, lua_ucl_object_unwrap); + lua_setfield (L, -2, "unwrap"); + + lua_pushcfunction (L, lua_ucl_object_unwrap); + lua_setfield (L, -2, "tolua"); + + lua_pushcfunction (L, lua_ucl_object_validate); + lua_setfield (L, -2, "validate"); + + lua_pushstring (L, OBJECT_META); + lua_setfield (L, -2, "class"); + + lua_pop (L, 1); +} + +static int +lua_ucl_to_json (lua_State *L) +{ + ucl_object_t *obj; + int format = UCL_EMIT_JSON; + + if (lua_gettop (L) > 1) { + if (lua_toboolean (L, 2)) { + format = UCL_EMIT_JSON_COMPACT; + } + } + + obj = ucl_object_lua_import (L, 1); + if (obj != NULL) { + lua_ucl_to_string (L, obj, format); + ucl_object_unref (obj); + } + else { + lua_pushnil (L); + } + + return 1; +} + +static int +lua_ucl_to_config (lua_State *L) +{ + ucl_object_t *obj; + + obj = ucl_object_lua_import (L, 1); + if (obj != NULL) { + lua_ucl_to_string (L, obj, UCL_EMIT_CONFIG); + ucl_object_unref (obj); + } + else { + lua_pushnil (L); + } + + return 1; +} + +/*** + * @function ucl.to_format(var, format) + * Converts lua variable `var` to the specified `format`. Formats supported are: + * + * - `json` - fine printed json + * - `json-compact` - compacted json + * - `config` - fine printed configuration + * - `ucl` - same as `config` + * - `yaml` - embedded yaml + * + * If `var` contains function, they are called during output formatting and if + * they return string value, then this value is used for ouptut. + * @param {variant} var any sort of lua variable (if userdata then metafield `__to_ucl` is searched for output) + * @param {string} format any available format + * @return {string} string representation of `var` in the specific `format`. + * @example +local table = { + str = 'value', + num = 100500, + null = ucl.null, + func = function () + return 'huh' + end +} + +print(ucl.to_format(table, 'ucl')) +-- Output: +--[[ +num = 100500; +str = "value"; +null = null; +func = "huh"; +--]] + */ +static int +lua_ucl_to_format (lua_State *L) +{ + ucl_object_t *obj; + int format = UCL_EMIT_JSON; + + if (lua_gettop (L) > 1) { + if (lua_type (L, 2) == LUA_TNUMBER) { + format = lua_tonumber (L, 2); + if (format < 0 || format >= UCL_EMIT_YAML) { + lua_pushnil (L); + return 1; + } + } + else if (lua_type (L, 2) == LUA_TSTRING) { + const char *strtype = lua_tostring (L, 2); + + if (strcasecmp (strtype, "json") == 0) { + format = UCL_EMIT_JSON; + } + else if (strcasecmp (strtype, "json-compact") == 0) { + format = UCL_EMIT_JSON_COMPACT; + } + else if (strcasecmp (strtype, "yaml") == 0) { + format = UCL_EMIT_YAML; + } + else if (strcasecmp (strtype, "config") == 0 || + strcasecmp (strtype, "ucl") == 0) { + format = UCL_EMIT_CONFIG; + } + } + } + + obj = ucl_object_lua_import (L, 1); + if (obj != NULL) { + lua_ucl_to_string (L, obj, format); + ucl_object_unref (obj); + } + else { + lua_pushnil (L); + } + + return 1; +} + +static int +lua_ucl_null_tostring (lua_State* L) +{ + lua_pushstring (L, "null"); + return 1; +} + +static void +lua_ucl_null_mt (lua_State *L) +{ + luaL_newmetatable (L, NULL_META); + + lua_pushcfunction (L, lua_ucl_null_tostring); + lua_setfield (L, -2, "__tostring"); + + lua_pop (L, 1); +} + +int +luaopen_ucl (lua_State *L) +{ + lua_ucl_parser_mt (L); + lua_ucl_null_mt (L); + lua_ucl_object_mt (L); + + /* Create the refs weak table: */ + lua_createtable (L, 0, 2); + lua_pushliteral (L, "v"); /* tbl, "v" */ + lua_setfield (L, -2, "__mode"); + lua_pushvalue (L, -1); /* tbl, tbl */ + lua_setmetatable (L, -2); /* tbl */ + lua_setfield (L, LUA_REGISTRYINDEX, "ucl.refs"); + + lua_newtable (L); + + lua_pushcfunction (L, lua_ucl_parser_init); + lua_setfield (L, -2, "parser"); + + lua_pushcfunction (L, lua_ucl_to_json); + lua_setfield (L, -2, "to_json"); + + lua_pushcfunction (L, lua_ucl_to_config); + lua_setfield (L, -2, "to_config"); + + lua_pushcfunction (L, lua_ucl_to_format); + lua_setfield (L, -2, "to_format"); + + ucl_null = lua_newuserdata (L, 0); + luaL_getmetatable (L, NULL_META); + lua_setmetatable (L, -2); + + lua_pushvalue (L, -1); + lua_setfield (L, LUA_REGISTRYINDEX, "ucl.null"); + + lua_setfield (L, -2, "null"); + + return 1; +} + +struct ucl_lua_funcdata* +ucl_object_toclosure (const ucl_object_t *obj) +{ + if (obj == NULL || obj->type != UCL_USERDATA) { + return NULL; + } + + return (struct ucl_lua_funcdata*)obj->value.ud; +} diff --git a/contrib/libucl/lua/test.lua b/contrib/libucl/lua/test.lua new file mode 100644 index 0000000..1e7b3b3 --- /dev/null +++ b/contrib/libucl/lua/test.lua @@ -0,0 +1,48 @@ +local ucl = require("ucl") + +function test_simple() + local expect = + '['.. + '"float",1.5,'.. + '"integer",5,'.. + '"true",true,'.. + '"false",false,'.. + '"null",null,'.. + '"string","hello",'.. + '"array",[1,2],'.. + '"object",{"key":"value"}'.. + ']' + + -- Input to to_value matches the output of to_string: + local parser = ucl.parser() + local res,err = parser:parse_string(expect) + if not res then + print('parser error: ' .. err) + return 1 + end + + local obj = parser:get_object() + local got = ucl.to_json(obj, true) + if expect == got then + return 0 + else + print(expect .. " == " .. tostring(got)) + return 1 + end +end + +test_simple() + +local table = { + str = 'value', + num = 100500, + null = ucl.null, + func = function () + return 'huh' + end, + badfunc = function() + print("I'm bad") + end +} + +print(ucl.to_format(table, 'ucl')) diff --git a/contrib/libucl/m4/ax_lua.m4 b/contrib/libucl/m4/ax_lua.m4 new file mode 100644 index 0000000..f8e2fd4 --- /dev/null +++ b/contrib/libucl/m4/ax_lua.m4 @@ -0,0 +1,664 @@ +# =========================================================================== +# http://www.gnu.org/software/autoconf-archive/ax_lua.html +# =========================================================================== +# +# SYNOPSIS +# +# AX_PROG_LUA[([MINIMUM-VERSION], [TOO-BIG-VERSION], [ACTION-IF-FOUND], [ACTION-IF-NOT-FOUND])] +# AX_LUA_HEADERS[([ACTION-IF-FOUND], [ACTION-IF-NOT-FOUND])] +# AX_LUA_LIBS[([ACTION-IF-FOUND], [ACTION-IF-NOT-FOUND])] +# AX_LUA_READLINE[([ACTION-IF-FOUND], [ACTION-IF-NOT-FOUND])] +# +# DESCRIPTION +# +# Detect a Lua interpreter, optionally specifying a minimum and maximum +# version number. Set up important Lua paths, such as the directories in +# which to install scripts and modules (shared libraries). +# +# Also detect Lua headers and libraries. The Lua version contained in the +# header is checked to match the Lua interpreter version exactly. When +# searching for Lua libraries, the version number is used as a suffix. +# This is done with the goal of supporting multiple Lua installs (5.1, +# 5.2, and 5.3 side-by-side). +# +# A note on compatibility with previous versions: This file has been +# mostly rewritten for serial 18. Most developers should be able to use +# these macros without needing to modify configure.ac. Care has been taken +# to preserve each macro's behavior, but there are some differences: +# +# 1) AX_WITH_LUA is deprecated; it now expands to the exact same thing as +# AX_PROG_LUA with no arguments. +# +# 2) AX_LUA_HEADERS now checks that the version number defined in lua.h +# matches the interpreter version. AX_LUA_HEADERS_VERSION is therefore +# unnecessary, so it is deprecated and does not expand to anything. +# +# 3) The configure flag --with-lua-suffix no longer exists; the user +# should instead specify the LUA precious variable on the command line. +# See the AX_PROG_LUA description for details. +# +# Please read the macro descriptions below for more information. +# +# This file was inspired by Andrew Dalke's and James Henstridge's +# python.m4 and Tom Payne's, Matthieu Moy's, and Reuben Thomas's ax_lua.m4 +# (serial 17). Basically, this file is a mash-up of those two files. I +# like to think it combines the best of the two! +# +# AX_PROG_LUA: Search for the Lua interpreter, and set up important Lua +# paths. Adds precious variable LUA, which may contain the path of the Lua +# interpreter. If LUA is blank, the user's path is searched for an +# suitable interpreter. +# +# If MINIMUM-VERSION is supplied, then only Lua interpreters with a +# version number greater or equal to MINIMUM-VERSION will be accepted. If +# TOO-BIG-VERSION is also supplied, then only Lua interpreters with a +# version number greater or equal to MINIMUM-VERSION and less than +# TOO-BIG-VERSION will be accepted. +# +# The Lua version number, LUA_VERSION, is found from the interpreter, and +# substituted. LUA_PLATFORM is also found, but not currently supported (no +# standard representation). +# +# Finally, the macro finds four paths: +# +# luadir Directory to install Lua scripts. +# pkgluadir $luadir/$PACKAGE +# luaexecdir Directory to install Lua modules. +# pkgluaexecdir $luaexecdir/$PACKAGE +# +# These paths are found based on $prefix, $exec_prefix, Lua's +# package.path, and package.cpath. The first path of package.path +# beginning with $prefix is selected as luadir. The first path of +# package.cpath beginning with $exec_prefix is used as luaexecdir. This +# should work on all reasonable Lua installations. If a path cannot be +# determined, a default path is used. Of course, the user can override +# these later when invoking make. +# +# luadir Default: $prefix/share/lua/$LUA_VERSION +# luaexecdir Default: $exec_prefix/lib/lua/$LUA_VERSION +# +# These directories can be used by Automake as install destinations. The +# variable name minus 'dir' needs to be used as a prefix to the +# appropriate Automake primary, e.g. lua_SCRIPS or luaexec_LIBRARIES. +# +# If an acceptable Lua interpreter is found, then ACTION-IF-FOUND is +# performed, otherwise ACTION-IF-NOT-FOUND is preformed. If ACTION-IF-NOT- +# FOUND is blank, then it will default to printing an error. To prevent +# the default behavior, give ':' as an action. +# +# AX_LUA_HEADERS: Search for Lua headers. Requires that AX_PROG_LUA be +# expanded before this macro. Adds precious variable LUA_INCLUDE, which +# may contain Lua specific include flags, e.g. -I/usr/include/lua5.1. If +# LUA_INCLUDE is blank, then this macro will attempt to find suitable +# flags. +# +# LUA_INCLUDE can be used by Automake to compile Lua modules or +# executables with embedded interpreters. The *_CPPFLAGS variables should +# be used for this purpose, e.g. myprog_CPPFLAGS = $(LUA_INCLUDE). +# +# This macro searches for the header lua.h (and others). The search is +# performed with a combination of CPPFLAGS, CPATH, etc, and LUA_INCLUDE. +# If the search is unsuccessful, then some common directories are tried. +# If the headers are then found, then LUA_INCLUDE is set accordingly. +# +# The paths automatically searched are: +# +# * /usr/include/luaX.Y +# * /usr/include/lua/X.Y +# * /usr/include/luaXY +# * /usr/local/include/luaX.Y +# * /usr/local/include/lua-X.Y +# * /usr/local/include/lua/X.Y +# * /usr/local/include/luaXY +# +# (Where X.Y is the Lua version number, e.g. 5.1.) +# +# The Lua version number found in the headers is always checked to match +# the Lua interpreter's version number. Lua headers with mismatched +# version numbers are not accepted. +# +# If headers are found, then ACTION-IF-FOUND is performed, otherwise +# ACTION-IF-NOT-FOUND is performed. If ACTION-IF-NOT-FOUND is blank, then +# it will default to printing an error. To prevent the default behavior, +# set the action to ':'. +# +# AX_LUA_LIBS: Search for Lua libraries. Requires that AX_PROG_LUA be +# expanded before this macro. Adds precious variable LUA_LIB, which may +# contain Lua specific linker flags, e.g. -llua5.1. If LUA_LIB is blank, +# then this macro will attempt to find suitable flags. +# +# LUA_LIB can be used by Automake to link Lua modules or executables with +# embedded interpreters. The *_LIBADD and *_LDADD variables should be used +# for this purpose, e.g. mymod_LIBADD = $(LUA_LIB). +# +# This macro searches for the Lua library. More technically, it searches +# for a library containing the function lua_load. The search is performed +# with a combination of LIBS, LIBRARY_PATH, and LUA_LIB. +# +# If the search determines that some linker flags are missing, then those +# flags will be added to LUA_LIB. +# +# If libraries are found, then ACTION-IF-FOUND is performed, otherwise +# ACTION-IF-NOT-FOUND is performed. If ACTION-IF-NOT-FOUND is blank, then +# it will default to printing an error. To prevent the default behavior, +# set the action to ':'. +# +# AX_LUA_READLINE: Search for readline headers and libraries. Requires the +# AX_LIB_READLINE macro, which is provided by ax_lib_readline.m4 from the +# Autoconf Archive. +# +# If a readline compatible library is found, then ACTION-IF-FOUND is +# performed, otherwise ACTION-IF-NOT-FOUND is performed. +# +# LICENSE +# +# Copyright (c) 2015 Reuben Thomas <rrt@sc3d.org> +# Copyright (c) 2014 Tim Perkins <tprk77@gmail.com> +# +# This program is free software: you can redistribute it and/or modify it +# under the terms of the GNU General Public License as published by the +# Free Software Foundation, either version 3 of the License, or (at your +# option) any later version. +# +# This program is distributed in the hope that it will be useful, but +# WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General +# Public License for more details. +# +# You should have received a copy of the GNU General Public License along +# with this program. If not, see <http://www.gnu.org/licenses/>. +# +# As a special exception, the respective Autoconf Macro's copyright owner +# gives unlimited permission to copy, distribute and modify the configure +# scripts that are the output of Autoconf when processing the Macro. You +# need not follow the terms of the GNU General Public License when using +# or distributing such scripts, even though portions of the text of the +# Macro appear in them. The GNU General Public License (GPL) does govern +# all other use of the material that constitutes the Autoconf Macro. +# +# This special exception to the GPL applies to versions of the Autoconf +# Macro released by the Autoconf Archive. When you make and distribute a +# modified version of the Autoconf Macro, you may extend this special +# exception to the GPL to apply to your modified version as well. + +#serial 39 + +dnl ========================================================================= +dnl AX_PROG_LUA([MINIMUM-VERSION], [TOO-BIG-VERSION], +dnl [ACTION-IF-FOUND], [ACTION-IF-NOT-FOUND]) +dnl ========================================================================= +AC_DEFUN([AX_PROG_LUA], +[ + dnl Check for required tools. + AC_REQUIRE([AC_PROG_GREP]) + AC_REQUIRE([AC_PROG_SED]) + + dnl Make LUA a precious variable. + AC_ARG_VAR([LUA], [The Lua interpreter, e.g. /usr/bin/lua5.1]) + + dnl Find a Lua interpreter. + m4_define_default([_AX_LUA_INTERPRETER_LIST], + [lua lua5.3 lua53 lua5.2 lua52 lua5.1 lua51 lua50]) + + m4_if([$1], [], + [ dnl No version check is needed. Find any Lua interpreter. + AS_IF([test "x$LUA" = 'x'], + [AC_PATH_PROGS([LUA], [_AX_LUA_INTERPRETER_LIST], [:])]) + ax_display_LUA='lua' + + AS_IF([test "x$LUA" != 'x:'], + [ dnl At least check if this is a Lua interpreter. + AC_MSG_CHECKING([if $LUA is a Lua interpreter]) + _AX_LUA_CHK_IS_INTRP([$LUA], + [AC_MSG_RESULT([yes])], + [ AC_MSG_RESULT([no]) + AC_MSG_ERROR([not a Lua interpreter]) + ]) + ]) + ], + [ dnl A version check is needed. + AS_IF([test "x$LUA" != 'x'], + [ dnl Check if this is a Lua interpreter. + AC_MSG_CHECKING([if $LUA is a Lua interpreter]) + _AX_LUA_CHK_IS_INTRP([$LUA], + [AC_MSG_RESULT([yes])], + [ AC_MSG_RESULT([no]) + AC_MSG_ERROR([not a Lua interpreter]) + ]) + dnl Check the version. + m4_if([$2], [], + [_ax_check_text="whether $LUA version >= $1"], + [_ax_check_text="whether $LUA version >= $1, < $2"]) + AC_MSG_CHECKING([$_ax_check_text]) + _AX_LUA_CHK_VER([$LUA], [$1], [$2], + [AC_MSG_RESULT([yes])], + [ AC_MSG_RESULT([no]) + AC_MSG_ERROR([version is out of range for specified LUA])]) + ax_display_LUA=$LUA + ], + [ dnl Try each interpreter until we find one that satisfies VERSION. + m4_if([$2], [], + [_ax_check_text="for a Lua interpreter with version >= $1"], + [_ax_check_text="for a Lua interpreter with version >= $1, < $2"]) + AC_CACHE_CHECK([$_ax_check_text], + [ax_cv_pathless_LUA], + [ for ax_cv_pathless_LUA in _AX_LUA_INTERPRETER_LIST none; do + test "x$ax_cv_pathless_LUA" = 'xnone' && break + _AX_LUA_CHK_IS_INTRP([$ax_cv_pathless_LUA], [], [continue]) + _AX_LUA_CHK_VER([$ax_cv_pathless_LUA], [$1], [$2], [break]) + done + ]) + dnl Set $LUA to the absolute path of $ax_cv_pathless_LUA. + AS_IF([test "x$ax_cv_pathless_LUA" = 'xnone'], + [LUA=':'], + [AC_PATH_PROG([LUA], [$ax_cv_pathless_LUA])]) + ax_display_LUA=$ax_cv_pathless_LUA + ]) + ]) + + AS_IF([test "x$LUA" = 'x:'], + [ dnl Run any user-specified action, or abort. + m4_default([$4], [AC_MSG_ERROR([cannot find suitable Lua interpreter])]) + ], + [ dnl Query Lua for its version number. + AC_CACHE_CHECK([for $ax_display_LUA version], + [ax_cv_lua_version], + [ dnl Get the interpreter version in X.Y format. This should work for + dnl interpreters version 5.0 and beyond. + ax_cv_lua_version=[`$LUA -e ' + -- return a version number in X.Y format + local _, _, ver = string.find(_VERSION, "^Lua (%d+%.%d+)") + print(ver)'`] + ]) + AS_IF([test "x$ax_cv_lua_version" = 'x'], + [AC_MSG_ERROR([invalid Lua version number])]) + AC_SUBST([LUA_VERSION], [$ax_cv_lua_version]) + AC_SUBST([LUA_SHORT_VERSION], [`echo "$LUA_VERSION" | $SED 's|\.||'`]) + + dnl The following check is not supported: + dnl At times (like when building shared libraries) you may want to know + dnl which OS platform Lua thinks this is. + AC_CACHE_CHECK([for $ax_display_LUA platform], + [ax_cv_lua_platform], + [ax_cv_lua_platform=[`$LUA -e 'print("unknown")'`]]) + AC_SUBST([LUA_PLATFORM], [$ax_cv_lua_platform]) + + dnl Use the values of $prefix and $exec_prefix for the corresponding + dnl values of LUA_PREFIX and LUA_EXEC_PREFIX. These are made distinct + dnl variables so they can be overridden if need be. However, the general + dnl consensus is that you shouldn't need this ability. + AC_SUBST([LUA_PREFIX], ['${prefix}']) + AC_SUBST([LUA_EXEC_PREFIX], ['${exec_prefix}']) + + dnl Lua provides no way to query the script directory, and instead + dnl provides LUA_PATH. However, we should be able to make a safe educated + dnl guess. If the built-in search path contains a directory which is + dnl prefixed by $prefix, then we can store scripts there. The first + dnl matching path will be used. + AC_CACHE_CHECK([for $ax_display_LUA script directory], + [ax_cv_lua_luadir], + [ AS_IF([test "x$prefix" = 'xNONE'], + [ax_lua_prefix=$ac_default_prefix], + [ax_lua_prefix=$prefix]) + + dnl Initialize to the default path. + ax_cv_lua_luadir="$LUA_PREFIX/share/lua/$LUA_VERSION" + + dnl Try to find a path with the prefix. + _AX_LUA_FND_PRFX_PTH([$LUA], [$ax_lua_prefix], [script]) + AS_IF([test "x$ax_lua_prefixed_path" != 'x'], + [ dnl Fix the prefix. + _ax_strip_prefix=`echo "$ax_lua_prefix" | $SED 's|.|.|g'` + ax_cv_lua_luadir=`echo "$ax_lua_prefixed_path" | \ + $SED "s|^$_ax_strip_prefix|$LUA_PREFIX|"` + ]) + ]) + AC_SUBST([luadir], [$ax_cv_lua_luadir]) + AC_SUBST([pkgluadir], [\${luadir}/$PACKAGE]) + + dnl Lua provides no way to query the module directory, and instead + dnl provides LUA_PATH. However, we should be able to make a safe educated + dnl guess. If the built-in search path contains a directory which is + dnl prefixed by $exec_prefix, then we can store modules there. The first + dnl matching path will be used. + AC_CACHE_CHECK([for $ax_display_LUA module directory], + [ax_cv_lua_luaexecdir], + [ AS_IF([test "x$exec_prefix" = 'xNONE'], + [ax_lua_exec_prefix=$ax_lua_prefix], + [ax_lua_exec_prefix=$exec_prefix]) + + dnl Initialize to the default path. + ax_cv_lua_luaexecdir="$LUA_EXEC_PREFIX/lib/lua/$LUA_VERSION" + + dnl Try to find a path with the prefix. + _AX_LUA_FND_PRFX_PTH([$LUA], + [$ax_lua_exec_prefix], [module]) + AS_IF([test "x$ax_lua_prefixed_path" != 'x'], + [ dnl Fix the prefix. + _ax_strip_prefix=`echo "$ax_lua_exec_prefix" | $SED 's|.|.|g'` + ax_cv_lua_luaexecdir=`echo "$ax_lua_prefixed_path" | \ + $SED "s|^$_ax_strip_prefix|$LUA_EXEC_PREFIX|"` + ]) + ]) + AC_SUBST([luaexecdir], [$ax_cv_lua_luaexecdir]) + AC_SUBST([pkgluaexecdir], [\${luaexecdir}/$PACKAGE]) + + dnl Run any user specified action. + $3 + ]) +]) + +dnl AX_WITH_LUA is now the same thing as AX_PROG_LUA. +AC_DEFUN([AX_WITH_LUA], +[ + AC_MSG_WARN([[$0 is deprecated, please use AX_PROG_LUA instead]]) + AX_PROG_LUA +]) + + +dnl ========================================================================= +dnl _AX_LUA_CHK_IS_INTRP(PROG, [ACTION-IF-TRUE], [ACTION-IF-FALSE]) +dnl ========================================================================= +AC_DEFUN([_AX_LUA_CHK_IS_INTRP], +[ + dnl A minimal Lua factorial to prove this is an interpreter. This should work + dnl for Lua interpreters version 5.0 and beyond. + _ax_lua_factorial=[`$1 2>/dev/null -e ' + -- a simple factorial + function fact (n) + if n == 0 then + return 1 + else + return n * fact(n-1) + end + end + print("fact(5) is " .. fact(5))'`] + AS_IF([test "$_ax_lua_factorial" = 'fact(5) is 120'], + [$2], [$3]) +]) + + +dnl ========================================================================= +dnl _AX_LUA_CHK_VER(PROG, MINIMUM-VERSION, [TOO-BIG-VERSION], +dnl [ACTION-IF-TRUE], [ACTION-IF-FALSE]) +dnl ========================================================================= +AC_DEFUN([_AX_LUA_CHK_VER], +[ + dnl Check that the Lua version is within the bounds. Only the major and minor + dnl version numbers are considered. This should work for Lua interpreters + dnl version 5.0 and beyond. + _ax_lua_good_version=[`$1 -e ' + -- a script to compare versions + function verstr2num(verstr) + local _, _, majorver, minorver = string.find(verstr, "^(%d+)%.(%d+)") + if majorver and minorver then + return tonumber(majorver) * 100 + tonumber(minorver) + end + end + local minver = verstr2num("$2") + local _, _, trimver = string.find(_VERSION, "^Lua (.*)") + local ver = verstr2num(trimver) + local maxver = verstr2num("$3") or 1e9 + if minver <= ver and ver < maxver then + print("yes") + else + print("no") + end'`] + AS_IF([test "x$_ax_lua_good_version" = "xyes"], + [$4], [$5]) +]) + + +dnl ========================================================================= +dnl _AX_LUA_FND_PRFX_PTH(PROG, PREFIX, SCRIPT-OR-MODULE-DIR) +dnl ========================================================================= +AC_DEFUN([_AX_LUA_FND_PRFX_PTH], +[ + dnl Get the script or module directory by querying the Lua interpreter, + dnl filtering on the given prefix, and selecting the shallowest path. If no + dnl path is found matching the prefix, the result will be an empty string. + dnl The third argument determines the type of search, it can be 'script' or + dnl 'module'. Supplying 'script' will perform the search with package.path + dnl and LUA_PATH, and supplying 'module' will search with package.cpath and + dnl LUA_CPATH. This is done for compatibility with Lua 5.0. + + ax_lua_prefixed_path=[`$1 -e ' + -- get the path based on search type + local searchtype = "$3" + local paths = "" + if searchtype == "script" then + paths = (package and package.path) or LUA_PATH + elseif searchtype == "module" then + paths = (package and package.cpath) or LUA_CPATH + end + -- search for the prefix + local prefix = "'$2'" + local minpath = "" + local mindepth = 1e9 + string.gsub(paths, "(@<:@^;@:>@+)", + function (path) + path = string.gsub(path, "%?.*$", "") + path = string.gsub(path, "/@<:@^/@:>@*$", "") + if string.find(path, prefix) then + local depth = string.len(string.gsub(path, "@<:@^/@:>@", "")) + if depth < mindepth then + minpath = path + mindepth = depth + end + end + end) + print(minpath)'`] +]) + + +dnl ========================================================================= +dnl AX_LUA_HEADERS([ACTION-IF-FOUND], [ACTION-IF-NOT-FOUND]) +dnl ========================================================================= +AC_DEFUN([AX_LUA_HEADERS], +[ + dnl Check for LUA_VERSION. + AC_MSG_CHECKING([if LUA_VERSION is defined]) + AS_IF([test "x$LUA_VERSION" != 'x'], + [AC_MSG_RESULT([yes])], + [ AC_MSG_RESULT([no]) + AC_MSG_ERROR([cannot check Lua headers without knowing LUA_VERSION]) + ]) + + dnl Make LUA_INCLUDE a precious variable. + AC_ARG_VAR([LUA_INCLUDE], [The Lua includes, e.g. -I/usr/include/lua5.1]) + + dnl Some default directories to search. + LUA_SHORT_VERSION=`echo "$LUA_VERSION" | $SED 's|\.||'` + m4_define_default([_AX_LUA_INCLUDE_LIST], + [ /usr/include/lua$LUA_VERSION \ + /usr/include/lua-$LUA_VERSION \ + /usr/include/lua/$LUA_VERSION \ + /usr/include/lua$LUA_SHORT_VERSION \ + /usr/local/include/lua$LUA_VERSION \ + /usr/local/include/lua-$LUA_VERSION \ + /usr/local/include/lua/$LUA_VERSION \ + /usr/local/include/lua$LUA_SHORT_VERSION \ + ]) + + dnl Try to find the headers. + _ax_lua_saved_cppflags=$CPPFLAGS + CPPFLAGS="$CPPFLAGS $LUA_INCLUDE" + AC_CHECK_HEADERS([lua.h lualib.h lauxlib.h luaconf.h]) + CPPFLAGS=$_ax_lua_saved_cppflags + + dnl Try some other directories if LUA_INCLUDE was not set. + AS_IF([test "x$LUA_INCLUDE" = 'x' && + test "x$ac_cv_header_lua_h" != 'xyes'], + [ dnl Try some common include paths. + for _ax_include_path in _AX_LUA_INCLUDE_LIST; do + test ! -d "$_ax_include_path" && continue + + AC_MSG_CHECKING([for Lua headers in]) + AC_MSG_RESULT([$_ax_include_path]) + + AS_UNSET([ac_cv_header_lua_h]) + AS_UNSET([ac_cv_header_lualib_h]) + AS_UNSET([ac_cv_header_lauxlib_h]) + AS_UNSET([ac_cv_header_luaconf_h]) + + _ax_lua_saved_cppflags=$CPPFLAGS + CPPFLAGS="$CPPFLAGS -I$_ax_include_path" + AC_CHECK_HEADERS([lua.h lualib.h lauxlib.h luaconf.h]) + CPPFLAGS=$_ax_lua_saved_cppflags + + AS_IF([test "x$ac_cv_header_lua_h" = 'xyes'], + [ LUA_INCLUDE="-I$_ax_include_path" + break + ]) + done + ]) + + AS_IF([test "x$ac_cv_header_lua_h" = 'xyes'], + [ dnl Make a program to print LUA_VERSION defined in the header. + dnl TODO It would be really nice if we could do this without compiling a + dnl program, then it would work when cross compiling. But I'm not sure how + dnl to do this reliably. For now, assume versions match when cross compiling. + + AS_IF([test "x$cross_compiling" != 'xyes'], + [ AC_CACHE_CHECK([for Lua header version], + [ax_cv_lua_header_version], + [ _ax_lua_saved_cppflags=$CPPFLAGS + CPPFLAGS="$CPPFLAGS $LUA_INCLUDE" + AC_RUN_IFELSE( + [ AC_LANG_SOURCE([[ +#include <lua.h> +#include <stdlib.h> +#include <stdio.h> +int main(int argc, char ** argv) +{ + if(argc > 1) printf("%s", LUA_VERSION); + exit(EXIT_SUCCESS); +} +]]) + ], + [ ax_cv_lua_header_version=`./conftest$EXEEXT p | \ + $SED -n "s|^Lua \(@<:@0-9@:>@\{1,\}\.@<:@0-9@:>@\{1,\}\).\{0,\}|\1|p"` + ], + [ax_cv_lua_header_version='unknown']) + CPPFLAGS=$_ax_lua_saved_cppflags + ]) + + dnl Compare this to the previously found LUA_VERSION. + AC_MSG_CHECKING([if Lua header version matches $LUA_VERSION]) + AS_IF([test "x$ax_cv_lua_header_version" = "x$LUA_VERSION"], + [ AC_MSG_RESULT([yes]) + ax_header_version_match='yes' + ], + [ AC_MSG_RESULT([no]) + ax_header_version_match='no' + ]) + ], + [ AC_MSG_WARN([cross compiling so assuming header version number matches]) + ax_header_version_match='yes' + ]) + ]) + + dnl Was LUA_INCLUDE specified? + AS_IF([test "x$ax_header_version_match" != 'xyes' && + test "x$LUA_INCLUDE" != 'x'], + [AC_MSG_ERROR([cannot find headers for specified LUA_INCLUDE])]) + + dnl Test the final result and run user code. + AS_IF([test "x$ax_header_version_match" = 'xyes'], [$1], + [m4_default([$2], [AC_MSG_ERROR([cannot find Lua includes])])]) +]) + +dnl AX_LUA_HEADERS_VERSION no longer exists, use AX_LUA_HEADERS. +AC_DEFUN([AX_LUA_HEADERS_VERSION], +[ + AC_MSG_WARN([[$0 is deprecated, please use AX_LUA_HEADERS instead]]) +]) + + +dnl ========================================================================= +dnl AX_LUA_LIBS([ACTION-IF-FOUND], [ACTION-IF-NOT-FOUND]) +dnl ========================================================================= +AC_DEFUN([AX_LUA_LIBS], +[ + dnl TODO Should this macro also check various -L flags? + + dnl Check for LUA_VERSION. + AC_MSG_CHECKING([if LUA_VERSION is defined]) + AS_IF([test "x$LUA_VERSION" != 'x'], + [AC_MSG_RESULT([yes])], + [ AC_MSG_RESULT([no]) + AC_MSG_ERROR([cannot check Lua libs without knowing LUA_VERSION]) + ]) + + dnl Make LUA_LIB a precious variable. + AC_ARG_VAR([LUA_LIB], [The Lua library, e.g. -llua5.1]) + + AS_IF([test "x$LUA_LIB" != 'x'], + [ dnl Check that LUA_LIBS works. + _ax_lua_saved_libs=$LIBS + LIBS="$LUA_LIB $LIBS" + AC_SEARCH_LIBS([lua_load], [], + [_ax_found_lua_libs='yes'], + [_ax_found_lua_libs='no']) + LIBS=$_ax_lua_saved_libs + + dnl Check the result. + AS_IF([test "x$_ax_found_lua_libs" != 'xyes'], + [AC_MSG_ERROR([cannot find libs for specified LUA_LIB])]) + ], + [ dnl First search for extra libs. + _ax_lua_extra_libs='' + + _ax_lua_saved_libs=$LIBS + LIBS="$LUA_LIB $LIBS" + AC_SEARCH_LIBS([exp], [m]) + AC_SEARCH_LIBS([dlopen], [dl]) + LIBS=$_ax_lua_saved_libs + + AS_IF([test "x$ac_cv_search_exp" != 'xno' && + test "x$ac_cv_search_exp" != 'xnone required'], + [_ax_lua_extra_libs="$_ax_lua_extra_libs $ac_cv_search_exp"]) + + AS_IF([test "x$ac_cv_search_dlopen" != 'xno' && + test "x$ac_cv_search_dlopen" != 'xnone required'], + [_ax_lua_extra_libs="$_ax_lua_extra_libs $ac_cv_search_dlopen"]) + + dnl Try to find the Lua libs. + _ax_lua_saved_libs=$LIBS + LIBS="$LUA_LIB $LIBS" + AC_SEARCH_LIBS([lua_load], + [ lua$LUA_VERSION \ + lua$LUA_SHORT_VERSION \ + lua-$LUA_VERSION \ + lua-$LUA_SHORT_VERSION \ + lua \ + ], + [_ax_found_lua_libs='yes'], + [_ax_found_lua_libs='no'], + [$_ax_lua_extra_libs]) + LIBS=$_ax_lua_saved_libs + + AS_IF([test "x$ac_cv_search_lua_load" != 'xno' && + test "x$ac_cv_search_lua_load" != 'xnone required'], + [LUA_LIB="$ac_cv_search_lua_load $_ax_lua_extra_libs"]) + ]) + + dnl Test the result and run user code. + AS_IF([test "x$_ax_found_lua_libs" = 'xyes'], [$1], + [m4_default([$2], [AC_MSG_ERROR([cannot find Lua libs])])]) +]) + + +dnl ========================================================================= +dnl AX_LUA_READLINE([ACTION-IF-FOUND], [ACTION-IF-NOT-FOUND]) +dnl ========================================================================= +AC_DEFUN([AX_LUA_READLINE], +[ + AX_LIB_READLINE + AS_IF([test "x$ac_cv_header_readline_readline_h" != 'x' && + test "x$ac_cv_header_readline_history_h" != 'x'], + [ LUA_LIBS_CFLAGS="-DLUA_USE_READLINE $LUA_LIBS_CFLAGS" + $1 + ], + [$2]) +]) diff --git a/contrib/libucl/python/setup.py b/contrib/libucl/python/setup.py new file mode 100644 index 0000000..b2b8981 --- /dev/null +++ b/contrib/libucl/python/setup.py @@ -0,0 +1,37 @@ +import distutils.ccompiler +import distutils.sysconfig +from distutils.core import setup, Extension +import os + + +compiler = distutils.ccompiler.new_compiler() +search_paths=[os.path.expanduser('~/{}'), '/opt/local/{}', '/usr/local/{}', '/usr/{}'] +lib_paths = [ a.format("lib") for a in search_paths] +inc_paths = [ a.format("include") for a in search_paths] + +uclmodule = Extension('ucl', + include_dirs = inc_paths, + library_dirs = lib_paths, + libraries = ['ucl'], + sources = ['src/uclmodule.c'], + runtime_library_dirs = lib_paths, + language='c') + +setup(name='ucl', + version='1.0', + description='ucl parser and emmitter', + ext_modules = [uclmodule], + author="Eitan Adler", + author_email="lists@eitanadler.com", + url="https://github.com/vstakhov/libucl/", + license="MIT", + classifiers=["Development Status :: 3 - Alpha", + "Intended Audience :: Developers", + "License :: DFSG approved", + "License :: OSI Approved :: MIT License", + "Programming Language :: C", + "Programming Language :: Python :: 3", + "Programming Language :: Python :: Implementation :: CPython", + "Topic :: Software Development :: Libraries", + ] + ) diff --git a/contrib/libucl/python/src/uclmodule.c b/contrib/libucl/python/src/uclmodule.c new file mode 100644 index 0000000..d63b73a --- /dev/null +++ b/contrib/libucl/python/src/uclmodule.c @@ -0,0 +1,156 @@ +// Attempts to load a UCL structure from a string +#include <ucl.h> +#include <Python.h> + +static PyObject* +_basic_ucl_type(ucl_object_t const * const obj) { + if (obj->type == UCL_INT) { + return Py_BuildValue("L", (long long)ucl_object_toint (obj)); + } + else if (obj->type == UCL_FLOAT) { + return Py_BuildValue("d", ucl_object_todouble (obj)); + } + else if (obj->type == UCL_STRING) { + return Py_BuildValue("s", ucl_object_tostring (obj)); + } + else if (obj->type == UCL_BOOLEAN) { + // maybe used 'p' here? + return Py_BuildValue("s", ucl_object_tostring_forced (obj)); + } + else if (obj->type == UCL_TIME) { + return Py_BuildValue("d", ucl_object_todouble (obj)); + } + return NULL; +} + +static PyObject* +_iterate_valid_ucl(ucl_object_t const * obj) { + const ucl_object_t *tmp; + ucl_object_iter_t it = NULL; + + tmp = obj; + + while ((obj = ucl_iterate_object (tmp, &it, false))) { + + PyObject* val; + + val = _basic_ucl_type(obj); + if (!val) { + PyObject* key = NULL; + if (obj->key != NULL) { + key = Py_BuildValue("s", ucl_object_key(obj)); + } + + PyObject* ret; + ret = PyDict_New(); + if (obj->type == UCL_OBJECT) { + val = PyDict_New(); + const ucl_object_t *cur; + ucl_object_iter_t it_obj = NULL; + while ((cur = ucl_iterate_object (obj, &it_obj, true))) { + PyObject* keyobj = Py_BuildValue("s",ucl_object_key(cur)); + PyDict_SetItem(val, keyobj, _iterate_valid_ucl(cur)); + } + } + else if (obj->type == UCL_ARRAY) { + val = PyList_New(0); + const ucl_object_t *cur; + ucl_object_iter_t it_obj = NULL; + while ((cur = ucl_iterate_object (obj, &it_obj, true))) { + PyList_Append(val, _iterate_valid_ucl(cur)); + } + } + else if (obj->type == UCL_USERDATA) { + // XXX: this should be + // PyBytes_FromStringAndSize; where is the + // length from? + val = PyBytes_FromString(obj->value.ud); + } + } + return val; + } + + PyErr_SetString(PyExc_SystemError, "unhandled type"); + return NULL; +} + +static PyObject* +_internal_load_ucl(char* uclstr) { + PyObject* ret; + + struct ucl_parser *parser = ucl_parser_new (UCL_PARSER_NO_TIME); + + bool r = ucl_parser_add_string(parser, uclstr, 0); + if (r) { + if (ucl_parser_get_error (parser)) { + PyErr_SetString(PyExc_ValueError, ucl_parser_get_error(parser)); + ucl_parser_free(parser); + ret = NULL; + goto return_with_parser; + } else { + ucl_object_t* uclobj = ucl_parser_get_object(parser); + ret = _iterate_valid_ucl(uclobj); + ucl_object_unref(uclobj); + goto return_with_parser; + } + + } else { + PyErr_SetString(PyExc_ValueError, ucl_parser_get_error (parser)); + ret = NULL; + goto return_with_parser; + } + +return_with_parser: + ucl_parser_free(parser); + return ret; +} + +static PyObject* +ucl_load(PyObject *self, PyObject *args) { + char* uclstr; + if (PyArg_ParseTuple(args, "z", &uclstr)) { + if (!uclstr) { + Py_RETURN_NONE; + } + return _internal_load_ucl(uclstr); + } + return NULL; +} + +static PyObject* +ucl_validate(PyObject *self, PyObject *args) { + char *uclstr, *schema; + if (PyArg_ParseTuple(args, "zz", &uclstr, &schema)) { + if (!uclstr || !schema) { + Py_RETURN_NONE; + } + PyErr_SetString(PyExc_NotImplementedError, "schema validation is not yet supported"); + return NULL; + } + return NULL; +} + +static PyMethodDef uclMethods[] = { + {"load", ucl_load, METH_VARARGS, "Load UCL from stream"}, + {"validate", ucl_validate, METH_VARARGS, "Validate ucl stream against schema"}, + {NULL, NULL, 0, NULL} +}; + +#if PY_MAJOR_VERSION >= 3 +static struct PyModuleDef uclmodule = { + PyModuleDef_HEAD_INIT, + "ucl", + NULL, + -1, + uclMethods +}; + +PyMODINIT_FUNC +PyInit_ucl(void) { + return PyModule_Create(&uclmodule); +} +#else +void initucl(void) { + Py_InitModule("ucl", uclMethods); +} +#endif diff --git a/contrib/libucl/python/test.sh b/contrib/libucl/python/test.sh new file mode 100755 index 0000000..53af6a3 --- /dev/null +++ b/contrib/libucl/python/test.sh @@ -0,0 +1,6 @@ +#!/bin/sh +set -xe +python3.4 setup.py build_ext --inplace +./test_uclmodule.py -v +rm -rfv build +rm ucl.so diff --git a/contrib/libucl/python/test_uclmodule.py b/contrib/libucl/python/test_uclmodule.py new file mode 100755 index 0000000..de295dc --- /dev/null +++ b/contrib/libucl/python/test_uclmodule.py @@ -0,0 +1,100 @@ +#!/usr/bin/env python +import json +import unittest +import ucl +import sys + +if sys.version_info[:2] == (2, 7): + unittest.TestCase.assertRaisesRegex = unittest.TestCase.assertRaisesRegexp + + +class TestUcl(unittest.TestCase): + def test_no_args(self): + with self.assertRaises(TypeError): + ucl.load() + + def test_multi_args(self): + with self.assertRaises(TypeError): + ucl.load(0,0) + + def test_none(self): + r = ucl.load(None) + self.assertEqual(r, None) + + def test_int(self): + r = ucl.load("a : 1") + self.assertEqual(ucl.load("a : 1"), { "a" : 1 } ) + + def test_braced_int(self): + self.assertEqual(ucl.load("{a : 1}"), { "a" : 1 } ) + + def test_nested_int(self): + self.assertEqual(ucl.load("a : { b : 1 }"), { "a" : { "b" : 1 } }) + + def test_str(self): + self.assertEqual(ucl.load("a : b"), {"a" : "b"}) + + def test_float(self): + self.assertEqual(ucl.load("a : 1.1"), {"a" : 1.1}) + + def test_empty_ucl(self): + r = ucl.load("{}") + self.assertEqual(r, {}) + + def test_single_brace(self): + self.assertEqual(ucl.load("{"), {}) + + def test_single_back_brace(self): + ucl.load("}") + + def test_single_square_forward(self): + self.assertEqual(ucl.load("["), []) + + def test_invalid_ucl(self): + with self.assertRaisesRegex(ValueError, "unfinished key$"): + ucl.load('{ "var"') + + def test_comment_ignored(self): + self.assertEqual(ucl.load("{/*1*/}"), {}) + + def test_1_in(self): + with open("../tests/basic/1.in", "r") as in1: + self.assertEqual(ucl.load(in1.read()), {'key1': 'value'}) + + def test_every_type(self): + totest="""{ + "key1": value; + "key2": value2; + "key3": "value;" + "key4": 1.0, + "key5": -0xdeadbeef + "key6": 0xdeadbeef.1 + "key7": 0xreadbeef + "key8": -1e-10, + "key9": 1 + "key10": true + "key11": no + "key12": yes + }""" + correct = { + 'key1': 'value', + 'key2': 'value2', + 'key3': 'value;', + 'key4': 1.0, + 'key5': -3735928559, + 'key6': '0xdeadbeef.1', + 'key7': '0xreadbeef', + 'key8': -1e-10, + 'key9': 1, + 'key10': 'true', + 'key11': 'false', + 'key12': 'true', + } + self.assertEqual(ucl.load(totest), correct) + + def test_validation_useless(self): + with self.assertRaises(NotImplementedError): + ucl.validate("","") + +if __name__ == '__main__': + unittest.main() diff --git a/contrib/libucl/src/Makefile.am b/contrib/libucl/src/Makefile.am new file mode 100644 index 0000000..c2f3a4b --- /dev/null +++ b/contrib/libucl/src/Makefile.am @@ -0,0 +1,31 @@ +libucl_common_cflags= -I$(top_srcdir)/src \ + -I$(top_srcdir)/include \ + -I$(top_srcdir)/uthash \ + -I$(top_srcdir)/klib \ + -Wall -W -Wno-unused-parameter -Wno-pointer-sign +lib_LTLIBRARIES= libucl.la +libucl_la_SOURCES= ucl_emitter.c \ + ucl_emitter_streamline.c \ + ucl_emitter_utils.c \ + ucl_hash.c \ + ucl_parser.c \ + ucl_schema.c \ + ucl_util.c \ + ucl_msgpack.c \ + ucl_sexp.c \ + xxhash.c +libucl_la_CFLAGS= $(libucl_common_cflags) \ + @CURL_CFLAGS@ +libucl_la_LDFLAGS = -version-info @SO_VERSION@ +libucl_la_LIBADD= @LIBFETCH_LIBS@ \ + @LIBCRYPTO_LIB@ \ + @LIBREGEX_LIB@ \ + @CURL_LIBS@ + +include_HEADERS= $(top_srcdir)/include/ucl.h \ + $(top_srcdir)/include/ucl++.h +noinst_HEADERS= ucl_internal.h \ + xxhash.h \ + ucl_hash.h \ + ucl_chartable.h \ + tree.h diff --git a/contrib/libucl/src/tree.h b/contrib/libucl/src/tree.h new file mode 100644 index 0000000..404b4a8 --- /dev/null +++ b/contrib/libucl/src/tree.h @@ -0,0 +1,219 @@ +/* tree.h -- AVL trees (in the spirit of BSD's 'queue.h') -*- C -*- */ + +/* Copyright (c) 2005 Ian Piumarta + * + * All rights reserved. + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the 'Software'), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, and/or sell copies of the + * Software, and to permit persons to whom the Software is furnished to do so, + * provided that the above copyright notice(s) and this permission notice appear + * in all copies of the Software and that both the above copyright notice(s) and + * this permission notice appear in supporting documentation. + * + * THE SOFTWARE IS PROVIDED 'AS IS'. USE ENTIRELY AT YOUR OWN RISK. + */ + +/* This file defines an AVL balanced binary tree [Georgii M. Adelson-Velskii and + * Evgenii M. Landis, 'An algorithm for the organization of information', + * Doklady Akademii Nauk SSSR, 146:263-266, 1962 (Russian). Also in Myron + * J. Ricci (trans.), Soviet Math, 3:1259-1263, 1962 (English)]. + * + * An AVL tree is headed by pointers to the root node and to a function defining + * the ordering relation between nodes. Each node contains an arbitrary payload + * plus three fields per tree entry: the depth of the subtree for which it forms + * the root and two pointers to child nodes (singly-linked for minimum space, at + * the expense of direct access to the parent node given a pointer to one of the + * children). The tree is rebalanced after every insertion or removal. The + * tree may be traversed in two directions: forward (in-order left-to-right) and + * reverse (in-order, right-to-left). + * + * Because of the recursive nature of many of the operations on trees it is + * necessary to define a number of helper functions for each type of tree node. + * The macro TREE_DEFINE(node_tag, entry_name) defines these functions with + * unique names according to the node_tag. This macro should be invoked, + * thereby defining the necessary functions, once per node tag in the program. + * + * For details on the use of these macros, see the tree(3) manual page. + */ + +#ifndef __tree_h +#define __tree_h + + +#define TREE_DELTA_MAX 1 +#ifndef _HU_FUNCTION +# if defined(__GNUC__) || defined(__clang__) +# define _HU_FUNCTION(x) __attribute__((__unused__)) x +# else +# define _HU_FUNCTION(x) x +# endif +#endif + +#define TREE_ENTRY(type) \ + struct { \ + struct type *avl_left; \ + struct type *avl_right; \ + int avl_height; \ + } + +#define TREE_HEAD(name, type) \ + struct name { \ + struct type *th_root; \ + int (*th_cmp)(struct type *lhs, struct type *rhs); \ + } + +#define TREE_INITIALIZER(cmp) { 0, cmp } + +#define TREE_DELTA(self, field) \ + (( (((self)->field.avl_left) ? (self)->field.avl_left->field.avl_height : 0)) \ + - (((self)->field.avl_right) ? (self)->field.avl_right->field.avl_height : 0)) + +/* Recursion prevents the following from being defined as macros. */ + +#define TREE_DEFINE(node, field) \ + \ + static struct node *_HU_FUNCTION(TREE_BALANCE_##node##_##field)(struct node *); \ + \ + static struct node *_HU_FUNCTION(TREE_ROTL_##node##_##field)(struct node *self) \ + { \ + struct node *r= self->field.avl_right; \ + self->field.avl_right= r->field.avl_left; \ + r->field.avl_left= TREE_BALANCE_##node##_##field(self); \ + return TREE_BALANCE_##node##_##field(r); \ + } \ + \ + static struct node *_HU_FUNCTION(TREE_ROTR_##node##_##field)(struct node *self) \ + { \ + struct node *l= self->field.avl_left; \ + self->field.avl_left= l->field.avl_right; \ + l->field.avl_right= TREE_BALANCE_##node##_##field(self); \ + return TREE_BALANCE_##node##_##field(l); \ + } \ + \ + static struct node *_HU_FUNCTION(TREE_BALANCE_##node##_##field)(struct node *self) \ + { \ + int delta= TREE_DELTA(self, field); \ + \ + if (delta < -TREE_DELTA_MAX) \ + { \ + if (TREE_DELTA(self->field.avl_right, field) > 0) \ + self->field.avl_right= TREE_ROTR_##node##_##field(self->field.avl_right); \ + return TREE_ROTL_##node##_##field(self); \ + } \ + else if (delta > TREE_DELTA_MAX) \ + { \ + if (TREE_DELTA(self->field.avl_left, field) < 0) \ + self->field.avl_left= TREE_ROTL_##node##_##field(self->field.avl_left); \ + return TREE_ROTR_##node##_##field(self); \ + } \ + self->field.avl_height= 0; \ + if (self->field.avl_left && (self->field.avl_left->field.avl_height > self->field.avl_height)) \ + self->field.avl_height= self->field.avl_left->field.avl_height; \ + if (self->field.avl_right && (self->field.avl_right->field.avl_height > self->field.avl_height)) \ + self->field.avl_height= self->field.avl_right->field.avl_height; \ + self->field.avl_height += 1; \ + return self; \ + } \ + \ + static struct node *_HU_FUNCTION(TREE_INSERT_##node##_##field) \ + (struct node *self, struct node *elm, int (*compare)(struct node *lhs, struct node *rhs)) \ + { \ + if (!self) \ + return elm; \ + if (compare(elm, self) < 0) \ + self->field.avl_left= TREE_INSERT_##node##_##field(self->field.avl_left, elm, compare); \ + else \ + self->field.avl_right= TREE_INSERT_##node##_##field(self->field.avl_right, elm, compare); \ + return TREE_BALANCE_##node##_##field(self); \ + } \ + \ + static struct node *_HU_FUNCTION(TREE_FIND_##node##_##field) \ + (struct node *self, struct node *elm, int (*compare)(struct node *lhs, struct node *rhs)) \ + { \ + if (!self) \ + return 0; \ + if (compare(elm, self) == 0) \ + return self; \ + if (compare(elm, self) < 0) \ + return TREE_FIND_##node##_##field(self->field.avl_left, elm, compare); \ + else \ + return TREE_FIND_##node##_##field(self->field.avl_right, elm, compare); \ + } \ + \ + static struct node *_HU_FUNCTION(TREE_MOVE_RIGHT)(struct node *self, struct node *rhs) \ + { \ + if (!self) \ + return rhs; \ + self->field.avl_right= TREE_MOVE_RIGHT(self->field.avl_right, rhs); \ + return TREE_BALANCE_##node##_##field(self); \ + } \ + \ + static struct node *_HU_FUNCTION(TREE_REMOVE_##node##_##field) \ + (struct node *self, struct node *elm, int (*compare)(struct node *lhs, struct node *rhs)) \ + { \ + if (!self) return 0; \ + \ + if (compare(elm, self) == 0) \ + { \ + struct node *tmp= TREE_MOVE_RIGHT(self->field.avl_left, self->field.avl_right); \ + self->field.avl_left= 0; \ + self->field.avl_right= 0; \ + return tmp; \ + } \ + if (compare(elm, self) < 0) \ + self->field.avl_left= TREE_REMOVE_##node##_##field(self->field.avl_left, elm, compare); \ + else \ + self->field.avl_right= TREE_REMOVE_##node##_##field(self->field.avl_right, elm, compare); \ + return TREE_BALANCE_##node##_##field(self); \ + } \ + \ + static void _HU_FUNCTION(TREE_FORWARD_APPLY_ALL_##node##_##field) \ + (struct node *self, void (*function)(struct node *node, void *data), void *data) \ + { \ + if (self) \ + { \ + TREE_FORWARD_APPLY_ALL_##node##_##field(self->field.avl_left, function, data); \ + function(self, data); \ + TREE_FORWARD_APPLY_ALL_##node##_##field(self->field.avl_right, function, data); \ + } \ + } \ + \ + static void _HU_FUNCTION(TREE_REVERSE_APPLY_ALL_##node##_##field) \ + (struct node *self, void (*function)(struct node *node, void *data), void *data) \ + { \ + if (self) \ + { \ + TREE_REVERSE_APPLY_ALL_##node##_##field(self->field.avl_right, function, data); \ + function(self, data); \ + TREE_REVERSE_APPLY_ALL_##node##_##field(self->field.avl_left, function, data); \ + } \ + } + +#define TREE_INSERT(head, node, field, elm) \ + ((head)->th_root= TREE_INSERT_##node##_##field((head)->th_root, (elm), (head)->th_cmp)) + +#define TREE_FIND(head, node, field, elm) \ + (TREE_FIND_##node##_##field((head)->th_root, (elm), (head)->th_cmp)) + +#define TREE_REMOVE(head, node, field, elm) \ + ((head)->th_root= TREE_REMOVE_##node##_##field((head)->th_root, (elm), (head)->th_cmp)) + +#define TREE_DEPTH(head, field) \ + ((head)->th_root->field.avl_height) + +#define TREE_FORWARD_APPLY(head, node, field, function, data) \ + TREE_FORWARD_APPLY_ALL_##node##_##field((head)->th_root, function, data) + +#define TREE_REVERSE_APPLY(head, node, field, function, data) \ + TREE_REVERSE_APPLY_ALL_##node##_##field((head)->th_root, function, data) + +#define TREE_INIT(head, cmp) do { \ + (head)->th_root= 0; \ + (head)->th_cmp= (cmp); \ + } while (0) + + +#endif /* __tree_h */ diff --git a/contrib/libucl/src/ucl_chartable.h b/contrib/libucl/src/ucl_chartable.h new file mode 100644 index 0000000..db9f029 --- /dev/null +++ b/contrib/libucl/src/ucl_chartable.h @@ -0,0 +1,268 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef UCL_CHARTABLE_H_ +#define UCL_CHARTABLE_H_ + +#include "ucl_internal.h" + +static const unsigned int ucl_chartable[256] = { +UCL_CHARACTER_VALUE_END, UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, +UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, +UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, +UCL_CHARACTER_JSON_UNSAFE|UCL_CHARACTER_UCL_UNSAFE, +UCL_CHARACTER_WHITESPACE|UCL_CHARACTER_WHITESPACE_UNSAFE|UCL_CHARACTER_KEY_SEP|UCL_CHARACTER_JSON_UNSAFE|UCL_CHARACTER_UCL_UNSAFE, +UCL_CHARACTER_WHITESPACE_UNSAFE|UCL_CHARACTER_VALUE_END|UCL_CHARACTER_JSON_UNSAFE|UCL_CHARACTER_UCL_UNSAFE, +UCL_CHARACTER_WHITESPACE_UNSAFE, +UCL_CHARACTER_WHITESPACE_UNSAFE|UCL_CHARACTER_JSON_UNSAFE|UCL_CHARACTER_UCL_UNSAFE, +UCL_CHARACTER_WHITESPACE_UNSAFE|UCL_CHARACTER_VALUE_END|UCL_CHARACTER_JSON_UNSAFE|UCL_CHARACTER_UCL_UNSAFE, +UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, +UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, +UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, +UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, +UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, +UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, UCL_CHARACTER_DENIED, +UCL_CHARACTER_WHITESPACE|UCL_CHARACTER_WHITESPACE_UNSAFE|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_KEY_SEP|UCL_CHARACTER_UCL_UNSAFE /* */, +UCL_CHARACTER_VALUE_STR /* ! */, +UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_ESCAPE|UCL_CHARACTER_JSON_UNSAFE|UCL_CHARACTER_UCL_UNSAFE /* " */, +UCL_CHARACTER_VALUE_END /* # */, UCL_CHARACTER_VALUE_STR /* $ */, +UCL_CHARACTER_VALUE_STR /* % */, UCL_CHARACTER_VALUE_STR /* & */, +UCL_CHARACTER_VALUE_STR /* ' */, UCL_CHARACTER_VALUE_STR /* ( */, +UCL_CHARACTER_VALUE_STR /* ) */, UCL_CHARACTER_VALUE_STR /* * */, +UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* + */, +UCL_CHARACTER_VALUE_END /* , */, +UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT_START|UCL_CHARACTER_VALUE_DIGIT /* - */, +UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* . */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_ESCAPE /* / */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT_START|UCL_CHARACTER_VALUE_DIGIT /* 0 */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT_START|UCL_CHARACTER_VALUE_DIGIT /* 1 */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT_START|UCL_CHARACTER_VALUE_DIGIT /* 2 */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT_START|UCL_CHARACTER_VALUE_DIGIT /* 3 */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT_START|UCL_CHARACTER_VALUE_DIGIT /* 4 */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT_START|UCL_CHARACTER_VALUE_DIGIT /* 5 */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT_START|UCL_CHARACTER_VALUE_DIGIT /* 6 */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT_START|UCL_CHARACTER_VALUE_DIGIT /* 7 */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT_START|UCL_CHARACTER_VALUE_DIGIT /* 8 */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT_START|UCL_CHARACTER_VALUE_DIGIT /* 9 */, +UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_KEY_SEP|UCL_CHARACTER_UCL_UNSAFE /* : */, +UCL_CHARACTER_VALUE_END /* ; */, UCL_CHARACTER_VALUE_STR /* < */, +UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_KEY_SEP|UCL_CHARACTER_UCL_UNSAFE /* = */, +UCL_CHARACTER_VALUE_STR /* > */, UCL_CHARACTER_VALUE_STR /* ? */, +UCL_CHARACTER_VALUE_STR /* @ */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* A */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* B */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* C */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* D */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* E */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* F */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* G */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* H */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* I */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* J */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* K */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* L */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* M */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* N */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* O */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* P */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* Q */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* R */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* S */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* T */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* U */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* V */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* W */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* X */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* Y */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* Z */, +UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_UCL_UNSAFE /* [ */, +UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_ESCAPE|UCL_CHARACTER_JSON_UNSAFE|UCL_CHARACTER_UCL_UNSAFE /* \ */, +UCL_CHARACTER_VALUE_END /* ] */, UCL_CHARACTER_VALUE_STR /* ^ */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR /* _ */, +UCL_CHARACTER_VALUE_STR /* ` */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* a */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT|UCL_CHARACTER_ESCAPE /* b */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* c */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* d */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* e */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT|UCL_CHARACTER_ESCAPE /* f */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* g */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* h */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* i */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* j */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* k */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* l */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* m */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT|UCL_CHARACTER_ESCAPE /* n */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* o */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* p */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* q */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT|UCL_CHARACTER_ESCAPE /* r */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* s */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT|UCL_CHARACTER_ESCAPE /* t */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT|UCL_CHARACTER_ESCAPE /* u */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* v */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* w */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* x */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* y */, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_VALUE_DIGIT /* z */, +UCL_CHARACTER_VALUE_STR|UCL_CHARACTER_UCL_UNSAFE /* { */, +UCL_CHARACTER_VALUE_STR /* | */, UCL_CHARACTER_VALUE_END /* } */, +UCL_CHARACTER_VALUE_STR /* ~ */, UCL_CHARACTER_DENIED, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR, +UCL_CHARACTER_KEY_START|UCL_CHARACTER_KEY|UCL_CHARACTER_VALUE_STR +}; + +static inline bool +ucl_test_character (unsigned char c, int type_flags) +{ + return (ucl_chartable[c] & type_flags) != 0; +} + +#endif /* UCL_CHARTABLE_H_ */ diff --git a/contrib/libucl/src/ucl_emitter.c b/contrib/libucl/src/ucl_emitter.c new file mode 100644 index 0000000..8bfbf09 --- /dev/null +++ b/contrib/libucl/src/ucl_emitter.c @@ -0,0 +1,643 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include "ucl.h" +#include "ucl_internal.h" +#include "ucl_chartable.h" +#ifdef HAVE_FLOAT_H +#include <float.h> +#endif +#ifdef HAVE_MATH_H +#include <math.h> +#endif + +/** + * @file ucl_emitter.c + * Serialise UCL object to various of output formats + */ + +static void ucl_emitter_common_elt (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool first, bool print_key, bool compact); + +#define UCL_EMIT_TYPE_OPS(type) \ + static void ucl_emit_ ## type ## _elt (struct ucl_emitter_context *ctx, \ + const ucl_object_t *obj, bool first, bool print_key); \ + static void ucl_emit_ ## type ## _start_obj (struct ucl_emitter_context *ctx, \ + const ucl_object_t *obj, bool print_key); \ + static void ucl_emit_ ## type## _start_array (struct ucl_emitter_context *ctx, \ + const ucl_object_t *obj, bool print_key); \ + static void ucl_emit_ ##type## _end_object (struct ucl_emitter_context *ctx, \ + const ucl_object_t *obj); \ + static void ucl_emit_ ##type## _end_array (struct ucl_emitter_context *ctx, \ + const ucl_object_t *obj) + +/* + * JSON format operations + */ +UCL_EMIT_TYPE_OPS(json); +UCL_EMIT_TYPE_OPS(json_compact); +UCL_EMIT_TYPE_OPS(config); +UCL_EMIT_TYPE_OPS(yaml); +UCL_EMIT_TYPE_OPS(msgpack); + +#define UCL_EMIT_TYPE_CONTENT(type) { \ + .ucl_emitter_write_elt = ucl_emit_ ## type ## _elt, \ + .ucl_emitter_start_object = ucl_emit_ ## type ##_start_obj, \ + .ucl_emitter_start_array = ucl_emit_ ## type ##_start_array, \ + .ucl_emitter_end_object = ucl_emit_ ## type ##_end_object, \ + .ucl_emitter_end_array = ucl_emit_ ## type ##_end_array \ +} + +const struct ucl_emitter_operations ucl_standartd_emitter_ops[] = { + [UCL_EMIT_JSON] = UCL_EMIT_TYPE_CONTENT(json), + [UCL_EMIT_JSON_COMPACT] = UCL_EMIT_TYPE_CONTENT(json_compact), + [UCL_EMIT_CONFIG] = UCL_EMIT_TYPE_CONTENT(config), + [UCL_EMIT_YAML] = UCL_EMIT_TYPE_CONTENT(yaml), + [UCL_EMIT_MSGPACK] = UCL_EMIT_TYPE_CONTENT(msgpack) +}; + +/* + * Utility to check whether we need a top object + */ +#define UCL_EMIT_IDENT_TOP_OBJ(ctx, obj) ((ctx)->top != (obj) || \ + ((ctx)->id == UCL_EMIT_JSON_COMPACT || (ctx)->id == UCL_EMIT_JSON)) + + +/** + * Add tabulation to the output buffer + * @param buf target buffer + * @param tabs number of tabs to add + */ +static inline void +ucl_add_tabs (const struct ucl_emitter_functions *func, unsigned int tabs, + bool compact) +{ + if (!compact && tabs > 0) { + func->ucl_emitter_append_character (' ', tabs * 4, func->ud); + } +} + +/** + * Print key for the element + * @param ctx + * @param obj + */ +static void +ucl_emitter_print_key (bool print_key, struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool compact) +{ + const struct ucl_emitter_functions *func = ctx->func; + + if (!print_key) { + return; + } + + if (ctx->id == UCL_EMIT_CONFIG) { + if (obj->flags & UCL_OBJECT_NEED_KEY_ESCAPE) { + ucl_elt_string_write_json (obj->key, obj->keylen, ctx); + } + else { + func->ucl_emitter_append_len (obj->key, obj->keylen, func->ud); + } + + if (obj->type != UCL_OBJECT && obj->type != UCL_ARRAY) { + func->ucl_emitter_append_len (" = ", 3, func->ud); + } + else { + func->ucl_emitter_append_character (' ', 1, func->ud); + } + } + else if (ctx->id == UCL_EMIT_YAML) { + if (obj->keylen > 0 && (obj->flags & UCL_OBJECT_NEED_KEY_ESCAPE)) { + ucl_elt_string_write_json (obj->key, obj->keylen, ctx); + } + else if (obj->keylen > 0) { + func->ucl_emitter_append_len (obj->key, obj->keylen, func->ud); + } + else { + func->ucl_emitter_append_len ("null", 4, func->ud); + } + + func->ucl_emitter_append_len (": ", 2, func->ud); + } + else { + if (obj->keylen > 0) { + ucl_elt_string_write_json (obj->key, obj->keylen, ctx); + } + else { + func->ucl_emitter_append_len ("null", 4, func->ud); + } + + if (compact) { + func->ucl_emitter_append_character (':', 1, func->ud); + } + else { + func->ucl_emitter_append_len (": ", 2, func->ud); + } + } +} + +static void +ucl_emitter_finish_object (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool compact, bool is_array) +{ + const struct ucl_emitter_functions *func = ctx->func; + + if (ctx->id == UCL_EMIT_CONFIG && obj != ctx->top) { + if (obj->type != UCL_OBJECT && obj->type != UCL_ARRAY) { + if (!is_array) { + /* Objects are split by ';' */ + func->ucl_emitter_append_len (";\n", 2, func->ud); + } + else { + /* Use commas for arrays */ + func->ucl_emitter_append_len (",\n", 2, func->ud); + } + } + else { + func->ucl_emitter_append_character ('\n', 1, func->ud); + } + } +} + +/** + * End standard ucl object + * @param ctx emitter context + * @param compact compact flag + */ +static void +ucl_emitter_common_end_object (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool compact) +{ + const struct ucl_emitter_functions *func = ctx->func; + + if (UCL_EMIT_IDENT_TOP_OBJ(ctx, obj)) { + ctx->indent --; + if (compact) { + func->ucl_emitter_append_character ('}', 1, func->ud); + } + else { + if (ctx->id != UCL_EMIT_CONFIG) { + /* newline is already added for this format */ + func->ucl_emitter_append_character ('\n', 1, func->ud); + } + ucl_add_tabs (func, ctx->indent, compact); + func->ucl_emitter_append_character ('}', 1, func->ud); + } + } + + ucl_emitter_finish_object (ctx, obj, compact, false); +} + +/** + * End standard ucl array + * @param ctx emitter context + * @param compact compact flag + */ +static void +ucl_emitter_common_end_array (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool compact) +{ + const struct ucl_emitter_functions *func = ctx->func; + + ctx->indent --; + if (compact) { + func->ucl_emitter_append_character (']', 1, func->ud); + } + else { + if (ctx->id != UCL_EMIT_CONFIG) { + /* newline is already added for this format */ + func->ucl_emitter_append_character ('\n', 1, func->ud); + } + ucl_add_tabs (func, ctx->indent, compact); + func->ucl_emitter_append_character (']', 1, func->ud); + } + + ucl_emitter_finish_object (ctx, obj, compact, true); +} + +/** + * Start emit standard UCL array + * @param ctx emitter context + * @param obj object to write + * @param compact compact flag + */ +static void +ucl_emitter_common_start_array (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool print_key, bool compact) +{ + const ucl_object_t *cur; + ucl_object_iter_t iter = NULL; + const struct ucl_emitter_functions *func = ctx->func; + bool first = true; + + ucl_emitter_print_key (print_key, ctx, obj, compact); + + if (compact) { + func->ucl_emitter_append_character ('[', 1, func->ud); + } + else { + func->ucl_emitter_append_len ("[\n", 2, func->ud); + } + + ctx->indent ++; + + if (obj->type == UCL_ARRAY) { + /* explicit array */ + while ((cur = ucl_iterate_object (obj, &iter, true)) != NULL) { + ucl_emitter_common_elt (ctx, cur, first, false, compact); + first = false; + } + } + else { + /* implicit array */ + cur = obj; + while (cur) { + ucl_emitter_common_elt (ctx, cur, first, false, compact); + first = false; + cur = cur->next; + } + } + + +} + +/** + * Start emit standard UCL object + * @param ctx emitter context + * @param obj object to write + * @param compact compact flag + */ +static void +ucl_emitter_common_start_object (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool print_key, bool compact) +{ + ucl_hash_iter_t it = NULL; + const ucl_object_t *cur, *elt; + const struct ucl_emitter_functions *func = ctx->func; + bool first = true; + + ucl_emitter_print_key (print_key, ctx, obj, compact); + /* + * Print <ident_level>{ + * <ident_level + 1><object content> + */ + if (UCL_EMIT_IDENT_TOP_OBJ(ctx, obj)) { + if (compact) { + func->ucl_emitter_append_character ('{', 1, func->ud); + } + else { + func->ucl_emitter_append_len ("{\n", 2, func->ud); + } + ctx->indent ++; + } + + while ((cur = ucl_hash_iterate (obj->value.ov, &it))) { + + if (ctx->id == UCL_EMIT_CONFIG) { + LL_FOREACH (cur, elt) { + ucl_emitter_common_elt (ctx, elt, first, true, compact); + } + } + else { + /* Expand implicit arrays */ + if (cur->next != NULL) { + if (!first) { + if (compact) { + func->ucl_emitter_append_character (',', 1, func->ud); + } + else { + func->ucl_emitter_append_len (",\n", 2, func->ud); + } + } + ucl_add_tabs (func, ctx->indent, compact); + ucl_emitter_common_start_array (ctx, cur, true, compact); + ucl_emitter_common_end_array (ctx, cur, compact); + } + else { + ucl_emitter_common_elt (ctx, cur, first, true, compact); + } + } + + first = false; + } +} + +/** + * Common choice of object emitting + * @param ctx emitter context + * @param obj object to print + * @param first flag to mark the first element + * @param print_key print key of an object + * @param compact compact output + */ +static void +ucl_emitter_common_elt (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool first, bool print_key, bool compact) +{ + const struct ucl_emitter_functions *func = ctx->func; + bool flag; + struct ucl_object_userdata *ud; + const char *ud_out = ""; + + if (ctx->id != UCL_EMIT_CONFIG && !first) { + if (compact) { + func->ucl_emitter_append_character (',', 1, func->ud); + } + else { + if (ctx->id == UCL_EMIT_YAML && ctx->indent == 0) { + func->ucl_emitter_append_len ("\n", 1, func->ud); + } else { + func->ucl_emitter_append_len (",\n", 2, func->ud); + } + } + } + + ucl_add_tabs (func, ctx->indent, compact); + + switch (obj->type) { + case UCL_INT: + ucl_emitter_print_key (print_key, ctx, obj, compact); + func->ucl_emitter_append_int (ucl_object_toint (obj), func->ud); + ucl_emitter_finish_object (ctx, obj, compact, !print_key); + break; + case UCL_FLOAT: + case UCL_TIME: + ucl_emitter_print_key (print_key, ctx, obj, compact); + func->ucl_emitter_append_double (ucl_object_todouble (obj), func->ud); + ucl_emitter_finish_object (ctx, obj, compact, !print_key); + break; + case UCL_BOOLEAN: + ucl_emitter_print_key (print_key, ctx, obj, compact); + flag = ucl_object_toboolean (obj); + if (flag) { + func->ucl_emitter_append_len ("true", 4, func->ud); + } + else { + func->ucl_emitter_append_len ("false", 5, func->ud); + } + ucl_emitter_finish_object (ctx, obj, compact, !print_key); + break; + case UCL_STRING: + ucl_emitter_print_key (print_key, ctx, obj, compact); + if (ctx->id == UCL_EMIT_CONFIG && ucl_maybe_long_string (obj)) { + ucl_elt_string_write_multiline (obj->value.sv, obj->len, ctx); + } + else { + ucl_elt_string_write_json (obj->value.sv, obj->len, ctx); + } + ucl_emitter_finish_object (ctx, obj, compact, !print_key); + break; + case UCL_NULL: + ucl_emitter_print_key (print_key, ctx, obj, compact); + func->ucl_emitter_append_len ("null", 4, func->ud); + ucl_emitter_finish_object (ctx, obj, compact, !print_key); + break; + case UCL_OBJECT: + ucl_emitter_common_start_object (ctx, obj, print_key, compact); + ucl_emitter_common_end_object (ctx, obj, compact); + break; + case UCL_ARRAY: + ucl_emitter_common_start_array (ctx, obj, print_key, compact); + ucl_emitter_common_end_array (ctx, obj, compact); + break; + case UCL_USERDATA: + ud = (struct ucl_object_userdata *)obj; + ucl_emitter_print_key (print_key, ctx, obj, compact); + if (ud->emitter) { + ud_out = ud->emitter (obj->value.ud); + if (ud_out == NULL) { + ud_out = "null"; + } + } + ucl_elt_string_write_json (ud_out, strlen (ud_out), ctx); + ucl_emitter_finish_object (ctx, obj, compact, !print_key); + break; + } +} + +/* + * Specific standard implementations of the emitter functions + */ +#define UCL_EMIT_TYPE_IMPL(type, compact) \ + static void ucl_emit_ ## type ## _elt (struct ucl_emitter_context *ctx, \ + const ucl_object_t *obj, bool first, bool print_key) { \ + ucl_emitter_common_elt (ctx, obj, first, print_key, (compact)); \ + } \ + static void ucl_emit_ ## type ## _start_obj (struct ucl_emitter_context *ctx, \ + const ucl_object_t *obj, bool print_key) { \ + ucl_emitter_common_start_object (ctx, obj, print_key, (compact)); \ + } \ + static void ucl_emit_ ## type## _start_array (struct ucl_emitter_context *ctx, \ + const ucl_object_t *obj, bool print_key) { \ + ucl_emitter_common_start_array (ctx, obj, print_key, (compact)); \ + } \ + static void ucl_emit_ ##type## _end_object (struct ucl_emitter_context *ctx, \ + const ucl_object_t *obj) { \ + ucl_emitter_common_end_object (ctx, obj, (compact)); \ + } \ + static void ucl_emit_ ##type## _end_array (struct ucl_emitter_context *ctx, \ + const ucl_object_t *obj) { \ + ucl_emitter_common_end_array (ctx, obj, (compact)); \ + } + +UCL_EMIT_TYPE_IMPL(json, false) +UCL_EMIT_TYPE_IMPL(json_compact, true) +UCL_EMIT_TYPE_IMPL(config, false) +UCL_EMIT_TYPE_IMPL(yaml, false) + +static void +ucl_emit_msgpack_elt (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool first, bool print_key) +{ + ucl_object_iter_t it; + struct ucl_object_userdata *ud; + const char *ud_out; + const ucl_object_t *cur, *celt; + + switch (obj->type) { + case UCL_INT: + ucl_emitter_print_key_msgpack (print_key, ctx, obj); + ucl_emitter_print_int_msgpack (ctx, ucl_object_toint (obj)); + break; + + case UCL_FLOAT: + case UCL_TIME: + ucl_emitter_print_key_msgpack (print_key, ctx, obj); + ucl_emitter_print_double_msgpack (ctx, ucl_object_todouble (obj)); + break; + + case UCL_BOOLEAN: + ucl_emitter_print_key_msgpack (print_key, ctx, obj); + ucl_emitter_print_bool_msgpack (ctx, ucl_object_toboolean (obj)); + break; + + case UCL_STRING: + ucl_emitter_print_key_msgpack (print_key, ctx, obj); + + if (obj->flags & UCL_OBJECT_BINARY) { + ucl_emitter_print_binary_string_msgpack (ctx, obj->value.sv, + obj->len); + } + else { + ucl_emitter_print_string_msgpack (ctx, obj->value.sv, obj->len); + } + break; + + case UCL_NULL: + ucl_emitter_print_key_msgpack (print_key, ctx, obj); + ucl_emitter_print_null_msgpack (ctx); + break; + + case UCL_OBJECT: + ucl_emitter_print_key_msgpack (print_key, ctx, obj); + ucl_emit_msgpack_start_obj (ctx, obj, print_key); + it = NULL; + + while ((cur = ucl_iterate_object (obj, &it, true)) != NULL) { + LL_FOREACH (cur, celt) { + ucl_emit_msgpack_elt (ctx, celt, false, true); + /* XXX: + * in msgpack the length of objects is encoded within a single elt + * so in case of multi-value keys we are using merely the first + * element ignoring others + */ + break; + } + } + + break; + + case UCL_ARRAY: + ucl_emitter_print_key_msgpack (print_key, ctx, obj); + ucl_emit_msgpack_start_array (ctx, obj, print_key); + it = NULL; + + while ((cur = ucl_iterate_object (obj, &it, true)) != NULL) { + ucl_emit_msgpack_elt (ctx, cur, false, false); + } + + break; + + case UCL_USERDATA: + ud = (struct ucl_object_userdata *)obj; + ucl_emitter_print_key_msgpack (print_key, ctx, obj); + + if (ud->emitter) { + ud_out = ud->emitter (obj->value.ud); + if (ud_out == NULL) { + ud_out = "null"; + } + } + ucl_emitter_print_string_msgpack (ctx, obj->value.sv, obj->len); + break; + } +} + +static void +ucl_emit_msgpack_start_obj (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool print_key) +{ + ucl_emitter_print_object_msgpack (ctx, obj->len); +} + +static void +ucl_emit_msgpack_start_array (struct ucl_emitter_context *ctx, + const ucl_object_t *obj, bool print_key) +{ + ucl_emitter_print_array_msgpack (ctx, obj->len); +} + +static void +ucl_emit_msgpack_end_object (struct ucl_emitter_context *ctx, + const ucl_object_t *obj) +{ + +} + +static void +ucl_emit_msgpack_end_array (struct ucl_emitter_context *ctx, + const ucl_object_t *obj) +{ + +} + +unsigned char * +ucl_object_emit (const ucl_object_t *obj, enum ucl_emitter emit_type) +{ + return ucl_object_emit_len (obj, emit_type, NULL); +} + +unsigned char * +ucl_object_emit_len (const ucl_object_t *obj, enum ucl_emitter emit_type, + size_t *outlen) +{ + unsigned char *res = NULL; + struct ucl_emitter_functions *func; + UT_string *s; + + if (obj == NULL) { + return NULL; + } + + func = ucl_object_emit_memory_funcs ((void **)&res); + s = func->ud; + + if (func != NULL) { + ucl_object_emit_full (obj, emit_type, func); + + if (outlen != NULL) { + *outlen = s->i; + } + + ucl_object_emit_funcs_free (func); + } + + return res; +} + +bool +ucl_object_emit_full (const ucl_object_t *obj, enum ucl_emitter emit_type, + struct ucl_emitter_functions *emitter) +{ + const struct ucl_emitter_context *ctx; + struct ucl_emitter_context my_ctx; + bool res = false; + + ctx = ucl_emit_get_standard_context (emit_type); + if (ctx != NULL) { + memcpy (&my_ctx, ctx, sizeof (my_ctx)); + my_ctx.func = emitter; + my_ctx.indent = 0; + my_ctx.top = obj; + + my_ctx.ops->ucl_emitter_write_elt (&my_ctx, obj, true, false); + res = true; + } + + return res; +} diff --git a/contrib/libucl/src/ucl_emitter_streamline.c b/contrib/libucl/src/ucl_emitter_streamline.c new file mode 100644 index 0000000..ff27c88 --- /dev/null +++ b/contrib/libucl/src/ucl_emitter_streamline.c @@ -0,0 +1,165 @@ +/* Copyright (c) 2014, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include "ucl.h" +#include "ucl_internal.h" +#include "ucl_chartable.h" + +struct ucl_emitter_streamline_stack { + bool is_array; + bool empty; + const ucl_object_t *obj; + struct ucl_emitter_streamline_stack *next; +}; + +struct ucl_emitter_context_streamline { + /* Inherited from the main context */ + const char *name; + int id; + const struct ucl_emitter_functions *func; + const struct ucl_emitter_operations *ops; + unsigned int ident; + const ucl_object_t *top; + + /* Streamline specific fields */ + struct ucl_emitter_streamline_stack *containers; +}; + +#define TO_STREAMLINE(ctx) (struct ucl_emitter_context_streamline *)(ctx) + +struct ucl_emitter_context* +ucl_object_emit_streamline_new (const ucl_object_t *obj, + enum ucl_emitter emit_type, + struct ucl_emitter_functions *emitter) +{ + const struct ucl_emitter_context *ctx; + struct ucl_emitter_context_streamline *sctx; + + ctx = ucl_emit_get_standard_context (emit_type); + if (ctx == NULL) { + return NULL; + } + + sctx = calloc (1, sizeof (*sctx)); + if (sctx == NULL) { + return NULL; + } + + memcpy (sctx, ctx, sizeof (*ctx)); + sctx->func = emitter; + sctx->top = obj; + + ucl_object_emit_streamline_start_container ((struct ucl_emitter_context *)sctx, + obj); + + return (struct ucl_emitter_context *)sctx; +} + +void +ucl_object_emit_streamline_start_container (struct ucl_emitter_context *ctx, + const ucl_object_t *obj) +{ + struct ucl_emitter_context_streamline *sctx = TO_STREAMLINE(ctx); + struct ucl_emitter_streamline_stack *st, *top; + bool print_key = false; + + /* Check top object presence */ + if (sctx->top == NULL) { + sctx->top = obj; + } + + top = sctx->containers; + st = malloc (sizeof (*st)); + if (st != NULL) { + if (top != NULL && !top->is_array) { + print_key = true; + } + st->empty = true; + st->obj = obj; + if (obj != NULL && obj->type == UCL_ARRAY) { + st->is_array = true; + sctx->ops->ucl_emitter_start_array (ctx, obj, print_key); + } + else { + st->is_array = false; + sctx->ops->ucl_emitter_start_object (ctx, obj, print_key); + } + LL_PREPEND (sctx->containers, st); + } +} + +void +ucl_object_emit_streamline_add_object ( + struct ucl_emitter_context *ctx, const ucl_object_t *obj) +{ + struct ucl_emitter_context_streamline *sctx = TO_STREAMLINE(ctx); + bool is_array = false, is_first = false; + + if (sctx->containers != NULL) { + if (sctx->containers->is_array) { + is_array = true; + } + if (sctx->containers->empty) { + is_first = true; + sctx->containers->empty = false; + } + } + + sctx->ops->ucl_emitter_write_elt (ctx, obj, is_first, !is_array); +} + +void +ucl_object_emit_streamline_end_container (struct ucl_emitter_context *ctx) +{ + struct ucl_emitter_context_streamline *sctx = TO_STREAMLINE(ctx); + struct ucl_emitter_streamline_stack *st; + + if (sctx->containers != NULL) { + st = sctx->containers; + + if (st->is_array) { + sctx->ops->ucl_emitter_end_array (ctx, st->obj); + } + else { + sctx->ops->ucl_emitter_end_object (ctx, st->obj); + } + sctx->containers = st->next; + free (st); + } +} + +void +ucl_object_emit_streamline_finish (struct ucl_emitter_context *ctx) +{ + struct ucl_emitter_context_streamline *sctx = TO_STREAMLINE(ctx); + + while (sctx->containers != NULL) { + ucl_object_emit_streamline_end_container (ctx); + } + + free (sctx); +} diff --git a/contrib/libucl/src/ucl_emitter_utils.c b/contrib/libucl/src/ucl_emitter_utils.c new file mode 100644 index 0000000..95ac9a5 --- /dev/null +++ b/contrib/libucl/src/ucl_emitter_utils.c @@ -0,0 +1,493 @@ +/* Copyright (c) 2014, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include "ucl.h" +#include "ucl_internal.h" +#include "ucl_chartable.h" + +#ifdef HAVE_FLOAT_H +#include <float.h> +#endif +#ifdef HAVE_MATH_H +#include <math.h> +#endif + +extern const struct ucl_emitter_operations ucl_standartd_emitter_ops[]; + +static const struct ucl_emitter_context ucl_standard_emitters[] = { + [UCL_EMIT_JSON] = { + .name = "json", + .id = UCL_EMIT_JSON, + .func = NULL, + .ops = &ucl_standartd_emitter_ops[UCL_EMIT_JSON] + }, + [UCL_EMIT_JSON_COMPACT] = { + .name = "json_compact", + .id = UCL_EMIT_JSON_COMPACT, + .func = NULL, + .ops = &ucl_standartd_emitter_ops[UCL_EMIT_JSON_COMPACT] + }, + [UCL_EMIT_CONFIG] = { + .name = "config", + .id = UCL_EMIT_CONFIG, + .func = NULL, + .ops = &ucl_standartd_emitter_ops[UCL_EMIT_CONFIG] + }, + [UCL_EMIT_YAML] = { + .name = "yaml", + .id = UCL_EMIT_YAML, + .func = NULL, + .ops = &ucl_standartd_emitter_ops[UCL_EMIT_YAML] + }, + [UCL_EMIT_MSGPACK] = { + .name = "msgpack", + .id = UCL_EMIT_MSGPACK, + .func = NULL, + .ops = &ucl_standartd_emitter_ops[UCL_EMIT_MSGPACK] + } +}; + +/** + * Get standard emitter context for a specified emit_type + * @param emit_type type of emitter + * @return context or NULL if input is invalid + */ +const struct ucl_emitter_context * +ucl_emit_get_standard_context (enum ucl_emitter emit_type) +{ + if (emit_type >= UCL_EMIT_JSON && emit_type < UCL_EMIT_MAX) { + return &ucl_standard_emitters[emit_type]; + } + + return NULL; +} + +/** + * Serialise string + * @param str string to emit + * @param buf target buffer + */ +void +ucl_elt_string_write_json (const char *str, size_t size, + struct ucl_emitter_context *ctx) +{ + const char *p = str, *c = str; + size_t len = 0; + const struct ucl_emitter_functions *func = ctx->func; + + func->ucl_emitter_append_character ('"', 1, func->ud); + + while (size) { + if (ucl_test_character (*p, UCL_CHARACTER_JSON_UNSAFE)) { + if (len > 0) { + func->ucl_emitter_append_len (c, len, func->ud); + } + switch (*p) { + case '\n': + func->ucl_emitter_append_len ("\\n", 2, func->ud); + break; + case '\r': + func->ucl_emitter_append_len ("\\r", 2, func->ud); + break; + case '\b': + func->ucl_emitter_append_len ("\\b", 2, func->ud); + break; + case '\t': + func->ucl_emitter_append_len ("\\t", 2, func->ud); + break; + case '\f': + func->ucl_emitter_append_len ("\\f", 2, func->ud); + break; + case '\\': + func->ucl_emitter_append_len ("\\\\", 2, func->ud); + break; + case '"': + func->ucl_emitter_append_len ("\\\"", 2, func->ud); + break; + } + len = 0; + c = ++p; + } + else { + p ++; + len ++; + } + size --; + } + if (len > 0) { + func->ucl_emitter_append_len (c, len, func->ud); + } + func->ucl_emitter_append_character ('"', 1, func->ud); +} + +void +ucl_elt_string_write_multiline (const char *str, size_t size, + struct ucl_emitter_context *ctx) +{ + const struct ucl_emitter_functions *func = ctx->func; + + func->ucl_emitter_append_len ("<<EOD\n", sizeof ("<<EOD\n") - 1, func->ud); + func->ucl_emitter_append_len (str, size, func->ud); + func->ucl_emitter_append_len ("\nEOD", sizeof ("\nEOD") - 1, func->ud); +} + +/* + * Generic utstring output + */ +static int +ucl_utstring_append_character (unsigned char c, size_t len, void *ud) +{ + UT_string *buf = ud; + + if (len == 1) { + utstring_append_c (buf, c); + } + else { + utstring_reserve (buf, len + 1); + memset (&buf->d[buf->i], c, len); + buf->i += len; + buf->d[buf->i] = '\0'; + } + + return 0; +} + +static int +ucl_utstring_append_len (const unsigned char *str, size_t len, void *ud) +{ + UT_string *buf = ud; + + utstring_append_len (buf, str, len); + + return 0; +} + +static int +ucl_utstring_append_int (int64_t val, void *ud) +{ + UT_string *buf = ud; + + utstring_printf (buf, "%jd", (intmax_t)val); + return 0; +} + +static int +ucl_utstring_append_double (double val, void *ud) +{ + UT_string *buf = ud; + const double delta = 0.0000001; + + if (val == (double)(int)val) { + utstring_printf (buf, "%.1lf", val); + } + else if (fabs (val - (double)(int)val) < delta) { + /* Write at maximum precision */ + utstring_printf (buf, "%.*lg", DBL_DIG, val); + } + else { + utstring_printf (buf, "%lf", val); + } + + return 0; +} + +/* + * Generic file output + */ +static int +ucl_file_append_character (unsigned char c, size_t len, void *ud) +{ + FILE *fp = ud; + + while (len --) { + fputc (c, fp); + } + + return 0; +} + +static int +ucl_file_append_len (const unsigned char *str, size_t len, void *ud) +{ + FILE *fp = ud; + + fwrite (str, len, 1, fp); + + return 0; +} + +static int +ucl_file_append_int (int64_t val, void *ud) +{ + FILE *fp = ud; + + fprintf (fp, "%jd", (intmax_t)val); + + return 0; +} + +static int +ucl_file_append_double (double val, void *ud) +{ + FILE *fp = ud; + const double delta = 0.0000001; + + if (val == (double)(int)val) { + fprintf (fp, "%.1lf", val); + } + else if (fabs (val - (double)(int)val) < delta) { + /* Write at maximum precision */ + fprintf (fp, "%.*lg", DBL_DIG, val); + } + else { + fprintf (fp, "%lf", val); + } + + return 0; +} + +/* + * Generic file descriptor writing functions + */ +static int +ucl_fd_append_character (unsigned char c, size_t len, void *ud) +{ + int fd = *(int *)ud; + unsigned char *buf; + + if (len == 1) { + return write (fd, &c, 1); + } + else { + buf = malloc (len); + if (buf == NULL) { + /* Fallback */ + while (len --) { + if (write (fd, &c, 1) == -1) { + return -1; + } + } + } + else { + memset (buf, c, len); + if (write (fd, buf, len) == -1) { + free(buf); + return -1; + } + free (buf); + } + } + + return 0; +} + +static int +ucl_fd_append_len (const unsigned char *str, size_t len, void *ud) +{ + int fd = *(int *)ud; + + return write (fd, str, len); +} + +static int +ucl_fd_append_int (int64_t val, void *ud) +{ + int fd = *(int *)ud; + char intbuf[64]; + + snprintf (intbuf, sizeof (intbuf), "%jd", (intmax_t)val); + return write (fd, intbuf, strlen (intbuf)); +} + +static int +ucl_fd_append_double (double val, void *ud) +{ + int fd = *(int *)ud; + const double delta = 0.0000001; + char nbuf[64]; + + if (val == (double)(int)val) { + snprintf (nbuf, sizeof (nbuf), "%.1lf", val); + } + else if (fabs (val - (double)(int)val) < delta) { + /* Write at maximum precision */ + snprintf (nbuf, sizeof (nbuf), "%.*lg", DBL_DIG, val); + } + else { + snprintf (nbuf, sizeof (nbuf), "%lf", val); + } + + return write (fd, nbuf, strlen (nbuf)); +} + +struct ucl_emitter_functions* +ucl_object_emit_memory_funcs (void **pmem) +{ + struct ucl_emitter_functions *f; + UT_string *s; + + f = calloc (1, sizeof (*f)); + + if (f != NULL) { + f->ucl_emitter_append_character = ucl_utstring_append_character; + f->ucl_emitter_append_double = ucl_utstring_append_double; + f->ucl_emitter_append_int = ucl_utstring_append_int; + f->ucl_emitter_append_len = ucl_utstring_append_len; + f->ucl_emitter_free_func = free; + utstring_new (s); + f->ud = s; + *pmem = s->d; + s->pd = pmem; + } + + return f; +} + +struct ucl_emitter_functions* +ucl_object_emit_file_funcs (FILE *fp) +{ + struct ucl_emitter_functions *f; + + f = calloc (1, sizeof (*f)); + + if (f != NULL) { + f->ucl_emitter_append_character = ucl_file_append_character; + f->ucl_emitter_append_double = ucl_file_append_double; + f->ucl_emitter_append_int = ucl_file_append_int; + f->ucl_emitter_append_len = ucl_file_append_len; + f->ucl_emitter_free_func = NULL; + f->ud = fp; + } + + return f; +} + +struct ucl_emitter_functions* +ucl_object_emit_fd_funcs (int fd) +{ + struct ucl_emitter_functions *f; + int *ip; + + f = calloc (1, sizeof (*f)); + + if (f != NULL) { + ip = malloc (sizeof (fd)); + if (ip == NULL) { + free (f); + return NULL; + } + + memcpy (ip, &fd, sizeof (fd)); + f->ucl_emitter_append_character = ucl_fd_append_character; + f->ucl_emitter_append_double = ucl_fd_append_double; + f->ucl_emitter_append_int = ucl_fd_append_int; + f->ucl_emitter_append_len = ucl_fd_append_len; + f->ucl_emitter_free_func = free; + f->ud = ip; + } + + return f; +} + +void +ucl_object_emit_funcs_free (struct ucl_emitter_functions *f) +{ + if (f != NULL) { + if (f->ucl_emitter_free_func != NULL) { + f->ucl_emitter_free_func (f->ud); + } + free (f); + } +} + + +unsigned char * +ucl_object_emit_single_json (const ucl_object_t *obj) +{ + UT_string *buf = NULL; + unsigned char *res = NULL; + + if (obj == NULL) { + return NULL; + } + + utstring_new (buf); + + if (buf != NULL) { + switch (obj->type) { + case UCL_OBJECT: + ucl_utstring_append_len ("object", 6, buf); + break; + case UCL_ARRAY: + ucl_utstring_append_len ("array", 5, buf); + break; + case UCL_INT: + ucl_utstring_append_int (obj->value.iv, buf); + break; + case UCL_FLOAT: + case UCL_TIME: + ucl_utstring_append_double (obj->value.dv, buf); + break; + case UCL_NULL: + ucl_utstring_append_len ("null", 4, buf); + break; + case UCL_BOOLEAN: + if (obj->value.iv) { + ucl_utstring_append_len ("true", 4, buf); + } + else { + ucl_utstring_append_len ("false", 5, buf); + } + break; + case UCL_STRING: + ucl_utstring_append_len (obj->value.sv, obj->len, buf); + break; + case UCL_USERDATA: + ucl_utstring_append_len ("userdata", 8, buf); + break; + } + res = utstring_body (buf); + free (buf); + } + + return res; +} + +#define LONG_STRING_LIMIT 80 + +bool +ucl_maybe_long_string (const ucl_object_t *obj) +{ + if (obj->len > LONG_STRING_LIMIT || (obj->flags & UCL_OBJECT_MULTILINE)) { + /* String is long enough, so search for newline characters in it */ + if (memchr (obj->value.sv, '\n', obj->len) != NULL) { + return true; + } + } + + return false; +} diff --git a/contrib/libucl/src/ucl_hash.c b/contrib/libucl/src/ucl_hash.c new file mode 100644 index 0000000..7de8196 --- /dev/null +++ b/contrib/libucl/src/ucl_hash.c @@ -0,0 +1,486 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include "ucl_internal.h" +#include "ucl_hash.h" +#include "khash.h" +#include "kvec.h" + +#include <time.h> +#include <limits.h> + +struct ucl_hash_elt { + const ucl_object_t *obj; + size_t ar_idx; +}; + +struct ucl_hash_struct { + void *hash; + kvec_t(const ucl_object_t *) ar; + bool caseless; +}; + +static uint64_t +ucl_hash_seed (void) +{ + static uint64_t seed; + + if (seed == 0) { +#ifdef UCL_RANDOM_FUNCTION + seed = UCL_RANDOM_FUNCTION; +#else + /* Not very random but can be useful for our purposes */ + seed = time (NULL); +#endif + } + + return seed; +} + +static const unsigned char lc_map[256] = { + 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, + 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, + 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, + 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f, + 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, + 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f, + 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, + 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f, + 0x40, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, + 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f, + 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, + 0x78, 0x79, 0x7a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f, + 0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, + 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f, + 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, + 0x78, 0x79, 0x7a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f, + 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, + 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f, + 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, + 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f, + 0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, + 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf, + 0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7, + 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf, + 0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, + 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, + 0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, + 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf, + 0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, + 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, + 0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, + 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff +}; + +#if (defined(WORD_BIT) && WORD_BIT == 64) || \ + (defined(__WORDSIZE) && __WORDSIZE == 64) || \ + defined(__x86_64__) || \ + defined(__amd64__) +#define UCL64_BIT_HASH 1 +#endif + +#ifdef UCL64_BIT_HASH +static inline uint32_t +ucl_hash_func (const ucl_object_t *o) +{ + return XXH64 (o->key, o->keylen, ucl_hash_seed ()); +} +#else +static inline uint32_t +ucl_hash_func (const ucl_object_t *o) +{ + return XXH32 (o->key, o->keylen, ucl_hash_seed ()); +} +#endif + +static inline int +ucl_hash_equal (const ucl_object_t *k1, const ucl_object_t *k2) +{ + if (k1->keylen == k2->keylen) { + return strncmp (k1->key, k2->key, k1->keylen) == 0; + } + + return 0; +} + +KHASH_INIT (ucl_hash_node, const ucl_object_t *, struct ucl_hash_elt, 1, + ucl_hash_func, ucl_hash_equal) + +#ifdef UCL64_BIT_HASH +static inline uint32_t +ucl_hash_caseless_func (const ucl_object_t *o) +{ + unsigned len = o->keylen; + unsigned leftover = o->keylen % 4; + unsigned fp, i; + const uint8_t* s = (const uint8_t*)o->key; + union { + struct { + unsigned char c1, c2, c3, c4; + } c; + uint32_t pp; + } u; + XXH64_state_t st; + + fp = len - leftover; + XXH64_reset (&st, ucl_hash_seed ()); + + for (i = 0; i != fp; i += 4) { + u.c.c1 = s[i], u.c.c2 = s[i + 1], u.c.c3 = s[i + 2], u.c.c4 = s[i + 3]; + u.c.c1 = lc_map[u.c.c1]; + u.c.c2 = lc_map[u.c.c2]; + u.c.c3 = lc_map[u.c.c3]; + u.c.c4 = lc_map[u.c.c4]; + XXH64_update (&st, &u.pp, sizeof (u)); + } + + u.pp = 0; + switch (leftover) { + case 3: + u.c.c3 = lc_map[(unsigned char)s[i++]]; + case 2: + u.c.c2 = lc_map[(unsigned char)s[i++]]; + case 1: + u.c.c1 = lc_map[(unsigned char)s[i]]; + XXH64_update (&st, &u.pp, leftover); + break; + } + + return XXH64_digest (&st); +} +#else +static inline uint32_t +ucl_hash_caseless_func (const ucl_object_t *o) +{ + unsigned len = o->keylen; + unsigned leftover = o->keylen % 4; + unsigned fp, i; + const uint8_t* s = (const uint8_t*)o->key; + union { + struct { + unsigned char c1, c2, c3, c4; + } c; + uint32_t pp; + } u; + XXH32_state_t st; + + fp = len - leftover; + XXH32_reset (&st, ucl_hash_seed ()); + + for (i = 0; i != fp; i += 4) { + u.c.c1 = s[i], u.c.c2 = s[i + 1], u.c.c3 = s[i + 2], u.c.c4 = s[i + 3]; + u.c.c1 = lc_map[u.c.c1]; + u.c.c2 = lc_map[u.c.c2]; + u.c.c3 = lc_map[u.c.c3]; + u.c.c4 = lc_map[u.c.c4]; + XXH32_update (&st, &u.pp, sizeof (u)); + } + + u.pp = 0; + switch (leftover) { + case 3: + u.c.c3 = lc_map[(unsigned char)s[i++]]; + case 2: + u.c.c2 = lc_map[(unsigned char)s[i++]]; + case 1: + u.c.c1 = lc_map[(unsigned char)s[i]]; + XXH32_update (&st, &u.pp, leftover); + break; + } + + return XXH32_digest (&st); +} +#endif + +static inline int +ucl_hash_caseless_equal (const ucl_object_t *k1, const ucl_object_t *k2) +{ + if (k1->keylen == k2->keylen) { + return strncasecmp (k1->key, k2->key, k1->keylen) == 0; + } + + return 0; +} + +KHASH_INIT (ucl_hash_caseless_node, const ucl_object_t *, struct ucl_hash_elt, 1, + ucl_hash_caseless_func, ucl_hash_caseless_equal) + +ucl_hash_t* +ucl_hash_create (bool ignore_case) +{ + ucl_hash_t *new; + + new = UCL_ALLOC (sizeof (ucl_hash_t)); + if (new != NULL) { + kv_init (new->ar); + + new->caseless = ignore_case; + if (ignore_case) { + khash_t(ucl_hash_caseless_node) *h = kh_init (ucl_hash_caseless_node); + new->hash = (void *)h; + } + else { + khash_t(ucl_hash_node) *h = kh_init (ucl_hash_node); + new->hash = (void *)h; + } + } + return new; +} + +void ucl_hash_destroy (ucl_hash_t* hashlin, ucl_hash_free_func *func) +{ + const ucl_object_t *cur, *tmp; + + if (hashlin == NULL) { + return; + } + + if (func != NULL) { + /* Iterate over the hash first */ + khash_t(ucl_hash_node) *h = (khash_t(ucl_hash_node) *) + hashlin->hash; + khiter_t k; + + for (k = kh_begin (h); k != kh_end (h); ++k) { + if (kh_exist (h, k)) { + cur = (kh_value (h, k)).obj; + while (cur != NULL) { + tmp = cur->next; + func (__DECONST (ucl_object_t *, cur)); + cur = tmp; + } + } + } + } + + if (hashlin->caseless) { + khash_t(ucl_hash_caseless_node) *h = (khash_t(ucl_hash_caseless_node) *) + hashlin->hash; + kh_destroy (ucl_hash_caseless_node, h); + } + else { + khash_t(ucl_hash_node) *h = (khash_t(ucl_hash_node) *) + hashlin->hash; + kh_destroy (ucl_hash_node, h); + } + + kv_destroy (hashlin->ar); + UCL_FREE (sizeof (*hashlin), hashlin); +} + +void +ucl_hash_insert (ucl_hash_t* hashlin, const ucl_object_t *obj, + const char *key, unsigned keylen) +{ + khiter_t k; + int ret; + struct ucl_hash_elt *elt; + + if (hashlin == NULL) { + return; + } + + if (hashlin->caseless) { + khash_t(ucl_hash_caseless_node) *h = (khash_t(ucl_hash_caseless_node) *) + hashlin->hash; + k = kh_put (ucl_hash_caseless_node, h, obj, &ret); + if (ret > 0) { + elt = &kh_value (h, k); + kv_push (const ucl_object_t *, hashlin->ar, obj); + elt->obj = obj; + elt->ar_idx = kv_size (hashlin->ar) - 1; + } + } + else { + khash_t(ucl_hash_node) *h = (khash_t(ucl_hash_node) *) + hashlin->hash; + k = kh_put (ucl_hash_node, h, obj, &ret); + if (ret > 0) { + elt = &kh_value (h, k); + kv_push (const ucl_object_t *, hashlin->ar, obj); + elt->obj = obj; + elt->ar_idx = kv_size (hashlin->ar) - 1; + } + } +} + +void ucl_hash_replace (ucl_hash_t* hashlin, const ucl_object_t *old, + const ucl_object_t *new) +{ + khiter_t k; + int ret; + struct ucl_hash_elt elt, *pelt; + + if (hashlin == NULL) { + return; + } + + if (hashlin->caseless) { + khash_t(ucl_hash_caseless_node) *h = (khash_t(ucl_hash_caseless_node) *) + hashlin->hash; + k = kh_put (ucl_hash_caseless_node, h, old, &ret); + if (ret == 0) { + elt = kh_value (h, k); + kh_del (ucl_hash_caseless_node, h, k); + k = kh_put (ucl_hash_caseless_node, h, new, &ret); + pelt = &kh_value (h, k); + pelt->obj = new; + pelt->ar_idx = elt.ar_idx; + kv_A (hashlin->ar, elt.ar_idx) = new; + } + } + else { + khash_t(ucl_hash_node) *h = (khash_t(ucl_hash_node) *) + hashlin->hash; + k = kh_put (ucl_hash_node, h, old, &ret); + if (ret == 0) { + elt = kh_value (h, k); + kh_del (ucl_hash_node, h, k); + k = kh_put (ucl_hash_node, h, new, &ret); + pelt = &kh_value (h, k); + pelt->obj = new; + pelt->ar_idx = elt.ar_idx; + kv_A (hashlin->ar, elt.ar_idx) = new; + } + } +} + +struct ucl_hash_real_iter { + const ucl_object_t **cur; + const ucl_object_t **end; +}; + +const void* +ucl_hash_iterate (ucl_hash_t *hashlin, ucl_hash_iter_t *iter) +{ + struct ucl_hash_real_iter *it = (struct ucl_hash_real_iter *)(*iter); + const ucl_object_t *ret = NULL; + + if (hashlin == NULL) { + return NULL; + } + + if (it == NULL) { + it = UCL_ALLOC (sizeof (*it)); + + if (it == NULL) { + return NULL; + } + + it->cur = &hashlin->ar.a[0]; + it->end = it->cur + hashlin->ar.n; + } + + if (it->cur < it->end) { + ret = *it->cur++; + } + else { + UCL_FREE (sizeof (*it), it); + *iter = NULL; + return NULL; + } + + *iter = it; + + return ret; +} + +bool +ucl_hash_iter_has_next (ucl_hash_t *hashlin, ucl_hash_iter_t iter) +{ + struct ucl_hash_real_iter *it = (struct ucl_hash_real_iter *)(iter); + + return it->cur < it->end - 1; +} + + +const ucl_object_t* +ucl_hash_search (ucl_hash_t* hashlin, const char *key, unsigned keylen) +{ + khiter_t k; + const ucl_object_t *ret = NULL; + ucl_object_t search; + struct ucl_hash_elt *elt; + + search.key = key; + search.keylen = keylen; + + if (hashlin == NULL) { + return NULL; + } + + if (hashlin->caseless) { + khash_t(ucl_hash_caseless_node) *h = (khash_t(ucl_hash_caseless_node) *) + hashlin->hash; + + k = kh_get (ucl_hash_caseless_node, h, &search); + if (k != kh_end (h)) { + elt = &kh_value (h, k); + ret = elt->obj; + } + } + else { + khash_t(ucl_hash_node) *h = (khash_t(ucl_hash_node) *) + hashlin->hash; + k = kh_get (ucl_hash_node, h, &search); + if (k != kh_end (h)) { + elt = &kh_value (h, k); + ret = elt->obj; + } + } + + return ret; +} + +void +ucl_hash_delete (ucl_hash_t* hashlin, const ucl_object_t *obj) +{ + khiter_t k; + struct ucl_hash_elt *elt; + + if (hashlin == NULL) { + return; + } + + if (hashlin->caseless) { + khash_t(ucl_hash_caseless_node) *h = (khash_t(ucl_hash_caseless_node) *) + hashlin->hash; + + k = kh_get (ucl_hash_caseless_node, h, obj); + if (k != kh_end (h)) { + elt = &kh_value (h, k); + kv_del (const ucl_object_t *, hashlin->ar, elt->ar_idx); + kh_del (ucl_hash_caseless_node, h, k); + } + } + else { + khash_t(ucl_hash_node) *h = (khash_t(ucl_hash_node) *) + hashlin->hash; + k = kh_get (ucl_hash_node, h, obj); + if (k != kh_end (h)) { + elt = &kh_value (h, k); + kv_del (const ucl_object_t *, hashlin->ar, elt->ar_idx); + kh_del (ucl_hash_node, h, k); + } + } +} diff --git a/contrib/libucl/src/ucl_hash.h b/contrib/libucl/src/ucl_hash.h new file mode 100644 index 0000000..64c83ea --- /dev/null +++ b/contrib/libucl/src/ucl_hash.h @@ -0,0 +1,93 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef __UCL_HASH_H +#define __UCL_HASH_H + +#include "ucl.h" + +/******************************************************************************/ + +struct ucl_hash_node_s; +typedef struct ucl_hash_node_s ucl_hash_node_t; + +typedef int ucl_hash_cmp_func (const void* void_a, const void* void_b); +typedef void ucl_hash_free_func (void *ptr); +typedef void* ucl_hash_iter_t; + + +/** + * Linear chained hashtable. + */ +struct ucl_hash_struct; +typedef struct ucl_hash_struct ucl_hash_t; + + +/** + * Initializes the hashtable. + */ +ucl_hash_t* ucl_hash_create (bool ignore_case); + +/** + * Deinitializes the hashtable. + */ +void ucl_hash_destroy (ucl_hash_t* hashlin, ucl_hash_free_func *func); + +/** + * Inserts an element in the the hashtable. + */ +void ucl_hash_insert (ucl_hash_t* hashlin, const ucl_object_t *obj, const char *key, + unsigned keylen); + +/** + * Replace element in the hash + */ +void ucl_hash_replace (ucl_hash_t* hashlin, const ucl_object_t *old, + const ucl_object_t *new); + +/** + * Delete an element from the the hashtable. + */ +void ucl_hash_delete (ucl_hash_t* hashlin, const ucl_object_t *obj); + +/** + * Searches an element in the hashtable. + */ +const ucl_object_t* ucl_hash_search (ucl_hash_t* hashlin, const char *key, + unsigned keylen); + + +/** + * Iterate over hash table + * @param hashlin hash + * @param iter iterator (must be NULL on first iteration) + * @return the next object + */ +const void* ucl_hash_iterate (ucl_hash_t *hashlin, ucl_hash_iter_t *iter); + +/** + * Check whether an iterator has next element + */ +bool ucl_hash_iter_has_next (ucl_hash_t *hashlin, ucl_hash_iter_t iter); + +#endif diff --git a/contrib/libucl/src/ucl_internal.h b/contrib/libucl/src/ucl_internal.h new file mode 100644 index 0000000..31d6b13 --- /dev/null +++ b/contrib/libucl/src/ucl_internal.h @@ -0,0 +1,542 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef UCL_INTERNAL_H_ +#define UCL_INTERNAL_H_ + +#ifdef HAVE_CONFIG_H +#include "config.h" +#else +/* Help embedded builds */ +#define HAVE_SYS_TYPES_H +#define HAVE_SYS_MMAN_H +#define HAVE_SYS_STAT_H +#define HAVE_SYS_PARAM_H +#define HAVE_LIMITS_H +#define HAVE_FCNTL_H +#define HAVE_ERRNO_H +#define HAVE_UNISTD_H +#define HAVE_CTYPE_H +#define HAVE_STDIO_H +#define HAVE_STRING_H +#define HAVE_FLOAT_H +#define HAVE_LIBGEN_H +#define HAVE_MATH_H +#define HAVE_STDBOOL_H +#define HAVE_STDINT_H +#define HAVE_STDARG_H +#ifndef _WIN32 +# define HAVE_REGEX_H +#endif +#endif + +#ifdef HAVE_SYS_TYPES_H +#include <sys/types.h> +#endif + +#ifdef HAVE_SYS_MMAN_H +# ifndef _WIN32 +# include <sys/mman.h> +# endif +#endif +#ifdef HAVE_SYS_STAT_H +#include <sys/stat.h> +#endif +#ifdef HAVE_SYS_PARAM_H +#include <sys/param.h> +#endif + +#ifdef HAVE_LIMITS_H +#include <limits.h> +#endif +#ifdef HAVE_FCNTL_H +#include <fcntl.h> +#endif +#ifdef HAVE_ERRNO_H +#include <errno.h> +#endif +#ifdef HAVE_UNISTD_H +#include <unistd.h> +#endif +#ifdef HAVE_CTYPE_H +#include <ctype.h> +#endif +#ifdef HAVE_STDIO_H +#include <stdio.h> +#endif +#ifdef HAVE_STRING_H +#include <string.h> +#endif + +#include "utlist.h" +#include "utstring.h" +#include "uthash.h" +#include "ucl.h" +#include "ucl_hash.h" +#include "xxhash.h" + +#ifdef HAVE_OPENSSL +#include <openssl/evp.h> +#endif + +#ifndef __DECONST +#define __DECONST(type, var) ((type)(uintptr_t)(const void *)(var)) +#endif + +/** + * @file rcl_internal.h + * Internal structures and functions of UCL library + */ + +#define UCL_MAX_RECURSION 16 +#define UCL_TRASH_KEY 0 +#define UCL_TRASH_VALUE 1 + +enum ucl_parser_state { + UCL_STATE_INIT = 0, + UCL_STATE_OBJECT, + UCL_STATE_ARRAY, + UCL_STATE_KEY, + UCL_STATE_VALUE, + UCL_STATE_AFTER_VALUE, + UCL_STATE_ARRAY_VALUE, + UCL_STATE_SCOMMENT, + UCL_STATE_MCOMMENT, + UCL_STATE_MACRO_NAME, + UCL_STATE_MACRO, + UCL_STATE_ERROR +}; + +enum ucl_character_type { + UCL_CHARACTER_DENIED = 0, + UCL_CHARACTER_KEY = 1, + UCL_CHARACTER_KEY_START = 1 << 1, + UCL_CHARACTER_WHITESPACE = 1 << 2, + UCL_CHARACTER_WHITESPACE_UNSAFE = 1 << 3, + UCL_CHARACTER_VALUE_END = 1 << 4, + UCL_CHARACTER_VALUE_STR = 1 << 5, + UCL_CHARACTER_VALUE_DIGIT = 1 << 6, + UCL_CHARACTER_VALUE_DIGIT_START = 1 << 7, + UCL_CHARACTER_ESCAPE = 1 << 8, + UCL_CHARACTER_KEY_SEP = 1 << 9, + UCL_CHARACTER_JSON_UNSAFE = 1 << 10, + UCL_CHARACTER_UCL_UNSAFE = 1 << 11 +}; + +struct ucl_macro { + char *name; + union { + ucl_macro_handler handler; + ucl_context_macro_handler context_handler; + } h; + void* ud; + bool is_context; + UT_hash_handle hh; +}; + +struct ucl_stack { + ucl_object_t *obj; + struct ucl_stack *next; + uint64_t level; +}; + +struct ucl_chunk { + const unsigned char *begin; + const unsigned char *end; + const unsigned char *pos; + size_t remain; + unsigned int line; + unsigned int column; + unsigned priority; + enum ucl_duplicate_strategy strategy; + enum ucl_parse_type parse_type; + struct ucl_chunk *next; +}; + +#ifdef HAVE_OPENSSL +struct ucl_pubkey { + EVP_PKEY *key; + struct ucl_pubkey *next; +}; +#else +struct ucl_pubkey { + struct ucl_pubkey *next; +}; +#endif + +struct ucl_variable { + char *var; + char *value; + size_t var_len; + size_t value_len; + struct ucl_variable *prev, *next; +}; + +struct ucl_parser { + enum ucl_parser_state state; + enum ucl_parser_state prev_state; + unsigned int recursion; + int flags; + unsigned default_priority; + int err_code; + ucl_object_t *top_obj; + ucl_object_t *cur_obj; + ucl_object_t *trash_objs; + ucl_object_t *includepaths; + char *cur_file; + struct ucl_macro *macroes; + struct ucl_stack *stack; + struct ucl_chunk *chunks; + struct ucl_pubkey *keys; + struct ucl_variable *variables; + ucl_variable_handler var_handler; + void *var_data; + UT_string *err; +}; + +struct ucl_object_userdata { + ucl_object_t obj; + ucl_userdata_dtor dtor; + ucl_userdata_emitter emitter; +}; + +/** + * Unescape json string inplace + * @param str + */ +size_t ucl_unescape_json_string (char *str, size_t len); + +/** + * Handle include macro + * @param data include data + * @param len length of data + * @param args UCL object representing arguments to the macro + * @param ud user data + * @return + */ +bool ucl_include_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, void* ud); + +/** + * Handle tryinclude macro + * @param data include data + * @param len length of data + * @param args UCL object representing arguments to the macro + * @param ud user data + * @return + */ +bool ucl_try_include_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, void* ud); + +/** + * Handle includes macro + * @param data include data + * @param len length of data + * @param args UCL object representing arguments to the macro + * @param ud user data + * @return + */ +bool ucl_includes_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, void* ud); + +/** + * Handle priority macro + * @param data include data + * @param len length of data + * @param args UCL object representing arguments to the macro + * @param ud user data + * @return + */ +bool ucl_priority_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, void* ud); + +/** + * Handle load macro + * @param data include data + * @param len length of data + * @param args UCL object representing arguments to the macro + * @param ud user data + * @return + */ +bool ucl_load_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, void* ud); +/** + * Handle inherit macro + * @param data include data + * @param len length of data + * @param args UCL object representing arguments to the macro + * @param ctx the current context object + * @param ud user data + * @return + */ +bool ucl_inherit_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, const ucl_object_t *ctx, void* ud); + +size_t ucl_strlcpy (char *dst, const char *src, size_t siz); +size_t ucl_strlcpy_unsafe (char *dst, const char *src, size_t siz); +size_t ucl_strlcpy_tolower (char *dst, const char *src, size_t siz); + +char *ucl_strnstr (const char *s, const char *find, int len); +char *ucl_strncasestr (const char *s, const char *find, int len); + +#ifdef __GNUC__ +static inline void +ucl_create_err (UT_string **err, const char *fmt, ...) +__attribute__ (( format( printf, 2, 3) )); +#endif + +#undef UCL_FATAL_ERRORS + +static inline void +ucl_create_err (UT_string **err, const char *fmt, ...) +{ + if (*err == NULL) { + utstring_new (*err); + va_list ap; + va_start (ap, fmt); + utstring_printf_va (*err, fmt, ap); + va_end (ap); + } + +#ifdef UCL_FATAL_ERRORS + assert (0); +#endif +} + +/** + * Check whether a given string contains a boolean value + * @param obj object to set + * @param start start of a string + * @param len length of a string + * @return true if a string is a boolean value + */ +static inline bool +ucl_maybe_parse_boolean (ucl_object_t *obj, const unsigned char *start, size_t len) +{ + const char *p = (const char *)start; + bool ret = false, val = false; + + if (len == 5) { + if ((p[0] == 'f' || p[0] == 'F') && strncasecmp (p, "false", 5) == 0) { + ret = true; + val = false; + } + } + else if (len == 4) { + if ((p[0] == 't' || p[0] == 'T') && strncasecmp (p, "true", 4) == 0) { + ret = true; + val = true; + } + } + else if (len == 3) { + if ((p[0] == 'y' || p[0] == 'Y') && strncasecmp (p, "yes", 3) == 0) { + ret = true; + val = true; + } + else if ((p[0] == 'o' || p[0] == 'O') && strncasecmp (p, "off", 3) == 0) { + ret = true; + val = false; + } + } + else if (len == 2) { + if ((p[0] == 'n' || p[0] == 'N') && strncasecmp (p, "no", 2) == 0) { + ret = true; + val = false; + } + else if ((p[0] == 'o' || p[0] == 'O') && strncasecmp (p, "on", 2) == 0) { + ret = true; + val = true; + } + } + + if (ret && obj != NULL) { + obj->type = UCL_BOOLEAN; + obj->value.iv = val; + } + + return ret; +} + +/** + * Check numeric string + * @param obj object to set if a string is numeric + * @param start start of string + * @param end end of string + * @param pos position where parsing has stopped + * @param allow_double allow parsing of floating point values + * @return 0 if string is numeric and error code (EINVAL or ERANGE) in case of conversion error + */ +int ucl_maybe_parse_number (ucl_object_t *obj, + const char *start, const char *end, const char **pos, + bool allow_double, bool number_bytes, bool allow_time); + + +static inline const ucl_object_t * +ucl_hash_search_obj (ucl_hash_t* hashlin, ucl_object_t *obj) +{ + return (const ucl_object_t *)ucl_hash_search (hashlin, obj->key, obj->keylen); +} + +static inline ucl_hash_t * ucl_hash_insert_object (ucl_hash_t *hashlin, + const ucl_object_t *obj, + bool ignore_case) UCL_WARN_UNUSED_RESULT; + +static inline ucl_hash_t * +ucl_hash_insert_object (ucl_hash_t *hashlin, + const ucl_object_t *obj, + bool ignore_case) +{ + if (hashlin == NULL) { + hashlin = ucl_hash_create (ignore_case); + } + ucl_hash_insert (hashlin, obj, obj->key, obj->keylen); + + return hashlin; +} + +/** + * Get standard emitter context for a specified emit_type + * @param emit_type type of emitter + * @return context or NULL if input is invalid + */ +const struct ucl_emitter_context * +ucl_emit_get_standard_context (enum ucl_emitter emit_type); + +/** + * Serialize string as JSON string + * @param str string to emit + * @param buf target buffer + */ +void ucl_elt_string_write_json (const char *str, size_t size, + struct ucl_emitter_context *ctx); + +/** + * Write multiline string using `EOD` as string terminator + * @param str + * @param size + * @param ctx + */ +void ucl_elt_string_write_multiline (const char *str, size_t size, + struct ucl_emitter_context *ctx); + +/** + * Emit a single object to string + * @param obj + * @return + */ +unsigned char * ucl_object_emit_single_json (const ucl_object_t *obj); + +/** + * Check whether a specified string is long and should be likely printed in + * multiline mode + * @param obj + * @return + */ +bool ucl_maybe_long_string (const ucl_object_t *obj); + +/** + * Print integer to the msgpack output + * @param ctx + * @param val + */ +void ucl_emitter_print_int_msgpack (struct ucl_emitter_context *ctx, + int64_t val); +/** + * Print integer to the msgpack output + * @param ctx + * @param val + */ +void ucl_emitter_print_double_msgpack (struct ucl_emitter_context *ctx, + double val); +/** + * Print double to the msgpack output + * @param ctx + * @param val + */ +void ucl_emitter_print_bool_msgpack (struct ucl_emitter_context *ctx, + bool val); +/** + * Print string to the msgpack output + * @param ctx + * @param s + * @param len + */ +void ucl_emitter_print_string_msgpack (struct ucl_emitter_context *ctx, + const char *s, size_t len); + +/** + * Print binary string to the msgpack output + * @param ctx + * @param s + * @param len + */ +void ucl_emitter_print_binary_string_msgpack (struct ucl_emitter_context *ctx, + const char *s, size_t len); + +/** + * Print array preamble for msgpack + * @param ctx + * @param len + */ +void ucl_emitter_print_array_msgpack (struct ucl_emitter_context *ctx, + size_t len); + +/** + * Print object preamble for msgpack + * @param ctx + * @param len + */ +void ucl_emitter_print_object_msgpack (struct ucl_emitter_context *ctx, + size_t len); +/** + * Print NULL to the msgpack output + * @param ctx + */ +void ucl_emitter_print_null_msgpack (struct ucl_emitter_context *ctx); +/** + * Print object's key if needed to the msgpack output + * @param print_key + * @param ctx + * @param obj + */ +void ucl_emitter_print_key_msgpack (bool print_key, + struct ucl_emitter_context *ctx, + const ucl_object_t *obj); + +/** + * Add new element to an object using the current merge strategy and priority + * @param parser + * @param nobj + * @return + */ +bool ucl_parser_process_object_element (struct ucl_parser *parser, + ucl_object_t *nobj); + +/** + * Parse msgpack chunk + * @param parser + * @return + */ +bool ucl_parse_msgpack (struct ucl_parser *parser); + +#endif /* UCL_INTERNAL_H_ */ diff --git a/contrib/libucl/src/ucl_msgpack.c b/contrib/libucl/src/ucl_msgpack.c new file mode 100644 index 0000000..e8ebfba --- /dev/null +++ b/contrib/libucl/src/ucl_msgpack.c @@ -0,0 +1,1600 @@ +/* + * Copyright (c) 2015, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY AUTHOR ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include "ucl.h" +#include "ucl_internal.h" + +#ifdef HAVE_ENDIAN_H +#include <endian.h> +#elif defined(HAVE_SYS_ENDIAN_H) +#include <sys/endian.h> +#elif defined(HAVE_MACHINE_ENDIAN_H) +#include <machine/endian.h> +#endif + +#if !defined(__LITTLE_ENDIAN__) && !defined(__BIG_ENDIAN__) + #if __BYTE_ORDER == __LITTLE_ENDIAN + #define __LITTLE_ENDIAN__ + #elif __BYTE_ORDER == __BIG_ENDIAN + #define __BIG_ENDIAN__ + #elif _WIN32 + #define __LITTLE_ENDIAN__ + #endif +#endif + +#define SWAP_LE_BE16(val) ((uint16_t) ( \ + (uint16_t) ((uint16_t) (val) >> 8) | \ + (uint16_t) ((uint16_t) (val) << 8))) + +#if defined(__clang__) || (defined(__GNUC__) && __GNUC__ >= 4 && defined (__GNUC_MINOR__) && __GNUC_MINOR__ >= 3) +# define SWAP_LE_BE32(val) ((uint32_t)__builtin_bswap32 ((uint32_t)(val))) +# define SWAP_LE_BE64(val) ((uint64_t)__builtin_bswap64 ((uint64_t)(val))) +#else + #define SWAP_LE_BE32(val) ((uint32_t)( \ + (((uint32_t)(val) & (uint32_t)0x000000ffU) << 24) | \ + (((uint32_t)(val) & (uint32_t)0x0000ff00U) << 8) | \ + (((uint32_t)(val) & (uint32_t)0x00ff0000U) >> 8) | \ + (((uint32_t)(val) & (uint32_t)0xff000000U) >> 24))) + + #define SWAP_LE_BE64(val) ((uint64_t)( \ + (((uint64_t)(val) & \ + (uint64_t)(0x00000000000000ffULL)) << 56) | \ + (((uint64_t)(val) & \ + (uint64_t)(0x000000000000ff00ULL)) << 40) | \ + (((uint64_t)(val) & \ + (uint64_t)(0x0000000000ff0000ULL)) << 24) | \ + (((uint64_t)(val) & \ + (uint64_t) (0x00000000ff000000ULL)) << 8) | \ + (((uint64_t)(val) & \ + (uint64_t)(0x000000ff00000000ULL)) >> 8) | \ + (((uint64_t)(val) & \ + (uint64_t)(0x0000ff0000000000ULL)) >> 24) | \ + (((uint64_t)(val) & \ + (uint64_t)(0x00ff000000000000ULL)) >> 40) | \ + (((uint64_t)(val) & \ + (uint64_t)(0xff00000000000000ULL)) >> 56))) +#endif + +#ifdef __LITTLE_ENDIAN__ +#define TO_BE16 SWAP_LE_BE16 +#define TO_BE32 SWAP_LE_BE32 +#define TO_BE64 SWAP_LE_BE64 +#define FROM_BE16 SWAP_LE_BE16 +#define FROM_BE32 SWAP_LE_BE32 +#define FROM_BE64 SWAP_LE_BE64 +#else +#define TO_BE16(val) (uint16_t)(val) +#define TO_BE32(val) (uint32_t)(val) +#define TO_BE64(val) (uint64_t)(val) +#define FROM_BE16(val) (uint16_t)(val) +#define FROM_BE32(val) (uint32_t)(val) +#define FROM_BE64(val) (uint64_t)(val) +#endif + +void +ucl_emitter_print_int_msgpack (struct ucl_emitter_context *ctx, int64_t val) +{ + const struct ucl_emitter_functions *func = ctx->func; + unsigned char buf[sizeof(uint64_t) + 1]; + const unsigned char mask_positive = 0x7f, mask_negative = 0xe0, + uint8_ch = 0xcc, uint16_ch = 0xcd, uint32_ch = 0xce, uint64_ch = 0xcf, + int8_ch = 0xd0, int16_ch = 0xd1, int32_ch = 0xd2, int64_ch = 0xd3; + unsigned len; + + if (val >= 0) { + if (val <= 0x7f) { + /* Fixed num 7 bits */ + len = 1; + buf[0] = mask_positive & val; + } + else if (val <= 0xff) { + len = 2; + buf[0] = uint8_ch; + buf[1] = val & 0xff; + } + else if (val <= 0xffff) { + uint16_t v = TO_BE16 (val); + + len = 3; + buf[0] = uint16_ch; + memcpy (&buf[1], &v, sizeof (v)); + } + else if (val <= 0xffffffff) { + uint32_t v = TO_BE32 (val); + + len = 5; + buf[0] = uint32_ch; + memcpy (&buf[1], &v, sizeof (v)); + } + else { + uint64_t v = TO_BE64 (val); + + len = 9; + buf[0] = uint64_ch; + memcpy (&buf[1], &v, sizeof (v)); + } + } + else { + uint64_t uval; + /* Bithack abs */ + uval = ((val ^ (val >> 63)) - (val >> 63)); + + if (val > -(1 << 5)) { + len = 1; + buf[0] = (mask_negative | uval) & 0xff; + } + else if (uval <= 0xff) { + len = 2; + buf[0] = int8_ch; + buf[1] = (unsigned char)val; + } + else if (uval <= 0xffff) { + uint16_t v = TO_BE16 (val); + + len = 3; + buf[0] = int16_ch; + memcpy (&buf[1], &v, sizeof (v)); + } + else if (uval <= 0xffffffff) { + uint32_t v = TO_BE32 (val); + + len = 5; + buf[0] = int32_ch; + memcpy (&buf[1], &v, sizeof (v)); + } + else { + uint64_t v = TO_BE64 (val); + + len = 9; + buf[0] = int64_ch; + memcpy (&buf[1], &v, sizeof (v)); + } + } + + func->ucl_emitter_append_len (buf, len, func->ud); +} + +void +ucl_emitter_print_double_msgpack (struct ucl_emitter_context *ctx, double val) +{ + const struct ucl_emitter_functions *func = ctx->func; + union { + double d; + uint64_t i; + } u; + const unsigned char dbl_ch = 0xcb; + unsigned char buf[sizeof(double) + 1]; + + /* Convert to big endian */ + u.d = val; + u.i = TO_BE64 (u.i); + + buf[0] = dbl_ch; + memcpy (&buf[1], &u.d, sizeof (double)); + func->ucl_emitter_append_len (buf, sizeof (buf), func->ud); +} + +void +ucl_emitter_print_bool_msgpack (struct ucl_emitter_context *ctx, bool val) +{ + const struct ucl_emitter_functions *func = ctx->func; + const unsigned char true_ch = 0xc3, false_ch = 0xc2; + + func->ucl_emitter_append_character (val ? true_ch : false_ch, 1, func->ud); +} + +void +ucl_emitter_print_string_msgpack (struct ucl_emitter_context *ctx, + const char *s, size_t len) +{ + const struct ucl_emitter_functions *func = ctx->func; + const unsigned char fix_mask = 0xA0, l8_ch = 0xd9, l16_ch = 0xda, l32_ch = 0xdb; + unsigned char buf[5]; + unsigned blen; + + if (len <= 0x1F) { + blen = 1; + buf[0] = (len | fix_mask) & 0xff; + } + else if (len <= 0xff) { + blen = 2; + buf[0] = l8_ch; + buf[1] = len & 0xff; + } + else if (len <= 0xffff) { + uint16_t bl = TO_BE16 (len); + + blen = 3; + buf[0] = l16_ch; + memcpy (&buf[1], &bl, sizeof (bl)); + } + else { + uint32_t bl = TO_BE32 (len); + + blen = 5; + buf[0] = l32_ch; + memcpy (&buf[1], &bl, sizeof (bl)); + } + + func->ucl_emitter_append_len (buf, blen, func->ud); + func->ucl_emitter_append_len (s, len, func->ud); +} + +void +ucl_emitter_print_binary_string_msgpack (struct ucl_emitter_context *ctx, + const char *s, size_t len) +{ + const struct ucl_emitter_functions *func = ctx->func; + const unsigned char l8_ch = 0xc4, l16_ch = 0xc5, l32_ch = 0xc6; + unsigned char buf[5]; + unsigned blen; + + if (len <= 0xff) { + blen = 2; + buf[0] = l8_ch; + buf[1] = len & 0xff; + } + else if (len <= 0xffff) { + uint16_t bl = TO_BE16 (len); + + blen = 3; + buf[0] = l16_ch; + memcpy (&buf[1], &bl, sizeof (bl)); + } + else { + uint32_t bl = TO_BE32 (len); + + blen = 5; + buf[0] = l32_ch; + memcpy (&buf[1], &bl, sizeof (bl)); + } + + func->ucl_emitter_append_len (buf, blen, func->ud); + func->ucl_emitter_append_len (s, len, func->ud); +} + +void +ucl_emitter_print_null_msgpack (struct ucl_emitter_context *ctx) +{ + const struct ucl_emitter_functions *func = ctx->func; + const unsigned char nil = 0xc0; + + func->ucl_emitter_append_character (nil, 1, func->ud); +} + +void +ucl_emitter_print_key_msgpack (bool print_key, struct ucl_emitter_context *ctx, + const ucl_object_t *obj) +{ + if (print_key) { + ucl_emitter_print_string_msgpack (ctx, obj->key, obj->keylen); + } +} + +void +ucl_emitter_print_array_msgpack (struct ucl_emitter_context *ctx, size_t len) +{ + const struct ucl_emitter_functions *func = ctx->func; + const unsigned char fix_mask = 0x90, l16_ch = 0xdc, l32_ch = 0xdd; + unsigned char buf[5]; + unsigned blen; + + if (len <= 0xF) { + blen = 1; + buf[0] = (len | fix_mask) & 0xff; + } + else if (len <= 0xffff) { + uint16_t bl = TO_BE16 (len); + + blen = 3; + buf[0] = l16_ch; + memcpy (&buf[1], &bl, sizeof (bl)); + } + else { + uint32_t bl = TO_BE32 (len); + + blen = 5; + buf[0] = l32_ch; + memcpy (&buf[1], &bl, sizeof (bl)); + } + + func->ucl_emitter_append_len (buf, blen, func->ud); +} + +void +ucl_emitter_print_object_msgpack (struct ucl_emitter_context *ctx, size_t len) +{ + const struct ucl_emitter_functions *func = ctx->func; + const unsigned char fix_mask = 0x80, l16_ch = 0xde, l32_ch = 0xdf; + unsigned char buf[5]; + unsigned blen; + + if (len <= 0xF) { + blen = 1; + buf[0] = (len | fix_mask) & 0xff; + } + else if (len <= 0xffff) { + uint16_t bl = TO_BE16 (len); + + blen = 3; + buf[0] = l16_ch; + memcpy (&buf[1], &bl, sizeof (bl)); + } + else { + uint32_t bl = TO_BE32 (len); + + blen = 5; + buf[0] = l32_ch; + memcpy (&buf[1], &bl, sizeof (bl)); + } + + func->ucl_emitter_append_len (buf, blen, func->ud); +} + + +enum ucl_msgpack_format { + msgpack_positive_fixint = 0, + msgpack_fixmap, + msgpack_fixarray, + msgpack_fixstr, + msgpack_nil, + msgpack_false, + msgpack_true, + msgpack_bin8, + msgpack_bin16, + msgpack_bin32, + msgpack_ext8, + msgpack_ext16, + msgpack_ext32, + msgpack_float32, + msgpack_float64, + msgpack_uint8, + msgpack_uint16, + msgpack_uint32, + msgpack_uint64, + msgpack_int8, + msgpack_int16, + msgpack_int32, + msgpack_int64, + msgpack_fixext1, + msgpack_fixext2, + msgpack_fixext4, + msgpack_fixext8, + msgpack_fixext16, + msgpack_str8, + msgpack_str16, + msgpack_str32, + msgpack_array16, + msgpack_array32, + msgpack_map16, + msgpack_map32, + msgpack_negative_fixint, + msgpack_invalid +}; + +typedef ssize_t (*ucl_msgpack_parse_function)(struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain); + +static ssize_t ucl_msgpack_parse_map (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain); +static ssize_t ucl_msgpack_parse_array (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain); +static ssize_t ucl_msgpack_parse_string (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain); +static ssize_t ucl_msgpack_parse_int (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain); +static ssize_t ucl_msgpack_parse_float (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain); +static ssize_t ucl_msgpack_parse_bool (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain); +static ssize_t ucl_msgpack_parse_null (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain); +static ssize_t ucl_msgpack_parse_ignore (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain); + +#define MSGPACK_FLAG_FIXED (1 << 0) +#define MSGPACK_FLAG_CONTAINER (1 << 1) +#define MSGPACK_FLAG_TYPEVALUE (1 << 2) +#define MSGPACK_FLAG_EXT (1 << 3) +#define MSGPACK_FLAG_ASSOC (1 << 4) +#define MSGPACK_FLAG_KEY (1 << 5) +#define MSGPACK_CONTAINER_BIT (1ULL << 62) + +/* + * Search tree packed in array + */ +struct ucl_msgpack_parser { + uint8_t prefix; /* Prefix byte */ + uint8_t prefixlen; /* Length of prefix in bits */ + uint8_t fmt; /* The desired format */ + uint8_t len; /* Length of the object + (either length bytes + or length of value in case + of fixed objects */ + uint8_t flags; /* Flags of the specified type */ + ucl_msgpack_parse_function func; /* Parser function */ +} parsers[] = { + { + 0xa0, + 3, + msgpack_fixstr, + 0, + MSGPACK_FLAG_FIXED|MSGPACK_FLAG_KEY, + ucl_msgpack_parse_string + }, + { + 0x0, + 1, + msgpack_positive_fixint, + 0, + MSGPACK_FLAG_FIXED|MSGPACK_FLAG_TYPEVALUE, + ucl_msgpack_parse_int + }, + { + 0xe0, + 3, + msgpack_negative_fixint, + 0, + MSGPACK_FLAG_FIXED|MSGPACK_FLAG_TYPEVALUE, + ucl_msgpack_parse_int + }, + { + 0x80, + 4, + msgpack_fixmap, + 0, + MSGPACK_FLAG_FIXED|MSGPACK_FLAG_CONTAINER|MSGPACK_FLAG_ASSOC, + ucl_msgpack_parse_map + }, + { + 0x90, + 4, + msgpack_fixarray, + 0, + MSGPACK_FLAG_FIXED|MSGPACK_FLAG_CONTAINER, + ucl_msgpack_parse_array + }, + { + 0xd9, + 8, + msgpack_str8, + 1, + MSGPACK_FLAG_KEY, + ucl_msgpack_parse_string + }, + { + 0xc4, + 8, + msgpack_bin8, + 1, + MSGPACK_FLAG_KEY, + ucl_msgpack_parse_string + }, + { + 0xcf, + 8, + msgpack_uint64, + 8, + MSGPACK_FLAG_FIXED, + ucl_msgpack_parse_int + }, + { + 0xd3, + 8, + msgpack_int64, + 8, + MSGPACK_FLAG_FIXED, + ucl_msgpack_parse_int + }, + { + 0xce, + 8, + msgpack_uint32, + 4, + MSGPACK_FLAG_FIXED, + ucl_msgpack_parse_int + }, + { + 0xd2, + 8, + msgpack_int32, + 4, + MSGPACK_FLAG_FIXED, + ucl_msgpack_parse_int + }, + { + 0xcb, + 8, + msgpack_float64, + 8, + MSGPACK_FLAG_FIXED, + ucl_msgpack_parse_float + }, + { + 0xca, + 8, + msgpack_float32, + 4, + MSGPACK_FLAG_FIXED, + ucl_msgpack_parse_float + }, + { + 0xc2, + 8, + msgpack_false, + 1, + MSGPACK_FLAG_FIXED | MSGPACK_FLAG_TYPEVALUE, + ucl_msgpack_parse_bool + }, + { + 0xc3, + 8, + msgpack_true, + 1, + MSGPACK_FLAG_FIXED | MSGPACK_FLAG_TYPEVALUE, + ucl_msgpack_parse_bool + }, + { + 0xcc, + 8, + msgpack_uint8, + 1, + MSGPACK_FLAG_FIXED, + ucl_msgpack_parse_int + }, + { + 0xcd, + 8, + msgpack_uint16, + 2, + MSGPACK_FLAG_FIXED, + ucl_msgpack_parse_int + }, + { + 0xd0, + 8, + msgpack_int8, + 1, + MSGPACK_FLAG_FIXED, + ucl_msgpack_parse_int + }, + { + 0xd1, + 8, + msgpack_int16, + 2, + MSGPACK_FLAG_FIXED, + ucl_msgpack_parse_int + }, + { + 0xc0, + 8, + msgpack_nil, + 0, + MSGPACK_FLAG_FIXED | MSGPACK_FLAG_TYPEVALUE, + ucl_msgpack_parse_null + }, + { + 0xda, + 8, + msgpack_str16, + 2, + MSGPACK_FLAG_KEY, + ucl_msgpack_parse_string + }, + { + 0xdb, + 8, + msgpack_str32, + 4, + MSGPACK_FLAG_KEY, + ucl_msgpack_parse_string + }, + { + 0xc5, + 8, + msgpack_bin16, + 2, + MSGPACK_FLAG_KEY, + ucl_msgpack_parse_string + }, + { + 0xc6, + 8, + msgpack_bin32, + 4, + MSGPACK_FLAG_KEY, + ucl_msgpack_parse_string + }, + { + 0xdc, + 8, + msgpack_array16, + 2, + MSGPACK_FLAG_CONTAINER, + ucl_msgpack_parse_array + }, + { + 0xdd, + 8, + msgpack_array32, + 4, + MSGPACK_FLAG_CONTAINER, + ucl_msgpack_parse_array + }, + { + 0xde, + 8, + msgpack_map16, + 2, + MSGPACK_FLAG_CONTAINER|MSGPACK_FLAG_ASSOC, + ucl_msgpack_parse_map + }, + { + 0xdf, + 8, + msgpack_map32, + 4, + MSGPACK_FLAG_CONTAINER|MSGPACK_FLAG_ASSOC, + ucl_msgpack_parse_map + }, + { + 0xc7, + 8, + msgpack_ext8, + 1, + MSGPACK_FLAG_EXT, + ucl_msgpack_parse_ignore + }, + { + 0xc8, + 8, + msgpack_ext16, + 2, + MSGPACK_FLAG_EXT, + ucl_msgpack_parse_ignore + }, + { + 0xc9, + 8, + msgpack_ext32, + 4, + MSGPACK_FLAG_EXT, + ucl_msgpack_parse_ignore + }, + { + 0xd4, + 8, + msgpack_fixext1, + 1, + MSGPACK_FLAG_FIXED | MSGPACK_FLAG_EXT, + ucl_msgpack_parse_ignore + }, + { + 0xd5, + 8, + msgpack_fixext2, + 2, + MSGPACK_FLAG_FIXED | MSGPACK_FLAG_EXT, + ucl_msgpack_parse_ignore + }, + { + 0xd6, + 8, + msgpack_fixext4, + 4, + MSGPACK_FLAG_FIXED | MSGPACK_FLAG_EXT, + ucl_msgpack_parse_ignore + }, + { + 0xd7, + 8, + msgpack_fixext8, + 8, + MSGPACK_FLAG_FIXED | MSGPACK_FLAG_EXT, + ucl_msgpack_parse_ignore + }, + { + 0xd8, + 8, + msgpack_fixext16, + 16, + MSGPACK_FLAG_FIXED | MSGPACK_FLAG_EXT, + ucl_msgpack_parse_ignore + } +}; + +#undef MSGPACK_DEBUG_PARSER + +static inline struct ucl_msgpack_parser * +ucl_msgpack_get_parser_from_type (unsigned char t) +{ + unsigned int i, shift, mask; + + for (i = 0; i < sizeof (parsers) / sizeof (parsers[0]); i ++) { + shift = CHAR_BIT - parsers[i].prefixlen; + mask = parsers[i].prefix >> shift; + + if (mask == (t >> shift)) { + return &parsers[i]; + } + } + + return NULL; +} + +static inline struct ucl_stack * +ucl_msgpack_get_container (struct ucl_parser *parser, + struct ucl_msgpack_parser *obj_parser, uint64_t len) +{ + struct ucl_stack *stack; + + assert (obj_parser != NULL); + + if (obj_parser->flags & MSGPACK_FLAG_CONTAINER) { + assert ((len & MSGPACK_CONTAINER_BIT) == 0); + /* + * Insert new container to the stack + */ + if (parser->stack == NULL) { + parser->stack = calloc (1, sizeof (struct ucl_stack)); + + if (parser->stack == NULL) { + ucl_create_err (&parser->err, "no memory"); + return NULL; + } + } + else { + stack = calloc (1, sizeof (struct ucl_stack)); + + if (stack == NULL) { + ucl_create_err (&parser->err, "no memory"); + return NULL; + } + + stack->next = parser->stack; + parser->stack = stack; + } + + parser->stack->level = len | MSGPACK_CONTAINER_BIT; + +#ifdef MSGPACK_DEBUG_PARSER + stack = parser->stack; + while (stack) { + fprintf(stderr, "+"); + stack = stack->next; + } + + fprintf(stderr, "%s -> %d\n", obj_parser->flags & MSGPACK_FLAG_ASSOC ? "object" : "array", (int)len); +#endif + } + else { + /* + * Get the current stack top + */ + if (parser->stack) { + return parser->stack; + } + else { + ucl_create_err (&parser->err, "bad top level object for msgpack"); + return NULL; + } + } + + return parser->stack; +} + +static bool +ucl_msgpack_is_container_finished (struct ucl_stack *container) +{ + uint64_t level; + + assert (container != NULL); + + if (container->level & MSGPACK_CONTAINER_BIT) { + level = container->level & ~MSGPACK_CONTAINER_BIT; + + if (level == 0) { + return true; + } + } + + return false; +} + +static bool +ucl_msgpack_insert_object (struct ucl_parser *parser, + const unsigned char *key, + size_t keylen, ucl_object_t *obj) +{ + uint64_t level; + struct ucl_stack *container; + + container = parser->stack; + assert (container != NULL); + assert (container->level > 0); + assert (obj != NULL); + assert (container->obj != NULL); + + if (container->obj->type == UCL_ARRAY) { + ucl_array_append (container->obj, obj); + } + else if (container->obj->type == UCL_OBJECT) { + if (key == NULL || keylen == 0) { + ucl_create_err (&parser->err, "cannot insert object with no key"); + return false; + } + + obj->key = key; + obj->keylen = keylen; + + if (!(parser->flags & UCL_PARSER_ZEROCOPY)) { + ucl_copy_key_trash (obj); + } + + ucl_parser_process_object_element (parser, obj); + } + else { + ucl_create_err (&parser->err, "bad container type"); + return false; + } + + if (container->level & MSGPACK_CONTAINER_BIT) { + level = container->level & ~MSGPACK_CONTAINER_BIT; + container->level = (level - 1) | MSGPACK_CONTAINER_BIT; + } + + return true; +} + +static struct ucl_stack * +ucl_msgpack_get_next_container (struct ucl_parser *parser) +{ + struct ucl_stack *cur = NULL; + uint64_t level; + + cur = parser->stack; + + if (cur == NULL) { + return NULL; + } + + if (cur->level & MSGPACK_CONTAINER_BIT) { + level = cur->level & ~MSGPACK_CONTAINER_BIT; + + if (level == 0) { + /* We need to switch to the previous container */ + parser->stack = cur->next; + parser->cur_obj = cur->obj; + free (cur); + +#ifdef MSGPACK_DEBUG_PARSER + cur = parser->stack; + while (cur) { + fprintf(stderr, "-"); + cur = cur->next; + } + fprintf(stderr, "-%s -> %d\n", parser->cur_obj->type == UCL_OBJECT ? "object" : "array", (int)parser->cur_obj->len); +#endif + + return ucl_msgpack_get_next_container (parser); + } + } + + /* + * For UCL containers we don't know length, so we just insert the whole + * message pack blob into the top level container + */ + + assert (cur->obj != NULL); + + return cur; +} + +#define CONSUME_RET do { \ + if (ret != -1) { \ + p += ret; \ + remain -= ret; \ + obj_parser = NULL; \ + assert (remain >= 0); \ + } \ + else { \ + ucl_create_err (&parser->err, \ + "cannot parse type %d of len %u", \ + (int)obj_parser->fmt, \ + (unsigned)len); \ + return false; \ + } \ +} while(0) + +#define GET_NEXT_STATE do { \ + container = ucl_msgpack_get_next_container (parser); \ + if (container == NULL) { \ + ucl_create_err (&parser->err, \ + "empty container"); \ + return false; \ + } \ + next_state = container->obj->type == UCL_OBJECT ? \ + read_assoc_key : read_array_value; \ +} while(0) + +static bool +ucl_msgpack_consume (struct ucl_parser *parser) +{ + const unsigned char *p, *end, *key = NULL; + struct ucl_stack *container; + enum e_msgpack_parser_state { + read_type, + start_assoc, + start_array, + read_assoc_key, + read_assoc_value, + finish_assoc_value, + read_array_value, + finish_array_value, + error_state + } state = read_type, next_state = error_state; + struct ucl_msgpack_parser *obj_parser; + uint64_t len; + ssize_t ret, remain, keylen = 0; +#ifdef MSGPACK_DEBUG_PARSER + uint64_t i; + enum e_msgpack_parser_state hist[256]; +#endif + + p = parser->chunks->begin; + remain = parser->chunks->remain; + end = p + remain; + + + while (p < end) { +#ifdef MSGPACK_DEBUG_PARSER + hist[i++ % 256] = state; +#endif + switch (state) { + case read_type: + obj_parser = ucl_msgpack_get_parser_from_type (*p); + + if (obj_parser == NULL) { + ucl_create_err (&parser->err, "unknown msgpack format: %x", + (unsigned int)*p); + + return false; + } + /* Now check length sanity */ + if (obj_parser->flags & MSGPACK_FLAG_FIXED) { + if (obj_parser->len == 0) { + /* We have an embedded size */ + len = *p & ~obj_parser->prefix; + } + else { + if (remain < obj_parser->len) { + ucl_create_err (&parser->err, "not enough data remain to " + "read object's length: %u remain, %u needed", + (unsigned)remain, obj_parser->len); + + return false; + } + + len = obj_parser->len; + } + + if (!(obj_parser->flags & MSGPACK_FLAG_TYPEVALUE)) { + /* We must pass value as the second byte */ + if (remain > 0) { + p ++; + remain --; + } + } + else { + /* Len is irrelevant now */ + len = 0; + } + } + else { + /* Length is not embedded */ + if (remain < obj_parser->len) { + ucl_create_err (&parser->err, "not enough data remain to " + "read object's length: %u remain, %u needed", + (unsigned)remain, obj_parser->len); + + return false; + } + + p ++; + remain --; + + switch (obj_parser->len) { + case 1: + len = *p; + break; + case 2: + len = FROM_BE16 (*(uint16_t *)p); + break; + case 4: + len = FROM_BE32 (*(uint32_t *)p); + break; + case 8: + len = FROM_BE64 (*(uint64_t *)p); + break; + default: + assert (0); + break; + } + + p += obj_parser->len; + remain -= obj_parser->len; + } + + if (obj_parser->flags & MSGPACK_FLAG_ASSOC) { + /* We have just read the new associative map */ + state = start_assoc; + } + else if (obj_parser->flags & MSGPACK_FLAG_CONTAINER){ + state = start_array; + } + else { + state = next_state; + } + + break; + case start_assoc: + parser->cur_obj = ucl_object_new_full (UCL_OBJECT, + parser->chunks->priority); + /* Insert to the previous level container */ + if (parser->stack && !ucl_msgpack_insert_object (parser, + key, keylen, parser->cur_obj)) { + return false; + } + /* Get new container */ + container = ucl_msgpack_get_container (parser, obj_parser, len); + + if (container == NULL) { + return false; + } + + ret = obj_parser->func (parser, container, len, obj_parser->fmt, + p, remain); + CONSUME_RET; + key = NULL; + keylen = 0; + + if (len > 0) { + state = read_type; + next_state = read_assoc_key; + } + else { + /* Empty object */ + state = finish_assoc_value; + } + break; + + case start_array: + parser->cur_obj = ucl_object_new_full (UCL_ARRAY, + parser->chunks->priority); + /* Insert to the previous level container */ + if (parser->stack && !ucl_msgpack_insert_object (parser, + key, keylen, parser->cur_obj)) { + return false; + } + /* Get new container */ + container = ucl_msgpack_get_container (parser, obj_parser, len); + + if (container == NULL) { + return false; + } + + ret = obj_parser->func (parser, container, len, obj_parser->fmt, + p, remain); + CONSUME_RET; + + if (len > 0) { + state = read_type; + next_state = read_array_value; + } + else { + /* Empty array */ + state = finish_array_value; + } + break; + + case read_array_value: + /* + * p is now at the value start, len now contains length read and + * obj_parser contains the corresponding specific parser + */ + container = parser->stack; + + if (container == NULL) { + return false; + } + + ret = obj_parser->func (parser, container, len, obj_parser->fmt, + p, remain); + CONSUME_RET; + + + /* Insert value to the container and check if we have finished array */ + if (!ucl_msgpack_insert_object (parser, NULL, 0, + parser->cur_obj)) { + return false; + } + + if (ucl_msgpack_is_container_finished (container)) { + state = finish_array_value; + } + else { + /* Read more elements */ + state = read_type; + next_state = read_array_value; + } + + break; + + case read_assoc_key: + /* + * Keys must have string type for ucl msgpack + */ + if (!(obj_parser->flags & MSGPACK_FLAG_KEY)) { + ucl_create_err (&parser->err, "bad type for key: %u, expected " + "string", (unsigned)obj_parser->fmt); + + return false; + } + + key = p; + keylen = len; + + if (keylen > remain || keylen == 0) { + ucl_create_err (&parser->err, "too long or empty key"); + return false; + } + + p += len; + remain -= len; + + state = read_type; + next_state = read_assoc_value; + break; + + case read_assoc_value: + /* + * p is now at the value start, len now contains length read and + * obj_parser contains the corresponding specific parser + */ + container = parser->stack; + + if (container == NULL) { + return false; + } + + ret = obj_parser->func (parser, container, len, obj_parser->fmt, + p, remain); + CONSUME_RET; + + assert (key != NULL && keylen > 0); + + if (!ucl_msgpack_insert_object (parser, key, keylen, + parser->cur_obj)) { + return false; + } + + key = NULL; + keylen = 0; + + if (ucl_msgpack_is_container_finished (container)) { + state = finish_assoc_value; + } + else { + /* Read more elements */ + state = read_type; + next_state = read_assoc_key; + } + break; + + case finish_array_value: + case finish_assoc_value: + GET_NEXT_STATE; + state = read_type; + break; + + case error_state: + ucl_create_err (&parser->err, "invalid state machine state"); + + return false; + } + } + + /* Check the finishing state */ + switch (state) { + case start_array: + case start_assoc: + /* Empty container at the end */ + if (len != 0) { + ucl_create_err (&parser->err, "invalid non-empty container at the end"); + + return false; + } + + parser->cur_obj = ucl_object_new_full ( + state == start_array ? UCL_ARRAY : UCL_OBJECT, + parser->chunks->priority); + /* Insert to the previous level container */ + if (!ucl_msgpack_insert_object (parser, + key, keylen, parser->cur_obj)) { + return false; + } + /* Get new container */ + container = ucl_msgpack_get_container (parser, obj_parser, len); + + if (container == NULL) { + return false; + } + + ret = obj_parser->func (parser, container, len, obj_parser->fmt, + p, remain); + break; + + case read_array_value: + case read_assoc_value: + if (len != 0) { + ucl_create_err (&parser->err, "unfinished value at the end"); + + return false; + } + + container = parser->stack; + + if (container == NULL) { + return false; + } + + ret = obj_parser->func (parser, container, len, obj_parser->fmt, + p, remain); + CONSUME_RET; + + + /* Insert value to the container and check if we have finished array */ + if (!ucl_msgpack_insert_object (parser, NULL, 0, + parser->cur_obj)) { + return false; + } + break; + case finish_array_value: + case finish_assoc_value: + case read_type: + /* Valid finishing state */ + break; + default: + /* Invalid finishing state */ + ucl_create_err (&parser->err, "invalid state machine finishing state: %d", + state); + + return false; + } + + /* Rewind to the top level container */ + ucl_msgpack_get_next_container (parser); + assert (parser->stack == NULL || + (parser->stack->level & MSGPACK_CONTAINER_BIT) == 0); + + return true; +} + +bool +ucl_parse_msgpack (struct ucl_parser *parser) +{ + ucl_object_t *container = NULL; + const unsigned char *p; + bool ret; + + assert (parser != NULL); + assert (parser->chunks != NULL); + assert (parser->chunks->begin != NULL); + assert (parser->chunks->remain != 0); + + p = parser->chunks->begin; + + if (parser->stack) { + container = parser->stack->obj; + } + + /* + * When we start parsing message pack chunk, we must ensure that we + * have either a valid container or the top object inside message pack is + * of container type + */ + if (container == NULL) { + if ((*p & 0x80) != 0x80 && !(*p >= 0xdc && *p <= 0xdf)) { + ucl_create_err (&parser->err, "bad top level object for msgpack"); + return false; + } + } + + ret = ucl_msgpack_consume (parser); + + if (ret && parser->top_obj == NULL) { + parser->top_obj = parser->cur_obj; + } + + return ret; +} + +static ssize_t +ucl_msgpack_parse_map (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain) +{ + container->obj = parser->cur_obj; + + return 0; +} + +static ssize_t +ucl_msgpack_parse_array (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain) +{ + container->obj = parser->cur_obj; + + return 0; +} + +static ssize_t +ucl_msgpack_parse_string (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain) +{ + ucl_object_t *obj; + + if (len > remain) { + return -1; + } + + obj = ucl_object_new_full (UCL_STRING, parser->chunks->priority); + obj->value.sv = pos; + obj->len = len; + + if (fmt >= msgpack_bin8 && fmt <= msgpack_bin32) { + obj->flags |= UCL_OBJECT_BINARY; + } + + if (!(parser->flags & UCL_PARSER_ZEROCOPY)) { + if (obj->flags & UCL_OBJECT_BINARY) { + obj->trash_stack[UCL_TRASH_VALUE] = malloc (len); + + if (obj->trash_stack[UCL_TRASH_VALUE] != NULL) { + memcpy (obj->trash_stack[UCL_TRASH_VALUE], pos, len); + } + } + else { + ucl_copy_value_trash (obj); + } + } + + parser->cur_obj = obj; + + return len; +} + +static ssize_t +ucl_msgpack_parse_int (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain) +{ + ucl_object_t *obj; + + if (len > remain) { + return -1; + } + + obj = ucl_object_new_full (UCL_INT, parser->chunks->priority); + + switch (fmt) { + case msgpack_positive_fixint: + obj->value.iv = (*pos & 0x7f); + len = 1; + break; + case msgpack_negative_fixint: + obj->value.iv = - (*pos & 0x1f); + len = 1; + break; + case msgpack_uint8: + obj->value.iv = (unsigned char)*pos; + len = 1; + break; + case msgpack_int8: + obj->value.iv = (signed char)*pos; + len = 1; + break; + case msgpack_int16: + obj->value.iv = FROM_BE16 (*(int16_t *)pos); + len = 2; + break; + case msgpack_uint16: + obj->value.iv = FROM_BE16 (*(uint16_t *)pos); + len = 2; + break; + case msgpack_int32: + obj->value.iv = FROM_BE32 (*(int32_t *)pos); + len = 4; + break; + case msgpack_uint32: + obj->value.iv = FROM_BE32 (*(uint32_t *)pos); + len = 4; + break; + case msgpack_int64: + obj->value.iv = FROM_BE64 (*(int64_t *)pos); + len = 8; + break; + case msgpack_uint64: + obj->value.iv = FROM_BE64 (*(uint64_t *)pos); + len = 8; + break; + default: + assert (0); + break; + } + + parser->cur_obj = obj; + + return len; +} + +static ssize_t +ucl_msgpack_parse_float (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain) +{ + ucl_object_t *obj; + union { + uint32_t i; + float f; + } d; + + if (len > remain) { + return -1; + } + + obj = ucl_object_new_full (UCL_FLOAT, parser->chunks->priority); + + switch (fmt) { + case msgpack_float32: + d.i = FROM_BE32 (*(uint32_t *)pos); + /* XXX: can be slow */ + obj->value.dv = d.f; + len = 4; + break; + case msgpack_float64: + obj->value.iv = FROM_BE64 (*(uint64_t *)pos); + len = 8; + break; + default: + assert (0); + break; + } + + parser->cur_obj = obj; + + return len; +} + +static ssize_t +ucl_msgpack_parse_bool (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain) +{ + ucl_object_t *obj; + + if (len > remain) { + return -1; + } + + obj = ucl_object_new_full (UCL_BOOLEAN, parser->chunks->priority); + + switch (fmt) { + case msgpack_true: + obj->value.iv = true; + break; + case msgpack_false: + obj->value.iv = false; + break; + default: + assert (0); + break; + } + + parser->cur_obj = obj; + + return 1; +} + +static ssize_t +ucl_msgpack_parse_null (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain) +{ + ucl_object_t *obj; + + if (len > remain) { + return -1; + } + + obj = ucl_object_new_full (UCL_NULL, parser->chunks->priority); + parser->cur_obj = obj; + + return 1; +} + +static ssize_t +ucl_msgpack_parse_ignore (struct ucl_parser *parser, + struct ucl_stack *container, size_t len, enum ucl_msgpack_format fmt, + const unsigned char *pos, size_t remain) +{ + if (len > remain) { + return -1; + } + + switch (fmt) { + case msgpack_fixext1: + len = 2; + break; + case msgpack_fixext2: + len = 3; + break; + case msgpack_fixext4: + len = 5; + break; + case msgpack_fixext8: + len = 9; + break; + case msgpack_fixext16: + len = 17; + break; + case msgpack_ext8: + case msgpack_ext16: + case msgpack_ext32: + len = len + 1; + break; + default: + ucl_create_err (&parser->err, "bad type: %x", (unsigned)fmt); + return -1; + } + + return len; +} diff --git a/contrib/libucl/src/ucl_parser.c b/contrib/libucl/src/ucl_parser.c new file mode 100644 index 0000000..9bd4139 --- /dev/null +++ b/contrib/libucl/src/ucl_parser.c @@ -0,0 +1,2485 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include "ucl.h" +#include "ucl_internal.h" +#include "ucl_chartable.h" + +/** + * @file ucl_parser.c + * The implementation of ucl parser + */ + +struct ucl_parser_saved_state { + unsigned int line; + unsigned int column; + size_t remain; + const unsigned char *pos; +}; + +/** + * Move up to len characters + * @param parser + * @param begin + * @param len + * @return new position in chunk + */ +#define ucl_chunk_skipc(chunk, p) do{ \ + if (*(p) == '\n') { \ + (chunk)->line ++; \ + (chunk)->column = 0; \ + } \ + else (chunk)->column ++; \ + (p++); \ + (chunk)->pos ++; \ + (chunk)->remain --; \ + } while (0) + +static inline void +ucl_set_err (struct ucl_parser *parser, int code, const char *str, UT_string **err) +{ + const char *fmt_string, *filename; + struct ucl_chunk *chunk = parser->chunks; + + if (parser->cur_file) { + filename = parser->cur_file; + } + else { + filename = "<unknown>"; + } + + if (chunk->pos < chunk->end) { + if (isgraph (*chunk->pos)) { + fmt_string = "error while parsing %s: " + "line: %d, column: %d - '%s', character: '%c'"; + } + else { + fmt_string = "error while parsing %s: " + "line: %d, column: %d - '%s', character: '0x%02x'"; + } + ucl_create_err (err, fmt_string, + filename, chunk->line, chunk->column, + str, *chunk->pos); + } + else { + ucl_create_err (err, "error while parsing %s: at the end of chunk: %s", + filename, str); + } + + parser->err_code = code; +} + +/** + * Skip all comments from the current pos resolving nested and multiline comments + * @param parser + * @return + */ +static bool +ucl_skip_comments (struct ucl_parser *parser) +{ + struct ucl_chunk *chunk = parser->chunks; + const unsigned char *p; + int comments_nested = 0; + bool quoted = false; + + p = chunk->pos; + +start: + if (chunk->remain > 0 && *p == '#') { + if (parser->state != UCL_STATE_SCOMMENT && + parser->state != UCL_STATE_MCOMMENT) { + while (p < chunk->end) { + if (*p == '\n') { + ucl_chunk_skipc (chunk, p); + goto start; + } + ucl_chunk_skipc (chunk, p); + } + } + } + else if (chunk->remain >= 2 && *p == '/') { + if (p[1] == '*') { + ucl_chunk_skipc (chunk, p); + comments_nested ++; + ucl_chunk_skipc (chunk, p); + + while (p < chunk->end) { + if (*p == '"' && *(p - 1) != '\\') { + quoted = !quoted; + } + + if (!quoted) { + if (*p == '*') { + ucl_chunk_skipc (chunk, p); + if (*p == '/') { + comments_nested --; + if (comments_nested == 0) { + ucl_chunk_skipc (chunk, p); + goto start; + } + } + ucl_chunk_skipc (chunk, p); + } + else if (p[0] == '/' && chunk->remain >= 2 && p[1] == '*') { + comments_nested ++; + ucl_chunk_skipc (chunk, p); + ucl_chunk_skipc (chunk, p); + continue; + } + } + ucl_chunk_skipc (chunk, p); + } + if (comments_nested != 0) { + ucl_set_err (parser, UCL_ENESTED, + "unfinished multiline comment", &parser->err); + return false; + } + } + } + + return true; +} + +/** + * Return multiplier for a character + * @param c multiplier character + * @param is_bytes if true use 1024 multiplier + * @return multiplier + */ +static inline unsigned long +ucl_lex_num_multiplier (const unsigned char c, bool is_bytes) { + const struct { + char c; + long mult_normal; + long mult_bytes; + } multipliers[] = { + {'m', 1000 * 1000, 1024 * 1024}, + {'k', 1000, 1024}, + {'g', 1000 * 1000 * 1000, 1024 * 1024 * 1024} + }; + int i; + + for (i = 0; i < 3; i ++) { + if (tolower (c) == multipliers[i].c) { + if (is_bytes) { + return multipliers[i].mult_bytes; + } + return multipliers[i].mult_normal; + } + } + + return 1; +} + + +/** + * Return multiplier for time scaling + * @param c + * @return + */ +static inline double +ucl_lex_time_multiplier (const unsigned char c) { + const struct { + char c; + double mult; + } multipliers[] = { + {'m', 60}, + {'h', 60 * 60}, + {'d', 60 * 60 * 24}, + {'w', 60 * 60 * 24 * 7}, + {'y', 60 * 60 * 24 * 7 * 365} + }; + int i; + + for (i = 0; i < 5; i ++) { + if (tolower (c) == multipliers[i].c) { + return multipliers[i].mult; + } + } + + return 1; +} + +/** + * Return true if a character is a end of an atom + * @param c + * @return + */ +static inline bool +ucl_lex_is_atom_end (const unsigned char c) +{ + return ucl_test_character (c, UCL_CHARACTER_VALUE_END); +} + +static inline bool +ucl_lex_is_comment (const unsigned char c1, const unsigned char c2) +{ + if (c1 == '/') { + if (c2 == '*') { + return true; + } + } + else if (c1 == '#') { + return true; + } + return false; +} + +/** + * Check variable found + * @param parser + * @param ptr + * @param remain + * @param out_len + * @param strict + * @param found + * @return + */ +static inline const char * +ucl_check_variable_safe (struct ucl_parser *parser, const char *ptr, size_t remain, + size_t *out_len, bool strict, bool *found) +{ + struct ucl_variable *var; + unsigned char *dst; + size_t dstlen; + bool need_free = false; + + LL_FOREACH (parser->variables, var) { + if (strict) { + if (remain == var->var_len) { + if (memcmp (ptr, var->var, var->var_len) == 0) { + *out_len += var->value_len; + *found = true; + return (ptr + var->var_len); + } + } + } + else { + if (remain >= var->var_len) { + if (memcmp (ptr, var->var, var->var_len) == 0) { + *out_len += var->value_len; + *found = true; + return (ptr + var->var_len); + } + } + } + } + + /* XXX: can only handle ${VAR} */ + if (!(*found) && parser->var_handler != NULL && strict) { + /* Call generic handler */ + if (parser->var_handler (ptr, remain, &dst, &dstlen, &need_free, + parser->var_data)) { + *found = true; + if (need_free) { + free (dst); + } + return (ptr + remain); + } + } + + return ptr; +} + +/** + * Check for a variable in a given string + * @param parser + * @param ptr + * @param remain + * @param out_len + * @param vars_found + * @return + */ +static const char * +ucl_check_variable (struct ucl_parser *parser, const char *ptr, + size_t remain, size_t *out_len, bool *vars_found) +{ + const char *p, *end, *ret = ptr; + bool found = false; + + if (*ptr == '{') { + /* We need to match the variable enclosed in braces */ + p = ptr + 1; + end = ptr + remain; + while (p < end) { + if (*p == '}') { + ret = ucl_check_variable_safe (parser, ptr + 1, p - ptr - 1, + out_len, true, &found); + if (found) { + /* {} must be excluded actually */ + ret ++; + if (!*vars_found) { + *vars_found = true; + } + } + else { + *out_len += 2; + } + break; + } + p ++; + } + } + else if (*ptr != '$') { + /* Not count escaped dollar sign */ + ret = ucl_check_variable_safe (parser, ptr, remain, out_len, false, &found); + if (found && !*vars_found) { + *vars_found = true; + } + if (!found) { + (*out_len) ++; + } + } + else { + ret ++; + (*out_len) ++; + } + + return ret; +} + +/** + * Expand a single variable + * @param parser + * @param ptr + * @param remain + * @param dest + * @return + */ +static const char * +ucl_expand_single_variable (struct ucl_parser *parser, const char *ptr, + size_t remain, unsigned char **dest) +{ + unsigned char *d = *dest, *dst; + const char *p = ptr + 1, *ret; + struct ucl_variable *var; + size_t dstlen; + bool need_free = false; + bool found = false; + bool strict = false; + + ret = ptr + 1; + remain --; + + if (*p == '$') { + *d++ = *p++; + *dest = d; + return p; + } + else if (*p == '{') { + p ++; + strict = true; + ret += 2; + remain -= 2; + } + + LL_FOREACH (parser->variables, var) { + if (remain >= var->var_len) { + if (memcmp (p, var->var, var->var_len) == 0) { + memcpy (d, var->value, var->value_len); + ret += var->var_len; + d += var->value_len; + found = true; + break; + } + } + } + if (!found) { + if (strict && parser->var_handler != NULL) { + if (parser->var_handler (ptr, remain, &dst, &dstlen, &need_free, + parser->var_data)) { + memcpy (d, dst, dstlen); + ret += dstlen; + d += remain; + found = true; + } + } + + /* Leave variable as is */ + if (!found) { + if (strict) { + /* Copy '${' */ + memcpy (d, ptr, 2); + d += 2; + ret --; + } + else { + memcpy (d, ptr, 1); + d ++; + } + } + } + + *dest = d; + return ret; +} + +/** + * Expand variables in string + * @param parser + * @param dst + * @param src + * @param in_len + * @return + */ +static ssize_t +ucl_expand_variable (struct ucl_parser *parser, unsigned char **dst, + const char *src, size_t in_len) +{ + const char *p, *end = src + in_len; + unsigned char *d; + size_t out_len = 0; + bool vars_found = false; + + p = src; + while (p != end) { + if (*p == '$') { + p = ucl_check_variable (parser, p + 1, end - p - 1, &out_len, &vars_found); + } + else { + p ++; + out_len ++; + } + } + + if (!vars_found) { + /* Trivial case */ + *dst = NULL; + return in_len; + } + + *dst = UCL_ALLOC (out_len + 1); + if (*dst == NULL) { + return in_len; + } + + d = *dst; + p = src; + while (p != end) { + if (*p == '$') { + p = ucl_expand_single_variable (parser, p, end - p, &d); + } + else { + *d++ = *p++; + } + } + + *d = '\0'; + + return out_len; +} + +/** + * Store or copy pointer to the trash stack + * @param parser parser object + * @param src src string + * @param dst destination buffer (trash stack pointer) + * @param dst_const const destination pointer (e.g. value of object) + * @param in_len input length + * @param need_unescape need to unescape source (and copy it) + * @param need_lowercase need to lowercase value (and copy) + * @param need_expand need to expand variables (and copy as well) + * @return output length (excluding \0 symbol) + */ +static inline ssize_t +ucl_copy_or_store_ptr (struct ucl_parser *parser, + const unsigned char *src, unsigned char **dst, + const char **dst_const, size_t in_len, + bool need_unescape, bool need_lowercase, bool need_expand) +{ + ssize_t ret = -1, tret; + unsigned char *tmp; + + if (need_unescape || need_lowercase || + (need_expand && parser->variables != NULL) || + !(parser->flags & UCL_PARSER_ZEROCOPY)) { + /* Copy string */ + *dst = UCL_ALLOC (in_len + 1); + if (*dst == NULL) { + ucl_set_err (parser, UCL_EINTERNAL, "cannot allocate memory for a string", + &parser->err); + return false; + } + if (need_lowercase) { + ret = ucl_strlcpy_tolower (*dst, src, in_len + 1); + } + else { + ret = ucl_strlcpy_unsafe (*dst, src, in_len + 1); + } + + if (need_unescape) { + ret = ucl_unescape_json_string (*dst, ret); + } + if (need_expand) { + tmp = *dst; + tret = ret; + ret = ucl_expand_variable (parser, dst, tmp, ret); + if (*dst == NULL) { + /* Nothing to expand */ + *dst = tmp; + ret = tret; + } + else { + /* Free unexpanded value */ + UCL_FREE (in_len + 1, tmp); + } + } + *dst_const = *dst; + } + else { + *dst_const = src; + ret = in_len; + } + + return ret; +} + +/** + * Create and append an object at the specified level + * @param parser + * @param is_array + * @param level + * @return + */ +static inline ucl_object_t * +ucl_parser_add_container (ucl_object_t *obj, struct ucl_parser *parser, + bool is_array, int level) +{ + struct ucl_stack *st; + + if (!is_array) { + if (obj == NULL) { + obj = ucl_object_new_full (UCL_OBJECT, parser->chunks->priority); + } + else { + obj->type = UCL_OBJECT; + } + if (obj->value.ov == NULL) { + obj->value.ov = ucl_hash_create (parser->flags & UCL_PARSER_KEY_LOWERCASE); + } + parser->state = UCL_STATE_KEY; + } + else { + if (obj == NULL) { + obj = ucl_object_new_full (UCL_ARRAY, parser->chunks->priority); + } + else { + obj->type = UCL_ARRAY; + } + parser->state = UCL_STATE_VALUE; + } + + st = UCL_ALLOC (sizeof (struct ucl_stack)); + if (st == NULL) { + ucl_set_err (parser, UCL_EINTERNAL, "cannot allocate memory for an object", + &parser->err); + ucl_object_unref (obj); + return NULL; + } + st->obj = obj; + st->level = level; + LL_PREPEND (parser->stack, st); + parser->cur_obj = obj; + + return obj; +} + +int +ucl_maybe_parse_number (ucl_object_t *obj, + const char *start, const char *end, const char **pos, + bool allow_double, bool number_bytes, bool allow_time) +{ + const char *p = start, *c = start; + char *endptr; + bool got_dot = false, got_exp = false, need_double = false, + is_time = false, valid_start = false, is_hex = false, + is_neg = false; + double dv = 0; + int64_t lv = 0; + + if (*p == '-') { + is_neg = true; + c ++; + p ++; + } + while (p < end) { + if (is_hex && isxdigit (*p)) { + p ++; + } + else if (isdigit (*p)) { + valid_start = true; + p ++; + } + else if (!is_hex && (*p == 'x' || *p == 'X')) { + is_hex = true; + allow_double = false; + c = p + 1; + } + else if (allow_double) { + if (p == c) { + /* Empty digits sequence, not a number */ + *pos = start; + return EINVAL; + } + else if (*p == '.') { + if (got_dot) { + /* Double dots, not a number */ + *pos = start; + return EINVAL; + } + else { + got_dot = true; + need_double = true; + p ++; + } + } + else if (*p == 'e' || *p == 'E') { + if (got_exp) { + /* Double exp, not a number */ + *pos = start; + return EINVAL; + } + else { + got_exp = true; + need_double = true; + p ++; + if (p >= end) { + *pos = start; + return EINVAL; + } + if (!isdigit (*p) && *p != '+' && *p != '-') { + /* Wrong exponent sign */ + *pos = start; + return EINVAL; + } + else { + p ++; + } + } + } + else { + /* Got the end of the number, need to check */ + break; + } + } + else { + break; + } + } + + if (!valid_start) { + *pos = start; + return EINVAL; + } + + errno = 0; + if (need_double) { + dv = strtod (c, &endptr); + } + else { + if (is_hex) { + lv = strtoimax (c, &endptr, 16); + } + else { + lv = strtoimax (c, &endptr, 10); + } + } + if (errno == ERANGE) { + *pos = start; + return ERANGE; + } + + /* Now check endptr */ + if (endptr == NULL || ucl_lex_is_atom_end (*endptr) || *endptr == '\0') { + p = endptr; + goto set_obj; + } + + if (endptr < end && endptr != start) { + p = endptr; + switch (*p) { + case 'm': + case 'M': + case 'g': + case 'G': + case 'k': + case 'K': + if (end - p >= 2) { + if (p[1] == 's' || p[1] == 'S') { + /* Milliseconds */ + if (!need_double) { + need_double = true; + dv = lv; + } + is_time = true; + if (p[0] == 'm' || p[0] == 'M') { + dv /= 1000.; + } + else { + dv *= ucl_lex_num_multiplier (*p, false); + } + p += 2; + goto set_obj; + } + else if (number_bytes || (p[1] == 'b' || p[1] == 'B')) { + /* Bytes */ + if (need_double) { + need_double = false; + lv = dv; + } + lv *= ucl_lex_num_multiplier (*p, true); + p += 2; + goto set_obj; + } + else if (ucl_lex_is_atom_end (p[1])) { + if (need_double) { + dv *= ucl_lex_num_multiplier (*p, false); + } + else { + lv *= ucl_lex_num_multiplier (*p, number_bytes); + } + p ++; + goto set_obj; + } + else if (allow_time && end - p >= 3) { + if (tolower (p[0]) == 'm' && + tolower (p[1]) == 'i' && + tolower (p[2]) == 'n') { + /* Minutes */ + if (!need_double) { + need_double = true; + dv = lv; + } + is_time = true; + dv *= 60.; + p += 3; + goto set_obj; + } + } + } + else { + if (need_double) { + dv *= ucl_lex_num_multiplier (*p, false); + } + else { + lv *= ucl_lex_num_multiplier (*p, number_bytes); + } + p ++; + goto set_obj; + } + break; + case 'S': + case 's': + if (allow_time && + (p == end - 1 || ucl_lex_is_atom_end (p[1]))) { + if (!need_double) { + need_double = true; + dv = lv; + } + p ++; + is_time = true; + goto set_obj; + } + break; + case 'h': + case 'H': + case 'd': + case 'D': + case 'w': + case 'W': + case 'Y': + case 'y': + if (allow_time && + (p == end - 1 || ucl_lex_is_atom_end (p[1]))) { + if (!need_double) { + need_double = true; + dv = lv; + } + is_time = true; + dv *= ucl_lex_time_multiplier (*p); + p ++; + goto set_obj; + } + break; + case '\t': + case ' ': + while (p < end && ucl_test_character(*p, UCL_CHARACTER_WHITESPACE)) { + p++; + } + if (ucl_lex_is_atom_end(*p)) + goto set_obj; + break; + } + } + else if (endptr == end) { + /* Just a number at the end of chunk */ + p = endptr; + goto set_obj; + } + + *pos = c; + return EINVAL; + +set_obj: + if (obj != NULL) { + if (allow_double && (need_double || is_time)) { + if (!is_time) { + obj->type = UCL_FLOAT; + } + else { + obj->type = UCL_TIME; + } + obj->value.dv = is_neg ? (-dv) : dv; + } + else { + obj->type = UCL_INT; + obj->value.iv = is_neg ? (-lv) : lv; + } + } + *pos = p; + return 0; +} + +/** + * Parse possible number + * @param parser + * @param chunk + * @param obj + * @return true if a number has been parsed + */ +static bool +ucl_lex_number (struct ucl_parser *parser, + struct ucl_chunk *chunk, ucl_object_t *obj) +{ + const unsigned char *pos; + int ret; + + ret = ucl_maybe_parse_number (obj, chunk->pos, chunk->end, (const char **)&pos, + true, false, ((parser->flags & UCL_PARSER_NO_TIME) == 0)); + + if (ret == 0) { + chunk->remain -= pos - chunk->pos; + chunk->column += pos - chunk->pos; + chunk->pos = pos; + return true; + } + else if (ret == ERANGE) { + ucl_set_err (parser, UCL_ESYNTAX, "numeric value out of range", + &parser->err); + } + + return false; +} + +/** + * Parse quoted string with possible escapes + * @param parser + * @param chunk + * @param need_unescape + * @param ucl_escape + * @param var_expand + * @return true if a string has been parsed + */ +static bool +ucl_lex_json_string (struct ucl_parser *parser, + struct ucl_chunk *chunk, bool *need_unescape, bool *ucl_escape, bool *var_expand) +{ + const unsigned char *p = chunk->pos; + unsigned char c; + int i; + + while (p < chunk->end) { + c = *p; + if (c < 0x1F) { + /* Unmasked control character */ + if (c == '\n') { + ucl_set_err (parser, UCL_ESYNTAX, "unexpected newline", + &parser->err); + } + else { + ucl_set_err (parser, UCL_ESYNTAX, "unexpected control character", + &parser->err); + } + return false; + } + else if (c == '\\') { + ucl_chunk_skipc (chunk, p); + c = *p; + if (p >= chunk->end) { + ucl_set_err (parser, UCL_ESYNTAX, "unfinished escape character", + &parser->err); + return false; + } + else if (ucl_test_character (c, UCL_CHARACTER_ESCAPE)) { + if (c == 'u') { + ucl_chunk_skipc (chunk, p); + for (i = 0; i < 4 && p < chunk->end; i ++) { + if (!isxdigit (*p)) { + ucl_set_err (parser, UCL_ESYNTAX, "invalid utf escape", + &parser->err); + return false; + } + ucl_chunk_skipc (chunk, p); + } + if (p >= chunk->end) { + ucl_set_err (parser, UCL_ESYNTAX, "unfinished escape character", + &parser->err); + return false; + } + } + else { + ucl_chunk_skipc (chunk, p); + } + } + *need_unescape = true; + *ucl_escape = true; + continue; + } + else if (c == '"') { + ucl_chunk_skipc (chunk, p); + return true; + } + else if (ucl_test_character (c, UCL_CHARACTER_UCL_UNSAFE)) { + *ucl_escape = true; + } + else if (c == '$') { + *var_expand = true; + } + ucl_chunk_skipc (chunk, p); + } + + ucl_set_err (parser, UCL_ESYNTAX, "no quote at the end of json string", + &parser->err); + return false; +} + +static void +ucl_parser_append_elt (struct ucl_parser *parser, ucl_hash_t *cont, + ucl_object_t *top, + ucl_object_t *elt) +{ + ucl_object_t *nobj; + + if ((parser->flags & UCL_PARSER_NO_IMPLICIT_ARRAYS) == 0) { + /* Implicit array */ + top->flags |= UCL_OBJECT_MULTIVALUE; + DL_APPEND (top, elt); + parser->stack->obj->len ++; + } + else { + if ((top->flags & UCL_OBJECT_MULTIVALUE) != 0) { + /* Just add to the explicit array */ + ucl_array_append (top, elt); + } + else { + /* Convert to an array */ + nobj = ucl_object_typed_new (UCL_ARRAY); + nobj->key = top->key; + nobj->keylen = top->keylen; + nobj->flags |= UCL_OBJECT_MULTIVALUE; + ucl_array_append (nobj, top); + ucl_array_append (nobj, elt); + ucl_hash_replace (cont, top, nobj); + } + } +} + +bool +ucl_parser_process_object_element (struct ucl_parser *parser, ucl_object_t *nobj) +{ + ucl_hash_t *container; + ucl_object_t *tobj; + + container = parser->stack->obj->value.ov; + + tobj = __DECONST (ucl_object_t *, ucl_hash_search_obj (container, nobj)); + if (tobj == NULL) { + container = ucl_hash_insert_object (container, nobj, + parser->flags & UCL_PARSER_KEY_LOWERCASE); + nobj->prev = nobj; + nobj->next = NULL; + parser->stack->obj->len ++; + } + else { + unsigned priold = ucl_object_get_priority (tobj), + prinew = ucl_object_get_priority (nobj); + switch (parser->chunks->strategy) { + + case UCL_DUPLICATE_APPEND: + /* + * The logic here is the following: + * + * - if we have two objects with the same priority, then we form an + * implicit or explicit array + * - if a new object has bigger priority, then we overwrite an old one + * - if a new object has lower priority, then we ignore it + */ + + + /* Special case for inherited objects */ + if (tobj->flags & UCL_OBJECT_INHERITED) { + prinew = priold + 1; + } + + if (priold == prinew) { + ucl_parser_append_elt (parser, container, tobj, nobj); + } + else if (priold > prinew) { + /* + * We add this new object to a list of trash objects just to ensure + * that it won't come to any real object + * XXX: rather inefficient approach + */ + DL_APPEND (parser->trash_objs, nobj); + } + else { + ucl_hash_replace (container, tobj, nobj); + ucl_object_unref (tobj); + } + + break; + + case UCL_DUPLICATE_REWRITE: + /* We just rewrite old values regardless of priority */ + ucl_hash_replace (container, tobj, nobj); + ucl_object_unref (tobj); + + break; + + case UCL_DUPLICATE_ERROR: + ucl_create_err (&parser->err, "error while parsing %s: " + "line: %d, column: %d: duplicate element for key '%s' " + "has been found", + parser->cur_file ? parser->cur_file : "<unknown>", + parser->chunks->line, parser->chunks->column, nobj->key); + return false; + + case UCL_DUPLICATE_MERGE: + /* + * Here we do have some old object so we just push it on top of objects stack + */ + if (tobj->type == UCL_OBJECT || tobj->type == UCL_ARRAY) { + ucl_object_unref (nobj); + nobj = tobj; + } + else { + /* For other types we create implicit array as usual */ + ucl_parser_append_elt (parser, container, tobj, nobj); + } + break; + } + } + + parser->stack->obj->value.ov = container; + parser->cur_obj = nobj; + + return true; +} + +/** + * Parse a key in an object + * @param parser + * @param chunk + * @param next_key + * @param end_of_object + * @return true if a key has been parsed + */ +static bool +ucl_parse_key (struct ucl_parser *parser, struct ucl_chunk *chunk, + bool *next_key, bool *end_of_object) +{ + const unsigned char *p, *c = NULL, *end, *t; + const char *key = NULL; + bool got_quote = false, got_eq = false, got_semicolon = false, + need_unescape = false, ucl_escape = false, var_expand = false, + got_content = false, got_sep = false; + ucl_object_t *nobj; + ssize_t keylen; + + p = chunk->pos; + + if (*p == '.') { + /* It is macro actually */ + ucl_chunk_skipc (chunk, p); + parser->prev_state = parser->state; + parser->state = UCL_STATE_MACRO_NAME; + *end_of_object = false; + return true; + } + while (p < chunk->end) { + /* + * A key must start with alpha, number, '/' or '_' and end with space character + */ + if (c == NULL) { + if (chunk->remain >= 2 && ucl_lex_is_comment (p[0], p[1])) { + if (!ucl_skip_comments (parser)) { + return false; + } + p = chunk->pos; + } + else if (ucl_test_character (*p, UCL_CHARACTER_WHITESPACE_UNSAFE)) { + ucl_chunk_skipc (chunk, p); + } + else if (ucl_test_character (*p, UCL_CHARACTER_KEY_START)) { + /* The first symbol */ + c = p; + ucl_chunk_skipc (chunk, p); + got_content = true; + } + else if (*p == '"') { + /* JSON style key */ + c = p + 1; + got_quote = true; + got_content = true; + ucl_chunk_skipc (chunk, p); + } + else if (*p == '}') { + /* We have actually end of an object */ + *end_of_object = true; + return true; + } + else if (*p == '.') { + ucl_chunk_skipc (chunk, p); + parser->prev_state = parser->state; + parser->state = UCL_STATE_MACRO_NAME; + return true; + } + else { + /* Invalid identifier */ + ucl_set_err (parser, UCL_ESYNTAX, "key must begin with a letter", + &parser->err); + return false; + } + } + else { + /* Parse the body of a key */ + if (!got_quote) { + if (ucl_test_character (*p, UCL_CHARACTER_KEY)) { + got_content = true; + ucl_chunk_skipc (chunk, p); + } + else if (ucl_test_character (*p, UCL_CHARACTER_KEY_SEP)) { + end = p; + break; + } + else { + ucl_set_err (parser, UCL_ESYNTAX, "invalid character in a key", + &parser->err); + return false; + } + } + else { + /* We need to parse json like quoted string */ + if (!ucl_lex_json_string (parser, chunk, &need_unescape, &ucl_escape, &var_expand)) { + return false; + } + /* Always escape keys obtained via json */ + end = chunk->pos - 1; + p = chunk->pos; + break; + } + } + } + + if (p >= chunk->end && got_content) { + ucl_set_err (parser, UCL_ESYNTAX, "unfinished key", &parser->err); + return false; + } + else if (!got_content) { + return true; + } + *end_of_object = false; + /* We are now at the end of the key, need to parse the rest */ + while (p < chunk->end) { + if (ucl_test_character (*p, UCL_CHARACTER_WHITESPACE)) { + ucl_chunk_skipc (chunk, p); + } + else if (*p == '=') { + if (!got_eq && !got_semicolon) { + ucl_chunk_skipc (chunk, p); + got_eq = true; + } + else { + ucl_set_err (parser, UCL_ESYNTAX, "unexpected '=' character", + &parser->err); + return false; + } + } + else if (*p == ':') { + if (!got_eq && !got_semicolon) { + ucl_chunk_skipc (chunk, p); + got_semicolon = true; + } + else { + ucl_set_err (parser, UCL_ESYNTAX, "unexpected ':' character", + &parser->err); + return false; + } + } + else if (chunk->remain >= 2 && ucl_lex_is_comment (p[0], p[1])) { + /* Check for comment */ + if (!ucl_skip_comments (parser)) { + return false; + } + p = chunk->pos; + } + else { + /* Start value */ + break; + } + } + + if (p >= chunk->end && got_content) { + ucl_set_err (parser, UCL_ESYNTAX, "unfinished key", &parser->err); + return false; + } + + got_sep = got_semicolon || got_eq; + + if (!got_sep) { + /* + * Maybe we have more keys nested, so search for termination character. + * Possible choices: + * 1) key1 key2 ... keyN [:=] value <- we treat that as error + * 2) key1 ... keyN {} or [] <- we treat that as nested objects + * 3) key1 value[;,\n] <- we treat that as linear object + */ + t = p; + *next_key = false; + while (ucl_test_character (*t, UCL_CHARACTER_WHITESPACE)) { + t ++; + } + /* Check first non-space character after a key */ + if (*t != '{' && *t != '[') { + while (t < chunk->end) { + if (*t == ',' || *t == ';' || *t == '\n' || *t == '\r') { + break; + } + else if (*t == '{' || *t == '[') { + *next_key = true; + break; + } + t ++; + } + } + } + + /* Create a new object */ + nobj = ucl_object_new_full (UCL_NULL, parser->chunks->priority); + keylen = ucl_copy_or_store_ptr (parser, c, &nobj->trash_stack[UCL_TRASH_KEY], + &key, end - c, need_unescape, parser->flags & UCL_PARSER_KEY_LOWERCASE, false); + if (keylen == -1) { + ucl_object_unref (nobj); + return false; + } + else if (keylen == 0) { + ucl_set_err (parser, UCL_ESYNTAX, "empty keys are not allowed", &parser->err); + ucl_object_unref (nobj); + return false; + } + + nobj->key = key; + nobj->keylen = keylen; + + if (!ucl_parser_process_object_element (parser, nobj)) { + return false; + } + + if (ucl_escape) { + nobj->flags |= UCL_OBJECT_NEED_KEY_ESCAPE; + } + + + return true; +} + +/** + * Parse a cl string + * @param parser + * @param chunk + * @param var_expand + * @param need_unescape + * @return true if a key has been parsed + */ +static bool +ucl_parse_string_value (struct ucl_parser *parser, + struct ucl_chunk *chunk, bool *var_expand, bool *need_unescape) +{ + const unsigned char *p; + enum { + UCL_BRACE_ROUND = 0, + UCL_BRACE_SQUARE, + UCL_BRACE_FIGURE + }; + int braces[3][2] = {{0, 0}, {0, 0}, {0, 0}}; + + p = chunk->pos; + + while (p < chunk->end) { + + /* Skip pairs of figure braces */ + if (*p == '{') { + braces[UCL_BRACE_FIGURE][0] ++; + } + else if (*p == '}') { + braces[UCL_BRACE_FIGURE][1] ++; + if (braces[UCL_BRACE_FIGURE][1] <= braces[UCL_BRACE_FIGURE][0]) { + /* This is not a termination symbol, continue */ + ucl_chunk_skipc (chunk, p); + continue; + } + } + /* Skip pairs of square braces */ + else if (*p == '[') { + braces[UCL_BRACE_SQUARE][0] ++; + } + else if (*p == ']') { + braces[UCL_BRACE_SQUARE][1] ++; + if (braces[UCL_BRACE_SQUARE][1] <= braces[UCL_BRACE_SQUARE][0]) { + /* This is not a termination symbol, continue */ + ucl_chunk_skipc (chunk, p); + continue; + } + } + else if (*p == '$') { + *var_expand = true; + } + else if (*p == '\\') { + *need_unescape = true; + ucl_chunk_skipc (chunk, p); + if (p < chunk->end) { + ucl_chunk_skipc (chunk, p); + } + continue; + } + + if (ucl_lex_is_atom_end (*p) || (chunk->remain >= 2 && ucl_lex_is_comment (p[0], p[1]))) { + break; + } + ucl_chunk_skipc (chunk, p); + } + + return true; +} + +/** + * Parse multiline string ending with \n{term}\n + * @param parser + * @param chunk + * @param term + * @param term_len + * @param beg + * @param var_expand + * @return size of multiline string or 0 in case of error + */ +static int +ucl_parse_multiline_string (struct ucl_parser *parser, + struct ucl_chunk *chunk, const unsigned char *term, + int term_len, unsigned char const **beg, + bool *var_expand) +{ + const unsigned char *p, *c, *tend; + bool newline = false; + int len = 0; + + p = chunk->pos; + + c = p; + + while (p < chunk->end) { + if (newline) { + if (chunk->end - p < term_len) { + return 0; + } + else if (memcmp (p, term, term_len) == 0) { + tend = p + term_len; + if (*tend != '\n' && *tend != ';' && *tend != ',') { + /* Incomplete terminator */ + ucl_chunk_skipc (chunk, p); + continue; + } + len = p - c; + chunk->remain -= term_len; + chunk->pos = p + term_len; + chunk->column = term_len; + *beg = c; + break; + } + } + if (*p == '\n') { + newline = true; + } + else { + if (*p == '$') { + *var_expand = true; + } + newline = false; + } + ucl_chunk_skipc (chunk, p); + } + + return len; +} + +static inline ucl_object_t* +ucl_parser_get_container (struct ucl_parser *parser) +{ + ucl_object_t *t, *obj = NULL; + + if (parser == NULL || parser->stack == NULL || parser->stack->obj == NULL) { + return NULL; + } + + if (parser->stack->obj->type == UCL_ARRAY) { + /* Object must be allocated */ + obj = ucl_object_new_full (UCL_NULL, parser->chunks->priority); + t = parser->stack->obj; + + if (!ucl_array_append (t, obj)) { + ucl_object_unref (obj); + return NULL; + } + + parser->cur_obj = obj; + } + else { + /* Object has been already allocated */ + obj = parser->cur_obj; + } + + return obj; +} + +/** + * Handle value data + * @param parser + * @param chunk + * @return + */ +static bool +ucl_parse_value (struct ucl_parser *parser, struct ucl_chunk *chunk) +{ + const unsigned char *p, *c; + ucl_object_t *obj = NULL; + unsigned int stripped_spaces; + int str_len; + bool need_unescape = false, ucl_escape = false, var_expand = false; + + p = chunk->pos; + + /* Skip any spaces and comments */ + if (ucl_test_character (*p, UCL_CHARACTER_WHITESPACE_UNSAFE) || + (chunk->remain >= 2 && ucl_lex_is_comment (p[0], p[1]))) { + while (p < chunk->end && ucl_test_character (*p, UCL_CHARACTER_WHITESPACE_UNSAFE)) { + ucl_chunk_skipc (chunk, p); + } + if (!ucl_skip_comments (parser)) { + return false; + } + p = chunk->pos; + } + + while (p < chunk->end) { + c = p; + switch (*p) { + case '"': + ucl_chunk_skipc (chunk, p); + + if (!ucl_lex_json_string (parser, chunk, &need_unescape, &ucl_escape, + &var_expand)) { + return false; + } + + obj = ucl_parser_get_container (parser); + str_len = chunk->pos - c - 2; + obj->type = UCL_STRING; + if ((str_len = ucl_copy_or_store_ptr (parser, c + 1, + &obj->trash_stack[UCL_TRASH_VALUE], + &obj->value.sv, str_len, need_unescape, false, + var_expand)) == -1) { + return false; + } + obj->len = str_len; + + parser->state = UCL_STATE_AFTER_VALUE; + p = chunk->pos; + + return true; + break; + case '{': + obj = ucl_parser_get_container (parser); + /* We have a new object */ + obj = ucl_parser_add_container (obj, parser, false, parser->stack->level); + if (obj == NULL) { + return false; + } + + ucl_chunk_skipc (chunk, p); + + return true; + break; + case '[': + obj = ucl_parser_get_container (parser); + /* We have a new array */ + obj = ucl_parser_add_container (obj, parser, true, parser->stack->level); + if (obj == NULL) { + return false; + } + + ucl_chunk_skipc (chunk, p); + + return true; + break; + case ']': + /* We have the array ending */ + if (parser->stack && parser->stack->obj->type == UCL_ARRAY) { + parser->state = UCL_STATE_AFTER_VALUE; + return true; + } + else { + goto parse_string; + } + break; + case '<': + obj = ucl_parser_get_container (parser); + /* We have something like multiline value, which must be <<[A-Z]+\n */ + if (chunk->end - p > 3) { + if (memcmp (p, "<<", 2) == 0) { + p += 2; + /* We allow only uppercase characters in multiline definitions */ + while (p < chunk->end && *p >= 'A' && *p <= 'Z') { + p ++; + } + if (*p =='\n') { + /* Set chunk positions and start multiline parsing */ + c += 2; + chunk->remain -= p - c; + chunk->pos = p + 1; + chunk->column = 0; + chunk->line ++; + if ((str_len = ucl_parse_multiline_string (parser, chunk, c, + p - c, &c, &var_expand)) == 0) { + ucl_set_err (parser, UCL_ESYNTAX, + "unterminated multiline value", &parser->err); + return false; + } + + obj->type = UCL_STRING; + obj->flags |= UCL_OBJECT_MULTILINE; + if ((str_len = ucl_copy_or_store_ptr (parser, c, + &obj->trash_stack[UCL_TRASH_VALUE], + &obj->value.sv, str_len - 1, false, + false, var_expand)) == -1) { + return false; + } + obj->len = str_len; + + parser->state = UCL_STATE_AFTER_VALUE; + + return true; + } + } + } + /* Fallback to ordinary strings */ + default: +parse_string: + if (obj == NULL) { + obj = ucl_parser_get_container (parser); + } + + /* Parse atom */ + if (ucl_test_character (*p, UCL_CHARACTER_VALUE_DIGIT_START)) { + if (!ucl_lex_number (parser, chunk, obj)) { + if (parser->state == UCL_STATE_ERROR) { + return false; + } + } + else { + parser->state = UCL_STATE_AFTER_VALUE; + return true; + } + /* Fallback to normal string */ + } + + if (!ucl_parse_string_value (parser, chunk, &var_expand, + &need_unescape)) { + return false; + } + /* Cut trailing spaces */ + stripped_spaces = 0; + while (ucl_test_character (*(chunk->pos - 1 - stripped_spaces), + UCL_CHARACTER_WHITESPACE)) { + stripped_spaces ++; + } + str_len = chunk->pos - c - stripped_spaces; + if (str_len <= 0) { + ucl_set_err (parser, UCL_ESYNTAX, "string value must not be empty", + &parser->err); + return false; + } + else if (str_len == 4 && memcmp (c, "null", 4) == 0) { + obj->len = 0; + obj->type = UCL_NULL; + } + else if (!ucl_maybe_parse_boolean (obj, c, str_len)) { + obj->type = UCL_STRING; + if ((str_len = ucl_copy_or_store_ptr (parser, c, + &obj->trash_stack[UCL_TRASH_VALUE], + &obj->value.sv, str_len, need_unescape, + false, var_expand)) == -1) { + return false; + } + obj->len = str_len; + } + parser->state = UCL_STATE_AFTER_VALUE; + p = chunk->pos; + + return true; + break; + } + } + + return true; +} + +/** + * Handle after value data + * @param parser + * @param chunk + * @return + */ +static bool +ucl_parse_after_value (struct ucl_parser *parser, struct ucl_chunk *chunk) +{ + const unsigned char *p; + bool got_sep = false; + struct ucl_stack *st; + + p = chunk->pos; + + while (p < chunk->end) { + if (ucl_test_character (*p, UCL_CHARACTER_WHITESPACE)) { + /* Skip whitespaces */ + ucl_chunk_skipc (chunk, p); + } + else if (chunk->remain >= 2 && ucl_lex_is_comment (p[0], p[1])) { + /* Skip comment */ + if (!ucl_skip_comments (parser)) { + return false; + } + /* Treat comment as a separator */ + got_sep = true; + p = chunk->pos; + } + else if (ucl_test_character (*p, UCL_CHARACTER_VALUE_END)) { + if (*p == '}' || *p == ']') { + if (parser->stack == NULL) { + ucl_set_err (parser, UCL_ESYNTAX, + "end of array or object detected without corresponding start", + &parser->err); + return false; + } + if ((*p == '}' && parser->stack->obj->type == UCL_OBJECT) || + (*p == ']' && parser->stack->obj->type == UCL_ARRAY)) { + + /* Pop all nested objects from a stack */ + st = parser->stack; + parser->stack = st->next; + UCL_FREE (sizeof (struct ucl_stack), st); + + while (parser->stack != NULL) { + st = parser->stack; + if (st->next == NULL || st->next->level == st->level) { + break; + } + parser->stack = st->next; + UCL_FREE (sizeof (struct ucl_stack), st); + } + } + else { + ucl_set_err (parser, UCL_ESYNTAX, + "unexpected terminating symbol detected", + &parser->err); + return false; + } + + if (parser->stack == NULL) { + /* Ignore everything after a top object */ + return true; + } + else { + ucl_chunk_skipc (chunk, p); + } + got_sep = true; + } + else { + /* Got a separator */ + got_sep = true; + ucl_chunk_skipc (chunk, p); + } + } + else { + /* Anything else */ + if (!got_sep) { + ucl_set_err (parser, UCL_ESYNTAX, "delimiter is missing", + &parser->err); + return false; + } + return true; + } + } + + return true; +} + +/** + * Handle macro data + * @param parser + * @param chunk + * @param marco + * @param macro_start + * @param macro_len + * @return + */ +static bool +ucl_parse_macro_value (struct ucl_parser *parser, + struct ucl_chunk *chunk, struct ucl_macro *macro, + unsigned char const **macro_start, size_t *macro_len) +{ + const unsigned char *p, *c; + bool need_unescape = false, ucl_escape = false, var_expand = false; + + p = chunk->pos; + + switch (*p) { + case '"': + /* We have macro value encoded in quotes */ + c = p; + ucl_chunk_skipc (chunk, p); + if (!ucl_lex_json_string (parser, chunk, &need_unescape, &ucl_escape, &var_expand)) { + return false; + } + + *macro_start = c + 1; + *macro_len = chunk->pos - c - 2; + p = chunk->pos; + break; + case '{': + /* We got a multiline macro body */ + ucl_chunk_skipc (chunk, p); + /* Skip spaces at the beginning */ + while (p < chunk->end) { + if (ucl_test_character (*p, UCL_CHARACTER_WHITESPACE_UNSAFE)) { + ucl_chunk_skipc (chunk, p); + } + else { + break; + } + } + c = p; + while (p < chunk->end) { + if (*p == '}') { + break; + } + ucl_chunk_skipc (chunk, p); + } + *macro_start = c; + *macro_len = p - c; + ucl_chunk_skipc (chunk, p); + break; + default: + /* Macro is not enclosed in quotes or braces */ + c = p; + while (p < chunk->end) { + if (ucl_lex_is_atom_end (*p)) { + break; + } + ucl_chunk_skipc (chunk, p); + } + *macro_start = c; + *macro_len = p - c; + break; + } + + /* We are at the end of a macro */ + /* Skip ';' and space characters and return to previous state */ + while (p < chunk->end) { + if (!ucl_test_character (*p, UCL_CHARACTER_WHITESPACE_UNSAFE) && *p != ';') { + break; + } + ucl_chunk_skipc (chunk, p); + } + return true; +} + +/** + * Parse macro arguments as UCL object + * @param parser parser structure + * @param chunk the current data chunk + * @return + */ +static ucl_object_t * +ucl_parse_macro_arguments (struct ucl_parser *parser, + struct ucl_chunk *chunk) +{ + ucl_object_t *res = NULL; + struct ucl_parser *params_parser; + int obraces = 1, ebraces = 0, state = 0; + const unsigned char *p, *c; + size_t args_len = 0; + struct ucl_parser_saved_state saved; + + saved.column = chunk->column; + saved.line = chunk->line; + saved.pos = chunk->pos; + saved.remain = chunk->remain; + p = chunk->pos; + + if (*p != '(' || chunk->remain < 2) { + return NULL; + } + + /* Set begin and start */ + ucl_chunk_skipc (chunk, p); + c = p; + + while ((p) < (chunk)->end) { + switch (state) { + case 0: + /* Parse symbols and check for '(', ')' and '"' */ + if (*p == '(') { + obraces ++; + } + else if (*p == ')') { + ebraces ++; + } + else if (*p == '"') { + state = 1; + } + /* Check pairing */ + if (obraces == ebraces) { + state = 99; + } + else { + args_len ++; + } + /* Check overflow */ + if (chunk->remain == 0) { + goto restore_chunk; + } + ucl_chunk_skipc (chunk, p); + break; + case 1: + /* We have quote character, so skip all but quotes */ + if (*p == '"' && *(p - 1) != '\\') { + state = 0; + } + if (chunk->remain == 0) { + goto restore_chunk; + } + args_len ++; + ucl_chunk_skipc (chunk, p); + break; + case 99: + /* + * We have read the full body of arguments, so we need to parse and set + * object from that + */ + params_parser = ucl_parser_new (parser->flags); + if (!ucl_parser_add_chunk (params_parser, c, args_len)) { + ucl_set_err (parser, UCL_ESYNTAX, "macro arguments parsing error", + &parser->err); + } + else { + res = ucl_parser_get_object (params_parser); + } + ucl_parser_free (params_parser); + + return res; + + break; + } + } + + return res; + +restore_chunk: + chunk->column = saved.column; + chunk->line = saved.line; + chunk->pos = saved.pos; + chunk->remain = saved.remain; + + return NULL; +} + +#define SKIP_SPACES_COMMENTS(parser, chunk, p) do { \ + while ((p) < (chunk)->end) { \ + if (!ucl_test_character (*(p), UCL_CHARACTER_WHITESPACE_UNSAFE)) { \ + if ((chunk)->remain >= 2 && ucl_lex_is_comment ((p)[0], (p)[1])) { \ + if (!ucl_skip_comments (parser)) { \ + return false; \ + } \ + p = (chunk)->pos; \ + } \ + break; \ + } \ + ucl_chunk_skipc (chunk, p); \ + } \ +} while(0) + +/** + * Handle the main states of rcl parser + * @param parser parser structure + * @return true if chunk has been parsed and false in case of error + */ +static bool +ucl_state_machine (struct ucl_parser *parser) +{ + ucl_object_t *obj, *macro_args; + struct ucl_chunk *chunk = parser->chunks; + const unsigned char *p, *c = NULL, *macro_start = NULL; + unsigned char *macro_escaped; + size_t macro_len = 0; + struct ucl_macro *macro = NULL; + bool next_key = false, end_of_object = false, ret; + + if (parser->top_obj == NULL) { + parser->state = UCL_STATE_INIT; + } + + p = chunk->pos; + while (chunk->pos < chunk->end) { + switch (parser->state) { + case UCL_STATE_INIT: + /* + * At the init state we can either go to the parse array or object + * if we got [ or { correspondingly or can just treat new data as + * a key of newly created object + */ + if (!ucl_skip_comments (parser)) { + parser->prev_state = parser->state; + parser->state = UCL_STATE_ERROR; + return false; + } + else { + /* Skip any spaces */ + while (p < chunk->end && ucl_test_character (*p, + UCL_CHARACTER_WHITESPACE_UNSAFE)) { + ucl_chunk_skipc (chunk, p); + } + + p = chunk->pos; + + if (*p == '[') { + parser->state = UCL_STATE_VALUE; + ucl_chunk_skipc (chunk, p); + } + else { + parser->state = UCL_STATE_KEY; + if (*p == '{') { + ucl_chunk_skipc (chunk, p); + } + } + + if (parser->top_obj == NULL) { + if (parser->state == UCL_STATE_VALUE) { + obj = ucl_parser_add_container (NULL, parser, true, 0); + } + else { + obj = ucl_parser_add_container (NULL, parser, false, 0); + } + + if (obj == NULL) { + return false; + } + + parser->top_obj = obj; + parser->cur_obj = obj; + } + + } + break; + case UCL_STATE_KEY: + /* Skip any spaces */ + while (p < chunk->end && ucl_test_character (*p, UCL_CHARACTER_WHITESPACE_UNSAFE)) { + ucl_chunk_skipc (chunk, p); + } + if (*p == '}') { + /* We have the end of an object */ + parser->state = UCL_STATE_AFTER_VALUE; + continue; + } + if (parser->stack == NULL) { + /* No objects are on stack, but we want to parse a key */ + ucl_set_err (parser, UCL_ESYNTAX, "top object is finished but the parser " + "expects a key", &parser->err); + parser->prev_state = parser->state; + parser->state = UCL_STATE_ERROR; + return false; + } + if (!ucl_parse_key (parser, chunk, &next_key, &end_of_object)) { + parser->prev_state = parser->state; + parser->state = UCL_STATE_ERROR; + return false; + } + if (end_of_object) { + p = chunk->pos; + parser->state = UCL_STATE_AFTER_VALUE; + continue; + } + else if (parser->state != UCL_STATE_MACRO_NAME) { + if (next_key && parser->stack->obj->type == UCL_OBJECT) { + /* Parse more keys and nest objects accordingly */ + obj = ucl_parser_add_container (parser->cur_obj, parser, false, + parser->stack->level + 1); + if (obj == NULL) { + return false; + } + } + else { + parser->state = UCL_STATE_VALUE; + } + } + else { + c = chunk->pos; + } + p = chunk->pos; + break; + case UCL_STATE_VALUE: + /* We need to check what we do have */ + if (!ucl_parse_value (parser, chunk)) { + parser->prev_state = parser->state; + parser->state = UCL_STATE_ERROR; + return false; + } + /* State is set in ucl_parse_value call */ + p = chunk->pos; + break; + case UCL_STATE_AFTER_VALUE: + if (!ucl_parse_after_value (parser, chunk)) { + parser->prev_state = parser->state; + parser->state = UCL_STATE_ERROR; + return false; + } + + if (parser->stack != NULL) { + if (parser->stack->obj->type == UCL_OBJECT) { + parser->state = UCL_STATE_KEY; + } + else { + /* Array */ + parser->state = UCL_STATE_VALUE; + } + } + else { + /* Skip everything at the end */ + return true; + } + p = chunk->pos; + break; + case UCL_STATE_MACRO_NAME: + if (!ucl_test_character (*p, UCL_CHARACTER_WHITESPACE_UNSAFE) && + *p != '(') { + ucl_chunk_skipc (chunk, p); + } + else { + if (p - c > 0) { + /* We got macro name */ + macro_len = (size_t) (p - c); + HASH_FIND (hh, parser->macroes, c, macro_len, macro); + if (macro == NULL) { + ucl_create_err (&parser->err, + "error on line %d at column %d: " + "unknown macro: '%.*s', character: '%c'", + chunk->line, + chunk->column, + (int) (p - c), + c, + *chunk->pos); + parser->state = UCL_STATE_ERROR; + return false; + } + /* Now we need to skip all spaces */ + SKIP_SPACES_COMMENTS(parser, chunk, p); + parser->state = UCL_STATE_MACRO; + } + else { + /* We have invalid macro name */ + ucl_create_err (&parser->err, + "error on line %d at column %d: invalid macro name", + chunk->line, + chunk->column); + parser->state = UCL_STATE_ERROR; + return false; + } + } + break; + case UCL_STATE_MACRO: + if (*chunk->pos == '(') { + macro_args = ucl_parse_macro_arguments (parser, chunk); + p = chunk->pos; + if (macro_args) { + SKIP_SPACES_COMMENTS(parser, chunk, p); + } + } + else { + macro_args = NULL; + } + if (!ucl_parse_macro_value (parser, chunk, macro, + ¯o_start, ¯o_len)) { + parser->prev_state = parser->state; + parser->state = UCL_STATE_ERROR; + return false; + } + macro_len = ucl_expand_variable (parser, ¯o_escaped, + macro_start, macro_len); + parser->state = parser->prev_state; + if (macro_escaped == NULL) { + if (macro->is_context) { + ret = macro->h.context_handler (macro_start, macro_len, + macro_args, + parser->top_obj, + macro->ud); + } + else { + ret = macro->h.handler (macro_start, macro_len, macro_args, + macro->ud); + } + } + else { + if (macro->is_context) { + ret = macro->h.context_handler (macro_escaped, macro_len, + macro_args, + parser->top_obj, + macro->ud); + } + else { + ret = macro->h.handler (macro_escaped, macro_len, macro_args, + macro->ud); + } + + UCL_FREE (macro_len + 1, macro_escaped); + } + + /* + * Chunk can be modified within macro handler + */ + chunk = parser->chunks; + p = chunk->pos; + if (macro_args) { + ucl_object_unref (macro_args); + } + if (!ret) { + return false; + } + break; + default: + /* TODO: add all states */ + ucl_set_err (parser, UCL_EINTERNAL, + "internal error: parser is in an unknown state", &parser->err); + parser->state = UCL_STATE_ERROR; + return false; + } + } + + return true; +} + +struct ucl_parser* +ucl_parser_new (int flags) +{ + struct ucl_parser *new; + + new = UCL_ALLOC (sizeof (struct ucl_parser)); + if (new == NULL) { + return NULL; + } + + memset (new, 0, sizeof (struct ucl_parser)); + + ucl_parser_register_macro (new, "include", ucl_include_handler, new); + ucl_parser_register_macro (new, "try_include", ucl_try_include_handler, new); + ucl_parser_register_macro (new, "includes", ucl_includes_handler, new); + ucl_parser_register_macro (new, "priority", ucl_priority_handler, new); + ucl_parser_register_macro (new, "load", ucl_load_handler, new); + ucl_parser_register_context_macro (new, "inherit", ucl_inherit_handler, new); + + new->flags = flags; + new->includepaths = NULL; + + /* Initial assumption about filevars */ + ucl_parser_set_filevars (new, NULL, false); + + return new; +} + +bool +ucl_parser_set_default_priority (struct ucl_parser *parser, unsigned prio) +{ + if (parser == NULL) { + return false; + } + + parser->default_priority = prio; + + return true; +} + +void +ucl_parser_register_macro (struct ucl_parser *parser, const char *macro, + ucl_macro_handler handler, void* ud) +{ + struct ucl_macro *new; + + if (macro == NULL || handler == NULL) { + return; + } + + new = UCL_ALLOC (sizeof (struct ucl_macro)); + if (new == NULL) { + return; + } + + memset (new, 0, sizeof (struct ucl_macro)); + new->h.handler = handler; + new->name = strdup (macro); + new->ud = ud; + HASH_ADD_KEYPTR (hh, parser->macroes, new->name, strlen (new->name), new); +} + +void +ucl_parser_register_context_macro (struct ucl_parser *parser, const char *macro, + ucl_context_macro_handler handler, void* ud) +{ + struct ucl_macro *new; + + if (macro == NULL || handler == NULL) { + return; + } + + new = UCL_ALLOC (sizeof (struct ucl_macro)); + if (new == NULL) { + return; + } + + memset (new, 0, sizeof (struct ucl_macro)); + new->h.context_handler = handler; + new->name = strdup (macro); + new->ud = ud; + new->is_context = true; + HASH_ADD_KEYPTR (hh, parser->macroes, new->name, strlen (new->name), new); +} + +void +ucl_parser_register_variable (struct ucl_parser *parser, const char *var, + const char *value) +{ + struct ucl_variable *new = NULL, *cur; + + if (var == NULL) { + return; + } + + /* Find whether a variable already exists */ + LL_FOREACH (parser->variables, cur) { + if (strcmp (cur->var, var) == 0) { + new = cur; + break; + } + } + + if (value == NULL) { + + if (new != NULL) { + /* Remove variable */ + DL_DELETE (parser->variables, new); + free (new->var); + free (new->value); + UCL_FREE (sizeof (struct ucl_variable), new); + } + else { + /* Do nothing */ + return; + } + } + else { + if (new == NULL) { + new = UCL_ALLOC (sizeof (struct ucl_variable)); + if (new == NULL) { + return; + } + memset (new, 0, sizeof (struct ucl_variable)); + new->var = strdup (var); + new->var_len = strlen (var); + new->value = strdup (value); + new->value_len = strlen (value); + + DL_APPEND (parser->variables, new); + } + else { + free (new->value); + new->value = strdup (value); + new->value_len = strlen (value); + } + } +} + +void +ucl_parser_set_variables_handler (struct ucl_parser *parser, + ucl_variable_handler handler, void *ud) +{ + parser->var_handler = handler; + parser->var_data = ud; +} + +bool +ucl_parser_add_chunk_full (struct ucl_parser *parser, const unsigned char *data, + size_t len, unsigned priority, enum ucl_duplicate_strategy strat, + enum ucl_parse_type parse_type) +{ + struct ucl_chunk *chunk; + + if (parser == NULL) { + return false; + } + + if (data == NULL) { + ucl_create_err (&parser->err, "invalid chunk added"); + return false; + } + if (len == 0) { + parser->top_obj = ucl_object_new_full (UCL_OBJECT, priority); + return true; + } + if (parser->state != UCL_STATE_ERROR) { + chunk = UCL_ALLOC (sizeof (struct ucl_chunk)); + if (chunk == NULL) { + ucl_create_err (&parser->err, "cannot allocate chunk structure"); + return false; + } + chunk->begin = data; + chunk->remain = len; + chunk->pos = chunk->begin; + chunk->end = chunk->begin + len; + chunk->line = 1; + chunk->column = 0; + chunk->priority = priority; + chunk->strategy = strat; + chunk->parse_type = parse_type; + LL_PREPEND (parser->chunks, chunk); + parser->recursion ++; + + if (parser->recursion > UCL_MAX_RECURSION) { + ucl_create_err (&parser->err, "maximum include nesting limit is reached: %d", + parser->recursion); + return false; + } + + switch (parse_type) { + default: + case UCL_PARSE_UCL: + return ucl_state_machine (parser); + case UCL_PARSE_MSGPACK: + return ucl_parse_msgpack (parser); + } + } + + ucl_create_err (&parser->err, "a parser is in an invalid state"); + + return false; +} + +bool +ucl_parser_add_chunk_priority (struct ucl_parser *parser, + const unsigned char *data, size_t len, unsigned priority) +{ + /* We dereference parser, so this check is essential */ + if (parser == NULL) { + return false; + } + + return ucl_parser_add_chunk_full (parser, data, len, + priority, UCL_DUPLICATE_APPEND, UCL_PARSE_UCL); +} + +bool +ucl_parser_add_chunk (struct ucl_parser *parser, const unsigned char *data, + size_t len) +{ + if (parser == NULL) { + return false; + } + + return ucl_parser_add_chunk_full (parser, data, len, + parser->default_priority, UCL_DUPLICATE_APPEND, UCL_PARSE_UCL); +} + +bool +ucl_parser_add_string_priority (struct ucl_parser *parser, const char *data, + size_t len, unsigned priority) +{ + if (data == NULL) { + ucl_create_err (&parser->err, "invalid string added"); + return false; + } + if (len == 0) { + len = strlen (data); + } + + return ucl_parser_add_chunk_priority (parser, + (const unsigned char *)data, len, priority); +} + +bool +ucl_parser_add_string (struct ucl_parser *parser, const char *data, + size_t len) +{ + if (parser == NULL) { + return false; + } + + return ucl_parser_add_string_priority (parser, + (const unsigned char *)data, len, parser->default_priority); +} + +bool +ucl_set_include_path (struct ucl_parser *parser, ucl_object_t *paths) +{ + if (parser == NULL || paths == NULL) { + return false; + } + + if (parser->includepaths == NULL) { + parser->includepaths = ucl_object_copy (paths); + } + else { + ucl_object_unref (parser->includepaths); + parser->includepaths = ucl_object_copy (paths); + } + + if (parser->includepaths == NULL) { + return false; + } + + return true; +} diff --git a/contrib/libucl/src/ucl_schema.c b/contrib/libucl/src/ucl_schema.c new file mode 100644 index 0000000..834b62a --- /dev/null +++ b/contrib/libucl/src/ucl_schema.c @@ -0,0 +1,1015 @@ +/* + * Copyright (c) 2014, Vsevolod Stakhov + * + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY AUTHOR ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include "ucl.h" +#include "ucl_internal.h" +#include "tree.h" +#include "utlist.h" +#ifdef HAVE_STDARG_H +#include <stdarg.h> +#endif +#ifdef HAVE_STDIO_H +#include <stdio.h> +#endif +#ifdef HAVE_REGEX_H +#include <regex.h> +#endif +#ifdef HAVE_MATH_H +#include <math.h> +#endif + +static bool ucl_schema_validate (const ucl_object_t *schema, + const ucl_object_t *obj, bool try_array, + struct ucl_schema_error *err, + const ucl_object_t *root); + +static bool +ucl_string_to_type (const char *input, ucl_type_t *res) +{ + if (strcasecmp (input, "object") == 0) { + *res = UCL_OBJECT; + } + else if (strcasecmp (input, "array") == 0) { + *res = UCL_ARRAY; + } + else if (strcasecmp (input, "integer") == 0) { + *res = UCL_INT; + } + else if (strcasecmp (input, "number") == 0) { + *res = UCL_FLOAT; + } + else if (strcasecmp (input, "string") == 0) { + *res = UCL_STRING; + } + else if (strcasecmp (input, "boolean") == 0) { + *res = UCL_BOOLEAN; + } + else if (strcasecmp (input, "null") == 0) { + *res = UCL_NULL; + } + else { + return false; + } + + return true; +} + +static const char * +ucl_object_type_to_string (ucl_type_t type) +{ + const char *res = "unknown"; + + switch (type) { + case UCL_OBJECT: + res = "object"; + break; + case UCL_ARRAY: + res = "array"; + break; + case UCL_INT: + res = "integer"; + break; + case UCL_FLOAT: + case UCL_TIME: + res = "number"; + break; + case UCL_STRING: + res = "string"; + break; + case UCL_BOOLEAN: + res = "boolean"; + break; + case UCL_NULL: + case UCL_USERDATA: + res = "null"; + break; + } + + return res; +} + +/* + * Create validation error + */ +static void +ucl_schema_create_error (struct ucl_schema_error *err, + enum ucl_schema_error_code code, const ucl_object_t *obj, + const char *fmt, ...) +{ + va_list va; + + if (err != NULL) { + err->code = code; + err->obj = obj; + va_start (va, fmt); + vsnprintf (err->msg, sizeof (err->msg), fmt, va); + va_end (va); + } +} + +/* + * Check whether we have a pattern specified + */ +static const ucl_object_t * +ucl_schema_test_pattern (const ucl_object_t *obj, const char *pattern) +{ + const ucl_object_t *res = NULL; +#ifdef HAVE_REGEX_H + regex_t reg; + const ucl_object_t *elt; + ucl_object_iter_t iter = NULL; + + if (regcomp (®, pattern, REG_EXTENDED | REG_NOSUB) == 0) { + while ((elt = ucl_iterate_object (obj, &iter, true)) != NULL) { + if (regexec (®, ucl_object_key (elt), 0, NULL, 0) == 0) { + res = elt; + break; + } + } + regfree (®); + } +#endif + return res; +} + +/* + * Check dependencies for an object + */ +static bool +ucl_schema_validate_dependencies (const ucl_object_t *deps, + const ucl_object_t *obj, struct ucl_schema_error *err, + const ucl_object_t *root) +{ + const ucl_object_t *elt, *cur, *cur_dep; + ucl_object_iter_t iter = NULL, piter; + bool ret = true; + + while (ret && (cur = ucl_iterate_object (deps, &iter, true)) != NULL) { + elt = ucl_object_find_key (obj, ucl_object_key (cur)); + if (elt != NULL) { + /* Need to check dependencies */ + if (cur->type == UCL_ARRAY) { + piter = NULL; + while (ret && (cur_dep = ucl_iterate_object (cur, &piter, true)) != NULL) { + if (ucl_object_find_key (obj, ucl_object_tostring (cur_dep)) == NULL) { + ucl_schema_create_error (err, UCL_SCHEMA_MISSING_DEPENDENCY, elt, + "dependency %s is missing for key %s", + ucl_object_tostring (cur_dep), ucl_object_key (cur)); + ret = false; + break; + } + } + } + else if (cur->type == UCL_OBJECT) { + ret = ucl_schema_validate (cur, obj, true, err, root); + } + } + } + + return ret; +} + +/* + * Validate object + */ +static bool +ucl_schema_validate_object (const ucl_object_t *schema, + const ucl_object_t *obj, struct ucl_schema_error *err, + const ucl_object_t *root) +{ + const ucl_object_t *elt, *prop, *found, *additional_schema = NULL, + *required = NULL, *pat, *pelt; + ucl_object_iter_t iter = NULL, piter = NULL; + bool ret = true, allow_additional = true; + int64_t minmax; + + while (ret && (elt = ucl_iterate_object (schema, &iter, true)) != NULL) { + if (elt->type == UCL_OBJECT && + strcmp (ucl_object_key (elt), "properties") == 0) { + piter = NULL; + while (ret && (prop = ucl_iterate_object (elt, &piter, true)) != NULL) { + found = ucl_object_find_key (obj, ucl_object_key (prop)); + if (found) { + ret = ucl_schema_validate (prop, found, true, err, root); + } + } + } + else if (strcmp (ucl_object_key (elt), "additionalProperties") == 0) { + if (elt->type == UCL_BOOLEAN) { + if (!ucl_object_toboolean (elt)) { + /* Deny additional fields completely */ + allow_additional = false; + } + } + else if (elt->type == UCL_OBJECT) { + /* Define validator for additional fields */ + additional_schema = elt; + } + else { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, elt, + "additionalProperties attribute is invalid in schema"); + ret = false; + break; + } + } + else if (strcmp (ucl_object_key (elt), "required") == 0) { + if (elt->type == UCL_ARRAY) { + required = elt; + } + else { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, elt, + "required attribute is invalid in schema"); + ret = false; + break; + } + } + else if (strcmp (ucl_object_key (elt), "minProperties") == 0 + && ucl_object_toint_safe (elt, &minmax)) { + if (obj->len < minmax) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "object has not enough properties: %u, minimum is: %u", + obj->len, (unsigned)minmax); + ret = false; + break; + } + } + else if (strcmp (ucl_object_key (elt), "maxProperties") == 0 + && ucl_object_toint_safe (elt, &minmax)) { + if (obj->len > minmax) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "object has too many properties: %u, maximum is: %u", + obj->len, (unsigned)minmax); + ret = false; + break; + } + } + else if (strcmp (ucl_object_key (elt), "patternProperties") == 0) { + piter = NULL; + while (ret && (prop = ucl_iterate_object (elt, &piter, true)) != NULL) { + found = ucl_schema_test_pattern (obj, ucl_object_key (prop)); + if (found) { + ret = ucl_schema_validate (prop, found, true, err, root); + } + } + } + else if (elt->type == UCL_OBJECT && + strcmp (ucl_object_key (elt), "dependencies") == 0) { + ret = ucl_schema_validate_dependencies (elt, obj, err, root); + } + } + + if (ret) { + /* Additional properties */ + if (!allow_additional || additional_schema != NULL) { + /* Check if we have exactly the same properties in schema and object */ + iter = NULL; + prop = ucl_object_find_key (schema, "properties"); + while ((elt = ucl_iterate_object (obj, &iter, true)) != NULL) { + found = ucl_object_find_key (prop, ucl_object_key (elt)); + if (found == NULL) { + /* Try patternProperties */ + piter = NULL; + pat = ucl_object_find_key (schema, "patternProperties"); + while ((pelt = ucl_iterate_object (pat, &piter, true)) != NULL) { + found = ucl_schema_test_pattern (obj, ucl_object_key (pelt)); + if (found != NULL) { + break; + } + } + } + if (found == NULL) { + if (!allow_additional) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "object has non-allowed property %s", + ucl_object_key (elt)); + ret = false; + break; + } + else if (additional_schema != NULL) { + if (!ucl_schema_validate (additional_schema, elt, true, err, root)) { + ret = false; + break; + } + } + } + } + } + /* Required properties */ + if (required != NULL) { + iter = NULL; + while ((elt = ucl_iterate_object (required, &iter, true)) != NULL) { + if (ucl_object_find_key (obj, ucl_object_tostring (elt)) == NULL) { + ucl_schema_create_error (err, UCL_SCHEMA_MISSING_PROPERTY, obj, + "object has missing property %s", + ucl_object_tostring (elt)); + ret = false; + break; + } + } + } + } + + + return ret; +} + +static bool +ucl_schema_validate_number (const ucl_object_t *schema, + const ucl_object_t *obj, struct ucl_schema_error *err) +{ + const ucl_object_t *elt, *test; + ucl_object_iter_t iter = NULL; + bool ret = true, exclusive = false; + double constraint, val; + const double alpha = 1e-16; + + while (ret && (elt = ucl_iterate_object (schema, &iter, true)) != NULL) { + if ((elt->type == UCL_FLOAT || elt->type == UCL_INT) && + strcmp (ucl_object_key (elt), "multipleOf") == 0) { + constraint = ucl_object_todouble (elt); + if (constraint <= 0) { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, elt, + "multipleOf must be greater than zero"); + ret = false; + break; + } + val = ucl_object_todouble (obj); + if (fabs (remainder (val, constraint)) > alpha) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "number %.4f is not multiple of %.4f, remainder is %.7f", + val, constraint); + ret = false; + break; + } + } + else if ((elt->type == UCL_FLOAT || elt->type == UCL_INT) && + strcmp (ucl_object_key (elt), "maximum") == 0) { + constraint = ucl_object_todouble (elt); + test = ucl_object_find_key (schema, "exclusiveMaximum"); + if (test && test->type == UCL_BOOLEAN) { + exclusive = ucl_object_toboolean (test); + } + val = ucl_object_todouble (obj); + if (val > constraint || (exclusive && val >= constraint)) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "number is too big: %.3f, maximum is: %.3f", + val, constraint); + ret = false; + break; + } + } + else if ((elt->type == UCL_FLOAT || elt->type == UCL_INT) && + strcmp (ucl_object_key (elt), "minimum") == 0) { + constraint = ucl_object_todouble (elt); + test = ucl_object_find_key (schema, "exclusiveMinimum"); + if (test && test->type == UCL_BOOLEAN) { + exclusive = ucl_object_toboolean (test); + } + val = ucl_object_todouble (obj); + if (val < constraint || (exclusive && val <= constraint)) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "number is too small: %.3f, minimum is: %.3f", + val, constraint); + ret = false; + break; + } + } + } + + return ret; +} + +static bool +ucl_schema_validate_string (const ucl_object_t *schema, + const ucl_object_t *obj, struct ucl_schema_error *err) +{ + const ucl_object_t *elt; + ucl_object_iter_t iter = NULL; + bool ret = true; + int64_t constraint; +#ifdef HAVE_REGEX_H + regex_t re; +#endif + + while (ret && (elt = ucl_iterate_object (schema, &iter, true)) != NULL) { + if (elt->type == UCL_INT && + strcmp (ucl_object_key (elt), "maxLength") == 0) { + constraint = ucl_object_toint (elt); + if (obj->len > constraint) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "string is too big: %.3f, maximum is: %.3f", + obj->len, constraint); + ret = false; + break; + } + } + else if (elt->type == UCL_INT && + strcmp (ucl_object_key (elt), "minLength") == 0) { + constraint = ucl_object_toint (elt); + if (obj->len < constraint) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "string is too short: %.3f, minimum is: %.3f", + obj->len, constraint); + ret = false; + break; + } + } +#ifdef HAVE_REGEX_H + else if (elt->type == UCL_STRING && + strcmp (ucl_object_key (elt), "pattern") == 0) { + if (regcomp (&re, ucl_object_tostring (elt), + REG_EXTENDED | REG_NOSUB) != 0) { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, elt, + "cannot compile pattern %s", ucl_object_tostring (elt)); + ret = false; + break; + } + if (regexec (&re, ucl_object_tostring (obj), 0, NULL, 0) != 0) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "string doesn't match regexp %s", + ucl_object_tostring (elt)); + ret = false; + } + regfree (&re); + } +#endif + } + + return ret; +} + +struct ucl_compare_node { + const ucl_object_t *obj; + TREE_ENTRY(ucl_compare_node) link; + struct ucl_compare_node *next; +}; + +typedef TREE_HEAD(_tree, ucl_compare_node) ucl_compare_tree_t; + +TREE_DEFINE(ucl_compare_node, link) + +static int +ucl_schema_elt_compare (struct ucl_compare_node *n1, struct ucl_compare_node *n2) +{ + const ucl_object_t *o1 = n1->obj, *o2 = n2->obj; + + return ucl_object_compare (o1, o2); +} + +static bool +ucl_schema_array_is_unique (const ucl_object_t *obj, struct ucl_schema_error *err) +{ + ucl_compare_tree_t tree = TREE_INITIALIZER (ucl_schema_elt_compare); + ucl_object_iter_t iter = NULL; + const ucl_object_t *elt; + struct ucl_compare_node *node, test, *nodes = NULL, *tmp; + bool ret = true; + + while ((elt = ucl_iterate_object (obj, &iter, true)) != NULL) { + test.obj = elt; + node = TREE_FIND (&tree, ucl_compare_node, link, &test); + if (node != NULL) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, elt, + "duplicate values detected while uniqueItems is true"); + ret = false; + break; + } + node = calloc (1, sizeof (*node)); + if (node == NULL) { + ucl_schema_create_error (err, UCL_SCHEMA_UNKNOWN, elt, + "cannot allocate tree node"); + ret = false; + break; + } + node->obj = elt; + TREE_INSERT (&tree, ucl_compare_node, link, node); + LL_PREPEND (nodes, node); + } + + LL_FOREACH_SAFE (nodes, node, tmp) { + free (node); + } + + return ret; +} + +static bool +ucl_schema_validate_array (const ucl_object_t *schema, + const ucl_object_t *obj, struct ucl_schema_error *err, + const ucl_object_t *root) +{ + const ucl_object_t *elt, *it, *found, *additional_schema = NULL, + *first_unvalidated = NULL; + ucl_object_iter_t iter = NULL, piter = NULL; + bool ret = true, allow_additional = true, need_unique = false; + int64_t minmax; + unsigned int idx = 0; + + while (ret && (elt = ucl_iterate_object (schema, &iter, true)) != NULL) { + if (strcmp (ucl_object_key (elt), "items") == 0) { + if (elt->type == UCL_ARRAY) { + found = ucl_array_head (obj); + while (ret && (it = ucl_iterate_object (elt, &piter, true)) != NULL) { + if (found) { + ret = ucl_schema_validate (it, found, false, err, root); + found = ucl_array_find_index (obj, ++idx); + } + } + if (found != NULL) { + /* The first element that is not validated */ + first_unvalidated = found; + } + } + else if (elt->type == UCL_OBJECT) { + /* Validate all items using the specified schema */ + while (ret && (it = ucl_iterate_object (obj, &piter, true)) != NULL) { + ret = ucl_schema_validate (elt, it, false, err, root); + } + } + else { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, elt, + "items attribute is invalid in schema"); + ret = false; + break; + } + } + else if (strcmp (ucl_object_key (elt), "additionalItems") == 0) { + if (elt->type == UCL_BOOLEAN) { + if (!ucl_object_toboolean (elt)) { + /* Deny additional fields completely */ + allow_additional = false; + } + } + else if (elt->type == UCL_OBJECT) { + /* Define validator for additional fields */ + additional_schema = elt; + } + else { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, elt, + "additionalItems attribute is invalid in schema"); + ret = false; + break; + } + } + else if (elt->type == UCL_BOOLEAN && + strcmp (ucl_object_key (elt), "uniqueItems") == 0) { + need_unique = ucl_object_toboolean (elt); + } + else if (strcmp (ucl_object_key (elt), "minItems") == 0 + && ucl_object_toint_safe (elt, &minmax)) { + if (obj->len < minmax) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "array has not enough items: %u, minimum is: %u", + obj->len, (unsigned)minmax); + ret = false; + break; + } + } + else if (strcmp (ucl_object_key (elt), "maxItems") == 0 + && ucl_object_toint_safe (elt, &minmax)) { + if (obj->len > minmax) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "array has too many items: %u, maximum is: %u", + obj->len, (unsigned)minmax); + ret = false; + break; + } + } + } + + if (ret) { + /* Additional properties */ + if (!allow_additional || additional_schema != NULL) { + if (first_unvalidated != NULL) { + if (!allow_additional) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "array has undefined item"); + ret = false; + } + else if (additional_schema != NULL) { + elt = ucl_array_find_index (obj, idx); + while (elt) { + if (!ucl_schema_validate (additional_schema, elt, false, + err, root)) { + ret = false; + break; + } + elt = ucl_array_find_index (obj, idx ++); + } + } + } + } + /* Required properties */ + if (ret && need_unique) { + ret = ucl_schema_array_is_unique (obj, err); + } + } + + return ret; +} + +/* + * Returns whether this object is allowed for this type + */ +static bool +ucl_schema_type_is_allowed (const ucl_object_t *type, const ucl_object_t *obj, + struct ucl_schema_error *err) +{ + ucl_object_iter_t iter = NULL; + const ucl_object_t *elt; + const char *type_str; + ucl_type_t t; + + if (type == NULL) { + /* Any type is allowed */ + return true; + } + + if (type->type == UCL_ARRAY) { + /* One of allowed types */ + while ((elt = ucl_iterate_object (type, &iter, true)) != NULL) { + if (ucl_schema_type_is_allowed (elt, obj, err)) { + return true; + } + } + } + else if (type->type == UCL_STRING) { + type_str = ucl_object_tostring (type); + if (!ucl_string_to_type (type_str, &t)) { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, type, + "Type attribute is invalid in schema"); + return false; + } + if (obj->type != t) { + /* Some types are actually compatible */ + if (obj->type == UCL_TIME && t == UCL_FLOAT) { + return true; + } + else if (obj->type == UCL_INT && t == UCL_FLOAT) { + return true; + } + else { + ucl_schema_create_error (err, UCL_SCHEMA_TYPE_MISMATCH, obj, + "Invalid type of %s, expected %s", + ucl_object_type_to_string (obj->type), + ucl_object_type_to_string (t)); + } + } + else { + /* Types are equal */ + return true; + } + } + + return false; +} + +/* + * Check if object is equal to one of elements of enum + */ +static bool +ucl_schema_validate_enum (const ucl_object_t *en, const ucl_object_t *obj, + struct ucl_schema_error *err) +{ + ucl_object_iter_t iter = NULL; + const ucl_object_t *elt; + bool ret = false; + + while ((elt = ucl_iterate_object (en, &iter, true)) != NULL) { + if (ucl_object_compare (elt, obj) == 0) { + ret = true; + break; + } + } + + if (!ret) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "object is not one of enumerated patterns"); + } + + return ret; +} + + +/* + * Check a single ref component + */ +static const ucl_object_t * +ucl_schema_resolve_ref_component (const ucl_object_t *cur, + const char *refc, int len, + struct ucl_schema_error *err) +{ + const ucl_object_t *res = NULL; + char *err_str; + int num, i; + + if (cur->type == UCL_OBJECT) { + /* Find a key inside an object */ + res = ucl_object_find_keyl (cur, refc, len); + if (res == NULL) { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, cur, + "reference %s is invalid, missing path component", refc); + return NULL; + } + } + else if (cur->type == UCL_ARRAY) { + /* We must figure out a number inside array */ + num = strtoul (refc, &err_str, 10); + if (err_str != NULL && *err_str != '/' && *err_str != '\0') { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, cur, + "reference %s is invalid, invalid item number", refc); + return NULL; + } + res = ucl_array_head (cur); + i = 0; + while (res != NULL) { + if (i == num) { + break; + } + res = res->next; + } + if (res == NULL) { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, cur, + "reference %s is invalid, item number %d does not exist", + refc, num); + return NULL; + } + } + else { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, res, + "reference %s is invalid, contains primitive object in the path", + refc); + return NULL; + } + + return res; +} +/* + * Find reference schema + */ +static const ucl_object_t * +ucl_schema_resolve_ref (const ucl_object_t *root, const char *ref, + struct ucl_schema_error *err) +{ + const char *p, *c; + const ucl_object_t *res = NULL; + + + if (ref[0] != '#') { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, root, + "reference %s is invalid, not started with #", ref); + return NULL; + } + if (ref[1] == '/') { + p = &ref[2]; + } + else if (ref[1] == '\0') { + return root; + } + else { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, root, + "reference %s is invalid, not started with #/", ref); + return NULL; + } + + c = p; + res = root; + + while (*p != '\0') { + if (*p == '/') { + if (p - c == 0) { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, res, + "reference %s is invalid, empty path component", ref); + return NULL; + } + /* Now we have some url part, so we need to figure out where we are */ + res = ucl_schema_resolve_ref_component (res, c, p - c, err); + if (res == NULL) { + return NULL; + } + c = p + 1; + } + p ++; + } + + if (p - c != 0) { + res = ucl_schema_resolve_ref_component (res, c, p - c, err); + } + + if (res == NULL || res->type != UCL_OBJECT) { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, res, + "reference %s is invalid, cannot find specified object", + ref); + return NULL; + } + + return res; +} + +static bool +ucl_schema_validate_values (const ucl_object_t *schema, const ucl_object_t *obj, + struct ucl_schema_error *err) +{ + const ucl_object_t *elt, *cur; + int64_t constraint, i; + + elt = ucl_object_find_key (schema, "maxValues"); + if (elt != NULL && elt->type == UCL_INT) { + constraint = ucl_object_toint (elt); + cur = obj; + i = 0; + while (cur) { + if (i > constraint) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "object has more values than defined: %ld", + (long int)constraint); + return false; + } + i ++; + cur = cur->next; + } + } + elt = ucl_object_find_key (schema, "minValues"); + if (elt != NULL && elt->type == UCL_INT) { + constraint = ucl_object_toint (elt); + cur = obj; + i = 0; + while (cur) { + if (i >= constraint) { + break; + } + i ++; + cur = cur->next; + } + if (i < constraint) { + ucl_schema_create_error (err, UCL_SCHEMA_CONSTRAINT, obj, + "object has less values than defined: %ld", + (long int)constraint); + return false; + } + } + + return true; +} + +static bool +ucl_schema_validate (const ucl_object_t *schema, + const ucl_object_t *obj, bool try_array, + struct ucl_schema_error *err, + const ucl_object_t *root) +{ + const ucl_object_t *elt, *cur; + ucl_object_iter_t iter = NULL; + bool ret; + + if (schema->type != UCL_OBJECT) { + ucl_schema_create_error (err, UCL_SCHEMA_INVALID_SCHEMA, schema, + "schema is %s instead of object", ucl_object_type_to_string (schema->type)); + return false; + } + + if (try_array) { + /* + * Special case for multiple values + */ + if (!ucl_schema_validate_values (schema, obj, err)) { + return false; + } + LL_FOREACH (obj, cur) { + if (!ucl_schema_validate (schema, cur, false, err, root)) { + return false; + } + } + return true; + } + + elt = ucl_object_find_key (schema, "enum"); + if (elt != NULL && elt->type == UCL_ARRAY) { + if (!ucl_schema_validate_enum (elt, obj, err)) { + return false; + } + } + + elt = ucl_object_find_key (schema, "allOf"); + if (elt != NULL && elt->type == UCL_ARRAY) { + iter = NULL; + while ((cur = ucl_iterate_object (elt, &iter, true)) != NULL) { + ret = ucl_schema_validate (cur, obj, true, err, root); + if (!ret) { + return false; + } + } + } + + elt = ucl_object_find_key (schema, "anyOf"); + if (elt != NULL && elt->type == UCL_ARRAY) { + iter = NULL; + while ((cur = ucl_iterate_object (elt, &iter, true)) != NULL) { + ret = ucl_schema_validate (cur, obj, true, err, root); + if (ret) { + break; + } + } + if (!ret) { + return false; + } + else { + /* Reset error */ + err->code = UCL_SCHEMA_OK; + } + } + + elt = ucl_object_find_key (schema, "oneOf"); + if (elt != NULL && elt->type == UCL_ARRAY) { + iter = NULL; + ret = false; + while ((cur = ucl_iterate_object (elt, &iter, true)) != NULL) { + if (!ret) { + ret = ucl_schema_validate (cur, obj, true, err, root); + } + else if (ucl_schema_validate (cur, obj, true, err, root)) { + ret = false; + break; + } + } + if (!ret) { + return false; + } + } + + elt = ucl_object_find_key (schema, "not"); + if (elt != NULL && elt->type == UCL_OBJECT) { + if (ucl_schema_validate (elt, obj, true, err, root)) { + return false; + } + else { + /* Reset error */ + err->code = UCL_SCHEMA_OK; + } + } + + elt = ucl_object_find_key (schema, "$ref"); + if (elt != NULL) { + cur = ucl_schema_resolve_ref (root, ucl_object_tostring (elt), err); + if (cur == NULL) { + return false; + } + if (!ucl_schema_validate (cur, obj, try_array, err, root)) { + return false; + } + } + + elt = ucl_object_find_key (schema, "type"); + if (!ucl_schema_type_is_allowed (elt, obj, err)) { + return false; + } + + switch (obj->type) { + case UCL_OBJECT: + return ucl_schema_validate_object (schema, obj, err, root); + break; + case UCL_ARRAY: + return ucl_schema_validate_array (schema, obj, err, root); + break; + case UCL_INT: + case UCL_FLOAT: + return ucl_schema_validate_number (schema, obj, err); + break; + case UCL_STRING: + return ucl_schema_validate_string (schema, obj, err); + break; + default: + break; + } + + return true; +} + +bool +ucl_object_validate (const ucl_object_t *schema, + const ucl_object_t *obj, struct ucl_schema_error *err) +{ + return ucl_schema_validate (schema, obj, true, err, schema); +} diff --git a/contrib/libucl/src/ucl_sexp.c b/contrib/libucl/src/ucl_sexp.c new file mode 100644 index 0000000..3ca7eb1 --- /dev/null +++ b/contrib/libucl/src/ucl_sexp.c @@ -0,0 +1,224 @@ +/* + * Copyright (c) 2015, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY AUTHOR ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include <ucl.h> +#include "ucl.h" +#include "ucl_internal.h" +#include "utlist.h" + +#define NEXT_STATE do { \ +if (p >= end) { \ + if (state != read_ebrace) { \ + ucl_create_err (&parser->err,\ + "extra data");\ + state = parse_err; \ + } \ +} \ +else { \ +switch (*p) { \ + case '(': \ + state = read_obrace; \ + break; \ + case ')': \ + state = read_ebrace; \ + break; \ + default: \ + len = 0; \ + mult = 1; \ + state = read_length; \ + break; \ + } \ +} \ +} while(0) + +bool +ucl_parse_csexp (struct ucl_parser *parser) +{ + const unsigned char *p, *end; + ucl_object_t *obj; + struct ucl_stack *st; + uint64_t len = 0, mult = 1; + enum { + start_parse, + read_obrace, + read_length, + read_value, + read_ebrace, + parse_err + } state = start_parse; + + assert (parser != NULL); + assert (parser->chunks != NULL); + assert (parser->chunks->begin != NULL); + assert (parser->chunks->remain != 0); + + p = parser->chunks->begin; + end = p + parser->chunks->remain; + + while (p < end) { + switch (state) { + case start_parse: + /* At this point we expect open brace */ + if (*p == '(') { + state = read_obrace; + } + else { + ucl_create_err (&parser->err, "bad starting character for " + "sexp block: %x", (int)*p); + state = parse_err; + } + break; + + case read_obrace: + st = calloc (1, sizeof (*st)); + + if (st == NULL) { + ucl_create_err (&parser->err, "no memory"); + state = parse_err; + continue; + } + + st->obj = ucl_object_typed_new (UCL_ARRAY); + + if (st->obj == NULL) { + ucl_create_err (&parser->err, "no memory"); + state = parse_err; + continue; + } + + if (parser->stack == NULL) { + /* We have no stack */ + parser->stack = st; + + if (parser->top_obj == NULL) { + parser->top_obj = st->obj; + } + } + else { + /* Prepend new element to the stack */ + LL_PREPEND (parser->stack, st); + } + + p ++; + NEXT_STATE; + + break; + + case read_length: + if (*p == ':') { + if (len == 0) { + ucl_create_err (&parser->err, "zero length element"); + state = parse_err; + continue; + } + + state = read_value; + } + else if (*p >= '0' && *p <= '9') { + len += (*p - '0') * mult; + mult *= 10; + + if (len > UINT32_MAX) { + ucl_create_err (&parser->err, "too big length of an " + "element"); + state = parse_err; + continue; + } + } + else { + ucl_create_err (&parser->err, "bad length character: %x", + (int)*p); + state = parse_err; + continue; + } + + p ++; + break; + + case read_value: + if ((uint64_t)(end - p) > len || len == 0) { + ucl_create_err (&parser->err, "invalid length: %llu, %ld " + "remain", (long long unsigned)len, (long)(end - p)); + state = parse_err; + continue; + } + obj = ucl_object_typed_new (UCL_STRING); + + obj->value.sv = (const char*)p; + obj->len = len; + obj->flags |= UCL_OBJECT_BINARY; + + if (!(parser->flags & UCL_PARSER_ZEROCOPY)) { + ucl_copy_value_trash (obj); + } + + ucl_array_append (parser->stack->obj, obj); + p += len; + NEXT_STATE; + break; + + case read_ebrace: + if (parser->stack == NULL) { + /* We have an extra end brace */ + ucl_create_err (&parser->err, "invalid length: %llu, %ld " + "remain", (long long unsigned)len, (long)(end - p)); + state = parse_err; + continue; + } + /* Pop the container */ + st = parser->stack; + parser->stack = st->next; + + if (parser->stack->obj->type == UCL_ARRAY) { + ucl_array_append (parser->stack->obj, st->obj); + } + else { + ucl_create_err (&parser->err, "bad container object, array " + "expected"); + state = parse_err; + continue; + } + + free (st); + p++; + NEXT_STATE; + break; + + case parse_err: + default: + return false; + } + } + + if (state != read_ebrace) { + ucl_create_err (&parser->err, "invalid finishing state: %d", state); + return false; + } + + return true; +}
\ No newline at end of file diff --git a/contrib/libucl/src/ucl_util.c b/contrib/libucl/src/ucl_util.c new file mode 100644 index 0000000..730a5c4 --- /dev/null +++ b/contrib/libucl/src/ucl_util.c @@ -0,0 +1,3257 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * Copyright (c) 2015 Allan Jude <allanjude@freebsd.org> + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include "ucl.h" +#include "ucl_internal.h" +#include "ucl_chartable.h" +#include "kvec.h" +#include <stdarg.h> + +#ifndef _WIN32 +#include <glob.h> +#endif + +#ifdef HAVE_LIBGEN_H +#include <libgen.h> /* For dirname */ +#endif + +typedef kvec_t(ucl_object_t *) ucl_array_t; + +#define UCL_ARRAY_GET(ar, obj) ucl_array_t *ar = \ + (ucl_array_t *)((obj) != NULL ? (obj)->value.av : NULL) + +#ifdef HAVE_OPENSSL +#include <openssl/err.h> +#include <openssl/sha.h> +#include <openssl/rsa.h> +#include <openssl/ssl.h> +#include <openssl/evp.h> +#endif + +#ifdef CURL_FOUND +#include <curl/curl.h> +#endif +#ifdef HAVE_FETCH_H +#include <fetch.h> +#endif + +#ifdef _WIN32 +#include <windows.h> + +#ifndef PROT_READ +#define PROT_READ 1 +#endif +#ifndef PROT_WRITE +#define PROT_WRITE 2 +#endif +#ifndef PROT_READWRITE +#define PROT_READWRITE 3 +#endif +#ifndef MAP_SHARED +#define MAP_SHARED 1 +#endif +#ifndef MAP_PRIVATE +#define MAP_PRIVATE 2 +#endif +#ifndef MAP_FAILED +#define MAP_FAILED ((void *) -1) +#endif + +#ifdef _WIN32 +#include <limits.h> +#define NBBY CHAR_BIT +#endif + +static void *ucl_mmap(char *addr, size_t length, int prot, int access, int fd, off_t offset) +{ + void *map = NULL; + HANDLE handle = INVALID_HANDLE_VALUE; + + switch (prot) { + default: + case PROT_READ: + { + handle = CreateFileMapping((HANDLE) _get_osfhandle(fd), 0, PAGE_READONLY, 0, length, 0); + if (!handle) break; + map = (void *) MapViewOfFile(handle, FILE_MAP_READ, 0, 0, length); + CloseHandle(handle); + break; + } + case PROT_WRITE: + { + handle = CreateFileMapping((HANDLE) _get_osfhandle(fd), 0, PAGE_READWRITE, 0, length, 0); + if (!handle) break; + map = (void *) MapViewOfFile(handle, FILE_MAP_WRITE, 0, 0, length); + CloseHandle(handle); + break; + } + case PROT_READWRITE: + { + handle = CreateFileMapping((HANDLE) _get_osfhandle(fd), 0, PAGE_READWRITE, 0, length, 0); + if (!handle) break; + map = (void *) MapViewOfFile(handle, FILE_MAP_ALL_ACCESS, 0, 0, length); + CloseHandle(handle); + break; + } + } + if (map == (void *) NULL) { + return (void *) MAP_FAILED; + } + return (void *) ((char *) map + offset); +} + +static int ucl_munmap(void *map,size_t length) +{ + if (!UnmapViewOfFile(map)) { + return(-1); + } + return(0); +} + +static char* ucl_realpath(const char *path, char *resolved_path) { + char *p; + char tmp[MAX_PATH + 1]; + strncpy(tmp, path, sizeof(tmp)-1); + p = tmp; + while(*p) { + if (*p == '/') *p = '\\'; + p++; + } + return _fullpath(resolved_path, tmp, MAX_PATH); +} +#else +#define ucl_mmap mmap +#define ucl_munmap munmap +#define ucl_realpath realpath +#endif + +typedef void (*ucl_object_dtor) (ucl_object_t *obj); +static void ucl_object_free_internal (ucl_object_t *obj, bool allow_rec, + ucl_object_dtor dtor); +static void ucl_object_dtor_unref (ucl_object_t *obj); + +static void +ucl_object_dtor_free (ucl_object_t *obj) +{ + if (obj->trash_stack[UCL_TRASH_KEY] != NULL) { + UCL_FREE (obj->hh.keylen, obj->trash_stack[UCL_TRASH_KEY]); + } + if (obj->trash_stack[UCL_TRASH_VALUE] != NULL) { + UCL_FREE (obj->len, obj->trash_stack[UCL_TRASH_VALUE]); + } + /* Do not free ephemeral objects */ + if ((obj->flags & UCL_OBJECT_EPHEMERAL) == 0) { + if (obj->type != UCL_USERDATA) { + UCL_FREE (sizeof (ucl_object_t), obj); + } + else { + struct ucl_object_userdata *ud = (struct ucl_object_userdata *)obj; + if (ud->dtor) { + ud->dtor (obj->value.ud); + } + UCL_FREE (sizeof (*ud), obj); + } + } +} + +/* + * This is a helper function that performs exactly the same as + * `ucl_object_unref` but it doesn't iterate over elements allowing + * to use it for individual elements of arrays and multiple values + */ +static void +ucl_object_dtor_unref_single (ucl_object_t *obj) +{ + if (obj != NULL) { +#ifdef HAVE_ATOMIC_BUILTINS + unsigned int rc = __sync_sub_and_fetch (&obj->ref, 1); + if (rc == 0) { +#else + if (--obj->ref == 0) { +#endif + ucl_object_free_internal (obj, false, ucl_object_dtor_unref); + } + } +} + +static void +ucl_object_dtor_unref (ucl_object_t *obj) +{ + if (obj->ref == 0) { + ucl_object_dtor_free (obj); + } + else { + /* This may cause dtor unref being called one more time */ + ucl_object_dtor_unref_single (obj); + } +} + +static void +ucl_object_free_internal (ucl_object_t *obj, bool allow_rec, ucl_object_dtor dtor) +{ + ucl_object_t *tmp, *sub; + + while (obj != NULL) { + if (obj->type == UCL_ARRAY) { + UCL_ARRAY_GET (vec, obj); + unsigned int i; + + if (vec != NULL) { + for (i = 0; i < vec->n; i ++) { + sub = kv_A (*vec, i); + if (sub != NULL) { + tmp = sub; + while (sub) { + tmp = sub->next; + dtor (sub); + sub = tmp; + } + } + } + kv_destroy (*vec); + UCL_FREE (sizeof (*vec), vec); + } + obj->value.av = NULL; + } + else if (obj->type == UCL_OBJECT) { + if (obj->value.ov != NULL) { + ucl_hash_destroy (obj->value.ov, (ucl_hash_free_func *)dtor); + } + obj->value.ov = NULL; + } + tmp = obj->next; + dtor (obj); + obj = tmp; + + if (!allow_rec) { + break; + } + } +} + +void +ucl_object_free (ucl_object_t *obj) +{ + ucl_object_free_internal (obj, true, ucl_object_dtor_free); +} + +size_t +ucl_unescape_json_string (char *str, size_t len) +{ + char *t = str, *h = str; + int i, uval; + + if (len <= 1) { + return len; + } + /* t is target (tortoise), h is source (hare) */ + + while (len) { + if (*h == '\\') { + h ++; + + if (len == 1) { + /* + * If \ is last, then do not try to go further + * Issue: #74 + */ + len --; + *t++ = '\\'; + continue; + } + + switch (*h) { + case 'n': + *t++ = '\n'; + break; + case 'r': + *t++ = '\r'; + break; + case 'b': + *t++ = '\b'; + break; + case 't': + *t++ = '\t'; + break; + case 'f': + *t++ = '\f'; + break; + case '\\': + *t++ = '\\'; + break; + case '"': + *t++ = '"'; + break; + case 'u': + /* Unicode escape */ + uval = 0; + if (len > 3) { + for (i = 0; i < 4; i++) { + uval <<= 4; + if (isdigit (h[i])) { + uval += h[i] - '0'; + } + else if (h[i] >= 'a' && h[i] <= 'f') { + uval += h[i] - 'a' + 10; + } + else if (h[i] >= 'A' && h[i] <= 'F') { + uval += h[i] - 'A' + 10; + } + else { + break; + } + } + h += 3; + len -= 3; + /* Encode */ + if(uval < 0x80) { + t[0] = (char)uval; + t ++; + } + else if(uval < 0x800) { + t[0] = 0xC0 + ((uval & 0x7C0) >> 6); + t[1] = 0x80 + ((uval & 0x03F)); + t += 2; + } + else if(uval < 0x10000) { + t[0] = 0xE0 + ((uval & 0xF000) >> 12); + t[1] = 0x80 + ((uval & 0x0FC0) >> 6); + t[2] = 0x80 + ((uval & 0x003F)); + t += 3; + } + else if(uval <= 0x10FFFF) { + t[0] = 0xF0 + ((uval & 0x1C0000) >> 18); + t[1] = 0x80 + ((uval & 0x03F000) >> 12); + t[2] = 0x80 + ((uval & 0x000FC0) >> 6); + t[3] = 0x80 + ((uval & 0x00003F)); + t += 4; + } + else { + *t++ = '?'; + } + } + else { + *t++ = 'u'; + } + break; + default: + *t++ = *h; + break; + } + h ++; + len --; + } + else { + *t++ = *h++; + } + + if (len > 0) { + len --; + } + } + *t = '\0'; + + return (t - str); +} + +char * +ucl_copy_key_trash (const ucl_object_t *obj) +{ + ucl_object_t *deconst; + + if (obj == NULL) { + return NULL; + } + if (obj->trash_stack[UCL_TRASH_KEY] == NULL && obj->key != NULL) { + deconst = __DECONST (ucl_object_t *, obj); + deconst->trash_stack[UCL_TRASH_KEY] = malloc (obj->keylen + 1); + if (deconst->trash_stack[UCL_TRASH_KEY] != NULL) { + memcpy (deconst->trash_stack[UCL_TRASH_KEY], obj->key, obj->keylen); + deconst->trash_stack[UCL_TRASH_KEY][obj->keylen] = '\0'; + } + deconst->key = obj->trash_stack[UCL_TRASH_KEY]; + deconst->flags |= UCL_OBJECT_ALLOCATED_KEY; + } + + return obj->trash_stack[UCL_TRASH_KEY]; +} + +char * +ucl_copy_value_trash (const ucl_object_t *obj) +{ + ucl_object_t *deconst; + + if (obj == NULL) { + return NULL; + } + if (obj->trash_stack[UCL_TRASH_VALUE] == NULL) { + deconst = __DECONST (ucl_object_t *, obj); + if (obj->type == UCL_STRING) { + + /* Special case for strings */ + if (obj->flags & UCL_OBJECT_BINARY) { + deconst->trash_stack[UCL_TRASH_VALUE] = malloc (obj->len); + if (deconst->trash_stack[UCL_TRASH_VALUE] != NULL) { + memcpy (deconst->trash_stack[UCL_TRASH_VALUE], + obj->value.sv, + obj->len); + deconst->value.sv = obj->trash_stack[UCL_TRASH_VALUE]; + } + } + else { + deconst->trash_stack[UCL_TRASH_VALUE] = malloc (obj->len + 1); + if (deconst->trash_stack[UCL_TRASH_VALUE] != NULL) { + memcpy (deconst->trash_stack[UCL_TRASH_VALUE], + obj->value.sv, + obj->len); + deconst->trash_stack[UCL_TRASH_VALUE][obj->len] = '\0'; + deconst->value.sv = obj->trash_stack[UCL_TRASH_VALUE]; + } + } + } + else { + /* Just emit value in json notation */ + deconst->trash_stack[UCL_TRASH_VALUE] = ucl_object_emit_single_json (obj); + deconst->len = strlen (obj->trash_stack[UCL_TRASH_VALUE]); + } + deconst->flags |= UCL_OBJECT_ALLOCATED_VALUE; + } + + return obj->trash_stack[UCL_TRASH_VALUE]; +} + +ucl_object_t* +ucl_parser_get_object (struct ucl_parser *parser) +{ + if (parser->state != UCL_STATE_ERROR && parser->top_obj != NULL) { + return ucl_object_ref (parser->top_obj); + } + + return NULL; +} + +void +ucl_parser_free (struct ucl_parser *parser) +{ + struct ucl_stack *stack, *stmp; + struct ucl_macro *macro, *mtmp; + struct ucl_chunk *chunk, *ctmp; + struct ucl_pubkey *key, *ktmp; + struct ucl_variable *var, *vtmp; + ucl_object_t *tr, *trtmp; + + if (parser == NULL) { + return; + } + + if (parser->top_obj != NULL) { + ucl_object_unref (parser->top_obj); + } + + if (parser->includepaths != NULL) { + ucl_object_unref (parser->includepaths); + } + + LL_FOREACH_SAFE (parser->stack, stack, stmp) { + free (stack); + } + HASH_ITER (hh, parser->macroes, macro, mtmp) { + free (macro->name); + HASH_DEL (parser->macroes, macro); + UCL_FREE (sizeof (struct ucl_macro), macro); + } + LL_FOREACH_SAFE (parser->chunks, chunk, ctmp) { + UCL_FREE (sizeof (struct ucl_chunk), chunk); + } + LL_FOREACH_SAFE (parser->keys, key, ktmp) { + UCL_FREE (sizeof (struct ucl_pubkey), key); + } + LL_FOREACH_SAFE (parser->variables, var, vtmp) { + free (var->value); + free (var->var); + UCL_FREE (sizeof (struct ucl_variable), var); + } + LL_FOREACH_SAFE (parser->trash_objs, tr, trtmp) { + ucl_object_free_internal (tr, false, ucl_object_dtor_free); + } + + if (parser->err != NULL) { + utstring_free (parser->err); + } + + if (parser->cur_file) { + free (parser->cur_file); + } + + UCL_FREE (sizeof (struct ucl_parser), parser); +} + +const char * +ucl_parser_get_error(struct ucl_parser *parser) +{ + if (parser == NULL) { + return NULL; + } + + if (parser->err == NULL) { + return NULL; + } + + return utstring_body (parser->err); +} + +int +ucl_parser_get_error_code(struct ucl_parser *parser) +{ + if (parser == NULL) { + return 0; + } + + return parser->err_code; +} + +unsigned +ucl_parser_get_column(struct ucl_parser *parser) +{ + if (parser == NULL || parser->chunks == NULL) { + return 0; + } + + return parser->chunks->column; +} + +unsigned +ucl_parser_get_linenum(struct ucl_parser *parser) +{ + if (parser == NULL || parser->chunks == NULL) { + return 0; + } + + return parser->chunks->line; +} + +void +ucl_parser_clear_error(struct ucl_parser *parser) +{ + if (parser != NULL && parser->err != NULL) { + utstring_free(parser->err); + parser->err = NULL; + parser->err_code = 0; + } +} + +bool +ucl_pubkey_add (struct ucl_parser *parser, const unsigned char *key, size_t len) +{ +#ifndef HAVE_OPENSSL + ucl_create_err (&parser->err, "cannot check signatures without openssl"); + return false; +#else +# if (OPENSSL_VERSION_NUMBER < 0x10000000L) + ucl_create_err (&parser->err, "cannot check signatures, openssl version is unsupported"); + return EXIT_FAILURE; +# else + struct ucl_pubkey *nkey; + BIO *mem; + + mem = BIO_new_mem_buf ((void *)key, len); + nkey = UCL_ALLOC (sizeof (struct ucl_pubkey)); + if (nkey == NULL) { + ucl_create_err (&parser->err, "cannot allocate memory for key"); + return false; + } + nkey->key = PEM_read_bio_PUBKEY (mem, &nkey->key, NULL, NULL); + BIO_free (mem); + if (nkey->key == NULL) { + UCL_FREE (sizeof (struct ucl_pubkey), nkey); + ucl_create_err (&parser->err, "%s", + ERR_error_string (ERR_get_error (), NULL)); + return false; + } + LL_PREPEND (parser->keys, nkey); +# endif +#endif + return true; +} + +#ifdef CURL_FOUND +struct ucl_curl_cbdata { + unsigned char *buf; + size_t buflen; +}; + +static size_t +ucl_curl_write_callback (void* contents, size_t size, size_t nmemb, void* ud) +{ + struct ucl_curl_cbdata *cbdata = ud; + size_t realsize = size * nmemb; + + cbdata->buf = realloc (cbdata->buf, cbdata->buflen + realsize + 1); + if (cbdata->buf == NULL) { + return 0; + } + + memcpy (&(cbdata->buf[cbdata->buflen]), contents, realsize); + cbdata->buflen += realsize; + cbdata->buf[cbdata->buflen] = 0; + + return realsize; +} +#endif + +/** + * Fetch a url and save results to the memory buffer + * @param url url to fetch + * @param len length of url + * @param buf target buffer + * @param buflen target length + * @return + */ +static bool +ucl_fetch_url (const unsigned char *url, unsigned char **buf, size_t *buflen, + UT_string **err, bool must_exist) +{ + +#ifdef HAVE_FETCH_H + struct url *fetch_url; + struct url_stat us; + FILE *in; + + fetch_url = fetchParseURL (url); + if (fetch_url == NULL) { + ucl_create_err (err, "invalid URL %s: %s", + url, strerror (errno)); + return false; + } + if ((in = fetchXGet (fetch_url, &us, "")) == NULL) { + if (!must_exist) { + ucl_create_err (err, "cannot fetch URL %s: %s", + url, strerror (errno)); + } + fetchFreeURL (fetch_url); + return false; + } + + *buflen = us.size; + *buf = malloc (*buflen); + if (*buf == NULL) { + ucl_create_err (err, "cannot allocate buffer for URL %s: %s", + url, strerror (errno)); + fclose (in); + fetchFreeURL (fetch_url); + return false; + } + + if (fread (*buf, *buflen, 1, in) != 1) { + ucl_create_err (err, "cannot read URL %s: %s", + url, strerror (errno)); + fclose (in); + fetchFreeURL (fetch_url); + return false; + } + + fetchFreeURL (fetch_url); + return true; +#elif defined(CURL_FOUND) + CURL *curl; + int r; + struct ucl_curl_cbdata cbdata; + + curl = curl_easy_init (); + if (curl == NULL) { + ucl_create_err (err, "CURL interface is broken"); + return false; + } + if ((r = curl_easy_setopt (curl, CURLOPT_URL, url)) != CURLE_OK) { + ucl_create_err (err, "invalid URL %s: %s", + url, curl_easy_strerror (r)); + curl_easy_cleanup (curl); + return false; + } + curl_easy_setopt (curl, CURLOPT_WRITEFUNCTION, ucl_curl_write_callback); + cbdata.buf = *buf; + cbdata.buflen = *buflen; + curl_easy_setopt (curl, CURLOPT_WRITEDATA, &cbdata); + + if ((r = curl_easy_perform (curl)) != CURLE_OK) { + if (!must_exist) { + ucl_create_err (err, "error fetching URL %s: %s", + url, curl_easy_strerror (r)); + } + curl_easy_cleanup (curl); + if (cbdata.buf) { + free (cbdata.buf); + } + return false; + } + *buf = cbdata.buf; + *buflen = cbdata.buflen; + + return true; +#else + ucl_create_err (err, "URL support is disabled"); + return false; +#endif +} + +/** + * Fetch a file and save results to the memory buffer + * @param filename filename to fetch + * @param len length of filename + * @param buf target buffer + * @param buflen target length + * @return + */ +static bool +ucl_fetch_file (const unsigned char *filename, unsigned char **buf, size_t *buflen, + UT_string **err, bool must_exist) +{ + int fd; + struct stat st; + + if (stat (filename, &st) == -1 || !S_ISREG (st.st_mode)) { + if (must_exist) { + ucl_create_err (err, "cannot stat file %s: %s", + filename, strerror (errno)); + } + return false; + } + if (st.st_size == 0) { + /* Do not map empty files */ + *buf = ""; + *buflen = 0; + } + else { + if ((fd = open (filename, O_RDONLY)) == -1) { + ucl_create_err (err, "cannot open file %s: %s", + filename, strerror (errno)); + return false; + } + if ((*buf = ucl_mmap (NULL, st.st_size, PROT_READ, MAP_SHARED, fd, 0)) == MAP_FAILED) { + close (fd); + ucl_create_err (err, "cannot mmap file %s: %s", + filename, strerror (errno)); + return false; + } + *buflen = st.st_size; + close (fd); + } + + return true; +} + + +#if (defined(HAVE_OPENSSL) && OPENSSL_VERSION_NUMBER >= 0x10000000L) +static inline bool +ucl_sig_check (const unsigned char *data, size_t datalen, + const unsigned char *sig, size_t siglen, struct ucl_parser *parser) +{ + struct ucl_pubkey *key; + char dig[EVP_MAX_MD_SIZE]; + unsigned int diglen; + EVP_PKEY_CTX *key_ctx; + EVP_MD_CTX *sign_ctx = NULL; + + sign_ctx = EVP_MD_CTX_create (); + + LL_FOREACH (parser->keys, key) { + key_ctx = EVP_PKEY_CTX_new (key->key, NULL); + if (key_ctx != NULL) { + if (EVP_PKEY_verify_init (key_ctx) <= 0) { + EVP_PKEY_CTX_free (key_ctx); + continue; + } + if (EVP_PKEY_CTX_set_rsa_padding (key_ctx, RSA_PKCS1_PADDING) <= 0) { + EVP_PKEY_CTX_free (key_ctx); + continue; + } + if (EVP_PKEY_CTX_set_signature_md (key_ctx, EVP_sha256 ()) <= 0) { + EVP_PKEY_CTX_free (key_ctx); + continue; + } + EVP_DigestInit (sign_ctx, EVP_sha256 ()); + EVP_DigestUpdate (sign_ctx, data, datalen); + EVP_DigestFinal (sign_ctx, dig, &diglen); + + if (EVP_PKEY_verify (key_ctx, sig, siglen, dig, diglen) == 1) { + EVP_MD_CTX_destroy (sign_ctx); + EVP_PKEY_CTX_free (key_ctx); + return true; + } + + EVP_PKEY_CTX_free (key_ctx); + } + } + + EVP_MD_CTX_destroy (sign_ctx); + + return false; +} +#endif + +struct ucl_include_params { + bool check_signature; + bool must_exist; + bool use_glob; + bool use_prefix; + bool soft_fail; + bool allow_glob; + unsigned priority; + enum ucl_duplicate_strategy strat; + enum ucl_parse_type parse_type; + const char *prefix; + const char *target; +}; + +/** + * Include an url to configuration + * @param data + * @param len + * @param parser + * @param err + * @return + */ +static bool +ucl_include_url (const unsigned char *data, size_t len, + struct ucl_parser *parser, + struct ucl_include_params *params) +{ + + bool res; + unsigned char *buf = NULL; + size_t buflen = 0; + struct ucl_chunk *chunk; + char urlbuf[PATH_MAX]; + int prev_state; + + snprintf (urlbuf, sizeof (urlbuf), "%.*s", (int)len, data); + + if (!ucl_fetch_url (urlbuf, &buf, &buflen, &parser->err, params->must_exist)) { + return (!params->must_exist || false); + } + + if (params->check_signature) { +#if (defined(HAVE_OPENSSL) && OPENSSL_VERSION_NUMBER >= 0x10000000L) + unsigned char *sigbuf = NULL; + size_t siglen = 0; + /* We need to check signature first */ + snprintf (urlbuf, sizeof (urlbuf), "%.*s.sig", (int)len, data); + if (!ucl_fetch_url (urlbuf, &sigbuf, &siglen, &parser->err, true)) { + return false; + } + if (!ucl_sig_check (buf, buflen, sigbuf, siglen, parser)) { + ucl_create_err (&parser->err, "cannot verify url %s: %s", + urlbuf, + ERR_error_string (ERR_get_error (), NULL)); + if (siglen > 0) { + ucl_munmap (sigbuf, siglen); + } + return false; + } + if (siglen > 0) { + ucl_munmap (sigbuf, siglen); + } +#endif + } + + prev_state = parser->state; + parser->state = UCL_STATE_INIT; + + res = ucl_parser_add_chunk_full (parser, buf, buflen, params->priority, + params->strat, params->parse_type); + if (res == true) { + /* Remove chunk from the stack */ + chunk = parser->chunks; + if (chunk != NULL) { + parser->chunks = chunk->next; + UCL_FREE (sizeof (struct ucl_chunk), chunk); + } + } + + parser->state = prev_state; + free (buf); + + return res; +} + +/** + * Include a single file to the parser + * @param data + * @param len + * @param parser + * @param check_signature + * @param must_exist + * @param allow_glob + * @param priority + * @return + */ +static bool +ucl_include_file_single (const unsigned char *data, size_t len, + struct ucl_parser *parser, struct ucl_include_params *params) +{ + bool res; + struct ucl_chunk *chunk; + unsigned char *buf = NULL; + char *old_curfile, *ext; + size_t buflen = 0; + char filebuf[PATH_MAX], realbuf[PATH_MAX]; + int prev_state; + struct ucl_variable *cur_var, *tmp_var, *old_curdir = NULL, + *old_filename = NULL; + ucl_object_t *nest_obj = NULL, *old_obj = NULL, *new_obj = NULL; + ucl_hash_t *container = NULL; + struct ucl_stack *st = NULL; + + snprintf (filebuf, sizeof (filebuf), "%.*s", (int)len, data); + if (ucl_realpath (filebuf, realbuf) == NULL) { + if (params->soft_fail) { + return false; + } + if (!params->must_exist) { + return true; + } + ucl_create_err (&parser->err, "cannot open file %s: %s", + filebuf, + strerror (errno)); + return false; + } + + if (parser->cur_file && strcmp (realbuf, parser->cur_file) == 0) { + /* We are likely including the file itself */ + if (params->soft_fail) { + return false; + } + + ucl_create_err (&parser->err, "trying to include the file %s from itself", + realbuf); + return false; + } + + if (!ucl_fetch_file (realbuf, &buf, &buflen, &parser->err, params->must_exist)) { + if (params->soft_fail) { + return false; + } + return (!params->must_exist || false); + } + + if (params->check_signature) { +#if (defined(HAVE_OPENSSL) && OPENSSL_VERSION_NUMBER >= 0x10000000L) + unsigned char *sigbuf = NULL; + size_t siglen = 0; + /* We need to check signature first */ + snprintf (filebuf, sizeof (filebuf), "%s.sig", realbuf); + if (!ucl_fetch_file (filebuf, &sigbuf, &siglen, &parser->err, true)) { + return false; + } + if (!ucl_sig_check (buf, buflen, sigbuf, siglen, parser)) { + ucl_create_err (&parser->err, "cannot verify file %s: %s", + filebuf, + ERR_error_string (ERR_get_error (), NULL)); + if (siglen > 0) { + ucl_munmap (sigbuf, siglen); + } + return false; + } + if (siglen > 0) { + ucl_munmap (sigbuf, siglen); + } +#endif + } + + old_curfile = parser->cur_file; + parser->cur_file = strdup (realbuf); + + /* Store old file vars */ + DL_FOREACH_SAFE (parser->variables, cur_var, tmp_var) { + if (strcmp (cur_var->var, "CURDIR") == 0) { + old_curdir = cur_var; + DL_DELETE (parser->variables, cur_var); + } + else if (strcmp (cur_var->var, "FILENAME") == 0) { + old_filename = cur_var; + DL_DELETE (parser->variables, cur_var); + } + } + + ucl_parser_set_filevars (parser, realbuf, false); + + prev_state = parser->state; + parser->state = UCL_STATE_INIT; + + if (params->use_prefix && params->prefix == NULL) { + /* Auto generate a key name based on the included filename */ + params->prefix = basename (realbuf); + ext = strrchr (params->prefix, '.'); + if (ext != NULL && (strcmp (ext, ".conf") == 0 || strcmp (ext, ".ucl") == 0)) { + /* Strip off .conf or .ucl */ + *ext = '\0'; + } + } + if (params->prefix != NULL) { + /* This is a prefixed include */ + container = parser->stack->obj->value.ov; + + old_obj = __DECONST (ucl_object_t *, ucl_hash_search (container, + params->prefix, strlen (params->prefix))); + + if (strcasecmp (params->target, "array") == 0 && old_obj == NULL) { + /* Create an array with key: prefix */ + old_obj = ucl_object_new_full (UCL_ARRAY, params->priority); + old_obj->key = params->prefix; + old_obj->keylen = strlen (params->prefix); + ucl_copy_key_trash(old_obj); + old_obj->prev = old_obj; + old_obj->next = NULL; + + container = ucl_hash_insert_object (container, old_obj, + parser->flags & UCL_PARSER_KEY_LOWERCASE); + parser->stack->obj->len ++; + + nest_obj = ucl_object_new_full (UCL_OBJECT, params->priority); + nest_obj->prev = nest_obj; + nest_obj->next = NULL; + + ucl_array_append (old_obj, nest_obj); + } + else if (old_obj == NULL) { + /* Create an object with key: prefix */ + nest_obj = ucl_object_new_full (UCL_OBJECT, params->priority); + nest_obj->key = params->prefix; + nest_obj->keylen = strlen (params->prefix); + ucl_copy_key_trash(nest_obj); + nest_obj->prev = nest_obj; + nest_obj->next = NULL; + + container = ucl_hash_insert_object (container, nest_obj, + parser->flags & UCL_PARSER_KEY_LOWERCASE); + parser->stack->obj->len ++; + } + else if (strcasecmp (params->target, "array") == 0 || + ucl_object_type(old_obj) == UCL_ARRAY) { + if (ucl_object_type(old_obj) == UCL_ARRAY) { + /* Append to the existing array */ + nest_obj = ucl_object_new_full (UCL_OBJECT, params->priority); + nest_obj->prev = nest_obj; + nest_obj->next = NULL; + + ucl_array_append (old_obj, nest_obj); + } + else { + /* Convert the object to an array */ + new_obj = ucl_object_typed_new (UCL_ARRAY); + new_obj->key = old_obj->key; + new_obj->keylen = old_obj->keylen; + new_obj->flags |= UCL_OBJECT_MULTIVALUE; + new_obj->prev = new_obj; + new_obj->next = NULL; + + nest_obj = ucl_object_new_full (UCL_OBJECT, params->priority); + nest_obj->prev = nest_obj; + nest_obj->next = NULL; + + ucl_array_append (new_obj, old_obj); + ucl_array_append (new_obj, nest_obj); + ucl_hash_replace (container, old_obj, new_obj); + } + } + else { + if (ucl_object_type (old_obj) == UCL_OBJECT) { + /* Append to existing Object*/ + nest_obj = old_obj; + } + else { + /* The key is not an object */ + ucl_create_err (&parser->err, + "Conflicting type for key: %s", + params->prefix); + return false; + } + } + + /* Put all of the content of the include inside that object */ + parser->stack->obj->value.ov = container; + + if (nest_obj != NULL) { + st = UCL_ALLOC (sizeof (struct ucl_stack)); + if (st == NULL) { + ucl_create_err (&parser->err, "cannot allocate memory for an object"); + ucl_object_unref (nest_obj); + return NULL; + } + st->obj = nest_obj; + st->level = parser->stack->level; + LL_PREPEND (parser->stack, st); + parser->cur_obj = nest_obj; + } + } + + res = ucl_parser_add_chunk_full (parser, buf, buflen, params->priority, + params->strat, params->parse_type); + if (!res && !params->must_exist) { + /* Free error */ + utstring_free (parser->err); + parser->err = NULL; + parser->state = UCL_STATE_AFTER_VALUE; + } + + /* Stop nesting the include, take 1 level off the stack */ + if (params->prefix != NULL && nest_obj != NULL) { + parser->stack = st->next; + UCL_FREE (sizeof (struct ucl_stack), st); + } + + /* Remove chunk from the stack */ + chunk = parser->chunks; + if (chunk != NULL) { + parser->chunks = chunk->next; + UCL_FREE (sizeof (struct ucl_chunk), chunk); + parser->recursion --; + } + + /* Restore old file vars */ + if (parser->cur_file) { + free (parser->cur_file); + } + + parser->cur_file = old_curfile; + DL_FOREACH_SAFE (parser->variables, cur_var, tmp_var) { + if (strcmp (cur_var->var, "CURDIR") == 0 && old_curdir) { + DL_DELETE (parser->variables, cur_var); + free (cur_var->var); + free (cur_var->value); + UCL_FREE (sizeof (struct ucl_variable), cur_var); + } + else if (strcmp (cur_var->var, "FILENAME") == 0 && old_filename) { + DL_DELETE (parser->variables, cur_var); + free (cur_var->var); + free (cur_var->value); + UCL_FREE (sizeof (struct ucl_variable), cur_var); + } + } + if (old_filename) { + DL_APPEND (parser->variables, old_filename); + } + if (old_curdir) { + DL_APPEND (parser->variables, old_curdir); + } + + parser->state = prev_state; + + if (buflen > 0) { + ucl_munmap (buf, buflen); + } + + return res; +} + +/** + * Include a file to configuration + * @param data + * @param len + * @param parser + * @param err + * @return + */ +static bool +ucl_include_file (const unsigned char *data, size_t len, + struct ucl_parser *parser, struct ucl_include_params *params) +{ + const unsigned char *p = data, *end = data + len; + bool need_glob = false; + int cnt = 0; + char glob_pattern[PATH_MAX]; + size_t i; + +#ifndef _WIN32 + if (!params->allow_glob) { + return ucl_include_file_single (data, len, parser, params); + } + else { + /* Check for special symbols in a filename */ + while (p != end) { + if (*p == '*' || *p == '?') { + need_glob = true; + break; + } + p ++; + } + if (need_glob) { + glob_t globbuf; + memset (&globbuf, 0, sizeof (globbuf)); + ucl_strlcpy (glob_pattern, (const char *)data, + (len + 1 < sizeof (glob_pattern) ? len + 1 : sizeof (glob_pattern))); + if (glob (glob_pattern, 0, NULL, &globbuf) != 0) { + return (!params->must_exist || false); + } + for (i = 0; i < globbuf.gl_pathc; i ++) { + if (!ucl_include_file_single ((unsigned char *)globbuf.gl_pathv[i], + strlen (globbuf.gl_pathv[i]), parser, params)) { + if (params->soft_fail) { + continue; + } + globfree (&globbuf); + return false; + } + cnt ++; + } + globfree (&globbuf); + + if (cnt == 0 && params->must_exist) { + ucl_create_err (&parser->err, "cannot match any files for pattern %s", + glob_pattern); + return false; + } + } + else { + return ucl_include_file_single (data, len, parser, params); + } + } +#else + /* Win32 compilers do not support globbing. Therefore, for Win32, + treat allow_glob/need_glob as a NOOP and just return */ + return ucl_include_file_single (data, len, parser, params); +#endif + + return true; +} + +/** + * Common function to handle .*include* macros + * @param data + * @param len + * @param args + * @param parser + * @param default_try + * @param default_sign + * @return + */ +static bool +ucl_include_common (const unsigned char *data, size_t len, + const ucl_object_t *args, struct ucl_parser *parser, + bool default_try, + bool default_sign) +{ + bool allow_url, search; + const char *duplicate; + const ucl_object_t *param; + ucl_object_iter_t it = NULL, ip = NULL; + char ipath[PATH_MAX]; + struct ucl_include_params params; + + /* Default values */ + params.soft_fail = default_try; + params.allow_glob = false; + params.check_signature = default_sign; + params.use_prefix = false; + params.target = "object"; + params.prefix = NULL; + params.priority = 0; + params.parse_type = UCL_PARSE_UCL; + params.strat = UCL_DUPLICATE_APPEND; + params.must_exist = !default_try; + + search = false; + + /* Process arguments */ + if (args != NULL && args->type == UCL_OBJECT) { + while ((param = ucl_iterate_object (args, &it, true)) != NULL) { + if (param->type == UCL_BOOLEAN) { + if (strncmp (param->key, "try", param->keylen) == 0) { + params.must_exist = !ucl_object_toboolean (param); + } + else if (strncmp (param->key, "sign", param->keylen) == 0) { + params.check_signature = ucl_object_toboolean (param); + } + else if (strncmp (param->key, "glob", param->keylen) == 0) { + params.allow_glob = ucl_object_toboolean (param); + } + else if (strncmp (param->key, "url", param->keylen) == 0) { + allow_url = ucl_object_toboolean (param); + } + else if (strncmp (param->key, "prefix", param->keylen) == 0) { + params.use_prefix = ucl_object_toboolean (param); + } + } + else if (param->type == UCL_STRING) { + if (strncmp (param->key, "key", param->keylen) == 0) { + params.prefix = ucl_object_tostring (param); + } + else if (strncmp (param->key, "target", param->keylen) == 0) { + params.target = ucl_object_tostring (param); + } + else if (strncmp (param->key, "duplicate", param->keylen) == 0) { + duplicate = ucl_object_tostring (param); + + if (strcmp (duplicate, "append") == 0) { + params.strat = UCL_DUPLICATE_APPEND; + } + else if (strcmp (duplicate, "merge") == 0) { + params.strat = UCL_DUPLICATE_MERGE; + } + else if (strcmp (duplicate, "rewrite") == 0) { + params.strat = UCL_DUPLICATE_REWRITE; + } + else if (strcmp (duplicate, "error") == 0) { + params.strat = UCL_DUPLICATE_ERROR; + } + } + } + else if (param->type == UCL_ARRAY) { + if (strncmp (param->key, "path", param->keylen) == 0) { + ucl_set_include_path (parser, __DECONST(ucl_object_t *, param)); + } + } + else if (param->type == UCL_INT) { + if (strncmp (param->key, "priority", param->keylen) == 0) { + params.priority = ucl_object_toint (param); + } + } + } + } + + if (parser->includepaths == NULL) { + if (allow_url && ucl_strnstr (data, "://", len) != NULL) { + /* Globbing is not used for URL's */ + return ucl_include_url (data, len, parser, ¶ms); + } + else if (data != NULL) { + /* Try to load a file */ + return ucl_include_file (data, len, parser, ¶ms); + } + } + else { + if (allow_url && ucl_strnstr (data, "://", len) != NULL) { + /* Globbing is not used for URL's */ + return ucl_include_url (data, len, parser, ¶ms); + } + + ip = ucl_object_iterate_new (parser->includepaths); + while ((param = ucl_object_iterate_safe (ip, true)) != NULL) { + if (ucl_object_type(param) == UCL_STRING) { + snprintf (ipath, sizeof (ipath), "%s/%.*s", ucl_object_tostring(param), + (int)len, data); + if ((search = ucl_include_file (ipath, strlen (ipath), + parser, ¶ms))) { + if (!params.allow_glob) { + break; + } + } + } + } + ucl_object_iterate_free (ip); + if (search == true) { + return true; + } + else { + ucl_create_err (&parser->err, + "cannot find file: %.*s in search path", + (int)len, data); + return false; + } + } + + return false; +} + +/** + * Handle include macro + * @param data include data + * @param len length of data + * @param args UCL object representing arguments to the macro + * @param ud user data + * @return + */ +bool +ucl_include_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, void* ud) +{ + struct ucl_parser *parser = ud; + + return ucl_include_common (data, len, args, parser, false, false); +} + +/** + * Handle includes macro + * @param data include data + * @param len length of data + * @param args UCL object representing arguments to the macro + * @param ud user data + * @return + */ +bool +ucl_includes_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, void* ud) +{ + struct ucl_parser *parser = ud; + + return ucl_include_common (data, len, args, parser, false, true); +} + +/** + * Handle tryinclude macro + * @param data include data + * @param len length of data + * @param args UCL object representing arguments to the macro + * @param ud user data + * @return + */ +bool +ucl_try_include_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, void* ud) +{ + struct ucl_parser *parser = ud; + + return ucl_include_common (data, len, args, parser, true, false); +} + +/** + * Handle priority macro + * @param data include data + * @param len length of data + * @param args UCL object representing arguments to the macro + * @param ud user data + * @return + */ +bool +ucl_priority_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, void* ud) +{ + struct ucl_parser *parser = ud; + unsigned priority = 255; + const ucl_object_t *param; + bool found = false; + char *value = NULL, *leftover = NULL; + ucl_object_iter_t it = NULL; + + if (parser == NULL) { + return false; + } + + /* Process arguments */ + if (args != NULL && args->type == UCL_OBJECT) { + while ((param = ucl_iterate_object (args, &it, true)) != NULL) { + if (param->type == UCL_INT) { + if (strncmp (param->key, "priority", param->keylen) == 0) { + priority = ucl_object_toint (param); + found = true; + } + } + } + } + + if (len > 0) { + value = malloc(len + 1); + ucl_strlcpy(value, (const char *)data, len + 1); + priority = strtol(value, &leftover, 10); + if (*leftover != '\0') { + ucl_create_err (&parser->err, "Invalid priority value in macro: %s", + value); + free(value); + return false; + } + free(value); + found = true; + } + + if (found == true) { + parser->chunks->priority = priority; + return true; + } + + ucl_create_err (&parser->err, "Unable to parse priority macro"); + return false; +} + +/** + * Handle load macro + * @param data include data + * @param len length of data + * @param args UCL object representing arguments to the macro + * @param ud user data + * @return + */ +bool +ucl_load_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, void* ud) +{ + struct ucl_parser *parser = ud; + const ucl_object_t *param; + ucl_object_t *obj, *old_obj; + ucl_object_iter_t it = NULL; + bool try_load, multiline, test; + const char *target, *prefix; + char *load_file, *tmp; + unsigned char *buf; + size_t buflen; + unsigned priority; + int64_t iv; + ucl_hash_t *container = NULL; + enum ucl_string_flags flags; + + /* Default values */ + try_load = false; + multiline = false; + test = false; + target = "string"; + prefix = NULL; + load_file = NULL; + buf = NULL; + buflen = 0; + priority = 0; + obj = NULL; + old_obj = NULL; + flags = 0; + + if (parser == NULL) { + return false; + } + + /* Process arguments */ + if (args != NULL && args->type == UCL_OBJECT) { + while ((param = ucl_iterate_object (args, &it, true)) != NULL) { + if (param->type == UCL_BOOLEAN) { + if (strncmp (param->key, "try", param->keylen) == 0) { + try_load = ucl_object_toboolean (param); + } + else if (strncmp (param->key, "multiline", param->keylen) == 0) { + multiline = ucl_object_toboolean (param); + } + else if (strncmp (param->key, "escape", param->keylen) == 0) { + test = ucl_object_toboolean (param); + if (test) { + flags |= UCL_STRING_ESCAPE; + } + } + else if (strncmp (param->key, "trim", param->keylen) == 0) { + test = ucl_object_toboolean (param); + if (test) { + flags |= UCL_STRING_TRIM; + } + } + } + else if (param->type == UCL_STRING) { + if (strncmp (param->key, "key", param->keylen) == 0) { + prefix = ucl_object_tostring (param); + } + else if (strncmp (param->key, "target", param->keylen) == 0) { + target = ucl_object_tostring (param); + } + } + else if (param->type == UCL_INT) { + if (strncmp (param->key, "priority", param->keylen) == 0) { + priority = ucl_object_toint (param); + } + } + } + } + + if (prefix == NULL || strlen(prefix) == 0) { + ucl_create_err (&parser->err, "No Key specified in load macro"); + return false; + } + + if (len > 0) { + asprintf (&load_file, "%.*s", (int)len, data); + if (!ucl_fetch_file (load_file, &buf, &buflen, &parser->err, !try_load)) { + return (try_load || false); + } + + container = parser->stack->obj->value.ov; + old_obj = __DECONST (ucl_object_t *, ucl_hash_search (container, prefix, strlen (prefix))); + if (old_obj != NULL) { + ucl_create_err (&parser->err, "Key %s already exists", prefix); + return false; + } + + if (strcasecmp (target, "string") == 0) { + obj = ucl_object_fromstring_common (buf, buflen, flags); + ucl_copy_value_trash (obj); + if (multiline) { + obj->flags |= UCL_OBJECT_MULTILINE; + } + } + else if (strcasecmp (target, "int") == 0) { + asprintf(&tmp, "%.*s", (int)buflen, buf); + iv = strtoll(tmp, NULL, 10); + obj = ucl_object_fromint(iv); + } + + if (buflen > 0) { + ucl_munmap (buf, buflen); + } + + if (obj != NULL) { + obj->key = prefix; + obj->keylen = strlen (prefix); + ucl_copy_key_trash(obj); + obj->prev = obj; + obj->next = NULL; + ucl_object_set_priority (obj, priority); + container = ucl_hash_insert_object (container, obj, + parser->flags & UCL_PARSER_KEY_LOWERCASE); + parser->stack->obj->value.ov = container; + } + return true; + } + + ucl_create_err (&parser->err, "Unable to parse load macro"); + return false; +} + +bool +ucl_inherit_handler (const unsigned char *data, size_t len, + const ucl_object_t *args, const ucl_object_t *ctx, void* ud) +{ + const ucl_object_t *parent, *cur; + ucl_object_t *target, *copy; + ucl_object_iter_t it = NULL; + bool replace = false; + struct ucl_parser *parser = ud; + + parent = ucl_object_find_keyl (ctx, data, len); + + /* Some sanity checks */ + if (parent == NULL || ucl_object_type (parent) != UCL_OBJECT) { + ucl_create_err (&parser->err, "Unable to find inherited object %*.s", + (int)len, data); + return false; + } + + if (parser->stack == NULL || parser->stack->obj == NULL || + ucl_object_type (parser->stack->obj) != UCL_OBJECT) { + ucl_create_err (&parser->err, "Invalid inherit context"); + return false; + } + + target = parser->stack->obj; + + if (args && (cur = ucl_object_find_key (args, "replace")) != NULL) { + replace = ucl_object_toboolean (cur); + } + + while ((cur = ucl_iterate_object (parent, &it, true))) { + /* We do not replace existing keys */ + if (!replace && ucl_object_find_keyl (target, cur->key, cur->keylen)) { + continue; + } + + copy = ucl_object_copy (cur); + + if (!replace) { + copy->flags |= UCL_OBJECT_INHERITED; + } + + ucl_object_insert_key (target, copy, copy->key, + copy->keylen, false); + } + + return true; +} + +bool +ucl_parser_set_filevars (struct ucl_parser *parser, const char *filename, bool need_expand) +{ + char realbuf[PATH_MAX], *curdir; + + if (filename != NULL) { + if (need_expand) { + if (ucl_realpath (filename, realbuf) == NULL) { + return false; + } + } + else { + ucl_strlcpy (realbuf, filename, sizeof (realbuf)); + } + + /* Define variables */ + ucl_parser_register_variable (parser, "FILENAME", realbuf); + curdir = dirname (realbuf); + ucl_parser_register_variable (parser, "CURDIR", curdir); + } + else { + /* Set everything from the current dir */ + curdir = getcwd (realbuf, sizeof (realbuf)); + ucl_parser_register_variable (parser, "FILENAME", "undef"); + ucl_parser_register_variable (parser, "CURDIR", curdir); + } + + return true; +} + +bool +ucl_parser_add_file_priority (struct ucl_parser *parser, const char *filename, + unsigned priority) +{ + unsigned char *buf; + size_t len; + bool ret; + char realbuf[PATH_MAX]; + + if (ucl_realpath (filename, realbuf) == NULL) { + ucl_create_err (&parser->err, "cannot open file %s: %s", + filename, + strerror (errno)); + return false; + } + + if (!ucl_fetch_file (realbuf, &buf, &len, &parser->err, true)) { + return false; + } + + if (parser->cur_file) { + free (parser->cur_file); + } + parser->cur_file = strdup (realbuf); + ucl_parser_set_filevars (parser, realbuf, false); + ret = ucl_parser_add_chunk_priority (parser, buf, len, priority); + + if (len > 0) { + ucl_munmap (buf, len); + } + + return ret; +} + +bool +ucl_parser_add_file (struct ucl_parser *parser, const char *filename) +{ + if (parser == NULL) { + return false; + } + + return ucl_parser_add_file_priority(parser, filename, + parser->default_priority); +} + +bool +ucl_parser_add_fd_priority (struct ucl_parser *parser, int fd, + unsigned priority) +{ + unsigned char *buf; + size_t len; + bool ret; + struct stat st; + + if (fstat (fd, &st) == -1) { + ucl_create_err (&parser->err, "cannot stat fd %d: %s", + fd, strerror (errno)); + return false; + } + if ((buf = ucl_mmap (NULL, st.st_size, PROT_READ, MAP_SHARED, fd, 0)) == MAP_FAILED) { + ucl_create_err (&parser->err, "cannot mmap fd %d: %s", + fd, strerror (errno)); + return false; + } + + if (parser->cur_file) { + free (parser->cur_file); + } + parser->cur_file = NULL; + len = st.st_size; + ret = ucl_parser_add_chunk_priority (parser, buf, len, priority); + + if (len > 0) { + ucl_munmap (buf, len); + } + + return ret; +} + +bool +ucl_parser_add_fd (struct ucl_parser *parser, int fd) +{ + if (parser == NULL) { + return false; + } + + return ucl_parser_add_fd_priority(parser, fd, parser->default_priority); +} + +size_t +ucl_strlcpy (char *dst, const char *src, size_t siz) +{ + char *d = dst; + const char *s = src; + size_t n = siz; + + /* Copy as many bytes as will fit */ + if (n != 0) { + while (--n != 0) { + if ((*d++ = *s++) == '\0') { + break; + } + } + } + + if (n == 0 && siz != 0) { + *d = '\0'; + } + + return (s - src - 1); /* count does not include NUL */ +} + +size_t +ucl_strlcpy_unsafe (char *dst, const char *src, size_t siz) +{ + memcpy (dst, src, siz - 1); + dst[siz - 1] = '\0'; + + return siz - 1; +} + +size_t +ucl_strlcpy_tolower (char *dst, const char *src, size_t siz) +{ + char *d = dst; + const char *s = src; + size_t n = siz; + + /* Copy as many bytes as will fit */ + if (n != 0) { + while (--n != 0) { + if ((*d++ = tolower (*s++)) == '\0') { + break; + } + } + } + + if (n == 0 && siz != 0) { + *d = '\0'; + } + + return (s - src); /* count does not include NUL */ +} + +/* + * Find the first occurrence of find in s + */ +char * +ucl_strnstr (const char *s, const char *find, int len) +{ + char c, sc; + int mlen; + + if ((c = *find++) != 0) { + mlen = strlen (find); + do { + do { + if ((sc = *s++) == 0 || len-- == 0) + return (NULL); + } while (sc != c); + } while (strncmp (s, find, mlen) != 0); + s--; + } + return ((char *)s); +} + +/* + * Find the first occurrence of find in s, ignore case. + */ +char * +ucl_strncasestr (const char *s, const char *find, int len) +{ + char c, sc; + int mlen; + + if ((c = *find++) != 0) { + c = tolower (c); + mlen = strlen (find); + do { + do { + if ((sc = *s++) == 0 || len-- == 0) + return (NULL); + } while (tolower (sc) != c); + } while (strncasecmp (s, find, mlen) != 0); + s--; + } + return ((char *)s); +} + +ucl_object_t * +ucl_object_fromstring_common (const char *str, size_t len, enum ucl_string_flags flags) +{ + ucl_object_t *obj; + const char *start, *end, *p, *pos; + char *dst, *d; + size_t escaped_len; + + if (str == NULL) { + return NULL; + } + + obj = ucl_object_new (); + if (obj) { + if (len == 0) { + len = strlen (str); + } + if (flags & UCL_STRING_TRIM) { + /* Skip leading spaces */ + for (start = str; (size_t)(start - str) < len; start ++) { + if (!ucl_test_character (*start, UCL_CHARACTER_WHITESPACE_UNSAFE)) { + break; + } + } + /* Skip trailing spaces */ + for (end = str + len - 1; end > start; end --) { + if (!ucl_test_character (*end, UCL_CHARACTER_WHITESPACE_UNSAFE)) { + break; + } + } + end ++; + } + else { + start = str; + end = str + len; + } + + obj->type = UCL_STRING; + if (flags & UCL_STRING_ESCAPE) { + for (p = start, escaped_len = 0; p < end; p ++, escaped_len ++) { + if (ucl_test_character (*p, UCL_CHARACTER_JSON_UNSAFE)) { + escaped_len ++; + } + } + dst = malloc (escaped_len + 1); + if (dst != NULL) { + for (p = start, d = dst; p < end; p ++, d ++) { + if (ucl_test_character (*p, UCL_CHARACTER_JSON_UNSAFE)) { + switch (*p) { + case '\n': + *d++ = '\\'; + *d = 'n'; + break; + case '\r': + *d++ = '\\'; + *d = 'r'; + break; + case '\b': + *d++ = '\\'; + *d = 'b'; + break; + case '\t': + *d++ = '\\'; + *d = 't'; + break; + case '\f': + *d++ = '\\'; + *d = 'f'; + break; + case '\\': + *d++ = '\\'; + *d = '\\'; + break; + case '"': + *d++ = '\\'; + *d = '"'; + break; + } + } + else { + *d = *p; + } + } + *d = '\0'; + obj->value.sv = dst; + obj->trash_stack[UCL_TRASH_VALUE] = dst; + obj->len = escaped_len; + } + } + else { + dst = malloc (end - start + 1); + if (dst != NULL) { + ucl_strlcpy_unsafe (dst, start, end - start + 1); + obj->value.sv = dst; + obj->trash_stack[UCL_TRASH_VALUE] = dst; + obj->len = end - start; + } + } + if ((flags & UCL_STRING_PARSE) && dst != NULL) { + /* Parse what we have */ + if (flags & UCL_STRING_PARSE_BOOLEAN) { + if (!ucl_maybe_parse_boolean (obj, dst, obj->len) && (flags & UCL_STRING_PARSE_NUMBER)) { + ucl_maybe_parse_number (obj, dst, dst + obj->len, &pos, + flags & UCL_STRING_PARSE_DOUBLE, + flags & UCL_STRING_PARSE_BYTES, + flags & UCL_STRING_PARSE_TIME); + } + } + else { + ucl_maybe_parse_number (obj, dst, dst + obj->len, &pos, + flags & UCL_STRING_PARSE_DOUBLE, + flags & UCL_STRING_PARSE_BYTES, + flags & UCL_STRING_PARSE_TIME); + } + } + } + + return obj; +} + +static bool +ucl_object_insert_key_common (ucl_object_t *top, ucl_object_t *elt, + const char *key, size_t keylen, bool copy_key, bool merge, bool replace) +{ + ucl_object_t *found, *tmp; + const ucl_object_t *cur; + ucl_object_iter_t it = NULL; + const char *p; + int ret = true; + + if (elt == NULL || key == NULL) { + return false; + } + + if (top == NULL) { + return false; + } + + if (top->type != UCL_OBJECT) { + /* It is possible to convert NULL type to an object */ + if (top->type == UCL_NULL) { + top->type = UCL_OBJECT; + } + else { + /* Refuse converting of other object types */ + return false; + } + } + + if (top->value.ov == NULL) { + top->value.ov = ucl_hash_create (false); + } + + if (keylen == 0) { + keylen = strlen (key); + } + + for (p = key; p < key + keylen; p ++) { + if (ucl_test_character (*p, UCL_CHARACTER_UCL_UNSAFE)) { + elt->flags |= UCL_OBJECT_NEED_KEY_ESCAPE; + break; + } + } + + /* workaround for some use cases */ + if (elt->trash_stack[UCL_TRASH_KEY] != NULL && + key != (const char *)elt->trash_stack[UCL_TRASH_KEY]) { + /* Remove copied key */ + free (elt->trash_stack[UCL_TRASH_KEY]); + elt->trash_stack[UCL_TRASH_KEY] = NULL; + elt->flags &= ~UCL_OBJECT_ALLOCATED_KEY; + } + + elt->key = key; + elt->keylen = keylen; + + if (copy_key) { + ucl_copy_key_trash (elt); + } + + found = __DECONST (ucl_object_t *, ucl_hash_search_obj (top->value.ov, elt)); + + if (found == NULL) { + top->value.ov = ucl_hash_insert_object (top->value.ov, elt, false); + top->len ++; + if (replace) { + ret = false; + } + } + else { + if (replace) { + ucl_hash_replace (top->value.ov, found, elt); + ucl_object_unref (found); + } + else if (merge) { + if (found->type != UCL_OBJECT && elt->type == UCL_OBJECT) { + /* Insert old elt to new one */ + ucl_object_insert_key_common (elt, found, found->key, + found->keylen, copy_key, false, false); + ucl_hash_delete (top->value.ov, found); + top->value.ov = ucl_hash_insert_object (top->value.ov, elt, false); + } + else if (found->type == UCL_OBJECT && elt->type != UCL_OBJECT) { + /* Insert new to old */ + ucl_object_insert_key_common (found, elt, elt->key, + elt->keylen, copy_key, false, false); + } + else if (found->type == UCL_OBJECT && elt->type == UCL_OBJECT) { + /* Mix two hashes */ + while ((cur = ucl_iterate_object (elt, &it, true)) != NULL) { + tmp = ucl_object_ref (cur); + ucl_object_insert_key_common (found, tmp, cur->key, + cur->keylen, copy_key, false, false); + } + ucl_object_unref (elt); + } + else { + /* Just make a list of scalars */ + DL_APPEND (found, elt); + } + } + else { + DL_APPEND (found, elt); + } + } + + return ret; +} + +bool +ucl_object_delete_keyl (ucl_object_t *top, const char *key, size_t keylen) +{ + ucl_object_t *found; + + if (top == NULL || key == NULL) { + return false; + } + + found = __DECONST (ucl_object_t *, ucl_object_find_keyl (top, key, keylen)); + + if (found == NULL) { + return false; + } + + ucl_hash_delete (top->value.ov, found); + ucl_object_unref (found); + top->len --; + + return true; +} + +bool +ucl_object_delete_key (ucl_object_t *top, const char *key) +{ + return ucl_object_delete_keyl (top, key, strlen (key)); +} + +ucl_object_t* +ucl_object_pop_keyl (ucl_object_t *top, const char *key, size_t keylen) +{ + const ucl_object_t *found; + + if (top == NULL || key == NULL) { + return false; + } + found = ucl_object_find_keyl (top, key, keylen); + + if (found == NULL) { + return NULL; + } + ucl_hash_delete (top->value.ov, found); + top->len --; + + return __DECONST (ucl_object_t *, found); +} + +ucl_object_t* +ucl_object_pop_key (ucl_object_t *top, const char *key) +{ + return ucl_object_pop_keyl (top, key, strlen (key)); +} + +bool +ucl_object_insert_key (ucl_object_t *top, ucl_object_t *elt, + const char *key, size_t keylen, bool copy_key) +{ + return ucl_object_insert_key_common (top, elt, key, keylen, copy_key, false, false); +} + +bool +ucl_object_insert_key_merged (ucl_object_t *top, ucl_object_t *elt, + const char *key, size_t keylen, bool copy_key) +{ + return ucl_object_insert_key_common (top, elt, key, keylen, copy_key, true, false); +} + +bool +ucl_object_replace_key (ucl_object_t *top, ucl_object_t *elt, + const char *key, size_t keylen, bool copy_key) +{ + return ucl_object_insert_key_common (top, elt, key, keylen, copy_key, false, true); +} + +bool +ucl_object_merge (ucl_object_t *top, ucl_object_t *elt, bool copy) +{ + ucl_object_t *cur = NULL, *cp = NULL, *found = NULL; + ucl_object_iter_t iter = NULL; + + if (top == NULL || top->type != UCL_OBJECT || elt == NULL || elt->type != UCL_OBJECT) { + return false; + } + + /* Mix two hashes */ + while ((cur = (ucl_object_t*)ucl_hash_iterate (elt->value.ov, &iter))) { + if (copy) { + cp = ucl_object_copy (cur); + } + else { + cp = ucl_object_ref (cur); + } + found = __DECONST(ucl_object_t *, ucl_hash_search (top->value.ov, cp->key, cp->keylen)); + if (found == NULL) { + /* The key does not exist */ + top->value.ov = ucl_hash_insert_object (top->value.ov, cp, false); + top->len ++; + } + else { + /* The key already exists, replace it */ + ucl_hash_replace (top->value.ov, found, cp); + ucl_object_unref (found); + } + } + + return true; +} + +const ucl_object_t * +ucl_object_find_keyl (const ucl_object_t *obj, const char *key, size_t klen) +{ + const ucl_object_t *ret; + ucl_object_t srch; + + if (obj == NULL || obj->type != UCL_OBJECT || key == NULL) { + return NULL; + } + + srch.key = key; + srch.keylen = klen; + ret = ucl_hash_search_obj (obj->value.ov, &srch); + + return ret; +} + +const ucl_object_t * +ucl_object_find_key (const ucl_object_t *obj, const char *key) +{ + if (key == NULL) { + return NULL; + } + + return ucl_object_find_keyl (obj, key, strlen (key)); +} + +const ucl_object_t* +ucl_object_find_any_key (const ucl_object_t *obj, + const char *key, ...) +{ + va_list ap; + const ucl_object_t *ret = NULL; + const char *nk = NULL; + + if (obj == NULL || key == NULL) { + return NULL; + } + + ret = ucl_object_find_keyl (obj, key, strlen (key)); + + if (ret == NULL) { + va_start (ap, key); + + while (ret == NULL) { + nk = va_arg (ap, const char *); + + if (nk == NULL) { + break; + } + else { + ret = ucl_object_find_keyl (obj, nk, strlen (nk)); + } + } + + va_end (ap); + } + + return ret; +} + +const ucl_object_t* +ucl_iterate_object (const ucl_object_t *obj, ucl_object_iter_t *iter, bool expand_values) +{ + const ucl_object_t *elt = NULL; + + if (obj == NULL || iter == NULL) { + return NULL; + } + + if (expand_values) { + switch (obj->type) { + case UCL_OBJECT: + return (const ucl_object_t*)ucl_hash_iterate (obj->value.ov, iter); + break; + case UCL_ARRAY: { + unsigned int idx; + UCL_ARRAY_GET (vec, obj); + idx = (unsigned int)(uintptr_t)(*iter); + + if (vec != NULL) { + while (idx < kv_size (*vec)) { + if ((elt = kv_A (*vec, idx)) != NULL) { + idx ++; + break; + } + idx ++; + } + *iter = (void *)(uintptr_t)idx; + } + + return elt; + break; + } + default: + /* Go to linear iteration */ + break; + } + } + /* Treat everything as a linear list */ + elt = *iter; + if (elt == NULL) { + elt = obj; + } + else if (elt == obj) { + return NULL; + } + *iter = __DECONST (void *, elt->next ? elt->next : obj); + return elt; + + /* Not reached */ + return NULL; +} + +const char safe_iter_magic[4] = {'u', 'i', 't', 'e'}; +struct ucl_object_safe_iter { + char magic[4]; /* safety check */ + const ucl_object_t *impl_it; /* implicit object iteration */ + ucl_object_iter_t expl_it; /* explicit iteration */ +}; + +#define UCL_SAFE_ITER(ptr) (struct ucl_object_safe_iter *)(ptr) +#define UCL_SAFE_ITER_CHECK(it) do { \ + assert (it != NULL); \ + assert (memcmp (it->magic, safe_iter_magic, sizeof (it->magic)) == 0); \ + } while (0) + +ucl_object_iter_t +ucl_object_iterate_new (const ucl_object_t *obj) +{ + struct ucl_object_safe_iter *it; + + it = UCL_ALLOC (sizeof (*it)); + if (it != NULL) { + memcpy (it->magic, safe_iter_magic, sizeof (it->magic)); + it->expl_it = NULL; + it->impl_it = obj; + } + + return (ucl_object_iter_t)it; +} + + +ucl_object_iter_t +ucl_object_iterate_reset (ucl_object_iter_t it, const ucl_object_t *obj) +{ + struct ucl_object_safe_iter *rit = UCL_SAFE_ITER (it); + + UCL_SAFE_ITER_CHECK (rit); + + rit->impl_it = obj; + rit->expl_it = NULL; + + return it; +} + +const ucl_object_t* +ucl_object_iterate_safe (ucl_object_iter_t it, bool expand_values) +{ + struct ucl_object_safe_iter *rit = UCL_SAFE_ITER (it); + const ucl_object_t *ret = NULL; + + UCL_SAFE_ITER_CHECK (rit); + + if (rit->impl_it == NULL) { + return NULL; + } + + if (rit->impl_it->type == UCL_OBJECT || rit->impl_it->type == UCL_ARRAY) { + ret = ucl_iterate_object (rit->impl_it, &rit->expl_it, true); + + if (ret == NULL) { + /* Need to switch to another implicit object in chain */ + rit->impl_it = rit->impl_it->next; + rit->expl_it = NULL; + return ucl_object_iterate_safe (it, expand_values); + } + } + else { + /* Just iterate over the implicit array */ + ret = rit->impl_it; + rit->impl_it = rit->impl_it->next; + if (expand_values) { + /* We flatten objects if need to expand values */ + if (ret->type == UCL_OBJECT || ret->type == UCL_ARRAY) { + return ucl_object_iterate_safe (it, expand_values); + } + } + } + + return ret; +} + +void +ucl_object_iterate_free (ucl_object_iter_t it) +{ + struct ucl_object_safe_iter *rit = UCL_SAFE_ITER (it); + + UCL_SAFE_ITER_CHECK (rit); + + UCL_FREE (sizeof (*rit), it); +} + +const ucl_object_t * +ucl_lookup_path (const ucl_object_t *top, const char *path_in) { + return ucl_lookup_path_char (top, path_in, '.'); +} + + +const ucl_object_t * +ucl_lookup_path_char (const ucl_object_t *top, const char *path_in, const char sep) { + const ucl_object_t *o = NULL, *found; + const char *p, *c; + char *err_str; + unsigned index; + + if (path_in == NULL || top == NULL) { + return NULL; + } + + found = NULL; + p = path_in; + + /* Skip leading dots */ + while (*p == sep) { + p ++; + } + + c = p; + while (*p != '\0') { + p ++; + if (*p == sep || *p == '\0') { + if (p > c) { + switch (top->type) { + case UCL_ARRAY: + /* Key should be an int */ + index = strtoul (c, &err_str, 10); + if (err_str != NULL && (*err_str != sep && *err_str != '\0')) { + return NULL; + } + o = ucl_array_find_index (top, index); + break; + default: + o = ucl_object_find_keyl (top, c, p - c); + break; + } + if (o == NULL) { + return NULL; + } + top = o; + } + if (*p != '\0') { + c = p + 1; + } + } + } + found = o; + + return found; +} + + +ucl_object_t * +ucl_object_new (void) +{ + return ucl_object_typed_new (UCL_NULL); +} + +ucl_object_t * +ucl_object_typed_new (ucl_type_t type) +{ + return ucl_object_new_full (type, 0); +} + +ucl_object_t * +ucl_object_new_full (ucl_type_t type, unsigned priority) +{ + ucl_object_t *new; + + if (type != UCL_USERDATA) { + new = UCL_ALLOC (sizeof (ucl_object_t)); + if (new != NULL) { + memset (new, 0, sizeof (ucl_object_t)); + new->ref = 1; + new->type = (type <= UCL_NULL ? type : UCL_NULL); + new->next = NULL; + new->prev = new; + ucl_object_set_priority (new, priority); + + if (type == UCL_ARRAY) { + new->value.av = UCL_ALLOC (sizeof (ucl_array_t)); + if (new->value.av) { + memset (new->value.av, 0, sizeof (ucl_array_t)); + UCL_ARRAY_GET (vec, new); + + /* Preallocate some space for arrays */ + kv_resize (ucl_object_t *, *vec, 8); + } + } + } + } + else { + new = ucl_object_new_userdata (NULL, NULL); + ucl_object_set_priority (new, priority); + } + + return new; +} + +ucl_object_t* +ucl_object_new_userdata (ucl_userdata_dtor dtor, ucl_userdata_emitter emitter) +{ + struct ucl_object_userdata *new; + size_t nsize = sizeof (*new); + + new = UCL_ALLOC (nsize); + if (new != NULL) { + memset (new, 0, nsize); + new->obj.ref = 1; + new->obj.type = UCL_USERDATA; + new->obj.next = NULL; + new->obj.prev = (ucl_object_t *)new; + new->dtor = dtor; + new->emitter = emitter; + } + + return (ucl_object_t *)new; +} + +ucl_type_t +ucl_object_type (const ucl_object_t *obj) +{ + if (obj == NULL) { + return UCL_NULL; + } + + return obj->type; +} + +ucl_object_t* +ucl_object_fromstring (const char *str) +{ + return ucl_object_fromstring_common (str, 0, UCL_STRING_ESCAPE); +} + +ucl_object_t * +ucl_object_fromlstring (const char *str, size_t len) +{ + return ucl_object_fromstring_common (str, len, UCL_STRING_ESCAPE); +} + +ucl_object_t * +ucl_object_fromint (int64_t iv) +{ + ucl_object_t *obj; + + obj = ucl_object_new (); + if (obj != NULL) { + obj->type = UCL_INT; + obj->value.iv = iv; + } + + return obj; +} + +ucl_object_t * +ucl_object_fromdouble (double dv) +{ + ucl_object_t *obj; + + obj = ucl_object_new (); + if (obj != NULL) { + obj->type = UCL_FLOAT; + obj->value.dv = dv; + } + + return obj; +} + +ucl_object_t* +ucl_object_frombool (bool bv) +{ + ucl_object_t *obj; + + obj = ucl_object_new (); + if (obj != NULL) { + obj->type = UCL_BOOLEAN; + obj->value.iv = bv; + } + + return obj; +} + +bool +ucl_array_append (ucl_object_t *top, ucl_object_t *elt) +{ + UCL_ARRAY_GET (vec, top); + + if (elt == NULL || top == NULL) { + return false; + } + + if (vec == NULL) { + vec = UCL_ALLOC (sizeof (*vec)); + + if (vec == NULL) { + return false; + } + + kv_init (*vec); + top->value.av = (void *)vec; + } + + kv_push (ucl_object_t *, *vec, elt); + + top->len ++; + + return true; +} + +bool +ucl_array_prepend (ucl_object_t *top, ucl_object_t *elt) +{ + UCL_ARRAY_GET (vec, top); + + if (elt == NULL || top == NULL) { + return false; + } + + if (vec == NULL) { + vec = UCL_ALLOC (sizeof (*vec)); + kv_init (*vec); + top->value.av = (void *)vec; + kv_push (ucl_object_t *, *vec, elt); + } + else { + /* Slow O(n) algorithm */ + kv_prepend (ucl_object_t *, *vec, elt); + } + + top->len ++; + + return true; +} + +bool +ucl_array_merge (ucl_object_t *top, ucl_object_t *elt, bool copy) +{ + unsigned i; + ucl_object_t *cp = NULL; + ucl_object_t **obj; + + if (elt == NULL || top == NULL || top->type != UCL_ARRAY || elt->type != UCL_ARRAY) { + return false; + } + + if (copy) { + cp = ucl_object_copy (elt); + } + else { + cp = ucl_object_ref (elt); + } + + UCL_ARRAY_GET (v1, top); + UCL_ARRAY_GET (v2, cp); + + kv_concat (ucl_object_t *, *v1, *v2); + + for (i = v2->n; i < v1->n; i ++) { + obj = &kv_A (*v1, i); + if (*obj == NULL) { + continue; + } + top->len ++; + } + + return true; +} + +ucl_object_t * +ucl_array_delete (ucl_object_t *top, ucl_object_t *elt) +{ + UCL_ARRAY_GET (vec, top); + ucl_object_t *ret = NULL; + unsigned i; + + if (vec == NULL) { + return NULL; + } + + for (i = 0; i < vec->n; i ++) { + if (kv_A (*vec, i) == elt) { + kv_del (ucl_object_t *, *vec, i); + ret = elt; + top->len --; + break; + } + } + + return ret; +} + +const ucl_object_t * +ucl_array_head (const ucl_object_t *top) +{ + UCL_ARRAY_GET (vec, top); + + if (vec == NULL || top == NULL || top->type != UCL_ARRAY || + top->value.av == NULL) { + return NULL; + } + + return (vec->n > 0 ? vec->a[0] : NULL); +} + +const ucl_object_t * +ucl_array_tail (const ucl_object_t *top) +{ + UCL_ARRAY_GET (vec, top); + + if (top == NULL || top->type != UCL_ARRAY || top->value.av == NULL) { + return NULL; + } + + return (vec->n > 0 ? vec->a[vec->n - 1] : NULL); +} + +ucl_object_t * +ucl_array_pop_last (ucl_object_t *top) +{ + UCL_ARRAY_GET (vec, top); + ucl_object_t **obj, *ret = NULL; + + if (vec != NULL && vec->n > 0) { + obj = &kv_A (*vec, vec->n - 1); + ret = *obj; + kv_del (ucl_object_t *, *vec, vec->n - 1); + top->len --; + } + + return ret; +} + +ucl_object_t * +ucl_array_pop_first (ucl_object_t *top) +{ + UCL_ARRAY_GET (vec, top); + ucl_object_t **obj, *ret = NULL; + + if (vec != NULL && vec->n > 0) { + obj = &kv_A (*vec, 0); + ret = *obj; + kv_del (ucl_object_t *, *vec, 0); + top->len --; + } + + return ret; +} + +const ucl_object_t * +ucl_array_find_index (const ucl_object_t *top, unsigned int index) +{ + UCL_ARRAY_GET (vec, top); + + if (vec != NULL && vec->n > 0 && index < vec->n) { + return kv_A (*vec, index); + } + + return NULL; +} + +unsigned int +ucl_array_index_of (ucl_object_t *top, ucl_object_t *elt) +{ + UCL_ARRAY_GET (vec, top); + unsigned i; + + if (vec == NULL) { + return (unsigned int)(-1); + } + + for (i = 0; i < vec->n; i ++) { + if (kv_A (*vec, i) == elt) { + return i; + } + } + + return (unsigned int)(-1); +} + +ucl_object_t * +ucl_array_replace_index (ucl_object_t *top, ucl_object_t *elt, + unsigned int index) +{ + UCL_ARRAY_GET (vec, top); + ucl_object_t *ret = NULL; + + if (vec != NULL && vec->n > 0 && index < vec->n) { + ret = kv_A (*vec, index); + kv_A (*vec, index) = elt; + } + + return ret; +} + +ucl_object_t * +ucl_elt_append (ucl_object_t *head, ucl_object_t *elt) +{ + + if (head == NULL) { + elt->next = NULL; + elt->prev = elt; + head = elt; + } + else { + elt->prev = head->prev; + head->prev->next = elt; + head->prev = elt; + elt->next = NULL; + } + + return head; +} + +bool +ucl_object_todouble_safe (const ucl_object_t *obj, double *target) +{ + if (obj == NULL || target == NULL) { + return false; + } + switch (obj->type) { + case UCL_INT: + *target = obj->value.iv; /* Probaly could cause overflow */ + break; + case UCL_FLOAT: + case UCL_TIME: + *target = obj->value.dv; + break; + default: + return false; + } + + return true; +} + +double +ucl_object_todouble (const ucl_object_t *obj) +{ + double result = 0.; + + ucl_object_todouble_safe (obj, &result); + return result; +} + +bool +ucl_object_toint_safe (const ucl_object_t *obj, int64_t *target) +{ + if (obj == NULL || target == NULL) { + return false; + } + switch (obj->type) { + case UCL_INT: + *target = obj->value.iv; + break; + case UCL_FLOAT: + case UCL_TIME: + *target = obj->value.dv; /* Loosing of decimal points */ + break; + default: + return false; + } + + return true; +} + +int64_t +ucl_object_toint (const ucl_object_t *obj) +{ + int64_t result = 0; + + ucl_object_toint_safe (obj, &result); + return result; +} + +bool +ucl_object_toboolean_safe (const ucl_object_t *obj, bool *target) +{ + if (obj == NULL || target == NULL) { + return false; + } + switch (obj->type) { + case UCL_BOOLEAN: + *target = (obj->value.iv == true); + break; + default: + return false; + } + + return true; +} + +bool +ucl_object_toboolean (const ucl_object_t *obj) +{ + bool result = false; + + ucl_object_toboolean_safe (obj, &result); + return result; +} + +bool +ucl_object_tostring_safe (const ucl_object_t *obj, const char **target) +{ + if (obj == NULL || target == NULL) { + return false; + } + + switch (obj->type) { + case UCL_STRING: + if (!(obj->flags & UCL_OBJECT_BINARY)) { + *target = ucl_copy_value_trash (obj); + } + break; + default: + return false; + } + + return true; +} + +const char * +ucl_object_tostring (const ucl_object_t *obj) +{ + const char *result = NULL; + + ucl_object_tostring_safe (obj, &result); + return result; +} + +const char * +ucl_object_tostring_forced (const ucl_object_t *obj) +{ + /* TODO: For binary strings we might encode string here */ + if (!(obj->flags & UCL_OBJECT_BINARY)) { + return ucl_copy_value_trash (obj); + } + + return NULL; +} + +bool +ucl_object_tolstring_safe (const ucl_object_t *obj, const char **target, size_t *tlen) +{ + if (obj == NULL || target == NULL) { + return false; + } + switch (obj->type) { + case UCL_STRING: + *target = obj->value.sv; + if (tlen != NULL) { + *tlen = obj->len; + } + break; + default: + return false; + } + + return true; +} + +const char * +ucl_object_tolstring (const ucl_object_t *obj, size_t *tlen) +{ + const char *result = NULL; + + ucl_object_tolstring_safe (obj, &result, tlen); + return result; +} + +const char * +ucl_object_key (const ucl_object_t *obj) +{ + return ucl_copy_key_trash (obj); +} + +const char * +ucl_object_keyl (const ucl_object_t *obj, size_t *len) +{ + if (len == NULL || obj == NULL) { + return NULL; + } + *len = obj->keylen; + return obj->key; +} + +ucl_object_t * +ucl_object_ref (const ucl_object_t *obj) +{ + ucl_object_t *res = NULL; + + if (obj != NULL) { + if (obj->flags & UCL_OBJECT_EPHEMERAL) { + /* + * Use deep copy for ephemeral objects, note that its refcount + * is NOT increased, since ephemeral objects does not need refcount + * at all + */ + res = ucl_object_copy (obj); + } + else { + res = __DECONST (ucl_object_t *, obj); +#ifdef HAVE_ATOMIC_BUILTINS + (void)__sync_add_and_fetch (&res->ref, 1); +#else + res->ref ++; +#endif + } + } + return res; +} + +static ucl_object_t * +ucl_object_copy_internal (const ucl_object_t *other, bool allow_array) +{ + + ucl_object_t *new; + ucl_object_iter_t it = NULL; + const ucl_object_t *cur; + + new = malloc (sizeof (*new)); + + if (new != NULL) { + memcpy (new, other, sizeof (*new)); + if (other->flags & UCL_OBJECT_EPHEMERAL) { + /* Copied object is always non ephemeral */ + new->flags &= ~UCL_OBJECT_EPHEMERAL; + } + new->ref = 1; + /* Unlink from others */ + new->next = NULL; + new->prev = new; + + /* deep copy of values stored */ + if (other->trash_stack[UCL_TRASH_KEY] != NULL) { + new->trash_stack[UCL_TRASH_KEY] = + strdup (other->trash_stack[UCL_TRASH_KEY]); + if (other->key == (const char *)other->trash_stack[UCL_TRASH_KEY]) { + new->key = new->trash_stack[UCL_TRASH_KEY]; + } + } + if (other->trash_stack[UCL_TRASH_VALUE] != NULL) { + new->trash_stack[UCL_TRASH_VALUE] = + strdup (other->trash_stack[UCL_TRASH_VALUE]); + if (new->type == UCL_STRING) { + new->value.sv = new->trash_stack[UCL_TRASH_VALUE]; + } + } + + if (other->type == UCL_ARRAY || other->type == UCL_OBJECT) { + /* reset old value */ + memset (&new->value, 0, sizeof (new->value)); + + while ((cur = ucl_iterate_object (other, &it, true)) != NULL) { + if (other->type == UCL_ARRAY) { + ucl_array_append (new, ucl_object_copy_internal (cur, false)); + } + else { + ucl_object_t *cp = ucl_object_copy_internal (cur, true); + if (cp != NULL) { + ucl_object_insert_key (new, cp, cp->key, cp->keylen, + false); + } + } + } + } + else if (allow_array && other->next != NULL) { + LL_FOREACH (other->next, cur) { + ucl_object_t *cp = ucl_object_copy_internal (cur, false); + if (cp != NULL) { + DL_APPEND (new, cp); + } + } + } + } + + return new; +} + +ucl_object_t * +ucl_object_copy (const ucl_object_t *other) +{ + return ucl_object_copy_internal (other, true); +} + +void +ucl_object_unref (ucl_object_t *obj) +{ + if (obj != NULL) { +#ifdef HAVE_ATOMIC_BUILTINS + unsigned int rc = __sync_sub_and_fetch (&obj->ref, 1); + if (rc == 0) { +#else + if (--obj->ref == 0) { +#endif + ucl_object_free_internal (obj, true, ucl_object_dtor_unref); + } + } +} + +int +ucl_object_compare (const ucl_object_t *o1, const ucl_object_t *o2) +{ + const ucl_object_t *it1, *it2; + ucl_object_iter_t iter = NULL; + int ret = 0; + + if (o1->type != o2->type) { + return (o1->type) - (o2->type); + } + + switch (o1->type) { + case UCL_STRING: + if (o1->len == o2->len && o1->len > 0) { + ret = strcmp (ucl_object_tostring(o1), ucl_object_tostring(o2)); + } + else { + ret = o1->len - o2->len; + } + break; + case UCL_FLOAT: + case UCL_INT: + case UCL_TIME: + ret = ucl_object_todouble (o1) - ucl_object_todouble (o2); + break; + case UCL_BOOLEAN: + ret = ucl_object_toboolean (o1) - ucl_object_toboolean (o2); + break; + case UCL_ARRAY: + if (o1->len == o2->len && o1->len > 0) { + UCL_ARRAY_GET (vec1, o1); + UCL_ARRAY_GET (vec2, o2); + unsigned i; + + /* Compare all elements in both arrays */ + for (i = 0; i < vec1->n; i ++) { + it1 = kv_A (*vec1, i); + it2 = kv_A (*vec2, i); + + if (it1 == NULL && it2 != NULL) { + return -1; + } + else if (it2 == NULL && it1 != NULL) { + return 1; + } + else if (it1 != NULL && it2 != NULL) { + ret = ucl_object_compare (it1, it2); + if (ret != 0) { + break; + } + } + } + } + else { + ret = o1->len - o2->len; + } + break; + case UCL_OBJECT: + if (o1->len == o2->len && o1->len > 0) { + while ((it1 = ucl_iterate_object (o1, &iter, true)) != NULL) { + it2 = ucl_object_find_key (o2, ucl_object_key (it1)); + if (it2 == NULL) { + ret = 1; + break; + } + ret = ucl_object_compare (it1, it2); + if (ret != 0) { + break; + } + } + } + else { + ret = o1->len - o2->len; + } + break; + default: + ret = 0; + break; + } + + return ret; +} + +void +ucl_object_array_sort (ucl_object_t *ar, + int (*cmp)(const ucl_object_t **o1, const ucl_object_t **o2)) +{ + UCL_ARRAY_GET (vec, ar); + + if (cmp == NULL || ar == NULL || ar->type != UCL_ARRAY) { + return; + } + + qsort (vec->a, vec->n, sizeof (ucl_object_t *), + (int (*)(const void *, const void *))cmp); +} + +#define PRIOBITS 4 + +unsigned int +ucl_object_get_priority (const ucl_object_t *obj) +{ + if (obj == NULL) { + return 0; + } + + return (obj->flags >> ((sizeof (obj->flags) * NBBY) - PRIOBITS)); +} + +void +ucl_object_set_priority (ucl_object_t *obj, + unsigned int priority) +{ + if (obj != NULL) { + priority &= (0x1 << PRIOBITS) - 1; + priority <<= ((sizeof (obj->flags) * NBBY) - PRIOBITS); + priority |= obj->flags & ((1 << ((sizeof (obj->flags) * NBBY) - + PRIOBITS)) - 1); + obj->flags = priority; + } +} diff --git a/contrib/libucl/src/xxhash.c b/contrib/libucl/src/xxhash.c new file mode 100644 index 0000000..3473eb4 --- /dev/null +++ b/contrib/libucl/src/xxhash.c @@ -0,0 +1,941 @@ +/* +xxHash - Fast Hash algorithm +Copyright (C) 2012-2014, Yann Collet. +BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php) + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are +met: + +* Redistributions of source code must retain the above copyright +notice, this list of conditions and the following disclaimer. +* Redistributions in binary form must reproduce the above +copyright notice, this list of conditions and the following disclaimer +in the documentation and/or other materials provided with the +distribution. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +You can contact the author at : +- xxHash source repository : http://code.google.com/p/xxhash/ +- public discussion board : https://groups.google.com/forum/#!forum/lz4c +*/ + + +//************************************** +// Tuning parameters +//************************************** +// Unaligned memory access is automatically enabled for "common" CPU, such as x86. +// For others CPU, the compiler will be more cautious, and insert extra code to ensure aligned access is respected. +// If you know your target CPU supports unaligned memory access, you want to force this option manually to improve performance. +// You can also enable this parameter if you know your input data will always be aligned (boundaries of 4, for U32). +#if defined(__ARM_FEATURE_UNALIGNED) || defined(__i386) || defined(_M_IX86) || defined(__x86_64__) || defined(_M_X64) +# define XXH_USE_UNALIGNED_ACCESS 1 +#endif + +// XXH_ACCEPT_NULL_INPUT_POINTER : +// If the input pointer is a null pointer, xxHash default behavior is to trigger a memory access error, since it is a bad pointer. +// When this option is enabled, xxHash output for null input pointers will be the same as a null-length input. +// This option has a very small performance cost (only measurable on small inputs). +// By default, this option is disabled. To enable it, uncomment below define : +// #define XXH_ACCEPT_NULL_INPUT_POINTER 1 + +// XXH_FORCE_NATIVE_FORMAT : +// By default, xxHash library provides endian-independant Hash values, based on little-endian convention. +// Results are therefore identical for little-endian and big-endian CPU. +// This comes at a performance cost for big-endian CPU, since some swapping is required to emulate little-endian format. +// Should endian-independance be of no importance for your application, you may set the #define below to 1. +// It will improve speed for Big-endian CPU. +// This option has no impact on Little_Endian CPU. +#define XXH_FORCE_NATIVE_FORMAT 0 + +//************************************** +// Compiler Specific Options +//************************************** +// Disable some Visual warning messages +#ifdef _MSC_VER // Visual Studio +# pragma warning(disable : 4127) // disable: C4127: conditional expression is constant +#endif + +#ifdef _MSC_VER // Visual Studio +# define FORCE_INLINE static __forceinline +#else +# ifdef __GNUC__ +# define FORCE_INLINE static inline __attribute__((always_inline)) +# else +# define FORCE_INLINE static inline +# endif +#endif + +//************************************** +// Includes & Memory related functions +//************************************** +#include "xxhash.h" +// Modify the local functions below should you wish to use some other memory routines +// for malloc(), free() +#include <stdlib.h> +static void* XXH_malloc(size_t s) { return malloc(s); } +static void XXH_free (void* p) { free(p); } +// for memcpy() +#include <string.h> +static void* XXH_memcpy(void* dest, const void* src, size_t size) +{ + return memcpy(dest,src,size); +} + + +//************************************** +// Basic Types +//************************************** +#if defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L // C99 +# include <stdint.h> +typedef uint8_t BYTE; +typedef uint16_t U16; +typedef uint32_t U32; +typedef int32_t S32; +typedef uint64_t U64; +#else +typedef unsigned char BYTE; +typedef unsigned short U16; +typedef unsigned int U32; +typedef signed int S32; +typedef uint64_t U64; +#endif + +#if defined(__GNUC__) && !defined(XXH_USE_UNALIGNED_ACCESS) +# define _PACKED __attribute__ ((packed)) +#else +# define _PACKED +#endif + +#if !defined(XXH_USE_UNALIGNED_ACCESS) && !defined(__GNUC__) +# ifdef __IBMC__ +# pragma pack(1) +# else +# pragma pack(push, 1) +# endif +#endif + +typedef struct _U32_S +{ + U32 v; +} _PACKED U32_S; +typedef struct _U64_S +{ + U64 v; +} _PACKED U64_S; + +#if !defined(XXH_USE_UNALIGNED_ACCESS) && !defined(__GNUC__) +# pragma pack(pop) +#endif + +#define A32(x) (((U32_S *)(x))->v) +#define A64(x) (((U64_S *)(x))->v) + + +//*************************************** +// Compiler-specific Functions and Macros +//*************************************** +#define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__) + +// Note : although _rotl exists for minGW (GCC under windows), performance seems poor +#if defined(_MSC_VER) +# define XXH_rotl32(x,r) _rotl(x,r) +# define XXH_rotl64(x,r) _rotl64(x,r) +#else +# define XXH_rotl32(x,r) ((x << r) | (x >> (32 - r))) +# define XXH_rotl64(x,r) ((x << r) | (x >> (64 - r))) +#endif + +#if defined(_MSC_VER) // Visual Studio +# define XXH_swap32 _byteswap_ulong +# define XXH_swap64 _byteswap_uint64 +#elif GCC_VERSION >= 403 || defined(__clang__) +# define XXH_swap32 __builtin_bswap32 +# define XXH_swap64 __builtin_bswap64 +#else +static inline U32 XXH_swap32 (U32 x) +{ + return ((x << 24) & 0xff000000 ) | + ((x << 8) & 0x00ff0000 ) | + ((x >> 8) & 0x0000ff00 ) | + ((x >> 24) & 0x000000ff ); +} +static inline U64 XXH_swap64 (U64 x) +{ + return ((x << 56) & 0xff00000000000000ULL) | + ((x << 40) & 0x00ff000000000000ULL) | + ((x << 24) & 0x0000ff0000000000ULL) | + ((x << 8) & 0x000000ff00000000ULL) | + ((x >> 8) & 0x00000000ff000000ULL) | + ((x >> 24) & 0x0000000000ff0000ULL) | + ((x >> 40) & 0x000000000000ff00ULL) | + ((x >> 56) & 0x00000000000000ffULL); +} +#endif + + +//************************************** +// Constants +//************************************** +#define PRIME32_1 2654435761U +#define PRIME32_2 2246822519U +#define PRIME32_3 3266489917U +#define PRIME32_4 668265263U +#define PRIME32_5 374761393U + +#define PRIME64_1 11400714785074694791ULL +#define PRIME64_2 14029467366897019727ULL +#define PRIME64_3 1609587929392839161ULL +#define PRIME64_4 9650029242287828579ULL +#define PRIME64_5 2870177450012600261ULL + +//************************************** +// Architecture Macros +//************************************** +typedef enum { XXH_bigEndian=0, XXH_littleEndian=1 } XXH_endianess; +#ifndef XXH_CPU_LITTLE_ENDIAN // It is possible to define XXH_CPU_LITTLE_ENDIAN externally, for example using a compiler switch +static const int one = 1; +# define XXH_CPU_LITTLE_ENDIAN (*(char*)(&one)) +#endif + + +//************************************** +// Macros +//************************************** +#define XXH_STATIC_ASSERT(c) { enum { XXH_static_assert = 1/(!!(c)) }; } // use only *after* variable declarations + + +//**************************** +// Memory reads +//**************************** +typedef enum { XXH_aligned, XXH_unaligned } XXH_alignment; + +FORCE_INLINE U32 XXH_readLE32_align(const void* ptr, XXH_endianess endian, XXH_alignment align) +{ + if (align==XXH_unaligned) + return endian==XXH_littleEndian ? A32(ptr) : XXH_swap32(A32(ptr)); + else + return endian==XXH_littleEndian ? *(U32*)ptr : XXH_swap32(*(U32*)ptr); +} + +FORCE_INLINE U32 XXH_readLE32(const void* ptr, XXH_endianess endian) +{ + return XXH_readLE32_align(ptr, endian, XXH_unaligned); +} + +FORCE_INLINE U64 XXH_readLE64_align(const void* ptr, XXH_endianess endian, XXH_alignment align) +{ + if (align==XXH_unaligned) + return endian==XXH_littleEndian ? A64(ptr) : XXH_swap64(A64(ptr)); + else + return endian==XXH_littleEndian ? *(U64*)ptr : XXH_swap64(*(U64*)ptr); +} + +FORCE_INLINE U64 XXH_readLE64(const void* ptr, XXH_endianess endian) +{ + return XXH_readLE64_align(ptr, endian, XXH_unaligned); +} + + +//**************************** +// Simple Hash Functions +//**************************** +FORCE_INLINE U32 XXH32_endian_align(const void* input, size_t len, U32 seed, XXH_endianess endian, XXH_alignment align) +{ + const BYTE* p = (const BYTE*)input; + const BYTE* bEnd = p + len; + U32 h32; +#define XXH_get32bits(p) XXH_readLE32_align(p, endian, align) + +#ifdef XXH_ACCEPT_NULL_INPUT_POINTER + if (p==NULL) + { + len=0; + bEnd=p=(const BYTE*)(size_t)16; + } +#endif + + if (len>=16) + { + const BYTE* const limit = bEnd - 16; + U32 v1 = seed + PRIME32_1 + PRIME32_2; + U32 v2 = seed + PRIME32_2; + U32 v3 = seed + 0; + U32 v4 = seed - PRIME32_1; + + do + { + v1 += XXH_get32bits(p) * PRIME32_2; + v1 = XXH_rotl32(v1, 13); + v1 *= PRIME32_1; + p+=4; + v2 += XXH_get32bits(p) * PRIME32_2; + v2 = XXH_rotl32(v2, 13); + v2 *= PRIME32_1; + p+=4; + v3 += XXH_get32bits(p) * PRIME32_2; + v3 = XXH_rotl32(v3, 13); + v3 *= PRIME32_1; + p+=4; + v4 += XXH_get32bits(p) * PRIME32_2; + v4 = XXH_rotl32(v4, 13); + v4 *= PRIME32_1; + p+=4; + } + while (p<=limit); + + h32 = XXH_rotl32(v1, 1) + XXH_rotl32(v2, 7) + XXH_rotl32(v3, 12) + XXH_rotl32(v4, 18); + } + else + { + h32 = seed + PRIME32_5; + } + + h32 += (U32) len; + + while (p+4<=bEnd) + { + h32 += XXH_get32bits(p) * PRIME32_3; + h32 = XXH_rotl32(h32, 17) * PRIME32_4 ; + p+=4; + } + + while (p<bEnd) + { + h32 += (*p) * PRIME32_5; + h32 = XXH_rotl32(h32, 11) * PRIME32_1 ; + p++; + } + + h32 ^= h32 >> 15; + h32 *= PRIME32_2; + h32 ^= h32 >> 13; + h32 *= PRIME32_3; + h32 ^= h32 >> 16; + + return h32; +} + + +unsigned int XXH32 (const void* input, size_t len, unsigned seed) +{ +#if 0 + // Simple version, good for code maintenance, but unfortunately slow for small inputs + XXH32_state_t state; + XXH32_reset(&state, seed); + XXH32_update(&state, input, len); + return XXH32_digest(&state); +#else + XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN; + +# if !defined(XXH_USE_UNALIGNED_ACCESS) + if ((((size_t)input) & 3) == 0) // Input is aligned, let's leverage the speed advantage + { + if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT) + return XXH32_endian_align(input, len, seed, XXH_littleEndian, XXH_aligned); + else + return XXH32_endian_align(input, len, seed, XXH_bigEndian, XXH_aligned); + } +# endif + + if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT) + return XXH32_endian_align(input, len, seed, XXH_littleEndian, XXH_unaligned); + else + return XXH32_endian_align(input, len, seed, XXH_bigEndian, XXH_unaligned); +#endif +} + +FORCE_INLINE U64 XXH64_endian_align(const void* input, size_t len, U64 seed, XXH_endianess endian, XXH_alignment align) +{ + const BYTE* p = (const BYTE*)input; + const BYTE* bEnd = p + len; + U64 h64; +#define XXH_get64bits(p) XXH_readLE64_align(p, endian, align) + +#ifdef XXH_ACCEPT_NULL_INPUT_POINTER + if (p==NULL) + { + len=0; + bEnd=p=(const BYTE*)(size_t)32; + } +#endif + + if (len>=32) + { + const BYTE* const limit = bEnd - 32; + U64 v1 = seed + PRIME64_1 + PRIME64_2; + U64 v2 = seed + PRIME64_2; + U64 v3 = seed + 0; + U64 v4 = seed - PRIME64_1; + + do + { + v1 += XXH_get64bits(p) * PRIME64_2; + p+=8; + v1 = XXH_rotl64(v1, 31); + v1 *= PRIME64_1; + v2 += XXH_get64bits(p) * PRIME64_2; + p+=8; + v2 = XXH_rotl64(v2, 31); + v2 *= PRIME64_1; + v3 += XXH_get64bits(p) * PRIME64_2; + p+=8; + v3 = XXH_rotl64(v3, 31); + v3 *= PRIME64_1; + v4 += XXH_get64bits(p) * PRIME64_2; + p+=8; + v4 = XXH_rotl64(v4, 31); + v4 *= PRIME64_1; + } + while (p<=limit); + + h64 = XXH_rotl64(v1, 1) + XXH_rotl64(v2, 7) + XXH_rotl64(v3, 12) + XXH_rotl64(v4, 18); + + v1 *= PRIME64_2; + v1 = XXH_rotl64(v1, 31); + v1 *= PRIME64_1; + h64 ^= v1; + h64 = h64 * PRIME64_1 + PRIME64_4; + + v2 *= PRIME64_2; + v2 = XXH_rotl64(v2, 31); + v2 *= PRIME64_1; + h64 ^= v2; + h64 = h64 * PRIME64_1 + PRIME64_4; + + v3 *= PRIME64_2; + v3 = XXH_rotl64(v3, 31); + v3 *= PRIME64_1; + h64 ^= v3; + h64 = h64 * PRIME64_1 + PRIME64_4; + + v4 *= PRIME64_2; + v4 = XXH_rotl64(v4, 31); + v4 *= PRIME64_1; + h64 ^= v4; + h64 = h64 * PRIME64_1 + PRIME64_4; + } + else + { + h64 = seed + PRIME64_5; + } + + h64 += (U64) len; + + while (p+8<=bEnd) + { + U64 k1 = XXH_get64bits(p); + k1 *= PRIME64_2; + k1 = XXH_rotl64(k1,31); + k1 *= PRIME64_1; + h64 ^= k1; + h64 = XXH_rotl64(h64,27) * PRIME64_1 + PRIME64_4; + p+=8; + } + + if (p+4<=bEnd) + { + h64 ^= (U64)(XXH_get32bits(p)) * PRIME64_1; + h64 = XXH_rotl64(h64, 23) * PRIME64_2 + PRIME64_3; + p+=4; + } + + while (p<bEnd) + { + h64 ^= (*p) * PRIME64_5; + h64 = XXH_rotl64(h64, 11) * PRIME64_1; + p++; + } + + h64 ^= h64 >> 33; + h64 *= PRIME64_2; + h64 ^= h64 >> 29; + h64 *= PRIME64_3; + h64 ^= h64 >> 32; + + return h64; +} + + +uint64_t XXH64 (const void* input, size_t len, uint64_t seed) +{ +#if 0 + // Simple version, good for code maintenance, but unfortunately slow for small inputs + XXH64_state_t state; + XXH64_reset(&state, seed); + XXH64_update(&state, input, len); + return XXH64_digest(&state); +#else + XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN; + +# if !defined(XXH_USE_UNALIGNED_ACCESS) + if ((((size_t)input) & 7)==0) // Input is aligned, let's leverage the speed advantage + { + if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT) + return XXH64_endian_align(input, len, seed, XXH_littleEndian, XXH_aligned); + else + return XXH64_endian_align(input, len, seed, XXH_bigEndian, XXH_aligned); + } +# endif + + if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT) + return XXH64_endian_align(input, len, seed, XXH_littleEndian, XXH_unaligned); + else + return XXH64_endian_align(input, len, seed, XXH_bigEndian, XXH_unaligned); +#endif +} + +/**************************************************** + * Advanced Hash Functions +****************************************************/ + +/*** Allocation ***/ +typedef struct +{ + U64 total_len; + U32 seed; + U32 v1; + U32 v2; + U32 v3; + U32 v4; + U32 mem32[4]; /* defined as U32 for alignment */ + U32 memsize; +} XXH_istate32_t; + +typedef struct +{ + U64 total_len; + U64 seed; + U64 v1; + U64 v2; + U64 v3; + U64 v4; + U64 mem64[4]; /* defined as U64 for alignment */ + U32 memsize; +} XXH_istate64_t; + + +XXH32_state_t* XXH32_createState(void) +{ + XXH_STATIC_ASSERT(sizeof(XXH32_state_t) >= sizeof(XXH_istate32_t)); // A compilation error here means XXH32_state_t is not large enough + return (XXH32_state_t*)XXH_malloc(sizeof(XXH32_state_t)); +} + +void* XXH32_init (unsigned seed) +{ + XXH32_state_t *st = XXH32_createState(); + XXH32_reset(st, seed); + + return st; +} + +XXH_errorcode XXH32_freeState(XXH32_state_t* statePtr) +{ + XXH_free(statePtr); + return XXH_OK; +}; + +XXH64_state_t* XXH64_createState(void) +{ + XXH_STATIC_ASSERT(sizeof(XXH64_state_t) >= sizeof(XXH_istate64_t)); // A compilation error here means XXH64_state_t is not large enough + return (XXH64_state_t*)XXH_malloc(sizeof(XXH64_state_t)); +} +XXH_errorcode XXH64_freeState(XXH64_state_t* statePtr) +{ + XXH_free(statePtr); + return XXH_OK; +}; + + +/*** Hash feed ***/ + +XXH_errorcode XXH32_reset(XXH32_state_t* state_in, U32 seed) +{ + XXH_istate32_t* state = (XXH_istate32_t*) state_in; + state->seed = seed; + state->v1 = seed + PRIME32_1 + PRIME32_2; + state->v2 = seed + PRIME32_2; + state->v3 = seed + 0; + state->v4 = seed - PRIME32_1; + state->total_len = 0; + state->memsize = 0; + return XXH_OK; +} + +XXH_errorcode XXH64_reset(XXH64_state_t* state_in, uint64_t seed) +{ + XXH_istate64_t* state = (XXH_istate64_t*) state_in; + state->seed = seed; + state->v1 = seed + PRIME64_1 + PRIME64_2; + state->v2 = seed + PRIME64_2; + state->v3 = seed + 0; + state->v4 = seed - PRIME64_1; + state->total_len = 0; + state->memsize = 0; + return XXH_OK; +} + + +FORCE_INLINE XXH_errorcode XXH32_update_endian (XXH32_state_t* state_in, const void* input, size_t len, XXH_endianess endian) +{ + XXH_istate32_t* state = (XXH_istate32_t *) state_in; + const BYTE* p = (const BYTE*)input; + const BYTE* const bEnd = p + len; + +#ifdef XXH_ACCEPT_NULL_INPUT_POINTER + if (input==NULL) return XXH_ERROR; +#endif + + state->total_len += len; + + if (state->memsize + len < 16) // fill in tmp buffer + { + XXH_memcpy((BYTE*)(state->mem32) + state->memsize, input, len); + state->memsize += (U32)len; + return XXH_OK; + } + + if (state->memsize) // some data left from previous update + { + XXH_memcpy((BYTE*)(state->mem32) + state->memsize, input, 16-state->memsize); + { + const U32* p32 = state->mem32; + state->v1 += XXH_readLE32(p32, endian) * PRIME32_2; + state->v1 = XXH_rotl32(state->v1, 13); + state->v1 *= PRIME32_1; + p32++; + state->v2 += XXH_readLE32(p32, endian) * PRIME32_2; + state->v2 = XXH_rotl32(state->v2, 13); + state->v2 *= PRIME32_1; + p32++; + state->v3 += XXH_readLE32(p32, endian) * PRIME32_2; + state->v3 = XXH_rotl32(state->v3, 13); + state->v3 *= PRIME32_1; + p32++; + state->v4 += XXH_readLE32(p32, endian) * PRIME32_2; + state->v4 = XXH_rotl32(state->v4, 13); + state->v4 *= PRIME32_1; + p32++; + } + p += 16-state->memsize; + state->memsize = 0; + } + + if (p <= bEnd-16) + { + const BYTE* const limit = bEnd - 16; + U32 v1 = state->v1; + U32 v2 = state->v2; + U32 v3 = state->v3; + U32 v4 = state->v4; + + do + { + v1 += XXH_readLE32(p, endian) * PRIME32_2; + v1 = XXH_rotl32(v1, 13); + v1 *= PRIME32_1; + p+=4; + v2 += XXH_readLE32(p, endian) * PRIME32_2; + v2 = XXH_rotl32(v2, 13); + v2 *= PRIME32_1; + p+=4; + v3 += XXH_readLE32(p, endian) * PRIME32_2; + v3 = XXH_rotl32(v3, 13); + v3 *= PRIME32_1; + p+=4; + v4 += XXH_readLE32(p, endian) * PRIME32_2; + v4 = XXH_rotl32(v4, 13); + v4 *= PRIME32_1; + p+=4; + } + while (p<=limit); + + state->v1 = v1; + state->v2 = v2; + state->v3 = v3; + state->v4 = v4; + } + + if (p < bEnd) + { + XXH_memcpy(state->mem32, p, bEnd-p); + state->memsize = (int)(bEnd-p); + } + + return XXH_OK; +} + +XXH_errorcode XXH32_update (XXH32_state_t* state_in, const void* input, size_t len) +{ + XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN; + + if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT) + return XXH32_update_endian(state_in, input, len, XXH_littleEndian); + else + return XXH32_update_endian(state_in, input, len, XXH_bigEndian); +} + + + +FORCE_INLINE U32 XXH32_digest_endian (const XXH32_state_t* state_in, XXH_endianess endian) +{ + XXH_istate32_t* state = (XXH_istate32_t*) state_in; + const BYTE * p = (const BYTE*)state->mem32; + BYTE* bEnd = (BYTE*)(state->mem32) + state->memsize; + U32 h32; + + if (state->total_len >= 16) + { + h32 = XXH_rotl32(state->v1, 1) + XXH_rotl32(state->v2, 7) + XXH_rotl32(state->v3, 12) + XXH_rotl32(state->v4, 18); + } + else + { + h32 = state->seed + PRIME32_5; + } + + h32 += (U32) state->total_len; + + while (p+4<=bEnd) + { + h32 += XXH_readLE32(p, endian) * PRIME32_3; + h32 = XXH_rotl32(h32, 17) * PRIME32_4; + p+=4; + } + + while (p<bEnd) + { + h32 += (*p) * PRIME32_5; + h32 = XXH_rotl32(h32, 11) * PRIME32_1; + p++; + } + + h32 ^= h32 >> 15; + h32 *= PRIME32_2; + h32 ^= h32 >> 13; + h32 *= PRIME32_3; + h32 ^= h32 >> 16; +#if 0 + XXH32_freeState((XXH32_state_t *)state_in); +#endif + return h32; +} + + +U32 XXH32_digest (const XXH32_state_t* state_in) +{ + XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN; + + if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT) + return XXH32_digest_endian(state_in, XXH_littleEndian); + else + return XXH32_digest_endian(state_in, XXH_bigEndian); +} + + +FORCE_INLINE XXH_errorcode XXH64_update_endian (XXH64_state_t* state_in, const void* input, size_t len, XXH_endianess endian) +{ + XXH_istate64_t * state = (XXH_istate64_t *) state_in; + const BYTE* p = (const BYTE*)input; + const BYTE* const bEnd = p + len; + +#ifdef XXH_ACCEPT_NULL_INPUT_POINTER + if (input==NULL) return XXH_ERROR; +#endif + + state->total_len += len; + + if (state->memsize + len < 32) // fill in tmp buffer + { + XXH_memcpy(((BYTE*)state->mem64) + state->memsize, input, len); + state->memsize += (U32)len; + return XXH_OK; + } + + if (state->memsize) // some data left from previous update + { + XXH_memcpy(((BYTE*)state->mem64) + state->memsize, input, 32-state->memsize); + { + const U64* p64 = state->mem64; + state->v1 += XXH_readLE64(p64, endian) * PRIME64_2; + state->v1 = XXH_rotl64(state->v1, 31); + state->v1 *= PRIME64_1; + p64++; + state->v2 += XXH_readLE64(p64, endian) * PRIME64_2; + state->v2 = XXH_rotl64(state->v2, 31); + state->v2 *= PRIME64_1; + p64++; + state->v3 += XXH_readLE64(p64, endian) * PRIME64_2; + state->v3 = XXH_rotl64(state->v3, 31); + state->v3 *= PRIME64_1; + p64++; + state->v4 += XXH_readLE64(p64, endian) * PRIME64_2; + state->v4 = XXH_rotl64(state->v4, 31); + state->v4 *= PRIME64_1; + p64++; + } + p += 32-state->memsize; + state->memsize = 0; + } + + if (p+32 <= bEnd) + { + const BYTE* const limit = bEnd - 32; + U64 v1 = state->v1; + U64 v2 = state->v2; + U64 v3 = state->v3; + U64 v4 = state->v4; + + do + { + v1 += XXH_readLE64(p, endian) * PRIME64_2; + v1 = XXH_rotl64(v1, 31); + v1 *= PRIME64_1; + p+=8; + v2 += XXH_readLE64(p, endian) * PRIME64_2; + v2 = XXH_rotl64(v2, 31); + v2 *= PRIME64_1; + p+=8; + v3 += XXH_readLE64(p, endian) * PRIME64_2; + v3 = XXH_rotl64(v3, 31); + v3 *= PRIME64_1; + p+=8; + v4 += XXH_readLE64(p, endian) * PRIME64_2; + v4 = XXH_rotl64(v4, 31); + v4 *= PRIME64_1; + p+=8; + } + while (p<=limit); + + state->v1 = v1; + state->v2 = v2; + state->v3 = v3; + state->v4 = v4; + } + + if (p < bEnd) + { + XXH_memcpy(state->mem64, p, bEnd-p); + state->memsize = (int)(bEnd-p); + } + + return XXH_OK; +} + +XXH_errorcode XXH64_update (XXH64_state_t* state_in, const void* input, size_t len) +{ + XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN; + + if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT) + return XXH64_update_endian(state_in, input, len, XXH_littleEndian); + else + return XXH64_update_endian(state_in, input, len, XXH_bigEndian); +} + + + +FORCE_INLINE U64 XXH64_digest_endian (const XXH64_state_t* state_in, XXH_endianess endian) +{ + XXH_istate64_t * state = (XXH_istate64_t *) state_in; + const BYTE * p = (const BYTE*)state->mem64; + BYTE* bEnd = (BYTE*)state->mem64 + state->memsize; + U64 h64; + + if (state->total_len >= 32) + { + U64 v1 = state->v1; + U64 v2 = state->v2; + U64 v3 = state->v3; + U64 v4 = state->v4; + + h64 = XXH_rotl64(v1, 1) + XXH_rotl64(v2, 7) + XXH_rotl64(v3, 12) + XXH_rotl64(v4, 18); + + v1 *= PRIME64_2; + v1 = XXH_rotl64(v1, 31); + v1 *= PRIME64_1; + h64 ^= v1; + h64 = h64*PRIME64_1 + PRIME64_4; + + v2 *= PRIME64_2; + v2 = XXH_rotl64(v2, 31); + v2 *= PRIME64_1; + h64 ^= v2; + h64 = h64*PRIME64_1 + PRIME64_4; + + v3 *= PRIME64_2; + v3 = XXH_rotl64(v3, 31); + v3 *= PRIME64_1; + h64 ^= v3; + h64 = h64*PRIME64_1 + PRIME64_4; + + v4 *= PRIME64_2; + v4 = XXH_rotl64(v4, 31); + v4 *= PRIME64_1; + h64 ^= v4; + h64 = h64*PRIME64_1 + PRIME64_4; + } + else + { + h64 = state->seed + PRIME64_5; + } + + h64 += (U64) state->total_len; + + while (p+8<=bEnd) + { + U64 k1 = XXH_readLE64(p, endian); + k1 *= PRIME64_2; + k1 = XXH_rotl64(k1,31); + k1 *= PRIME64_1; + h64 ^= k1; + h64 = XXH_rotl64(h64,27) * PRIME64_1 + PRIME64_4; + p+=8; + } + + if (p+4<=bEnd) + { + h64 ^= (U64)(XXH_readLE32(p, endian)) * PRIME64_1; + h64 = XXH_rotl64(h64, 23) * PRIME64_2 + PRIME64_3; + p+=4; + } + + while (p<bEnd) + { + h64 ^= (*p) * PRIME64_5; + h64 = XXH_rotl64(h64, 11) * PRIME64_1; + p++; + } + + h64 ^= h64 >> 33; + h64 *= PRIME64_2; + h64 ^= h64 >> 29; + h64 *= PRIME64_3; + h64 ^= h64 >> 32; +#if 0 + XXH64_freeState((XXH64_state_t *)state_in); +#endif + return h64; +} + + +uint64_t XXH64_digest (const XXH64_state_t* state_in) +{ + XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN; + + if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT) + return XXH64_digest_endian(state_in, XXH_littleEndian); + else + return XXH64_digest_endian(state_in, XXH_bigEndian); +} + + diff --git a/contrib/libucl/src/xxhash.h b/contrib/libucl/src/xxhash.h new file mode 100644 index 0000000..49b0239 --- /dev/null +++ b/contrib/libucl/src/xxhash.h @@ -0,0 +1,165 @@ +/* + xxHash - Extremely Fast Hash algorithm + Header File + Copyright (C) 2012-2014, Yann Collet. + BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php) + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions are + met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following disclaimer + in the documentation and/or other materials provided with the + distribution. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + You can contact the author at : + - xxHash source repository : http://code.google.com/p/xxhash/ +*/ + +/* Notice extracted from xxHash homepage : + +xxHash is an extremely fast Hash algorithm, running at RAM speed limits. +It also successfully passes all tests from the SMHasher suite. + +Comparison (single thread, Windows Seven 32 bits, using SMHasher on a Core 2 Duo @3GHz) + +Name Speed Q.Score Author +xxHash 5.4 GB/s 10 +CrapWow 3.2 GB/s 2 Andrew +MumurHash 3a 2.7 GB/s 10 Austin Appleby +SpookyHash 2.0 GB/s 10 Bob Jenkins +SBox 1.4 GB/s 9 Bret Mulvey +Lookup3 1.2 GB/s 9 Bob Jenkins +SuperFastHash 1.2 GB/s 1 Paul Hsieh +CityHash64 1.05 GB/s 10 Pike & Alakuijala +FNV 0.55 GB/s 5 Fowler, Noll, Vo +CRC32 0.43 GB/s 9 +MD5-32 0.33 GB/s 10 Ronald L. Rivest +SHA1-32 0.28 GB/s 10 + +Q.Score is a measure of quality of the hash function. +It depends on successfully passing SMHasher test set. +10 is a perfect score. +*/ +#ifndef LIBUCL_XXHASH_H +#define LIBUCL_XXHASH_H + +#if defined (__cplusplus) +extern "C" { +#endif + + +/***************************** + Includes +*****************************/ +#include <stddef.h> /* size_t */ +#include <stdint.h> + + +/***************************** + Type +*****************************/ +typedef enum { XXH_OK=0, XXH_ERROR } XXH_errorcode; + + + +/***************************** + Simple Hash Functions +*****************************/ + +unsigned int XXH32 (const void* input, size_t length, unsigned seed); +uint64_t XXH64 (const void* input, size_t length, uint64_t seed); + +/* +XXH32() : + Calculate the 32-bits hash of sequence "length" bytes stored at memory address "input". + The memory between input & input+length must be valid (allocated and read-accessible). + "seed" can be used to alter the result predictably. + This function successfully passes all SMHasher tests. + Speed on Core 2 Duo @ 3 GHz (single thread, SMHasher benchmark) : 5.4 GB/s +XXH64() : + Calculate the 64-bits hash of sequence of length "len" stored at memory address "input". +*/ + + + +/***************************** + Advanced Hash Functions +*****************************/ +typedef struct { int64_t ll[ 6]; } XXH32_state_t; +typedef struct { int64_t ll[11]; } XXH64_state_t; + +/* +These structures allow static allocation of XXH states. +States must then be initialized using XXHnn_reset() before first use. + +If you prefer dynamic allocation, please refer to functions below. +*/ + +/* + * !!! + * Rspamd specific: we use the legacy method to free state when digest is obtained + * !!! + */ +void * XXH32_init (unsigned seed); +XXH32_state_t* XXH32_createState(void); +XXH_errorcode XXH32_freeState(XXH32_state_t* statePtr); + +XXH64_state_t* XXH64_createState(void); +XXH_errorcode XXH64_freeState(XXH64_state_t* statePtr); + +/* +These functions create and release memory for XXH state. +States must then be initialized using XXHnn_reset() before first use. +*/ + + +XXH_errorcode XXH32_reset (XXH32_state_t* statePtr, unsigned seed); +XXH_errorcode XXH32_update (XXH32_state_t* statePtr, const void* input, size_t length); +unsigned int XXH32_digest (const XXH32_state_t* statePtr); + +XXH_errorcode XXH64_reset (XXH64_state_t* statePtr, uint64_t seed); +XXH_errorcode XXH64_update (XXH64_state_t* statePtr, const void* input, size_t length); +uint64_t XXH64_digest (const XXH64_state_t* statePtr); + +/* +These functions calculate the xxHash of an input provided in multiple smaller packets, +as opposed to an input provided as a single block. + +XXH state space must first be allocated, using either static or dynamic method provided above. + +Start a new hash by initializing state with a seed, using XXHnn_reset(). + +Then, feed the hash state by calling XXHnn_update() as many times as necessary. +Obviously, input must be valid, meaning allocated and read accessible. +The function returns an error code, with 0 meaning OK, and any other value meaning there is an error. + +Finally, you can produce a hash anytime, by using XXHnn_digest(). +This function returns the final nn-bits hash. +You can nonetheless continue feeding the hash state with more input, +and therefore get some new hashes, by calling again XXHnn_digest(). + +When you are done, don't forget to free XXH state space, using typically XXHnn_freeState(). +*/ + + +#if defined (__cplusplus) +} +#endif + +#endif diff --git a/contrib/libucl/stamp-h.in b/contrib/libucl/stamp-h.in new file mode 100644 index 0000000..9788f70 --- /dev/null +++ b/contrib/libucl/stamp-h.in @@ -0,0 +1 @@ +timestamp diff --git a/contrib/libucl/tests/.gitignore b/contrib/libucl/tests/.gitignore new file mode 100644 index 0000000..5a48681 --- /dev/null +++ b/contrib/libucl/tests/.gitignore @@ -0,0 +1,8 @@ +*.log +*.trs +*.plist + +test_basic +test_generate +test_schema +test_speed diff --git a/contrib/libucl/tests/Makefile.am b/contrib/libucl/tests/Makefile.am new file mode 100644 index 0000000..055eb8b --- /dev/null +++ b/contrib/libucl/tests/Makefile.am @@ -0,0 +1,45 @@ +EXTRA_DIST = $(TESTS) basic schema generate.res \ + streamline.res rcl_test.json.xz + +TESTS = basic.test \ + generate.test \ + schema.test \ + msgpack.test \ + speed.test \ + msgpack.test +TESTS_ENVIRONMENT = $(SH) \ + TEST_DIR=$(top_srcdir)/tests \ + TEST_OUT_DIR=$(top_builddir)/tests \ + TEST_BINARY_DIR=$(top_builddir)/tests + +common_test_cflags = -I$(top_srcdir)/include \ + -I$(top_srcdir)/src \ + -I$(top_srcdir)/uthash +common_test_ldadd = $(top_builddir)/src/libucl.la + +test_basic_SOURCES = test_basic.c +test_basic_LDADD = $(common_test_ldadd) +test_basic_CFLAGS = $(common_test_cflags) + +test_speed_SOURCES = test_speed.c +test_speed_LDADD = $(common_test_ldadd) +test_speed_CFLAGS = $(common_test_cflags) + +test_generate_SOURCES = test_generate.c +test_generate_LDADD = $(common_test_ldadd) +test_generate_CFLAGS = $(common_test_cflags) + +test_schema_SOURCES = test_schema.c +test_schema_LDADD = $(common_test_ldadd) +test_schema_CFLAGS = $(common_test_cflags) + +test_streamline_SOURCES = test_streamline.c +test_streamline_LDADD = $(common_test_ldadd) +test_streamline_CFLAGS = $(common_test_cflags) + +test_msgpack_SOURCES = test_msgpack.c +test_msgpack_LDADD = $(common_test_ldadd) +test_msgpack_CFLAGS = $(common_test_cflags) + +check_PROGRAMS = test_basic test_speed test_generate test_schema test_streamline \ + test_msgpack
\ No newline at end of file diff --git a/contrib/libucl/tests/basic.test b/contrib/libucl/tests/basic.test new file mode 100755 index 0000000..a4a07eb --- /dev/null +++ b/contrib/libucl/tests/basic.test @@ -0,0 +1,26 @@ +#!/bin/sh + +PROG=${TEST_BINARY_DIR}/test_basic + +for _tin in ${TEST_DIR}/basic/*.in ; do + _t=`echo $_tin | sed -e 's/.in$//'` + _out=${TEST_OUT_DIR}/basic.out + $PROG $_t.in $_out + if [ $? -ne 0 ] ; then + echo "Test: $_t failed, output:" + cat $_out + rm $_out + exit 1 + fi + if [ -f $_t.res ] ; then + diff -s $_out $_t.res -u 2>/dev/null + if [ $? -ne 0 ] ; then + rm $_out + echo "Test: $_t output missmatch" + exit 1 + fi + fi + rm $_out +done + + diff --git a/contrib/libucl/tests/basic/1.in b/contrib/libucl/tests/basic/1.in new file mode 100644 index 0000000..9c14df6 --- /dev/null +++ b/contrib/libucl/tests/basic/1.in @@ -0,0 +1,14 @@ +{ +"key1": value; +"key1": value2; +"key1": "value;" +"key1": 1.0, +"key1": -0xdeadbeef +"key1": 0xdeadbeef.1 +"key1": 0xreadbeef +"key1": -1e-10, +"key1": 1 +"key1": true +"key1": no +"key1": yes +} diff --git a/contrib/libucl/tests/basic/1.res b/contrib/libucl/tests/basic/1.res new file mode 100644 index 0000000..660af9f --- /dev/null +++ b/contrib/libucl/tests/basic/1.res @@ -0,0 +1,13 @@ +key1 = "value"; +key1 = "value2"; +key1 = "value;"; +key1 = 1.0; +key1 = -3735928559; +key1 = "0xdeadbeef.1"; +key1 = "0xreadbeef"; +key1 = -1e-10; +key1 = 1; +key1 = true; +key1 = false; +key1 = true; + diff --git a/contrib/libucl/tests/basic/10.in b/contrib/libucl/tests/basic/10.in new file mode 100644 index 0000000..e0910c1 --- /dev/null +++ b/contrib/libucl/tests/basic/10.in @@ -0,0 +1,9 @@ +section foo bar { + key = value +} +section foo baz { + key = value +} +section foo { + bar = lol /* removing this line makes parsing successful */ +} diff --git a/contrib/libucl/tests/basic/11.in b/contrib/libucl/tests/basic/11.in new file mode 100644 index 0000000..3131fc7 --- /dev/null +++ b/contrib/libucl/tests/basic/11.in @@ -0,0 +1,3 @@ + + + {"key": "value"} diff --git a/contrib/libucl/tests/basic/11.res b/contrib/libucl/tests/basic/11.res new file mode 100644 index 0000000..4b17c4b --- /dev/null +++ b/contrib/libucl/tests/basic/11.res @@ -0,0 +1,2 @@ +key = "value"; + diff --git a/contrib/libucl/tests/basic/12.in b/contrib/libucl/tests/basic/12.in new file mode 100644 index 0000000..b63a32f --- /dev/null +++ b/contrib/libucl/tests/basic/12.in @@ -0,0 +1,2 @@ +key1: 12 , +key2: 12 value diff --git a/contrib/libucl/tests/basic/12.res b/contrib/libucl/tests/basic/12.res new file mode 100644 index 0000000..163af46 --- /dev/null +++ b/contrib/libucl/tests/basic/12.res @@ -0,0 +1,3 @@ +key1 = 12; +key2 = "12 value"; + diff --git a/contrib/libucl/tests/basic/13.in b/contrib/libucl/tests/basic/13.in new file mode 100644 index 0000000..6e31e9c --- /dev/null +++ b/contrib/libucl/tests/basic/13.in @@ -0,0 +1,9 @@ +key = value_orig; + +# test glob +.include(glob=true,something="test") "${CURDIR}/include_dir/test*.conf" + +.include(priority=1) "${CURDIR}/include_dir/pri1.conf" +.include(priority=2) "${CURDIR}/include_dir/pri2.conf" + +.include(try=true) "${CURDIR}/include_dir/invalid.conf" diff --git a/contrib/libucl/tests/basic/13.res b/contrib/libucl/tests/basic/13.res new file mode 100644 index 0000000..aa13a54 --- /dev/null +++ b/contrib/libucl/tests/basic/13.res @@ -0,0 +1,8 @@ +key = "value_orig"; +key = "value1"; +key = "value2"; +key = "value3"; +key_pri = "priority2"; +key_trace1 = "pri1"; +key_trace2 = "pri2"; + diff --git a/contrib/libucl/tests/basic/14.in b/contrib/libucl/tests/basic/14.in new file mode 100644 index 0000000..62d2e60 --- /dev/null +++ b/contrib/libucl/tests/basic/14.in @@ -0,0 +1,8 @@ +# Bad comments case + +section { +# key = value; +} +.include(try=true) "./1.in" + +key = value; diff --git a/contrib/libucl/tests/basic/14.res b/contrib/libucl/tests/basic/14.res new file mode 100644 index 0000000..f803349 --- /dev/null +++ b/contrib/libucl/tests/basic/14.res @@ -0,0 +1,4 @@ +section { +} +key = "value"; + diff --git a/contrib/libucl/tests/basic/15.in b/contrib/libucl/tests/basic/15.in new file mode 100644 index 0000000..1329b8c --- /dev/null +++ b/contrib/libucl/tests/basic/15.in @@ -0,0 +1,24 @@ +# In this test we test include override bug + +.include(priority = 1) "${CURDIR}/15.inc" + +section = { + value = "test"; +} + +overrided = { + value = "not-to-be-shown"; +} + +/* + BUGGED UCL: + overrided { + key = "overrided"; + } + !!! So overrided has actually rewritten the previous key + section { + value { + value = "not-to-be-shown"; + } + } +*/ diff --git a/contrib/libucl/tests/basic/15.inc b/contrib/libucl/tests/basic/15.inc new file mode 100644 index 0000000..a9ab027 --- /dev/null +++ b/contrib/libucl/tests/basic/15.inc @@ -0,0 +1,3 @@ +overrided { + key = "overrided"; +} diff --git a/contrib/libucl/tests/basic/15.res b/contrib/libucl/tests/basic/15.res new file mode 100644 index 0000000..b6b4e26 --- /dev/null +++ b/contrib/libucl/tests/basic/15.res @@ -0,0 +1,7 @@ +overrided { + key = "overrided"; +} +section { + value = "test"; +} + diff --git a/contrib/libucl/tests/basic/16.in b/contrib/libucl/tests/basic/16.in new file mode 100644 index 0000000..0712252 --- /dev/null +++ b/contrib/libucl/tests/basic/16.in @@ -0,0 +1,12 @@ +.include(priority = 1) "${CURDIR}/16.inc" + +section = { + value = "test"; +} + +overrided = { + value = "not-to-be-shown"; +} +overrided = { + value2 = "implicit-array"; +} diff --git a/contrib/libucl/tests/basic/16.inc b/contrib/libucl/tests/basic/16.inc new file mode 100644 index 0000000..34b6085 --- /dev/null +++ b/contrib/libucl/tests/basic/16.inc @@ -0,0 +1,3 @@ +overrided { + key = "overrided"; +} diff --git a/contrib/libucl/tests/basic/16.res b/contrib/libucl/tests/basic/16.res new file mode 100644 index 0000000..b6b4e26 --- /dev/null +++ b/contrib/libucl/tests/basic/16.res @@ -0,0 +1,7 @@ +overrided { + key = "overrided"; +} +section { + value = "test"; +} + diff --git a/contrib/libucl/tests/basic/17.in b/contrib/libucl/tests/basic/17.in new file mode 100644 index 0000000..bae66fb --- /dev/null +++ b/contrib/libucl/tests/basic/17.in @@ -0,0 +1,2 @@ +# issue 74 +string that ends in slash\
\ No newline at end of file diff --git a/contrib/libucl/tests/basic/17.res b/contrib/libucl/tests/basic/17.res new file mode 100644 index 0000000..9870b04 --- /dev/null +++ b/contrib/libucl/tests/basic/17.res @@ -0,0 +1,2 @@ +string = "that ends in slash\\"; + diff --git a/contrib/libucl/tests/basic/18.in b/contrib/libucl/tests/basic/18.in new file mode 100644 index 0000000..bd12c01 --- /dev/null +++ b/contrib/libucl/tests/basic/18.in @@ -0,0 +1,10 @@ +defaults { + key = "val" + foo = "bar" + many = "values here" +} + +mything { + .inherit "defaults" + key = "newval" +} diff --git a/contrib/libucl/tests/basic/18.res b/contrib/libucl/tests/basic/18.res new file mode 100644 index 0000000..a6272d5 --- /dev/null +++ b/contrib/libucl/tests/basic/18.res @@ -0,0 +1,11 @@ +defaults { + key = "val"; + foo = "bar"; + many = "values here"; +} +mything { + key = "newval"; + foo = "bar"; + many = "values here"; +} + diff --git a/contrib/libucl/tests/basic/19-append.inc b/contrib/libucl/tests/basic/19-append.inc new file mode 100644 index 0000000..909dfce --- /dev/null +++ b/contrib/libucl/tests/basic/19-append.inc @@ -0,0 +1,8 @@ +okey_append = { + key = value1; + key1 = value2 +} + +akey_append = ["value3"]; + +skey_append = "value4"; diff --git a/contrib/libucl/tests/basic/19-merge.inc b/contrib/libucl/tests/basic/19-merge.inc new file mode 100644 index 0000000..46deccb --- /dev/null +++ b/contrib/libucl/tests/basic/19-merge.inc @@ -0,0 +1,8 @@ +okey_merge = { + key = value1; + key1 = value2; +} + +akey_merge = ["value3"]; + +skey_merge = "value4"; diff --git a/contrib/libucl/tests/basic/19-rewrite.inc b/contrib/libucl/tests/basic/19-rewrite.inc new file mode 100644 index 0000000..5a7094b --- /dev/null +++ b/contrib/libucl/tests/basic/19-rewrite.inc @@ -0,0 +1,8 @@ +okey_rewrite = { + key = value1; + key1 = value2; +} + +akey_rewrite = ["value3"]; + +skey_rewrite = "value4"; diff --git a/contrib/libucl/tests/basic/19.in b/contrib/libucl/tests/basic/19.in new file mode 100644 index 0000000..717aed8 --- /dev/null +++ b/contrib/libucl/tests/basic/19.in @@ -0,0 +1,28 @@ +okey_append = { + key = value; +} + +akey_append = ["value"]; + +skey_append = "value"; + +okey_merge = { + key = value; + source = original; +} + +akey_merge = ["value"]; + +skey_merge = "value"; + +okey_rewrite = { + key = value; +} + +akey_rewrite = ["value"]; + +skey_rewrite = "value"; + +.include(duplicate="append") "${CURDIR}/19-append.inc" +.include(duplicate="merge") "${CURDIR}/19-merge.inc" +.include(duplicate="rewrite") "${CURDIR}/19-rewrite.inc" diff --git a/contrib/libucl/tests/basic/19.res b/contrib/libucl/tests/basic/19.res new file mode 100644 index 0000000..ca7bc41 --- /dev/null +++ b/contrib/libucl/tests/basic/19.res @@ -0,0 +1,36 @@ +okey_append { + key = "value"; +} +okey_append { + key = "value1"; + key1 = "value2"; +} +akey_append [ + "value", +] +akey_append [ + "value3", +] +skey_append = "value"; +skey_append = "value4"; +okey_merge { + key = "value"; + key = "value1"; + source = "original"; + key1 = "value2"; +} +akey_merge [ + "value", + "value3", +] +skey_merge = "value"; +skey_merge = "value4"; +okey_rewrite { + key = "value1"; + key1 = "value2"; +} +akey_rewrite [ + "value3", +] +skey_rewrite = "value4"; + diff --git a/contrib/libucl/tests/basic/2.in b/contrib/libucl/tests/basic/2.in new file mode 100644 index 0000000..a4419a4 --- /dev/null +++ b/contrib/libucl/tests/basic/2.in @@ -0,0 +1,24 @@ +section1 { param1 = value; param2 = value, +section3 {param = value; param2 = value, param3 = ["value1", value2, 100500]}} +section2 { param1 = {key = value}, param1 = ["key"]} + +# Numbers +key1 = 1s +key2 = 1min +key3 = 1kb +key4 = 5M +key5 = 10mS +key6 = 10y + +# Strings +key1 = "some string"; +key2 = /some/path; +key3 = 111some, +key4: s1, +"key5": "\n\r123" + +# Variables +keyvar = "$ABItest"; +keyvar = "${ABI}$ABI${ABI}${$ABI}"; +keyvar = "${some}$no${}$$test$$$$$$$"; +keyvar = "$ABI$$ABI$$$ABI$$$$"; diff --git a/contrib/libucl/tests/basic/2.res b/contrib/libucl/tests/basic/2.res new file mode 100644 index 0000000..72e9570 --- /dev/null +++ b/contrib/libucl/tests/basic/2.res @@ -0,0 +1,37 @@ +section1 { + param1 = "value"; + param2 = "value"; + section3 { + param = "value"; + param2 = "value"; + param3 [ + "value1", + "value2", + 100500, + ] + } +} +section2 { + param1 { + key = "value"; + } + param1 [ + "key", + ] +} +key1 = 1.0; +key1 = "some string"; +key2 = 60.0; +key2 = "/some/path"; +key3 = 1024; +key3 = "111some"; +key4 = 5000000; +key4 = "s1"; +key5 = 0.010000; +key5 = "\n\r123"; +key6 = 2207520000.000000; +keyvar = "unknowntest"; +keyvar = "unknownunknownunknown${unknown}"; +keyvar = "${some}$no${}$$test$$$$$$$"; +keyvar = "unknown$ABI$unknown$$"; + diff --git a/contrib/libucl/tests/basic/20.in b/contrib/libucl/tests/basic/20.in new file mode 100644 index 0000000..f9d4088 --- /dev/null +++ b/contrib/libucl/tests/basic/20.in @@ -0,0 +1,2 @@ +# issue 112 +[[0
\ No newline at end of file diff --git a/contrib/libucl/tests/basic/20.res b/contrib/libucl/tests/basic/20.res new file mode 100644 index 0000000..abfbbf0 --- /dev/null +++ b/contrib/libucl/tests/basic/20.res @@ -0,0 +1,5 @@ +[ + [ + 0, + ] +] diff --git a/contrib/libucl/tests/basic/21.in b/contrib/libucl/tests/basic/21.in new file mode 100644 index 0000000..8f4b328 --- /dev/null +++ b/contrib/libucl/tests/basic/21.in @@ -0,0 +1,2 @@ + [9 +{0 [[0
\ No newline at end of file diff --git a/contrib/libucl/tests/basic/21.res b/contrib/libucl/tests/basic/21.res new file mode 100644 index 0000000..db091ce --- /dev/null +++ b/contrib/libucl/tests/basic/21.res @@ -0,0 +1,10 @@ +[ + 9, + { + 0 [ + [ + 0, + ] + ] + } +] diff --git a/contrib/libucl/tests/basic/22.in b/contrib/libucl/tests/basic/22.in new file mode 100644 index 0000000..244cea0 --- /dev/null +++ b/contrib/libucl/tests/basic/22.in @@ -0,0 +1,2 @@ +# issue 113 +ÿ=1
\ No newline at end of file diff --git a/contrib/libucl/tests/basic/22.res b/contrib/libucl/tests/basic/22.res new file mode 100644 index 0000000..d4bf94d --- /dev/null +++ b/contrib/libucl/tests/basic/22.res @@ -0,0 +1,2 @@ +ÿ = 1; + diff --git a/contrib/libucl/tests/basic/3.in b/contrib/libucl/tests/basic/3.in new file mode 100644 index 0000000..b3e3696 --- /dev/null +++ b/contrib/libucl/tests/basic/3.in @@ -0,0 +1,31 @@ +/* + * Pkg conf + */ + +#packagesite http//pkg.freebsd.org/freebsd-9-amd64/latest +#packagesite http//pkg.freebsd.org/freebsd-9-amd64/latest +packagesite: http://pkg-test.freebsd.org/pkg-test/${ABI}/latest +squaretest: some[]value +ALIAS : { + all-depends: query %dn-%dv, + annotations: info -A, + build-depends: info -qd, + download: fetch, + iinfo: info -i -g -x, + isearch: search -i -g -x, + leaf: query -e '%a == 0' '%n-%v', + leaf: query -e '%a == 0' '%n-%v', + list: info -ql, + origin: info -qo, + provided-depends: info -qb, + raw: info -R, + required-depends: info -qr, + shared-depends: info -qB, + show: info -f -k, + size: info -sq, + } + +repo_dirs : [ + /home/bapt, + /usr/local/etc +] diff --git a/contrib/libucl/tests/basic/3.res b/contrib/libucl/tests/basic/3.res new file mode 100644 index 0000000..bb18457 --- /dev/null +++ b/contrib/libucl/tests/basic/3.res @@ -0,0 +1,25 @@ +packagesite = "http://pkg-test.freebsd.org/pkg-test/unknown/latest"; +squaretest = "some[]value"; +alias { + all-depends = "query %dn-%dv"; + annotations = "info -A"; + build-depends = "info -qd"; + download = "fetch"; + iinfo = "info -i -g -x"; + isearch = "search -i -g -x"; + leaf = "query -e '%a == 0' '%n-%v'"; + leaf = "query -e '%a == 0' '%n-%v'"; + list = "info -ql"; + origin = "info -qo"; + provided-depends = "info -qb"; + raw = "info -R"; + required-depends = "info -qr"; + shared-depends = "info -qB"; + show = "info -f -k"; + size = "info -sq"; +} +repo_dirs [ + "/home/bapt", + "/usr/local/etc", +] + diff --git a/contrib/libucl/tests/basic/4.in b/contrib/libucl/tests/basic/4.in new file mode 100644 index 0000000..2b296ef --- /dev/null +++ b/contrib/libucl/tests/basic/4.in @@ -0,0 +1,47 @@ +name : "pkgconf" +version : "0.9.3" +origin : "devel/pkgconf" +comment : "Utility to help to configure compiler and linker flags" +arch : "freebsd:9:x86:64" +maintainer : "bapt@FreeBSD.org" +prefix : "/usr/local" +licenselogic : "single" +licenses : [ + "BSD", +] +flatsize : 60523 +desc : "pkgconf is a program which helps to configure compiler and linker flags for\ndevelopment frameworks. It is similar to pkg-config, but was written from\nscratch in Summer of 2011 to replace pkg-config, which now needs itself to build\nitself.\n\nWWW: https://github.com/pkgconf/pkgconf" +categories : [ + "devel", +] +files : { + /usr/local/bin/pkg-config : "-", + /usr/local/bin/pkgconf : "4a0fc53e5ad64e8085da2e61652d61c50b192a086421d865703f1de9f724da38", + /usr/local/share/aclocal/pkg.m4 : "cffab33d659adfe36497ec57665eec36fa6fb7b007e578e6ac2434cc28be8820", + /usr/local/share/licenses/pkgconf-0.9.3/BSD : "85e7a53b5e2d3e350e2d084fed2f94b7f63005f8e1168740e1e84aa9fa5d48ce", + /usr/local/share/licenses/pkgconf-0.9.3/LICENSE : "d9cce0db43502eb1bd8fbef7e960cfaa43b5647186f7f7379923b336209fd77b", + /usr/local/share/licenses/pkgconf-0.9.3/catalog.mk : "e7b131acce7c3d3c61f2214607b11b34526e03b05afe89a608f50586a898e2ef", +} +directories : { + /usr/local/share/licenses/pkgconf-0.9.3/ : false, + /usr/local/share/licenses/ : true, +} +scripts : { + post-install : "cd /usr/local\nn", + pre-deinstall : "cd /usr/local\nn", + post-deinstall : "cd /usr/local\nn", +} +multiline-key : <<EOD +test +test +test\n +/* comment like */ +# Some invalid endings + EOD +EOD +EOF +# Valid ending + empty string + +EOD + +normal-key : <<EODnot diff --git a/contrib/libucl/tests/basic/4.res b/contrib/libucl/tests/basic/4.res new file mode 100644 index 0000000..2b862fc --- /dev/null +++ b/contrib/libucl/tests/basic/4.res @@ -0,0 +1,54 @@ +name = "pkgconf"; +version = "0.9.3"; +origin = "devel/pkgconf"; +comment = "Utility to help to configure compiler and linker flags"; +arch = "freebsd:9:x86:64"; +maintainer = "bapt@FreeBSD.org"; +prefix = "/usr/local"; +licenselogic = "single"; +licenses [ + "BSD", +] +flatsize = 60523; +desc = <<EOD +pkgconf is a program which helps to configure compiler and linker flags for +development frameworks. It is similar to pkg-config, but was written from +scratch in Summer of 2011 to replace pkg-config, which now needs itself to build +itself. + +WWW: https://github.com/pkgconf/pkgconf +EOD; +categories [ + "devel", +] +files { + /usr/local/bin/pkg-config = "-"; + /usr/local/bin/pkgconf = "4a0fc53e5ad64e8085da2e61652d61c50b192a086421d865703f1de9f724da38"; + /usr/local/share/aclocal/pkg.m4 = "cffab33d659adfe36497ec57665eec36fa6fb7b007e578e6ac2434cc28be8820"; + /usr/local/share/licenses/pkgconf-0.9.3/bsd = "85e7a53b5e2d3e350e2d084fed2f94b7f63005f8e1168740e1e84aa9fa5d48ce"; + /usr/local/share/licenses/pkgconf-0.9.3/license = "d9cce0db43502eb1bd8fbef7e960cfaa43b5647186f7f7379923b336209fd77b"; + /usr/local/share/licenses/pkgconf-0.9.3/catalog.mk = "e7b131acce7c3d3c61f2214607b11b34526e03b05afe89a608f50586a898e2ef"; +} +directories { + /usr/local/share/licenses/pkgconf-0.9.3/ = false; + /usr/local/share/licenses/ = true; +} +scripts { + post-install = "cd /usr/local\nn"; + pre-deinstall = "cd /usr/local\nn"; + post-deinstall = "cd /usr/local\nn"; +} +multiline-key = <<EOD +test +test +test\n +/* comment like */ +# Some invalid endings + EOD +EOD +EOF +# Valid ending + empty string + +EOD; +normal-key = "<<EODnot"; + diff --git a/contrib/libucl/tests/basic/6.in b/contrib/libucl/tests/basic/6.in new file mode 100644 index 0000000..5b46088 --- /dev/null +++ b/contrib/libucl/tests/basic/6.in @@ -0,0 +1,4 @@ + +# test +# +key = value diff --git a/contrib/libucl/tests/basic/6.res b/contrib/libucl/tests/basic/6.res new file mode 100644 index 0000000..4b17c4b --- /dev/null +++ b/contrib/libucl/tests/basic/6.res @@ -0,0 +1,2 @@ +key = "value"; + diff --git a/contrib/libucl/tests/basic/8.in b/contrib/libucl/tests/basic/8.in new file mode 100644 index 0000000..605c0c9 --- /dev/null +++ b/contrib/libucl/tests/basic/8.in @@ -0,0 +1,23 @@ +section blah { # test + param = "value" +} +section test { + key = test; + subsection testsub { + flag on; + subsubsection testsubsub1 testsubsub2 { + key = [1, 2, 3]; + } + } +} + +section test { + /* Empty */ +} + + +section foo { # test + param = 123.2; +} + +array = [] diff --git a/contrib/libucl/tests/basic/8.res b/contrib/libucl/tests/basic/8.res new file mode 100644 index 0000000..54c904e --- /dev/null +++ b/contrib/libucl/tests/basic/8.res @@ -0,0 +1,38 @@ +section { + blah { + param = "value"; + } +} +section { + test { + key = "test"; + subsection { + testsub { + flag = true; + subsubsection { + testsubsub1 { + testsubsub2 { + key [ + 1, + 2, + 3, + ] + } + } + } + } + } + } +} +section { + test { + } +} +section { + foo { + param = 123.200000; + } +} +array [ +] + diff --git a/contrib/libucl/tests/basic/9-comment.inc b/contrib/libucl/tests/basic/9-comment.inc new file mode 100644 index 0000000..8939db1 --- /dev/null +++ b/contrib/libucl/tests/basic/9-comment.inc @@ -0,0 +1 @@ +#key = value diff --git a/contrib/libucl/tests/basic/9-empty.inc b/contrib/libucl/tests/basic/9-empty.inc new file mode 100644 index 0000000..e69de29 --- /dev/null +++ b/contrib/libucl/tests/basic/9-empty.inc diff --git a/contrib/libucl/tests/basic/9.in b/contrib/libucl/tests/basic/9.in new file mode 100644 index 0000000..4c06e39 --- /dev/null +++ b/contrib/libucl/tests/basic/9.in @@ -0,0 +1,12 @@ +.include "$CURDIR/9.inc" +.include "$CURDIR/9-comment.inc" +#.include "$CURDIR/9.inc" +.include "$CURDIR/9.inc" + +key = value; +.include "$CURDIR/9.inc" + +.try_include "/non/existent" +#.try_include "$CURDIR/9.incorrect.inc" +# 9.incorrect.inc contains '{}}' +#key = value; diff --git a/contrib/libucl/tests/basic/9.inc b/contrib/libucl/tests/basic/9.inc new file mode 100644 index 0000000..c5592b8 --- /dev/null +++ b/contrib/libucl/tests/basic/9.inc @@ -0,0 +1 @@ +key1 = value diff --git a/contrib/libucl/tests/basic/9.res b/contrib/libucl/tests/basic/9.res new file mode 100644 index 0000000..ec3f014 --- /dev/null +++ b/contrib/libucl/tests/basic/9.res @@ -0,0 +1,5 @@ +key1 = "value"; +key1 = "value"; +key1 = "value"; +key = "value"; + diff --git a/contrib/libucl/tests/basic/comments.in b/contrib/libucl/tests/basic/comments.in new file mode 100644 index 0000000..3144fa0 --- /dev/null +++ b/contrib/libucl/tests/basic/comments.in @@ -0,0 +1,25 @@ +# This test is intended to check various comments in ucl + +obj { + + key = value + key = "/* value" + /* + key = value + */ +# Nested comments + key = nested + /* + adasdasdads + /* asdasdasd */asjdasjldaskd + /* asdsadasd */ + /* /* /* /* /* */ */ */ */ */ +# some + */ + key = quotes # quoted +# Quotes + /* + key = "/* value" + key = "*/value" + */ +} diff --git a/contrib/libucl/tests/basic/comments.res b/contrib/libucl/tests/basic/comments.res new file mode 100644 index 0000000..4c3a797 --- /dev/null +++ b/contrib/libucl/tests/basic/comments.res @@ -0,0 +1,7 @@ +obj { + key = "value"; + key = "/* value"; + key = "nested"; + key = "quotes"; +} + diff --git a/contrib/libucl/tests/basic/include_dir/invalid.conf b/contrib/libucl/tests/basic/include_dir/invalid.conf new file mode 100644 index 0000000..c3267cb --- /dev/null +++ b/contrib/libucl/tests/basic/include_dir/invalid.conf @@ -0,0 +1 @@ +@@@@ BAD UCL ~~~~ diff --git a/contrib/libucl/tests/basic/include_dir/pri1.conf b/contrib/libucl/tests/basic/include_dir/pri1.conf new file mode 100644 index 0000000..d218433 --- /dev/null +++ b/contrib/libucl/tests/basic/include_dir/pri1.conf @@ -0,0 +1,2 @@ +key_pri = priority1; +key_trace1 = pri1; diff --git a/contrib/libucl/tests/basic/include_dir/pri2.conf b/contrib/libucl/tests/basic/include_dir/pri2.conf new file mode 100644 index 0000000..ee734a8 --- /dev/null +++ b/contrib/libucl/tests/basic/include_dir/pri2.conf @@ -0,0 +1,2 @@ +key_pri = priority2; +key_trace2 = pri2; diff --git a/contrib/libucl/tests/basic/include_dir/test1.conf b/contrib/libucl/tests/basic/include_dir/test1.conf new file mode 100644 index 0000000..61f2906 --- /dev/null +++ b/contrib/libucl/tests/basic/include_dir/test1.conf @@ -0,0 +1 @@ +key = value1; diff --git a/contrib/libucl/tests/basic/include_dir/test2.conf b/contrib/libucl/tests/basic/include_dir/test2.conf new file mode 100644 index 0000000..0b55427 --- /dev/null +++ b/contrib/libucl/tests/basic/include_dir/test2.conf @@ -0,0 +1 @@ +key = value2; diff --git a/contrib/libucl/tests/basic/include_dir/test3.conf b/contrib/libucl/tests/basic/include_dir/test3.conf new file mode 100644 index 0000000..3ceb7dd --- /dev/null +++ b/contrib/libucl/tests/basic/include_dir/test3.conf @@ -0,0 +1 @@ +key = value3; diff --git a/contrib/libucl/tests/generate.res b/contrib/libucl/tests/generate.res new file mode 100644 index 0000000..970b312 --- /dev/null +++ b/contrib/libucl/tests/generate.res @@ -0,0 +1,21 @@ +key0 = 0.100000; +key1 = "test string"; +key2 = "test \\nstring"; +key3 = " test string \n"; +key4 [ + 9.999000, + 10, + 10.100000, +] +key4 = true; +key5 = ""; +key6 = ""; +key7 = " \\n"; +key8 = 1048576; +key9 = 3.140000; +key10 = true; +key11 = false; +key12 = "gslin@gslin.org"; +key13 = "#test"; +"k=3" = true; + diff --git a/contrib/libucl/tests/generate.test b/contrib/libucl/tests/generate.test new file mode 100755 index 0000000..ed237a3 --- /dev/null +++ b/contrib/libucl/tests/generate.test @@ -0,0 +1,13 @@ +#!/bin/sh + +PROG=${TEST_BINARY_DIR}/test_generate + +$PROG ${TEST_OUT_DIR}/generate.out +diff -s ${TEST_OUT_DIR}/generate.out ${TEST_DIR}/generate.res -u 2>/dev/null +if [ $? -ne 0 ] ; then + rm ${TEST_OUT_DIR}/generate.out + echo "Test: generate.res output missmatch" + exit 1 +fi +rm ${TEST_OUT_DIR}/generate.out + diff --git a/contrib/libucl/tests/msgpack.test b/contrib/libucl/tests/msgpack.test new file mode 100755 index 0000000..67a3c93 --- /dev/null +++ b/contrib/libucl/tests/msgpack.test @@ -0,0 +1,3 @@ +#!/bin/sh + +${TEST_BINARY_DIR}/test_msgpack
\ No newline at end of file diff --git a/contrib/libucl/tests/run_tests.sh b/contrib/libucl/tests/run_tests.sh new file mode 100755 index 0000000..bc26160 --- /dev/null +++ b/contrib/libucl/tests/run_tests.sh @@ -0,0 +1,67 @@ +#!/bin/sh + +if [ $# -lt 1 ] ; then + echo 'Specify binary to run as the first argument' + exit 1 +fi + + +for _tin in ${TEST_DIR}/*.in ; do + _t=`echo $_tin | sed -e 's/.in$//'` + $1 $_t.in $_t.out + if [ $? -ne 0 ] ; then + echo "Test: $_t failed, output:" + cat $_t.out + rm $_t.out + exit 1 + fi + if [ -f $_t.res ] ; then + diff -s $_t.out $_t.res -u 2>/dev/null + if [ $? -ne 0 ] ; then + rm $_t.out + echo "Test: $_t output missmatch" + exit 1 + fi + fi + rm $_t.out +done + +if [ $# -gt 2 ] ; then + $3 ${TEST_DIR}/generate.out + diff -s ${TEST_DIR}/generate.out ${TEST_DIR}/generate.res -u 2>/dev/null + if [ $? -ne 0 ] ; then + rm ${TEST_DIR}/generate.out + echo "Test: generate.res output missmatch" + exit 1 + fi + rm ${TEST_DIR}/generate.out +fi + +if [ $# -gt 3 ] ; then + rm /tmp/_ucl_test_schema.out ||true + for i in ${TEST_DIR}/schema/*.json ; do + _name=`basename $i` + printf "running schema test suite $_name... " + cat $i | $4 >> /tmp/_ucl_test_schema.out && ( echo "OK" ) || ( echo "Fail" ) + done +fi + +sh -c "xz -c < /dev/null > /dev/null" +if [ $? -eq 0 -a $# -gt 1 ] ; then + echo 'Running speed tests' + for _tin in ${TEST_DIR}/*.xz ; do + echo "Unpacking $_tin..." + xz -cd < $_tin > ${TEST_DIR}/test_file + # Preread file to cheat benchmark! + cat ${TEST_DIR}/test_file > /dev/null + echo "Starting benchmarking for $_tin..." + $2 ${TEST_DIR}/test_file + if [ $? -ne 0 ] ; then + echo "Test: $_tin failed" + rm ${TEST_DIR}/test_file + exit 1 + fi + rm ${TEST_DIR}/test_file + done +fi + diff --git a/contrib/libucl/tests/schema.test b/contrib/libucl/tests/schema.test new file mode 100755 index 0000000..fd26804 --- /dev/null +++ b/contrib/libucl/tests/schema.test @@ -0,0 +1,9 @@ +#!/bin/sh + +PROG=${TEST_BINARY_DIR}/test_schema +rm /tmp/_ucl_test_schema.out ||true +for i in ${TEST_DIR}/schema/*.json ; do + _name=`basename $i` + printf "running schema test suite $_name... " + $PROG >> /tmp/_ucl_test_schema.out < $i && ( echo "OK" ) || ( echo "Fail" ; exit 1 ) +done diff --git a/contrib/libucl/tests/schema/additionalItems.json b/contrib/libucl/tests/schema/additionalItems.json new file mode 100644 index 0000000..521745c --- /dev/null +++ b/contrib/libucl/tests/schema/additionalItems.json @@ -0,0 +1,82 @@ +[ + { + "description": "additionalItems as schema", + "schema": { + "items": [{}], + "additionalItems": {"type": "integer"} + }, + "tests": [ + { + "description": "additional items match schema", + "data": [ null, 2, 3, 4 ], + "valid": true + }, + { + "description": "additional items do not match schema", + "data": [ null, 2, 3, "foo" ], + "valid": false + } + ] + }, + { + "description": "items is schema, no additionalItems", + "schema": { + "items": {}, + "additionalItems": false + }, + "tests": [ + { + "description": "all items match schema", + "data": [ 1, 2, 3, 4, 5 ], + "valid": true + } + ] + }, + { + "description": "array of items with no additionalItems", + "schema": { + "items": [{}, {}, {}], + "additionalItems": false + }, + "tests": [ + { + "description": "no additional items present", + "data": [ 1, 2, 3 ], + "valid": true + }, + { + "description": "additional items are not permitted", + "data": [ 1, 2, 3, 4 ], + "valid": false + } + ] + }, + { + "description": "additionalItems as false without items", + "schema": {"additionalItems": false}, + "tests": [ + { + "description": + "items defaults to empty schema so everything is valid", + "data": [ 1, 2, 3, 4, 5 ], + "valid": true + }, + { + "description": "ignores non-arrays", + "data": {"foo" : "bar"}, + "valid": true + } + ] + }, + { + "description": "additionalItems are allowed by default", + "schema": {"items": [{"type": "integer"}]}, + "tests": [ + { + "description": "only the first item is validated", + "data": [1, "foo", false], + "valid": true + } + ] + } +] diff --git a/contrib/libucl/tests/schema/additionalProperties.json b/contrib/libucl/tests/schema/additionalProperties.json new file mode 100644 index 0000000..eb334c9 --- /dev/null +++ b/contrib/libucl/tests/schema/additionalProperties.json @@ -0,0 +1,69 @@ +[ + { + "description": + "additionalProperties being false does not allow other properties", + "schema": { + "properties": {"foo": {}, "bar": {}}, + "patternProperties": { "^v": {} }, + "additionalProperties": false + }, + "tests": [ + { + "description": "no additional properties is valid", + "data": {"foo": 1}, + "valid": true + }, + { + "description": "an additional property is invalid", + "data": {"foo" : 1, "bar" : 2, "quux" : "boom"}, + "valid": false + }, + { + "description": "ignores non-objects", + "data": [1, 2, 3], + "valid": true + }, + { + "description": "patternProperties are not additional properties", + "data": {"foo":1, "vroom": 2}, + "valid": true + } + ] + }, + { + "description": + "additionalProperties allows a schema which should validate", + "schema": { + "properties": {"foo": {}, "bar": {}}, + "additionalProperties": {"type": "boolean"} + }, + "tests": [ + { + "description": "no additional properties is valid", + "data": {"foo": 1}, + "valid": true + }, + { + "description": "an additional valid property is valid", + "data": {"foo" : 1, "bar" : 2, "quux" : true}, + "valid": true + }, + { + "description": "an additional invalid property is invalid", + "data": {"foo" : 1, "bar" : 2, "quux" : 12}, + "valid": false + } + ] + }, + { + "description": "additionalProperties are allowed by default", + "schema": {"properties": {"foo": {}, "bar": {}}}, + "tests": [ + { + "description": "additional properties are allowed", + "data": {"foo": 1, "bar": 2, "quux": true}, + "valid": true + } + ] + } +] diff --git a/contrib/libucl/tests/schema/allOf.json b/contrib/libucl/tests/schema/allOf.json new file mode 100644 index 0000000..bbb5f89 --- /dev/null +++ b/contrib/libucl/tests/schema/allOf.json @@ -0,0 +1,112 @@ +[ + { + "description": "allOf", + "schema": { + "allOf": [ + { + "properties": { + "bar": {"type": "integer"} + }, + "required": ["bar"] + }, + { + "properties": { + "foo": {"type": "string"} + }, + "required": ["foo"] + } + ] + }, + "tests": [ + { + "description": "allOf", + "data": {"foo": "baz", "bar": 2}, + "valid": true + }, + { + "description": "mismatch second", + "data": {"foo": "baz"}, + "valid": false + }, + { + "description": "mismatch first", + "data": {"bar": 2}, + "valid": false + }, + { + "description": "wrong type", + "data": {"foo": "baz", "bar": "quux"}, + "valid": false + } + ] + }, + { + "description": "allOf with base schema", + "schema": { + "properties": {"bar": {"type": "integer"}}, + "required": ["bar"], + "allOf" : [ + { + "properties": { + "foo": {"type": "string"} + }, + "required": ["foo"] + }, + { + "properties": { + "baz": {"type": "null"} + }, + "required": ["baz"] + } + ] + }, + "tests": [ + { + "description": "valid", + "data": {"foo": "quux", "bar": 2, "baz": null}, + "valid": true + }, + { + "description": "mismatch base schema", + "data": {"foo": "quux", "baz": null}, + "valid": false + }, + { + "description": "mismatch first allOf", + "data": {"bar": 2, "baz": null}, + "valid": false + }, + { + "description": "mismatch second allOf", + "data": {"foo": "quux", "bar": 2}, + "valid": false + }, + { + "description": "mismatch both", + "data": {"bar": 2}, + "valid": false + } + ] + }, + { + "description": "allOf simple types", + "schema": { + "allOf": [ + {"maximum": 30}, + {"minimum": 20} + ] + }, + "tests": [ + { + "description": "valid", + "data": 25, + "valid": true + }, + { + "description": "mismatch one", + "data": 35, + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/anyOf.json b/contrib/libucl/tests/schema/anyOf.json new file mode 100644 index 0000000..a58714a --- /dev/null +++ b/contrib/libucl/tests/schema/anyOf.json @@ -0,0 +1,68 @@ +[ + { + "description": "anyOf", + "schema": { + "anyOf": [ + { + "type": "integer" + }, + { + "minimum": 2 + } + ] + }, + "tests": [ + { + "description": "first anyOf valid", + "data": 1, + "valid": true + }, + { + "description": "second anyOf valid", + "data": 2.5, + "valid": true + }, + { + "description": "both anyOf valid", + "data": 3, + "valid": true + }, + { + "description": "neither anyOf valid", + "data": 1.5, + "valid": false + } + ] + }, + { + "description": "anyOf with base schema", + "schema": { + "type": "string", + "anyOf" : [ + { + "maxLength": 2 + }, + { + "minLength": 4 + } + ] + }, + "tests": [ + { + "description": "mismatch base schema", + "data": 3, + "valid": false + }, + { + "description": "one anyOf valid", + "data": "foobar", + "valid": true + }, + { + "description": "both anyOf invalid", + "data": "foo", + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/definitions.json b/contrib/libucl/tests/schema/definitions.json new file mode 100644 index 0000000..cf935a3 --- /dev/null +++ b/contrib/libucl/tests/schema/definitions.json @@ -0,0 +1,32 @@ +[ + { + "description": "valid definition", + "schema": {"$ref": "http://json-schema.org/draft-04/schema#"}, + "tests": [ + { + "description": "valid definition schema", + "data": { + "definitions": { + "foo": {"type": "integer"} + } + }, + "valid": true + } + ] + }, + { + "description": "invalid definition", + "schema": {"$ref": "http://json-schema.org/draft-04/schema#"}, + "tests": [ + { + "description": "invalid definition schema", + "data": { + "definitions": { + "foo": {"type": 1} + } + }, + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/dependencies.json b/contrib/libucl/tests/schema/dependencies.json new file mode 100644 index 0000000..7b9b16a --- /dev/null +++ b/contrib/libucl/tests/schema/dependencies.json @@ -0,0 +1,113 @@ +[ + { + "description": "dependencies", + "schema": { + "dependencies": {"bar": ["foo"]} + }, + "tests": [ + { + "description": "neither", + "data": {}, + "valid": true + }, + { + "description": "nondependant", + "data": {"foo": 1}, + "valid": true + }, + { + "description": "with dependency", + "data": {"foo": 1, "bar": 2}, + "valid": true + }, + { + "description": "missing dependency", + "data": {"bar": 2}, + "valid": false + }, + { + "description": "ignores non-objects", + "data": "foo", + "valid": true + } + ] + }, + { + "description": "multiple dependencies", + "schema": { + "dependencies": {"quux": ["foo", "bar"]} + }, + "tests": [ + { + "description": "neither", + "data": {}, + "valid": true + }, + { + "description": "nondependants", + "data": {"foo": 1, "bar": 2}, + "valid": true + }, + { + "description": "with dependencies", + "data": {"foo": 1, "bar": 2, "quux": 3}, + "valid": true + }, + { + "description": "missing dependency", + "data": {"foo": 1, "quux": 2}, + "valid": false + }, + { + "description": "missing other dependency", + "data": {"bar": 1, "quux": 2}, + "valid": false + }, + { + "description": "missing both dependencies", + "data": {"quux": 1}, + "valid": false + } + ] + }, + { + "description": "multiple dependencies subschema", + "schema": { + "dependencies": { + "bar": { + "properties": { + "foo": {"type": "integer"}, + "bar": {"type": "integer"} + } + } + } + }, + "tests": [ + { + "description": "valid", + "data": {"foo": 1, "bar": 2}, + "valid": true + }, + { + "description": "no dependency", + "data": {"foo": "quux"}, + "valid": true + }, + { + "description": "wrong type", + "data": {"foo": "quux", "bar": 2}, + "valid": false + }, + { + "description": "wrong type other", + "data": {"foo": 2, "bar": "quux"}, + "valid": false + }, + { + "description": "wrong type both", + "data": {"foo": "quux", "bar": "quux"}, + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/enum.json b/contrib/libucl/tests/schema/enum.json new file mode 100644 index 0000000..f124436 --- /dev/null +++ b/contrib/libucl/tests/schema/enum.json @@ -0,0 +1,72 @@ +[ + { + "description": "simple enum validation", + "schema": {"enum": [1, 2, 3]}, + "tests": [ + { + "description": "one of the enum is valid", + "data": 1, + "valid": true + }, + { + "description": "something else is invalid", + "data": 4, + "valid": false + } + ] + }, + { + "description": "heterogeneous enum validation", + "schema": {"enum": [6, "foo", [], true, {"foo": 12}]}, + "tests": [ + { + "description": "one of the enum is valid", + "data": [], + "valid": true + }, + { + "description": "something else is invalid", + "data": null, + "valid": false + }, + { + "description": "objects are deep compared", + "data": {"foo": false}, + "valid": false + } + ] + }, + { + "description": "enums in properties", + "schema": { + "type":"object", + "properties": { + "foo": {"enum":["foo"]}, + "bar": {"enum":["bar"]} + }, + "required": ["bar"] + }, + "tests": [ + { + "description": "both properties are valid", + "data": {"foo":"foo", "bar":"bar"}, + "valid": true + }, + { + "description": "missing optional property is valid", + "data": {"bar":"bar"}, + "valid": true + }, + { + "description": "missing required property is invalid", + "data": {"foo":"foo"}, + "valid": false + }, + { + "description": "missing all properties is invalid", + "data": {}, + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/items.json b/contrib/libucl/tests/schema/items.json new file mode 100644 index 0000000..f5e18a1 --- /dev/null +++ b/contrib/libucl/tests/schema/items.json @@ -0,0 +1,46 @@ +[ + { + "description": "a schema given for items", + "schema": { + "items": {"type": "integer"} + }, + "tests": [ + { + "description": "valid items", + "data": [ 1, 2, 3 ], + "valid": true + }, + { + "description": "wrong type of items", + "data": [1, "x"], + "valid": false + }, + { + "description": "ignores non-arrays", + "data": {"foo" : "bar"}, + "valid": true + } + ] + }, + { + "description": "an array of schemas for items", + "schema": { + "items": [ + {"type": "integer"}, + {"type": "string"} + ] + }, + "tests": [ + { + "description": "correct types", + "data": [ 1, "foo" ], + "valid": true + }, + { + "description": "wrong types", + "data": [ "foo", 1 ], + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/maxItems.json b/contrib/libucl/tests/schema/maxItems.json new file mode 100644 index 0000000..3b53a6b --- /dev/null +++ b/contrib/libucl/tests/schema/maxItems.json @@ -0,0 +1,28 @@ +[ + { + "description": "maxItems validation", + "schema": {"maxItems": 2}, + "tests": [ + { + "description": "shorter is valid", + "data": [1], + "valid": true + }, + { + "description": "exact length is valid", + "data": [1, 2], + "valid": true + }, + { + "description": "too long is invalid", + "data": [1, 2, 3], + "valid": false + }, + { + "description": "ignores non-arrays", + "data": "foobar", + "valid": true + } + ] + } +] diff --git a/contrib/libucl/tests/schema/maxLength.json b/contrib/libucl/tests/schema/maxLength.json new file mode 100644 index 0000000..561767b --- /dev/null +++ b/contrib/libucl/tests/schema/maxLength.json @@ -0,0 +1,28 @@ +[ + { + "description": "maxLength validation", + "schema": {"maxLength": 2}, + "tests": [ + { + "description": "shorter is valid", + "data": "f", + "valid": true + }, + { + "description": "exact length is valid", + "data": "fo", + "valid": true + }, + { + "description": "too long is invalid", + "data": "foo", + "valid": false + }, + { + "description": "ignores non-strings", + "data": 10, + "valid": true + } + ] + } +] diff --git a/contrib/libucl/tests/schema/maxProperties.json b/contrib/libucl/tests/schema/maxProperties.json new file mode 100644 index 0000000..d282446 --- /dev/null +++ b/contrib/libucl/tests/schema/maxProperties.json @@ -0,0 +1,28 @@ +[ + { + "description": "maxProperties validation", + "schema": {"maxProperties": 2}, + "tests": [ + { + "description": "shorter is valid", + "data": {"foo": 1}, + "valid": true + }, + { + "description": "exact length is valid", + "data": {"foo": 1, "bar": 2}, + "valid": true + }, + { + "description": "too long is invalid", + "data": {"foo": 1, "bar": 2, "baz": 3}, + "valid": false + }, + { + "description": "ignores non-objects", + "data": "foobar", + "valid": true + } + ] + } +] diff --git a/contrib/libucl/tests/schema/maximum.json b/contrib/libucl/tests/schema/maximum.json new file mode 100644 index 0000000..86c7b89 --- /dev/null +++ b/contrib/libucl/tests/schema/maximum.json @@ -0,0 +1,42 @@ +[ + { + "description": "maximum validation", + "schema": {"maximum": 3.0}, + "tests": [ + { + "description": "below the maximum is valid", + "data": 2.6, + "valid": true + }, + { + "description": "above the maximum is invalid", + "data": 3.5, + "valid": false + }, + { + "description": "ignores non-numbers", + "data": "x", + "valid": true + } + ] + }, + { + "description": "exclusiveMaximum validation", + "schema": { + "maximum": 3.0, + "exclusiveMaximum": true + }, + "tests": [ + { + "description": "below the maximum is still valid", + "data": 2.2, + "valid": true + }, + { + "description": "boundary point is invalid", + "data": 3.0, + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/minItems.json b/contrib/libucl/tests/schema/minItems.json new file mode 100644 index 0000000..ed51188 --- /dev/null +++ b/contrib/libucl/tests/schema/minItems.json @@ -0,0 +1,28 @@ +[ + { + "description": "minItems validation", + "schema": {"minItems": 1}, + "tests": [ + { + "description": "longer is valid", + "data": [1, 2], + "valid": true + }, + { + "description": "exact length is valid", + "data": [1], + "valid": true + }, + { + "description": "too short is invalid", + "data": [], + "valid": false + }, + { + "description": "ignores non-arrays", + "data": "", + "valid": true + } + ] + } +] diff --git a/contrib/libucl/tests/schema/minLength.json b/contrib/libucl/tests/schema/minLength.json new file mode 100644 index 0000000..e9c14b1 --- /dev/null +++ b/contrib/libucl/tests/schema/minLength.json @@ -0,0 +1,28 @@ +[ + { + "description": "minLength validation", + "schema": {"minLength": 2}, + "tests": [ + { + "description": "longer is valid", + "data": "foo", + "valid": true + }, + { + "description": "exact length is valid", + "data": "fo", + "valid": true + }, + { + "description": "too short is invalid", + "data": "f", + "valid": false + }, + { + "description": "ignores non-strings", + "data": 1, + "valid": true + } + ] + } +] diff --git a/contrib/libucl/tests/schema/minProperties.json b/contrib/libucl/tests/schema/minProperties.json new file mode 100644 index 0000000..a72c7d2 --- /dev/null +++ b/contrib/libucl/tests/schema/minProperties.json @@ -0,0 +1,28 @@ +[ + { + "description": "minProperties validation", + "schema": {"minProperties": 1}, + "tests": [ + { + "description": "longer is valid", + "data": {"foo": 1, "bar": 2}, + "valid": true + }, + { + "description": "exact length is valid", + "data": {"foo": 1}, + "valid": true + }, + { + "description": "too short is invalid", + "data": {}, + "valid": false + }, + { + "description": "ignores non-objects", + "data": "", + "valid": true + } + ] + } +] diff --git a/contrib/libucl/tests/schema/minimum.json b/contrib/libucl/tests/schema/minimum.json new file mode 100644 index 0000000..d5bf000 --- /dev/null +++ b/contrib/libucl/tests/schema/minimum.json @@ -0,0 +1,42 @@ +[ + { + "description": "minimum validation", + "schema": {"minimum": 1.1}, + "tests": [ + { + "description": "above the minimum is valid", + "data": 2.6, + "valid": true + }, + { + "description": "below the minimum is invalid", + "data": 0.6, + "valid": false + }, + { + "description": "ignores non-numbers", + "data": "x", + "valid": true + } + ] + }, + { + "description": "exclusiveMinimum validation", + "schema": { + "minimum": 1.1, + "exclusiveMinimum": true + }, + "tests": [ + { + "description": "above the minimum is still valid", + "data": 1.2, + "valid": true + }, + { + "description": "boundary point is invalid", + "data": 1.1, + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/multipleOf.json b/contrib/libucl/tests/schema/multipleOf.json new file mode 100644 index 0000000..ca3b761 --- /dev/null +++ b/contrib/libucl/tests/schema/multipleOf.json @@ -0,0 +1,60 @@ +[ + { + "description": "by int", + "schema": {"multipleOf": 2}, + "tests": [ + { + "description": "int by int", + "data": 10, + "valid": true + }, + { + "description": "int by int fail", + "data": 7, + "valid": false + }, + { + "description": "ignores non-numbers", + "data": "foo", + "valid": true + } + ] + }, + { + "description": "by number", + "schema": {"multipleOf": 1.5}, + "tests": [ + { + "description": "zero is multiple of anything", + "data": 0, + "valid": true + }, + { + "description": "4.5 is multiple of 1.5", + "data": 4.5, + "valid": true + }, + { + "description": "35 is not multiple of 1.5", + "data": 35, + "valid": false + } + ] + }, + { + "description": "by small number", + "schema": {"multipleOf": 0.0001}, + "tests": [ + { + "description": "0.0075 is multiple of 0.0001", + "data": 0.0075, + "valid": true + }, + { + "description": "0.00751 is not multiple of 0.0001", + "data": 0.00751, + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/not.json b/contrib/libucl/tests/schema/not.json new file mode 100644 index 0000000..cbb7f46 --- /dev/null +++ b/contrib/libucl/tests/schema/not.json @@ -0,0 +1,96 @@ +[ + { + "description": "not", + "schema": { + "not": {"type": "integer"} + }, + "tests": [ + { + "description": "allowed", + "data": "foo", + "valid": true + }, + { + "description": "disallowed", + "data": 1, + "valid": false + } + ] + }, + { + "description": "not multiple types", + "schema": { + "not": {"type": ["integer", "boolean"]} + }, + "tests": [ + { + "description": "valid", + "data": "foo", + "valid": true + }, + { + "description": "mismatch", + "data": 1, + "valid": false + }, + { + "description": "other mismatch", + "data": true, + "valid": false + } + ] + }, + { + "description": "not more complex schema", + "schema": { + "not": { + "type": "object", + "properties": { + "foo": { + "type": "string" + } + } + } + }, + "tests": [ + { + "description": "match", + "data": 1, + "valid": true + }, + { + "description": "other match", + "data": {"foo": 1}, + "valid": true + }, + { + "description": "mismatch", + "data": {"foo": "bar"}, + "valid": false + } + ] + }, + { + "description": "forbidden property", + "schema": { + "properties": { + "foo": { + "not": {} + } + } + }, + "tests": [ + { + "description": "property present", + "data": {"foo": 1, "bar": 2}, + "valid": false + }, + { + "description": "property absent", + "data": {"bar": 1, "baz": 2}, + "valid": true + } + ] + } + +] diff --git a/contrib/libucl/tests/schema/oneOf.json b/contrib/libucl/tests/schema/oneOf.json new file mode 100644 index 0000000..1eaa4e4 --- /dev/null +++ b/contrib/libucl/tests/schema/oneOf.json @@ -0,0 +1,68 @@ +[ + { + "description": "oneOf", + "schema": { + "oneOf": [ + { + "type": "integer" + }, + { + "minimum": 2 + } + ] + }, + "tests": [ + { + "description": "first oneOf valid", + "data": 1, + "valid": true + }, + { + "description": "second oneOf valid", + "data": 2.5, + "valid": true + }, + { + "description": "both oneOf valid", + "data": 3, + "valid": false + }, + { + "description": "neither oneOf valid", + "data": 1.5, + "valid": false + } + ] + }, + { + "description": "oneOf with base schema", + "schema": { + "type": "string", + "oneOf" : [ + { + "minLength": 2 + }, + { + "maxLength": 4 + } + ] + }, + "tests": [ + { + "description": "mismatch base schema", + "data": 3, + "valid": false + }, + { + "description": "one oneOf valid", + "data": "foobar", + "valid": true + }, + { + "description": "both oneOf valid", + "data": "foo", + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/pattern.json b/contrib/libucl/tests/schema/pattern.json new file mode 100644 index 0000000..befc4b5 --- /dev/null +++ b/contrib/libucl/tests/schema/pattern.json @@ -0,0 +1,23 @@ +[ + { + "description": "pattern validation", + "schema": {"pattern": "^a*$"}, + "tests": [ + { + "description": "a matching pattern is valid", + "data": "aaa", + "valid": true + }, + { + "description": "a non-matching pattern is invalid", + "data": "abc", + "valid": false + }, + { + "description": "ignores non-strings", + "data": true, + "valid": true + } + ] + } +] diff --git a/contrib/libucl/tests/schema/patternProperties.json b/contrib/libucl/tests/schema/patternProperties.json new file mode 100644 index 0000000..18586e5 --- /dev/null +++ b/contrib/libucl/tests/schema/patternProperties.json @@ -0,0 +1,110 @@ +[ + { + "description": + "patternProperties validates properties matching a regex", + "schema": { + "patternProperties": { + "f.*o": {"type": "integer"} + } + }, + "tests": [ + { + "description": "a single valid match is valid", + "data": {"foo": 1}, + "valid": true + }, + { + "description": "multiple valid matches is valid", + "data": {"foo": 1, "foooooo" : 2}, + "valid": true + }, + { + "description": "a single invalid match is invalid", + "data": {"foo": "bar", "fooooo": 2}, + "valid": false + }, + { + "description": "multiple invalid matches is invalid", + "data": {"foo": "bar", "foooooo" : "baz"}, + "valid": false + }, + { + "description": "ignores non-objects", + "data": 12, + "valid": true + } + ] + }, + { + "description": "multiple simultaneous patternProperties are validated", + "schema": { + "patternProperties": { + "a*": {"type": "integer"}, + "aaa*": {"maximum": 20} + } + }, + "tests": [ + { + "description": "a single valid match is valid", + "data": {"a": 21}, + "valid": true + }, + { + "description": "a simultaneous match is valid", + "data": {"aaaa": 18}, + "valid": true + }, + { + "description": "multiple matches is valid", + "data": {"a": 21, "aaaa": 18}, + "valid": true + }, + { + "description": "an invalid due to one is invalid", + "data": {"a": "bar"}, + "valid": false + }, + { + "description": "an invalid due to the other is invalid", + "data": {"aaaa": 31}, + "valid": false + }, + { + "description": "an invalid due to both is invalid", + "data": {"aaa": "foo", "aaaa": 31}, + "valid": false + } + ] + }, + { + "description": "regexes are not anchored by default and are case sensitive", + "schema": { + "patternProperties": { + "[0-9]{2,}": { "type": "boolean" }, + "X_": { "type": "string" } + } + }, + "tests": [ + { + "description": "non recognized members are ignored", + "data": { "answer 1": "42" }, + "valid": true + }, + { + "description": "recognized members are accounted for", + "data": { "a31b": null }, + "valid": false + }, + { + "description": "regexes are case sensitive", + "data": { "a_x_3": 3 }, + "valid": true + }, + { + "description": "regexes are case sensitive, 2", + "data": { "a_X_3": 3 }, + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/properties.json b/contrib/libucl/tests/schema/properties.json new file mode 100644 index 0000000..cd1644d --- /dev/null +++ b/contrib/libucl/tests/schema/properties.json @@ -0,0 +1,92 @@ +[ + { + "description": "object properties validation", + "schema": { + "properties": { + "foo": {"type": "integer"}, + "bar": {"type": "string"} + } + }, + "tests": [ + { + "description": "both properties present and valid is valid", + "data": {"foo": 1, "bar": "baz"}, + "valid": true + }, + { + "description": "one property invalid is invalid", + "data": {"foo": 1, "bar": {}}, + "valid": false + }, + { + "description": "both properties invalid is invalid", + "data": {"foo": [], "bar": {}}, + "valid": false + }, + { + "description": "doesn't invalidate other properties", + "data": {"quux": []}, + "valid": true + }, + { + "description": "ignores non-objects", + "data": [], + "valid": true + } + ] + }, + { + "description": + "properties, patternProperties, additionalProperties interaction", + "schema": { + "properties": { + "foo": {"type": "array", "maxItems": 3}, + "bar": {"type": "array"} + }, + "patternProperties": {"f.o": {"minItems": 2}}, + "additionalProperties": {"type": "integer"} + }, + "tests": [ + { + "description": "property validates property", + "data": {"foo": [1, 2]}, + "valid": true + }, + { + "description": "property invalidates property", + "data": {"foo": [1, 2, 3, 4]}, + "valid": false + }, + { + "description": "patternProperty invalidates property", + "data": {"foo": []}, + "valid": false + }, + { + "description": "patternProperty validates nonproperty", + "data": {"fxo": [1, 2]}, + "valid": true + }, + { + "description": "patternProperty invalidates nonproperty", + "data": {"fxo": []}, + "valid": false + }, + { + "description": "additionalProperty ignores property", + "data": {"bar": []}, + "valid": true + }, + { + "description": "additionalProperty validates others", + "data": {"quux": 3}, + "valid": true + }, + { + "description": "additionalProperty invalidates others", + "data": {"quux": "foo"}, + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/ref.json b/contrib/libucl/tests/schema/ref.json new file mode 100644 index 0000000..1b88b27 --- /dev/null +++ b/contrib/libucl/tests/schema/ref.json @@ -0,0 +1,146 @@ +[ + { + "description": "root pointer ref", + "schema": { + "properties": { + "foo": {"$ref": "#"} + }, + "additionalProperties": false + }, + "tests": [ + { + "description": "match", + "data": {"foo": false}, + "valid": true + }, + { + "description": "recursive match", + "data": {"foo": {"foo": false}}, + "valid": true + }, + { + "description": "mismatch", + "data": {"bar": false}, + "valid": false + }, + { + "description": "recursive mismatch", + "data": {"foo": {"bar": false}}, + "valid": false + } + ] + }, + { + "description": "relative pointer ref to object", + "schema": { + "properties": { + "foo": {"type": "integer"}, + "bar": {"$ref": "#/properties/foo"} + } + }, + "tests": [ + { + "description": "match", + "data": {"bar": 3}, + "valid": true + }, + { + "description": "mismatch", + "data": {"bar": true}, + "valid": false + } + ] + }, + { + "description": "relative pointer ref to array", + "schema": { + "items": [ + {"type": "integer"}, + {"$ref": "#/items/0"} + ] + }, + "tests": [ + { + "description": "match array", + "data": [1, 2], + "valid": true + }, + { + "description": "mismatch array", + "data": [1, "foo"], + "valid": false + } + ] + }, + { + "description": "escaped pointer ref", + "schema": { + "tilda~field": {"type": "integer"}, + "slash/field": {"type": "integer"}, + "percent%field": {"type": "integer"}, + "properties": { + "tilda": {"$ref": "#/tilda~0field"}, + "slash": {"$ref": "#/slash~1field"}, + "percent": {"$ref": "#/percent%25field"} + } + }, + "tests": [ + { + "description": "slash", + "data": {"slash": "aoeu"}, + "valid": false + }, + { + "description": "tilda", + "data": {"tilda": "aoeu"}, + "valid": false + }, + { + "description": "percent", + "data": {"percent": "aoeu"}, + "valid": false + } + ] + }, + { + "description": "nested refs", + "schema": { + "definitions": { + "a": {"type": "integer"}, + "b": {"$ref": "#/definitions/a"}, + "c": {"$ref": "#/definitions/b"} + }, + "$ref": "#/definitions/c" + }, + "tests": [ + { + "description": "nested ref valid", + "data": 5, + "valid": true + }, + { + "description": "nested ref invalid", + "data": "a", + "valid": false + } + ] + }, +/* + { + "description": "remote ref, containing refs itself", + "schema": {"$ref": "http://json-schema.org/draft-04/schema#"}, + "tests": [ + { + "description": "remote ref valid", + "data": {"minLength": 1}, + "valid": true + }, + { + "description": "remote ref invalid", + "data": {"minLength": -1}, + "valid": false + } + ] + } +*/ +] diff --git a/contrib/libucl/tests/schema/refRemote.json b/contrib/libucl/tests/schema/refRemote.json new file mode 100644 index 0000000..4ca8047 --- /dev/null +++ b/contrib/libucl/tests/schema/refRemote.json @@ -0,0 +1,74 @@ +[ + { + "description": "remote ref", + "schema": {"$ref": "http://localhost:1234/integer.json"}, + "tests": [ + { + "description": "remote ref valid", + "data": 1, + "valid": true + }, + { + "description": "remote ref invalid", + "data": "a", + "valid": false + } + ] + }, + { + "description": "fragment within remote ref", + "schema": {"$ref": "http://localhost:1234/subSchemas.json#/integer"}, + "tests": [ + { + "description": "remote fragment valid", + "data": 1, + "valid": true + }, + { + "description": "remote fragment invalid", + "data": "a", + "valid": false + } + ] + }, + { + "description": "ref within remote ref", + "schema": { + "$ref": "http://localhost:1234/subSchemas.json#/refToInteger" + }, + "tests": [ + { + "description": "ref within ref valid", + "data": 1, + "valid": true + }, + { + "description": "ref within ref invalid", + "data": "a", + "valid": false + } + ] + }, + { + "description": "change resolution scope", + "schema": { + "id": "http://localhost:1234/", + "items": { + "id": "folder/", + "items": {"$ref": "folderInteger.json"} + } + }, + "tests": [ + { + "description": "changed scope ref valid", + "data": [[1]], + "valid": true + }, + { + "description": "changed scope ref invalid", + "data": [["a"]], + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/required.json b/contrib/libucl/tests/schema/required.json new file mode 100644 index 0000000..612f73f --- /dev/null +++ b/contrib/libucl/tests/schema/required.json @@ -0,0 +1,39 @@ +[ + { + "description": "required validation", + "schema": { + "properties": { + "foo": {}, + "bar": {} + }, + "required": ["foo"] + }, + "tests": [ + { + "description": "present required property is valid", + "data": {"foo": 1}, + "valid": true + }, + { + "description": "non-present required property is invalid", + "data": {"bar": 1}, + "valid": false + } + ] + }, + { + "description": "required default validation", + "schema": { + "properties": { + "foo": {} + } + }, + "tests": [ + { + "description": "not required by default", + "data": {}, + "valid": true + } + ] + } +] diff --git a/contrib/libucl/tests/schema/type.json b/contrib/libucl/tests/schema/type.json new file mode 100644 index 0000000..257f051 --- /dev/null +++ b/contrib/libucl/tests/schema/type.json @@ -0,0 +1,330 @@ +[ + { + "description": "integer type matches integers", + "schema": {"type": "integer"}, + "tests": [ + { + "description": "an integer is an integer", + "data": 1, + "valid": true + }, + { + "description": "a float is not an integer", + "data": 1.1, + "valid": false + }, + { + "description": "a string is not an integer", + "data": "foo", + "valid": false + }, + { + "description": "an object is not an integer", + "data": {}, + "valid": false + }, + { + "description": "an array is not an integer", + "data": [], + "valid": false + }, + { + "description": "a boolean is not an integer", + "data": true, + "valid": false + }, + { + "description": "null is not an integer", + "data": null, + "valid": false + } + ] + }, + { + "description": "number type matches numbers", + "schema": {"type": "number"}, + "tests": [ + { + "description": "an integer is a number", + "data": 1, + "valid": true + }, + { + "description": "a float is a number", + "data": 1.1, + "valid": true + }, + { + "description": "a string is not a number", + "data": "foo", + "valid": false + }, + { + "description": "an object is not a number", + "data": {}, + "valid": false + }, + { + "description": "an array is not a number", + "data": [], + "valid": false + }, + { + "description": "a boolean is not a number", + "data": true, + "valid": false + }, + { + "description": "null is not a number", + "data": null, + "valid": false + } + ] + }, + { + "description": "string type matches strings", + "schema": {"type": "string"}, + "tests": [ + { + "description": "1 is not a string", + "data": 1, + "valid": false + }, + { + "description": "a float is not a string", + "data": 1.1, + "valid": false + }, + { + "description": "a string is a string", + "data": "foo", + "valid": true + }, + { + "description": "an object is not a string", + "data": {}, + "valid": false + }, + { + "description": "an array is not a string", + "data": [], + "valid": false + }, + { + "description": "a boolean is not a string", + "data": true, + "valid": false + }, + { + "description": "null is not a string", + "data": null, + "valid": false + } + ] + }, + { + "description": "object type matches objects", + "schema": {"type": "object"}, + "tests": [ + { + "description": "an integer is not an object", + "data": 1, + "valid": false + }, + { + "description": "a float is not an object", + "data": 1.1, + "valid": false + }, + { + "description": "a string is not an object", + "data": "foo", + "valid": false + }, + { + "description": "an object is an object", + "data": {}, + "valid": true + }, + { + "description": "an array is not an object", + "data": [], + "valid": false + }, + { + "description": "a boolean is not an object", + "data": true, + "valid": false + }, + { + "description": "null is not an object", + "data": null, + "valid": false + } + ] + }, + { + "description": "array type matches arrays", + "schema": {"type": "array"}, + "tests": [ + { + "description": "an integer is not an array", + "data": 1, + "valid": false + }, + { + "description": "a float is not an array", + "data": 1.1, + "valid": false + }, + { + "description": "a string is not an array", + "data": "foo", + "valid": false + }, + { + "description": "an object is not an array", + "data": {}, + "valid": false + }, + { + "description": "an array is not an array", + "data": [], + "valid": true + }, + { + "description": "a boolean is not an array", + "data": true, + "valid": false + }, + { + "description": "null is not an array", + "data": null, + "valid": false + } + ] + }, + { + "description": "boolean type matches booleans", + "schema": {"type": "boolean"}, + "tests": [ + { + "description": "an integer is not a boolean", + "data": 1, + "valid": false + }, + { + "description": "a float is not a boolean", + "data": 1.1, + "valid": false + }, + { + "description": "a string is not a boolean", + "data": "foo", + "valid": false + }, + { + "description": "an object is not a boolean", + "data": {}, + "valid": false + }, + { + "description": "an array is not a boolean", + "data": [], + "valid": false + }, + { + "description": "a boolean is not a boolean", + "data": true, + "valid": true + }, + { + "description": "null is not a boolean", + "data": null, + "valid": false + } + ] + }, + { + "description": "null type matches only the null object", + "schema": {"type": "null"}, + "tests": [ + { + "description": "an integer is not null", + "data": 1, + "valid": false + }, + { + "description": "a float is not null", + "data": 1.1, + "valid": false + }, + { + "description": "a string is not null", + "data": "foo", + "valid": false + }, + { + "description": "an object is not null", + "data": {}, + "valid": false + }, + { + "description": "an array is not null", + "data": [], + "valid": false + }, + { + "description": "a boolean is not null", + "data": true, + "valid": false + }, + { + "description": "null is null", + "data": null, + "valid": true + } + ] + }, + { + "description": "multiple types can be specified in an array", + "schema": {"type": ["integer", "string"]}, + "tests": [ + { + "description": "an integer is valid", + "data": 1, + "valid": true + }, + { + "description": "a string is valid", + "data": "foo", + "valid": true + }, + { + "description": "a float is invalid", + "data": 1.1, + "valid": false + }, + { + "description": "an object is invalid", + "data": {}, + "valid": false + }, + { + "description": "an array is invalid", + "data": [], + "valid": false + }, + { + "description": "a boolean is invalid", + "data": true, + "valid": false + }, + { + "description": "null is invalid", + "data": null, + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/schema/uniqueItems.json b/contrib/libucl/tests/schema/uniqueItems.json new file mode 100644 index 0000000..c1f4ab9 --- /dev/null +++ b/contrib/libucl/tests/schema/uniqueItems.json @@ -0,0 +1,79 @@ +[ + { + "description": "uniqueItems validation", + "schema": {"uniqueItems": true}, + "tests": [ + { + "description": "unique array of integers is valid", + "data": [1, 2], + "valid": true + }, + { + "description": "non-unique array of integers is invalid", + "data": [1, 1], + "valid": false + }, + { + "description": "numbers are unique if mathematically unequal", + "data": [1.0, 1.00, 1], + "valid": false + }, + { + "description": "unique array of objects is valid", + "data": [{"foo": "bar"}, {"foo": "baz"}], + "valid": true + }, + { + "description": "non-unique array of objects is invalid", + "data": [{"foo": "bar"}, {"foo": "bar"}], + "valid": false + }, + { + "description": "unique array of nested objects is valid", + "data": [ + {"foo": {"bar" : {"baz" : true}}}, + {"foo": {"bar" : {"baz" : false}}} + ], + "valid": true + }, + { + "description": "non-unique array of nested objects is invalid", + "data": [ + {"foo": {"bar" : {"baz" : true}}}, + {"foo": {"bar" : {"baz" : true}}} + ], + "valid": false + }, + { + "description": "unique array of arrays is valid", + "data": [["foo"], ["bar"]], + "valid": true + }, + { + "description": "non-unique array of arrays is invalid", + "data": [["foo"], ["foo"]], + "valid": false + }, + { + "description": "1 and true are unique", + "data": [1, true], + "valid": true + }, + { + "description": "0 and false are unique", + "data": [0, false], + "valid": true + }, + { + "description": "unique heterogeneous types are valid", + "data": [{}, [1], true, null, 1], + "valid": true + }, + { + "description": "non-unique heterogeneous types are invalid", + "data": [{}, [1], true, null, {}, 1], + "valid": false + } + ] + } +] diff --git a/contrib/libucl/tests/speed.test b/contrib/libucl/tests/speed.test new file mode 100755 index 0000000..34db653 --- /dev/null +++ b/contrib/libucl/tests/speed.test @@ -0,0 +1,21 @@ +#!/bin/sh + +PROG=${TEST_BINARY_DIR}/test_speed + +sh -c "xz -c < /dev/null > /dev/null" +echo 'Running speed tests' +for _tin in ${TEST_DIR}/*.xz ; do + echo "Unpacking $_tin..." + xz -cd < $_tin > ${TEST_OUT_DIR}/test_file + # Preread file to cheat benchmark! + cat ${TEST_OUT_DIR}/test_file > /dev/null + echo "Starting benchmarking for $_tin..." + $PROG ${TEST_OUT_DIR}/test_file + if [ $? -ne 0 ] ; then + echo "Test: $_tin failed" + rm ${TEST_OUT_DIR}/test_file + exit 1 + fi + rm ${TEST_OUT_DIR}/test_file +done + diff --git a/contrib/libucl/tests/streamline.res b/contrib/libucl/tests/streamline.res new file mode 100644 index 0000000..5eb18d7 --- /dev/null +++ b/contrib/libucl/tests/streamline.res @@ -0,0 +1,8 @@ +key1 = "test string"; +key2 = "test \\nstring"; +key3 = " test string \n"; +key4 [ + 10, + 10.100000, + 9.999000, +] diff --git a/contrib/libucl/tests/streamline.test b/contrib/libucl/tests/streamline.test new file mode 100755 index 0000000..dbe5836 --- /dev/null +++ b/contrib/libucl/tests/streamline.test @@ -0,0 +1,12 @@ +#!/bin/sh + +PROG=${TEST_BINARY_DIR}/test_streamline + +$PROG ${TEST_OUT_DIR}/streamline.out +diff -s ${TEST_OUT_DIR}/streamline.out ${TEST_DIR}/streamline.res -u 2>/dev/null +if [ $? -ne 0 ] ; then + rm ${TEST_OUT_DIR}/streamline.out + echo "Test: streamline.res output missmatch" + exit 1 +fi +rm ${TEST_OUT_DIR}/streamline.out
\ No newline at end of file diff --git a/contrib/libucl/tests/test_basic.c b/contrib/libucl/tests/test_basic.c new file mode 100644 index 0000000..45a9c8b --- /dev/null +++ b/contrib/libucl/tests/test_basic.c @@ -0,0 +1,192 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include "ucl.h" +#include "ucl_internal.h" + +int +main (int argc, char **argv) +{ + char *inbuf; + struct ucl_parser *parser = NULL, *parser2 = NULL; + ucl_object_t *obj; + ssize_t bufsize, r; + FILE *in, *out; + unsigned char *emitted = NULL; + const char *fname_in = NULL, *fname_out = NULL; + int ret = 0, opt, json = 0, compact = 0, yaml = 0; + + while ((opt = getopt(argc, argv, "jcy")) != -1) { + switch (opt) { + case 'j': + json = 1; + break; + case 'c': + compact = 1; + break; + case 'y': + yaml = 1; + break; + default: /* '?' */ + fprintf (stderr, "Usage: %s [-jcy] [in] [out]\n", + argv[0]); + exit (EXIT_FAILURE); + } + } + + argc -= optind; + argv += optind; + + switch (argc) { + case 1: + fname_in = argv[0]; + break; + case 2: + fname_in = argv[0]; + fname_out = argv[1]; + break; + } + + if (fname_in != NULL) { + in = fopen (fname_in, "r"); + if (in == NULL) { + exit (-errno); + } + } + else { + in = stdin; + } + parser = ucl_parser_new (UCL_PARSER_KEY_LOWERCASE); + ucl_parser_register_variable (parser, "ABI", "unknown"); + + if (fname_in != NULL) { + ucl_parser_set_filevars (parser, fname_in, true); + } + + inbuf = malloc (BUFSIZ); + bufsize = BUFSIZ; + r = 0; + + while (!feof (in) && !ferror (in)) { + if (r == bufsize) { + inbuf = realloc (inbuf, bufsize * 2); + bufsize *= 2; + if (inbuf == NULL) { + perror ("realloc"); + exit (EXIT_FAILURE); + } + } + r += fread (inbuf + r, 1, bufsize - r, in); + } + + if (ferror (in)) { + fprintf (stderr, "Failed to read the input file.\n"); + exit (EXIT_FAILURE); + } + + ucl_parser_add_chunk (parser, (const unsigned char *)inbuf, r); + fclose (in); + + if (fname_out != NULL) { + out = fopen (fname_out, "w"); + if (out == NULL) { + exit (-errno); + } + } + else { + out = stdout; + } + + if (ucl_parser_get_error (parser) != NULL) { + fprintf (out, "Error occurred: %s\n", ucl_parser_get_error(parser)); + ret = 1; + goto end; + } + + obj = ucl_parser_get_object (parser); + + if (json) { + if (compact) { + emitted = ucl_object_emit (obj, UCL_EMIT_JSON_COMPACT); + } + else { + emitted = ucl_object_emit (obj, UCL_EMIT_JSON); + } + } + else if (yaml) { + emitted = ucl_object_emit (obj, UCL_EMIT_YAML); + } + else { + emitted = ucl_object_emit (obj, UCL_EMIT_CONFIG); + } + + ucl_parser_free (parser); + ucl_object_unref (obj); + parser2 = ucl_parser_new (UCL_PARSER_KEY_LOWERCASE); + ucl_parser_add_string (parser2, (const char *)emitted, 0); + + if (ucl_parser_get_error(parser2) != NULL) { + fprintf (out, "Error occurred: %s\n", ucl_parser_get_error(parser2)); + fprintf (out, "%s\n", emitted); + ret = 1; + goto end; + } + + if (emitted != NULL) { + free (emitted); + } + + obj = ucl_parser_get_object (parser2); + if (json) { + if (compact) { + emitted = ucl_object_emit (obj, UCL_EMIT_JSON_COMPACT); + } + else { + emitted = ucl_object_emit (obj, UCL_EMIT_JSON); + } + } + else if (yaml) { + emitted = ucl_object_emit (obj, UCL_EMIT_YAML); + } + else { + emitted = ucl_object_emit (obj, UCL_EMIT_CONFIG); + } + + fprintf (out, "%s\n", emitted); + ucl_object_unref (obj); + +end: + if (emitted != NULL) { + free (emitted); + } + if (parser2 != NULL) { + ucl_parser_free (parser2); + } + if (inbuf != NULL) { + free (inbuf); + } + + fclose (out); + + return ret; +} diff --git a/contrib/libucl/tests/test_generate.c b/contrib/libucl/tests/test_generate.c new file mode 100644 index 0000000..f09eabb --- /dev/null +++ b/contrib/libucl/tests/test_generate.c @@ -0,0 +1,185 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <stdio.h> +#include <errno.h> +#include <assert.h> +#include "ucl.h" + +int +main (int argc, char **argv) +{ + ucl_object_t *obj, *cur, *ar, *ref; + ucl_object_iter_t it; + const ucl_object_t *found, *it_obj; + FILE *out; + unsigned char *emitted; + const char *fname_out = NULL; + int ret = 0; + + switch (argc) { + case 2: + fname_out = argv[1]; + break; + } + + + if (fname_out != NULL) { + out = fopen (fname_out, "w"); + if (out == NULL) { + exit (-errno); + } + } + else { + out = stdout; + } + + obj = ucl_object_typed_new (UCL_OBJECT); + + /* Keys replacing */ + cur = ucl_object_fromstring_common ("value1", 0, UCL_STRING_TRIM); + ucl_object_insert_key (obj, cur, "key0", 0, false); + cur = ucl_object_fromdouble (0.1); + assert (ucl_object_replace_key (obj, cur, "key0", 0, false)); + + /* Create some strings */ + cur = ucl_object_fromstring_common (" test string ", 0, UCL_STRING_TRIM); + ucl_object_insert_key (obj, cur, "key1", 0, false); + cur = ucl_object_fromstring_common (" test \nstring\n ", 0, UCL_STRING_TRIM | UCL_STRING_ESCAPE); + ucl_object_insert_key (obj, cur, "key2", 0, false); + cur = ucl_object_fromstring_common (" test string \n", 0, 0); + ucl_object_insert_key (obj, cur, "key3", 0, false); + /* Array of numbers */ + ar = ucl_object_typed_new (UCL_ARRAY); + cur = ucl_object_fromint (10); + ucl_array_append (ar, cur); + cur = ucl_object_fromdouble (10.1); + ucl_array_append (ar, cur); + cur = ucl_object_fromdouble (9.999); + ucl_array_prepend (ar, cur); + + /* Removing from an array */ + cur = ucl_object_fromdouble (1.0); + ucl_array_append (ar, cur); + cur = ucl_array_delete (ar, cur); + assert (ucl_object_todouble (cur) == 1.0); + ucl_object_unref (cur); + cur = ucl_object_fromdouble (2.0); + ucl_array_append (ar, cur); + cur = ucl_array_pop_last (ar); + assert (ucl_object_todouble (cur) == 2.0); + ucl_object_unref (cur); + cur = ucl_object_fromdouble (3.0); + ucl_array_prepend (ar, cur); + cur = ucl_array_pop_first (ar); + assert (ucl_object_todouble (cur) == 3.0); + ucl_object_unref (cur); + + ucl_object_insert_key (obj, ar, "key4", 0, false); + cur = ucl_object_frombool (true); + /* Ref object to test refcounts */ + ref = ucl_object_ref (cur); + ucl_object_insert_key (obj, cur, "key4", 0, false); + /* Empty strings */ + cur = ucl_object_fromstring_common (" ", 0, UCL_STRING_TRIM); + ucl_object_insert_key (obj, cur, "key5", 0, false); + cur = ucl_object_fromstring_common ("", 0, UCL_STRING_ESCAPE); + ucl_object_insert_key (obj, cur, "key6", 0, false); + cur = ucl_object_fromstring_common (" \n", 0, UCL_STRING_ESCAPE); + ucl_object_insert_key (obj, cur, "key7", 0, false); + /* Numbers and booleans */ + cur = ucl_object_fromstring_common ("1mb", 0, UCL_STRING_ESCAPE | UCL_STRING_PARSE); + ucl_object_insert_key (obj, cur, "key8", 0, false); + cur = ucl_object_fromstring_common ("3.14", 0, UCL_STRING_PARSE); + ucl_object_insert_key (obj, cur, "key9", 0, false); + cur = ucl_object_fromstring_common ("true", 0, UCL_STRING_PARSE); + ucl_object_insert_key (obj, cur, "key10", 0, false); + cur = ucl_object_fromstring_common (" off ", 0, UCL_STRING_PARSE | UCL_STRING_TRIM); + ucl_object_insert_key (obj, cur, "key11", 0, false); + cur = ucl_object_fromstring_common ("gslin@gslin.org", 0, UCL_STRING_PARSE_INT); + ucl_object_insert_key (obj, cur, "key12", 0, false); + cur = ucl_object_fromstring_common ("#test", 0, UCL_STRING_PARSE_INT); + ucl_object_insert_key (obj, cur, "key13", 0, false); + cur = ucl_object_frombool (true); + ucl_object_insert_key (obj, cur, "k=3", 0, false); + + /* Try to find using path */ + /* Should exist */ + found = ucl_lookup_path (obj, "key4.1"); + assert (found != NULL && ucl_object_toint (found) == 10); + /* . should be ignored */ + found = ucl_lookup_path (obj, ".key4.1"); + assert (found != NULL && ucl_object_toint (found) == 10); + /* moar dots... */ + found = ucl_lookup_path (obj, ".key4........1..."); + assert (found != NULL && ucl_object_toint (found) == 10); + /* No such index */ + found = ucl_lookup_path (obj, ".key4.3"); + assert (found == NULL); + /* No such key */ + found = ucl_lookup_path (obj, "key9..key1"); + assert (found == NULL); + + /* Test iteration */ + it = ucl_object_iterate_new (obj); + it_obj = ucl_object_iterate_safe (it, true); + /* key0 = 0.1 */ + assert (ucl_object_type (it_obj) == UCL_FLOAT); + it_obj = ucl_object_iterate_safe (it, true); + /* key1 = "" */ + assert (ucl_object_type (it_obj) == UCL_STRING); + it_obj = ucl_object_iterate_safe (it, true); + /* key2 = "" */ + assert (ucl_object_type (it_obj) == UCL_STRING); + it_obj = ucl_object_iterate_safe (it, true); + /* key3 = "" */ + assert (ucl_object_type (it_obj) == UCL_STRING); + it_obj = ucl_object_iterate_safe (it, true); + /* key4 = ([float, int, float], boolean) */ + ucl_object_iterate_reset (it, it_obj); + it_obj = ucl_object_iterate_safe (it, true); + assert (ucl_object_type (it_obj) == UCL_FLOAT); + it_obj = ucl_object_iterate_safe (it, true); + assert (ucl_object_type (it_obj) == UCL_INT); + it_obj = ucl_object_iterate_safe (it, true); + assert (ucl_object_type (it_obj) == UCL_FLOAT); + it_obj = ucl_object_iterate_safe (it, true); + assert (ucl_object_type (it_obj) == UCL_BOOLEAN); + ucl_object_iterate_free (it); + + emitted = ucl_object_emit (obj, UCL_EMIT_CONFIG); + + fprintf (out, "%s\n", emitted); + ucl_object_unref (obj); + + if (emitted != NULL) { + free (emitted); + } + fclose (out); + + /* Ref should still be accessible */ + ref->value.iv = 100500; + ucl_object_unref (ref); + + return ret; +} diff --git a/contrib/libucl/tests/test_msgpack.c b/contrib/libucl/tests/test_msgpack.c new file mode 100644 index 0000000..dd5860e --- /dev/null +++ b/contrib/libucl/tests/test_msgpack.c @@ -0,0 +1,340 @@ +/* + * Copyright (c) 2015, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY AUTHOR ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include "ucl.h" +#include "ucl_internal.h" +#include <ctype.h> + +static const int niter = 1000; +static const int ntests = 100; +static const int nelt = 10; + +static int recursion = 0; + +typedef ucl_object_t* (*ucl_msgpack_test)(void); + +static ucl_object_t* ucl_test_integer (void); +static ucl_object_t* ucl_test_string (void); +static ucl_object_t* ucl_test_boolean (void); +static ucl_object_t* ucl_test_map (void); +static ucl_object_t* ucl_test_array (void); + +ucl_msgpack_test tests[] = { + ucl_test_integer, + ucl_test_string, + ucl_test_boolean, + ucl_test_map, + ucl_test_array, +}; + +#define NTESTS (sizeof(tests) / sizeof(tests[0])) + +typedef struct +{ + uint64_t state; + uint64_t inc; +} pcg32_random_t; + +pcg32_random_t rng; + +/* + * From http://www.pcg-random.org/ + */ +static uint32_t +pcg32_random (void) +{ + uint64_t oldstate = rng.state; + + rng.state = oldstate * 6364136223846793005ULL + (rng.inc | 1); + uint32_t xorshifted = ((oldstate >> 18u) ^ oldstate) >> 27u; + uint32_t rot = oldstate >> 59u; + return (xorshifted >> rot) | (xorshifted << ((-rot) & 31)); +} + +static const char * +random_key (size_t *lenptr) +{ + static char keybuf[512]; + int keylen, i; + char c; + + keylen = pcg32_random () % (sizeof (keybuf) - 1) + 1; + + for (i = 0; i < keylen; i ++) { + do { + c = pcg32_random () & 0xFF; + } while (!isgraph (c)); + + keybuf[i] = c; + } + + *lenptr = keylen; + return keybuf; +} + +int +main (int argc, char **argv) +{ + int fd, i, j; + uint32_t sel; + ucl_object_t *obj, *elt; + struct ucl_parser *parser; + size_t klen, elen, elen2; + const char *key; + unsigned char *emitted, *emitted2; + FILE *out; + const char *fname_out = NULL; + + switch (argc) { + case 2: + fname_out = argv[1]; + break; + } + + /* Seed prng */ + fd = open ("/dev/urandom", O_RDONLY); + assert (fd != -1); + assert (read (fd, &rng, sizeof (rng)) == sizeof (rng)); + close (fd); + + for (i = 0; i < niter; i ++) { + if (fname_out != NULL) { + out = fopen (fname_out, "w"); + if (out == NULL) { + exit (-errno); + } + } + else { + out = NULL; + } + + /* Generate phase */ + obj = ucl_object_typed_new (UCL_OBJECT); + + for (j = 0; j < ntests; j ++) { + sel = pcg32_random () % NTESTS; + + key = random_key (&klen); + recursion = 0; + elt = tests[sel](); + assert (elt != NULL); + assert (klen != 0); + + ucl_object_insert_key (obj, elt, key, klen, true); + } + + emitted = ucl_object_emit_len (obj, UCL_EMIT_MSGPACK, &elen); + + assert (emitted != NULL); + + if (out) { + fprintf (out, "%*.s\n", (int)elen, emitted); + + fclose (out); + } + ucl_object_unref (obj); + + parser = ucl_parser_new (0); + + if (!ucl_parser_add_chunk_full (parser, emitted, elen, 0, + UCL_DUPLICATE_APPEND, UCL_PARSE_MSGPACK)) { + fprintf (stderr, "error parsing input: %s", + ucl_parser_get_error (parser)); + assert (0); + } + + obj = ucl_parser_get_object (parser); + assert (obj != NULL); + + emitted2 = ucl_object_emit_len (obj, UCL_EMIT_MSGPACK, &elen2); + + assert (emitted2 != NULL); + assert (elen2 == elen); + assert (memcmp (emitted, emitted2, elen) == 0); + + ucl_parser_free (parser); + ucl_object_unref (obj); + free (emitted); + free (emitted2); + } + + return 0; +} + + +static ucl_object_t* +ucl_test_integer (void) +{ + ucl_object_t *res; + int count, i; + uint64_t cur; + + res = ucl_object_typed_new (UCL_ARRAY); + count = pcg32_random () % nelt; + + for (i = 0; i < count; i ++) { + cur = ((uint64_t)pcg32_random ()) << 32 | pcg32_random (); + ucl_array_append (res, ucl_object_fromint (cur % 128)); + cur = ((uint64_t)pcg32_random ()) << 32 | pcg32_random (); + ucl_array_append (res, ucl_object_fromint (-cur % 128)); + cur = ((uint64_t)pcg32_random ()) << 32 | pcg32_random (); + ucl_array_append (res, ucl_object_fromint (cur % 65536)); + cur = ((uint64_t)pcg32_random ()) << 32 | pcg32_random (); + ucl_array_append (res, ucl_object_fromint (cur % INT32_MAX)); + cur = ((uint64_t)pcg32_random ()) << 32 | pcg32_random (); + ucl_array_append (res, ucl_object_fromint (cur)); + } + + return res; +} + +static ucl_object_t* +ucl_test_string (void) +{ + ucl_object_t *res, *elt; + int count, i; + uint32_t cur_len; + char *str; + + res = ucl_object_typed_new (UCL_ARRAY); + count = pcg32_random () % nelt; + + for (i = 0; i < count; i ++) { + while ((cur_len = pcg32_random ()) % 128 == 0); + + str = malloc (cur_len % 128); + ucl_array_append (res, ucl_object_fromstring_common (str, cur_len % 128, + UCL_STRING_RAW)); + free (str); + + while ((cur_len = pcg32_random ()) % 512 == 0); + str = malloc (cur_len % 512); + ucl_array_append (res, ucl_object_fromstring_common (str, cur_len % 512, + UCL_STRING_RAW)); + free (str); + + while ((cur_len = pcg32_random ()) % 128 == 0); + str = malloc (cur_len % 128); + elt = ucl_object_fromstring_common (str, cur_len % 128, + UCL_STRING_RAW); + elt->flags |= UCL_OBJECT_BINARY; + ucl_array_append (res, elt); + free (str); + + while ((cur_len = pcg32_random ()) % 512 == 0); + str = malloc (cur_len % 512); + elt = ucl_object_fromstring_common (str, cur_len % 512, + UCL_STRING_RAW); + elt->flags |= UCL_OBJECT_BINARY; + ucl_array_append (res, elt); + free (str); + } + + return res; +} + +static ucl_object_t* +ucl_test_boolean (void) +{ + ucl_object_t *res; + int count, i; + + res = ucl_object_typed_new (UCL_ARRAY); + count = pcg32_random () % nelt; + + for (i = 0; i < count; i ++) { + ucl_array_append (res, ucl_object_frombool (pcg32_random () % 2)); + } + + return res; +} + +static ucl_object_t* +ucl_test_map (void) +{ + ucl_object_t *res, *cur; + int count, i; + uint32_t cur_len, sel; + size_t klen; + const char *key; + + res = ucl_object_typed_new (UCL_OBJECT); + count = pcg32_random () % nelt; + + recursion ++; + + for (i = 0; i < count; i ++) { + + if (recursion > 10) { + sel = pcg32_random () % (NTESTS - 2); + } + else { + sel = pcg32_random () % NTESTS; + } + + key = random_key (&klen); + cur = tests[sel](); + assert (cur != NULL); + assert (klen != 0); + + ucl_object_insert_key (res, cur, key, klen, true); + + /* Multi value key */ + cur = tests[sel](); + assert (cur != NULL); + + ucl_object_insert_key (res, cur, key, klen, true); + } + + return res; +} + +static ucl_object_t* +ucl_test_array (void) +{ + ucl_object_t *res, *cur; + int count, i; + uint32_t cur_len, sel; + + res = ucl_object_typed_new (UCL_ARRAY); + count = pcg32_random () % nelt; + + recursion ++; + + for (i = 0; i < count; i ++) { + if (recursion > 10) { + sel = pcg32_random () % (NTESTS - 2); + } + else { + sel = pcg32_random () % NTESTS; + } + + cur = tests[sel](); + assert (cur != NULL); + + ucl_array_append (res, cur); + } + + return res; +} diff --git a/contrib/libucl/tests/test_schema.c b/contrib/libucl/tests/test_schema.c new file mode 100644 index 0000000..39eb7f3 --- /dev/null +++ b/contrib/libucl/tests/test_schema.c @@ -0,0 +1,164 @@ +/* Copyright (c) 2014, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <stdio.h> +#include <errno.h> +#include <unistd.h> +#include "ucl.h" + +static int +read_stdin (char **buf) +{ + int size = BUFSIZ, remain, ret; + char *p; + + *buf = malloc (size); + if (*buf == NULL) { + return -1; + } + + p = *buf; + remain = size; + + while ((ret = read (STDIN_FILENO, p, remain - 1)) > 0) { + remain -= ret; + p += ret; + + if (remain <= 1) { + *buf = realloc (*buf, size * 2); + if (*buf == NULL) { + return -1; + } + + p = *buf + size - 1; + remain = size + 1; + size *= 2; + } + } + + *p = '\0'; + + return ret; +} + +static bool +perform_test (const ucl_object_t *schema, const ucl_object_t *obj, + struct ucl_schema_error *err) +{ + const ucl_object_t *valid, *data, *description; + bool match; + + data = ucl_object_find_key (obj, "data"); + description = ucl_object_find_key (obj, "description"); + valid = ucl_object_find_key (obj, "valid"); + + if (data == NULL || description == NULL || valid == NULL) { + fprintf (stdout, "Bad test case\n"); + return false; + } + + match = ucl_object_validate (schema, data, err); + if (match != ucl_object_toboolean (valid)) { + fprintf (stdout, "Test case '%s' failed (expected %s): '%s'\n", + ucl_object_tostring (description), + ucl_object_toboolean (valid) ? "valid" : "invalid", + err->msg); + fprintf (stdout, "%s\n", ucl_object_emit (data, UCL_EMIT_CONFIG)); + fprintf (stdout, "%s\n", ucl_object_emit (schema, UCL_EMIT_CONFIG)); + return false; + } + + return true; +} + +static int +perform_tests (const ucl_object_t *obj) +{ + struct ucl_schema_error err; + ucl_object_iter_t iter = NULL; + const ucl_object_t *schema, *tests, *description, *test; + + if (obj->type != UCL_OBJECT) { + fprintf (stdout, "Bad test case\n"); + return EXIT_FAILURE; + } + + schema = ucl_object_find_key (obj, "schema"); + tests = ucl_object_find_key (obj, "tests"); + description = ucl_object_find_key (obj, "description"); + + if (schema == NULL || tests == NULL || description == NULL) { + fprintf (stdout, "Bad test case\n"); + return EXIT_FAILURE; + } + + memset (&err, 0, sizeof (err)); + + while ((test = ucl_iterate_object (tests, &iter, true)) != NULL) { + if (!perform_test (schema, test, &err)) { + fprintf (stdout, "Test suite '%s' failed\n", + ucl_object_tostring (description)); + return EXIT_FAILURE; + } + } + + return 0; +} + +int +main (int argc, char **argv) +{ + char *buf = NULL; + struct ucl_parser *parser; + ucl_object_t *obj = NULL; + const ucl_object_t *elt; + ucl_object_iter_t iter = NULL; + int ret = 0; + + if (read_stdin (&buf) == -1) { + exit (EXIT_FAILURE); + } + + parser = ucl_parser_new (0); + + ucl_parser_add_string (parser, buf, 0); + + if (ucl_parser_get_error (parser) != NULL) { + fprintf (stdout, "Error occurred: %s\n", ucl_parser_get_error (parser)); + ret = 1; + return EXIT_FAILURE; + } + obj = ucl_parser_get_object (parser); + ucl_parser_free (parser); + + while ((elt = ucl_iterate_object (obj, &iter, true)) != NULL) { + ret = perform_tests (elt); + if (ret != 0) { + break; + } + } + + ucl_object_unref (obj); + + return ret; +} diff --git a/contrib/libucl/tests/test_speed.c b/contrib/libucl/tests/test_speed.c new file mode 100644 index 0000000..56f2e5a --- /dev/null +++ b/contrib/libucl/tests/test_speed.c @@ -0,0 +1,151 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <sys/types.h> +#include <sys/mman.h> +#include <sys/stat.h> +#include <sys/time.h> +#include <stdio.h> +#include <errno.h> +#include <unistd.h> +#include <fcntl.h> +#include <time.h> + +#ifdef __APPLE__ +#ifdef HAVE_MACH_MACH_TIME_H +#include <mach/mach_time.h> +#endif +#endif + +#include "ucl.h" + +static double +get_ticks (void) +{ + double res; + +#ifdef __APPLE__ + res = mach_absolute_time () / 1000000000.; +#else + struct timespec ts; + clock_gettime (CLOCK_MONOTONIC, &ts); + + res = (double)ts.tv_sec + ts.tv_nsec / 1000000000.; +#endif + + return res; +} + +int +main (int argc, char **argv) +{ + void *map; + struct ucl_parser *parser; + ucl_object_t *obj; + int fin; + unsigned char *emitted; + struct stat st; + const char *fname_in = NULL; + int ret = 0; + double start, end, seconds; + + switch (argc) { + case 2: + fname_in = argv[1]; + break; + } + + fin = open (fname_in, O_RDONLY); + if (fin == -1) { + perror ("open failed"); + exit (EXIT_FAILURE); + } + parser = ucl_parser_new (UCL_PARSER_ZEROCOPY); + + (void)fstat (fin, &st); + map = mmap (NULL, st.st_size, PROT_READ, MAP_SHARED, fin, 0); + if (map == MAP_FAILED) { + perror ("mmap failed"); + exit (EXIT_FAILURE); + } + + close (fin); + + start = get_ticks (); + ucl_parser_add_chunk (parser, map, st.st_size); + + obj = ucl_parser_get_object (parser); + end = get_ticks (); + + seconds = end - start; + printf ("ucl: parsed input in %.4f seconds\n", seconds); + if (ucl_parser_get_error(parser)) { + printf ("Error occurred: %s\n", ucl_parser_get_error(parser)); + ret = 1; + goto err; + } + + start = get_ticks (); + emitted = ucl_object_emit (obj, UCL_EMIT_CONFIG); + end = get_ticks (); + + seconds = end - start; + printf ("ucl: emitted config in %.4f seconds\n", seconds); + + free (emitted); + + start = get_ticks (); + emitted = ucl_object_emit (obj, UCL_EMIT_JSON); + end = get_ticks (); + + seconds = end - start; + printf ("ucl: emitted json in %.4f seconds\n", seconds); + + free (emitted); + + start = get_ticks (); + emitted = ucl_object_emit (obj, UCL_EMIT_JSON_COMPACT); + end = get_ticks (); + + seconds = end - start; + printf ("ucl: emitted compact json in %.4f seconds\n", seconds); + + free (emitted); + + start = get_ticks (); + emitted = ucl_object_emit (obj, UCL_EMIT_YAML); + end = get_ticks (); + + seconds = end - start; + printf ("ucl: emitted yaml in %.4f seconds\n", seconds); + + free (emitted); + + ucl_parser_free (parser); + ucl_object_unref (obj); + +err: + munmap (map, st.st_size); + + return ret; +} diff --git a/contrib/libucl/tests/test_streamline.c b/contrib/libucl/tests/test_streamline.c new file mode 100644 index 0000000..4c56c4c --- /dev/null +++ b/contrib/libucl/tests/test_streamline.c @@ -0,0 +1,92 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <stdio.h> +#include <errno.h> +#include <assert.h> +#include "ucl.h" + +int +main (int argc, char **argv) +{ + ucl_object_t *obj, *cur, *ar; + FILE *out; + const char *fname_out = NULL; + struct ucl_emitter_context *ctx; + struct ucl_emitter_functions *f; + int ret = 0; + + switch (argc) { + case 2: + fname_out = argv[1]; + break; + } + + if (fname_out != NULL) { + out = fopen (fname_out, "w"); + if (out == NULL) { + exit (-errno); + } + } + else { + out = stdout; + } + + obj = ucl_object_typed_new (UCL_OBJECT); + + /* Create some strings */ + cur = ucl_object_fromstring_common (" test string ", 0, UCL_STRING_TRIM); + ucl_object_insert_key (obj, cur, "key1", 0, false); + cur = ucl_object_fromstring_common (" test \nstring\n ", 0, UCL_STRING_TRIM | UCL_STRING_ESCAPE); + ucl_object_insert_key (obj, cur, "key2", 0, false); + cur = ucl_object_fromstring_common (" test string \n", 0, 0); + ucl_object_insert_key (obj, cur, "key3", 0, false); + + f = ucl_object_emit_file_funcs (out); + ctx = ucl_object_emit_streamline_new (obj, UCL_EMIT_CONFIG, f); + + assert (ctx != NULL); + + /* Array of numbers */ + ar = ucl_object_typed_new (UCL_ARRAY); + ar->key = "key4"; + ar->keylen = sizeof ("key4") - 1; + + ucl_object_emit_streamline_start_container (ctx, ar); + cur = ucl_object_fromint (10); + ucl_object_emit_streamline_add_object (ctx, cur); + cur = ucl_object_fromdouble (10.1); + ucl_object_emit_streamline_add_object (ctx, cur); + cur = ucl_object_fromdouble (9.999); + ucl_object_emit_streamline_add_object (ctx, cur); + + + ucl_object_emit_streamline_end_container (ctx); + ucl_object_emit_streamline_finish (ctx); + ucl_object_emit_funcs_free (f); + ucl_object_unref (obj); + + fclose (out); + + return ret; +} diff --git a/contrib/libucl/uthash/uthash.h b/contrib/libucl/uthash/uthash.h new file mode 100644 index 0000000..36b1cf4 --- /dev/null +++ b/contrib/libucl/uthash/uthash.h @@ -0,0 +1,720 @@ +/* +Copyright (c) 2003-2013, Troy D. Hanson http://troydhanson.github.com/uthash/ +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS +IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED +TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A +PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER +OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, +EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, +PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR +PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF +LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING +NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +*/ + +#ifndef UTHASH_H +#define UTHASH_H + +#include <string.h> /* memcmp,strlen */ +#include <stddef.h> /* ptrdiff_t */ +#include <stdlib.h> /* exit() */ +#include "xxhash.h" + +/* These macros use decltype or the earlier __typeof GNU extension. + As decltype is only available in newer compilers (VS2010 or gcc 4.3+ + when compiling c++ source) this code uses whatever method is needed + or, for VS2008 where neither is available, uses casting workarounds. */ +#ifdef _MSC_VER /* MS compiler */ +#if _MSC_VER >= 1600 && defined(__cplusplus) /* VS2010 or newer in C++ mode */ +#define DECLTYPE(x) (decltype(x)) +#else /* VS2008 or older (or VS2010 in C mode) */ +#define NO_DECLTYPE +#define DECLTYPE(x) +#endif +#else /* GNU, Sun and other compilers */ +#define DECLTYPE(x) (__typeof(x)) +#endif + +#ifdef NO_DECLTYPE +#define DECLTYPE_ASSIGN(dst,src) \ +do { \ + char **_da_dst = (char**)(&(dst)); \ + *_da_dst = (char*)(src); \ +} while(0) +#else +#define DECLTYPE_ASSIGN(dst,src) \ +do { \ + (dst) = DECLTYPE(dst)(src); \ +} while(0) +#endif + +/* a number of the hash function use uint32_t which isn't defined on win32 */ +#ifdef _MSC_VER +typedef unsigned int uint32_t; +typedef unsigned char uint8_t; +#else +#include <inttypes.h> /* uint32_t */ +#endif + +#define UTHASH_VERSION 1.9.8 + +#ifndef uthash_fatal +#define uthash_fatal(msg) exit(-1) /* fatal error (out of memory,etc) */ +#endif +#ifndef uthash_malloc +#define uthash_malloc(sz) malloc(sz) /* malloc fcn */ +#endif +#ifndef uthash_free +#define uthash_free(ptr,sz) free(ptr) /* free fcn */ +#endif + +#ifndef uthash_noexpand_fyi +#define uthash_noexpand_fyi(tbl) /* can be defined to log noexpand */ +#endif +#ifndef uthash_expand_fyi +#define uthash_expand_fyi(tbl) /* can be defined to log expands */ +#endif + +/* initial number of buckets */ +#define HASH_INITIAL_NUM_BUCKETS 32 /* initial number of buckets */ +#define HASH_INITIAL_NUM_BUCKETS_LOG2 5 /* lg2 of initial number of buckets */ +#define HASH_BKT_CAPACITY_THRESH 10 /* expand when bucket count reaches */ + +/* calculate the element whose hash handle address is hhe */ +#define ELMT_FROM_HH(tbl,hhp) ((void*)(((char*)(hhp)) - ((tbl)->hho))) + +#define HASH_FIND(hh,head,keyptr,keylen,out) \ +do { \ + unsigned _hf_bkt,_hf_hashv; \ + out=NULL; \ + if (head) { \ + HASH_FCN(keyptr,keylen, (head)->hh.tbl->num_buckets, _hf_hashv, _hf_bkt); \ + if (HASH_BLOOM_TEST((head)->hh.tbl, _hf_hashv)) { \ + HASH_FIND_IN_BKT((head)->hh.tbl, hh, (head)->hh.tbl->buckets[ _hf_bkt ], \ + keyptr,keylen,out); \ + } \ + } \ +} while (0) + +#ifdef HASH_BLOOM +#define HASH_BLOOM_BITLEN (1ULL << HASH_BLOOM) +#define HASH_BLOOM_BYTELEN (HASH_BLOOM_BITLEN/8) + ((HASH_BLOOM_BITLEN%8) ? 1:0) +#define HASH_BLOOM_MAKE(tbl) \ +do { \ + (tbl)->bloom_nbits = HASH_BLOOM; \ + (tbl)->bloom_bv = (uint8_t*)uthash_malloc(HASH_BLOOM_BYTELEN); \ + if (!((tbl)->bloom_bv)) { uthash_fatal( "out of memory"); } \ + memset((tbl)->bloom_bv, 0, HASH_BLOOM_BYTELEN); \ + (tbl)->bloom_sig = HASH_BLOOM_SIGNATURE; \ +} while (0) + +#define HASH_BLOOM_FREE(tbl) \ +do { \ + uthash_free((tbl)->bloom_bv, HASH_BLOOM_BYTELEN); \ +} while (0) + +#define HASH_BLOOM_BITSET(bv,idx) (bv[(idx)/8] |= (1U << ((idx)%8))) +#define HASH_BLOOM_BITTEST(bv,idx) (bv[(idx)/8] & (1U << ((idx)%8))) + +#define HASH_BLOOM_ADD(tbl,hashv) \ + HASH_BLOOM_BITSET((tbl)->bloom_bv, (hashv & (uint32_t)((1ULL << (tbl)->bloom_nbits) - 1))) + +#define HASH_BLOOM_TEST(tbl,hashv) \ + HASH_BLOOM_BITTEST((tbl)->bloom_bv, (hashv & (uint32_t)((1ULL << (tbl)->bloom_nbits) - 1))) + +#else +#define HASH_BLOOM_MAKE(tbl) +#define HASH_BLOOM_FREE(tbl) +#define HASH_BLOOM_ADD(tbl,hashv) +#define HASH_BLOOM_TEST(tbl,hashv) (1) +#define HASH_BLOOM_BYTELEN 0 +#endif + +#define HASH_MAKE_TABLE(hh,head) \ +do { \ + (head)->hh.tbl = (UT_hash_table*)uthash_malloc( \ + sizeof(UT_hash_table)); \ + if (!((head)->hh.tbl)) { uthash_fatal( "out of memory"); } \ + memset((head)->hh.tbl, 0, sizeof(UT_hash_table)); \ + (head)->hh.tbl->tail = &((head)->hh); \ + (head)->hh.tbl->num_buckets = HASH_INITIAL_NUM_BUCKETS; \ + (head)->hh.tbl->log2_num_buckets = HASH_INITIAL_NUM_BUCKETS_LOG2; \ + (head)->hh.tbl->hho = (char*)(&(head)->hh) - (char*)(head); \ + (head)->hh.tbl->buckets = (UT_hash_bucket*)uthash_malloc( \ + HASH_INITIAL_NUM_BUCKETS*sizeof(struct UT_hash_bucket)); \ + if (! (head)->hh.tbl->buckets) { uthash_fatal( "out of memory"); } \ + memset((head)->hh.tbl->buckets, 0, \ + HASH_INITIAL_NUM_BUCKETS*sizeof(struct UT_hash_bucket)); \ + HASH_BLOOM_MAKE((head)->hh.tbl); \ + (head)->hh.tbl->signature = HASH_SIGNATURE; \ +} while(0) + +#define HASH_ADD(hh,head,fieldname,keylen_in,add) \ + HASH_ADD_KEYPTR(hh,head,&((add)->fieldname),keylen_in,add) + +#define HASH_REPLACE(hh,head,fieldname,keylen_in,add,replaced) \ +do { \ + replaced=NULL; \ + HASH_FIND(hh,head,&((add)->fieldname),keylen_in,replaced); \ + if (replaced!=NULL) { \ + HASH_DELETE(hh,head,replaced); \ + }; \ + HASH_ADD(hh,head,fieldname,keylen_in,add); \ +} while(0) + +#define HASH_ADD_KEYPTR(hh,head,keyptr,keylen_in,add) \ +do { \ + unsigned _ha_bkt; \ + (add)->hh.next = NULL; \ + (add)->hh.key = (const char*)keyptr; \ + (add)->hh.keylen = (unsigned)keylen_in; \ + if (!(head)) { \ + head = (add); \ + (head)->hh.prev = NULL; \ + HASH_MAKE_TABLE(hh,head); \ + } else { \ + (head)->hh.tbl->tail->next = (add); \ + (add)->hh.prev = ELMT_FROM_HH((head)->hh.tbl, (head)->hh.tbl->tail); \ + (head)->hh.tbl->tail = &((add)->hh); \ + } \ + (head)->hh.tbl->num_items++; \ + (add)->hh.tbl = (head)->hh.tbl; \ + HASH_FCN(keyptr,keylen_in, (head)->hh.tbl->num_buckets, \ + (add)->hh.hashv, _ha_bkt); \ + HASH_ADD_TO_BKT((head)->hh.tbl->buckets[_ha_bkt],&(add)->hh); \ + HASH_BLOOM_ADD((head)->hh.tbl,(add)->hh.hashv); \ + HASH_EMIT_KEY(hh,head,keyptr,keylen_in); \ + HASH_FSCK(hh,head); \ +} while(0) + +#define HASH_TO_BKT( hashv, num_bkts, bkt ) \ +do { \ + bkt = ((hashv) & ((num_bkts) - 1)); \ +} while(0) + +/* delete "delptr" from the hash table. + * "the usual" patch-up process for the app-order doubly-linked-list. + * The use of _hd_hh_del below deserves special explanation. + * These used to be expressed using (delptr) but that led to a bug + * if someone used the same symbol for the head and deletee, like + * HASH_DELETE(hh,users,users); + * We want that to work, but by changing the head (users) below + * we were forfeiting our ability to further refer to the deletee (users) + * in the patch-up process. Solution: use scratch space to + * copy the deletee pointer, then the latter references are via that + * scratch pointer rather than through the repointed (users) symbol. + */ +#define HASH_DELETE(hh,head,delptr) \ +do { \ + unsigned _hd_bkt; \ + struct UT_hash_handle *_hd_hh_del; \ + if ( ((delptr)->hh.prev == NULL) && ((delptr)->hh.next == NULL) ) { \ + uthash_free((head)->hh.tbl->buckets, \ + (head)->hh.tbl->num_buckets*sizeof(struct UT_hash_bucket) ); \ + HASH_BLOOM_FREE((head)->hh.tbl); \ + uthash_free((head)->hh.tbl, sizeof(UT_hash_table)); \ + head = NULL; \ + } else { \ + _hd_hh_del = &((delptr)->hh); \ + if ((delptr) == ELMT_FROM_HH((head)->hh.tbl,(head)->hh.tbl->tail)) { \ + (head)->hh.tbl->tail = \ + (UT_hash_handle*)((ptrdiff_t)((delptr)->hh.prev) + \ + (head)->hh.tbl->hho); \ + } \ + if ((delptr)->hh.prev) { \ + ((UT_hash_handle*)((ptrdiff_t)((delptr)->hh.prev) + \ + (head)->hh.tbl->hho))->next = (delptr)->hh.next; \ + } else { \ + DECLTYPE_ASSIGN(head,(delptr)->hh.next); \ + } \ + if (_hd_hh_del->next) { \ + ((UT_hash_handle*)((ptrdiff_t)_hd_hh_del->next + \ + (head)->hh.tbl->hho))->prev = \ + _hd_hh_del->prev; \ + } \ + HASH_TO_BKT( _hd_hh_del->hashv, (head)->hh.tbl->num_buckets, _hd_bkt); \ + HASH_DEL_IN_BKT(hh,(head)->hh.tbl->buckets[_hd_bkt], _hd_hh_del); \ + (head)->hh.tbl->num_items--; \ + } \ + HASH_FSCK(hh,head); \ +} while (0) + + +/* convenience forms of HASH_FIND/HASH_ADD/HASH_DEL */ +#define HASH_FIND_STR(head,findstr,out) \ + HASH_FIND(hh,head,findstr,strlen(findstr),out) +#define HASH_ADD_STR(head,strfield,add) \ + HASH_ADD(hh,head,strfield,strlen(add->strfield),add) +#define HASH_REPLACE_STR(head,strfield,add,replaced) \ + HASH_REPLACE(hh,head,strfield,strlen(add->strfield),add,replaced) +#define HASH_FIND_INT(head,findint,out) \ + HASH_FIND(hh,head,findint,sizeof(int),out) +#define HASH_ADD_INT(head,intfield,add) \ + HASH_ADD(hh,head,intfield,sizeof(int),add) +#define HASH_REPLACE_INT(head,intfield,add,replaced) \ + HASH_REPLACE(hh,head,intfield,sizeof(int),add,replaced) +#define HASH_FIND_PTR(head,findptr,out) \ + HASH_FIND(hh,head,findptr,sizeof(void *),out) +#define HASH_ADD_PTR(head,ptrfield,add) \ + HASH_ADD(hh,head,ptrfield,sizeof(void *),add) +#define HASH_REPLACE_PTR(head,ptrfield,add) \ + HASH_REPLACE(hh,head,ptrfield,sizeof(void *),add,replaced) +#define HASH_DEL(head,delptr) \ + HASH_DELETE(hh,head,delptr) + +/* HASH_FSCK checks hash integrity on every add/delete when HASH_DEBUG is defined. + * This is for uthash developer only; it compiles away if HASH_DEBUG isn't defined. + */ +#ifdef HASH_DEBUG +#define HASH_OOPS(...) do { fprintf(stderr,__VA_ARGS__); exit(-1); } while (0) +#define HASH_FSCK(hh,head) \ +do { \ + unsigned _bkt_i; \ + unsigned _count, _bkt_count; \ + char *_prev; \ + struct UT_hash_handle *_thh; \ + if (head) { \ + _count = 0; \ + for( _bkt_i = 0; _bkt_i < (head)->hh.tbl->num_buckets; _bkt_i++) { \ + _bkt_count = 0; \ + _thh = (head)->hh.tbl->buckets[_bkt_i].hh_head; \ + _prev = NULL; \ + while (_thh) { \ + if (_prev != (char*)(_thh->hh_prev)) { \ + HASH_OOPS("invalid hh_prev %p, actual %p\n", \ + _thh->hh_prev, _prev ); \ + } \ + _bkt_count++; \ + _prev = (char*)(_thh); \ + _thh = _thh->hh_next; \ + } \ + _count += _bkt_count; \ + if ((head)->hh.tbl->buckets[_bkt_i].count != _bkt_count) { \ + HASH_OOPS("invalid bucket count %d, actual %d\n", \ + (head)->hh.tbl->buckets[_bkt_i].count, _bkt_count); \ + } \ + } \ + if (_count != (head)->hh.tbl->num_items) { \ + HASH_OOPS("invalid hh item count %d, actual %d\n", \ + (head)->hh.tbl->num_items, _count ); \ + } \ + /* traverse hh in app order; check next/prev integrity, count */ \ + _count = 0; \ + _prev = NULL; \ + _thh = &(head)->hh; \ + while (_thh) { \ + _count++; \ + if (_prev !=(char*)(_thh->prev)) { \ + HASH_OOPS("invalid prev %p, actual %p\n", \ + _thh->prev, _prev ); \ + } \ + _prev = (char*)ELMT_FROM_HH((head)->hh.tbl, _thh); \ + _thh = ( _thh->next ? (UT_hash_handle*)((char*)(_thh->next) + \ + (head)->hh.tbl->hho) : NULL ); \ + } \ + if (_count != (head)->hh.tbl->num_items) { \ + HASH_OOPS("invalid app item count %d, actual %d\n", \ + (head)->hh.tbl->num_items, _count ); \ + } \ + } \ +} while (0) +#else +#define HASH_FSCK(hh,head) +#endif + +/* When compiled with -DHASH_EMIT_KEYS, length-prefixed keys are emitted to + * the descriptor to which this macro is defined for tuning the hash function. + * The app can #include <unistd.h> to get the prototype for write(2). */ +#ifdef HASH_EMIT_KEYS +#define HASH_EMIT_KEY(hh,head,keyptr,fieldlen) \ +do { \ + unsigned _klen = fieldlen; \ + write(HASH_EMIT_KEYS, &_klen, sizeof(_klen)); \ + write(HASH_EMIT_KEYS, keyptr, fieldlen); \ +} while (0) +#else +#define HASH_EMIT_KEY(hh,head,keyptr,fieldlen) +#endif + +/* default to Jenkin's hash unless overridden e.g. DHASH_FUNCTION=HASH_SAX */ +#ifdef HASH_FUNCTION +#define HASH_FCN HASH_FUNCTION +#else +#define HASH_FCN HASH_XX +#endif + +#define XX_HASH_PRIME 2654435761U + +#define HASH_XX(key,keylen,num_bkts,hashv,bkt) \ +do { \ + hashv = XXH32 (key, keylen, XX_HASH_PRIME); \ + bkt = (hashv) & (num_bkts-1); \ +} while (0) + + + +/* key comparison function; return 0 if keys equal */ +#define HASH_KEYCMP(a,b,len) memcmp(a,b,len) + +/* iterate over items in a known bucket to find desired item */ +#define HASH_FIND_IN_BKT(tbl,hh,head,keyptr,keylen_in,out) \ +do { \ + if (head.hh_head) DECLTYPE_ASSIGN(out,ELMT_FROM_HH(tbl,head.hh_head)); \ + else out=NULL; \ + while (out) { \ + if ((out)->hh.keylen == keylen_in) { \ + if ((HASH_KEYCMP((out)->hh.key,keyptr,keylen_in)) == 0) break; \ + } \ + if ((out)->hh.hh_next) DECLTYPE_ASSIGN(out,ELMT_FROM_HH(tbl,(out)->hh.hh_next)); \ + else out = NULL; \ + } \ +} while(0) + +/* add an item to a bucket */ +#define HASH_ADD_TO_BKT(head,addhh) \ +do { \ + head.count++; \ + (addhh)->hh_next = head.hh_head; \ + (addhh)->hh_prev = NULL; \ + if (head.hh_head) { (head).hh_head->hh_prev = (addhh); } \ + (head).hh_head=addhh; \ + if (head.count >= ((head.expand_mult+1) * HASH_BKT_CAPACITY_THRESH) \ + && (addhh)->tbl->noexpand != 1) { \ + HASH_EXPAND_BUCKETS((addhh)->tbl); \ + } \ +} while(0) + +/* remove an item from a given bucket */ +#define HASH_DEL_IN_BKT(hh,head,hh_del) \ + (head).count--; \ + if ((head).hh_head == hh_del) { \ + (head).hh_head = hh_del->hh_next; \ + } \ + if (hh_del->hh_prev) { \ + hh_del->hh_prev->hh_next = hh_del->hh_next; \ + } \ + if (hh_del->hh_next) { \ + hh_del->hh_next->hh_prev = hh_del->hh_prev; \ + } + +/* Bucket expansion has the effect of doubling the number of buckets + * and redistributing the items into the new buckets. Ideally the + * items will distribute more or less evenly into the new buckets + * (the extent to which this is true is a measure of the quality of + * the hash function as it applies to the key domain). + * + * With the items distributed into more buckets, the chain length + * (item count) in each bucket is reduced. Thus by expanding buckets + * the hash keeps a bound on the chain length. This bounded chain + * length is the essence of how a hash provides constant time lookup. + * + * The calculation of tbl->ideal_chain_maxlen below deserves some + * explanation. First, keep in mind that we're calculating the ideal + * maximum chain length based on the *new* (doubled) bucket count. + * In fractions this is just n/b (n=number of items,b=new num buckets). + * Since the ideal chain length is an integer, we want to calculate + * ceil(n/b). We don't depend on floating point arithmetic in this + * hash, so to calculate ceil(n/b) with integers we could write + * + * ceil(n/b) = (n/b) + ((n%b)?1:0) + * + * and in fact a previous version of this hash did just that. + * But now we have improved things a bit by recognizing that b is + * always a power of two. We keep its base 2 log handy (call it lb), + * so now we can write this with a bit shift and logical AND: + * + * ceil(n/b) = (n>>lb) + ( (n & (b-1)) ? 1:0) + * + */ +#define HASH_EXPAND_BUCKETS(tbl) \ +do { \ + unsigned _he_bkt; \ + unsigned _he_bkt_i; \ + struct UT_hash_handle *_he_thh, *_he_hh_nxt; \ + UT_hash_bucket *_he_new_buckets, *_he_newbkt; \ + _he_new_buckets = (UT_hash_bucket*)uthash_malloc( \ + 2 * tbl->num_buckets * sizeof(struct UT_hash_bucket)); \ + if (!_he_new_buckets) { uthash_fatal( "out of memory"); } \ + memset(_he_new_buckets, 0, \ + 2 * tbl->num_buckets * sizeof(struct UT_hash_bucket)); \ + tbl->ideal_chain_maxlen = \ + (tbl->num_items >> (tbl->log2_num_buckets+1)) + \ + ((tbl->num_items & ((tbl->num_buckets*2)-1)) ? 1 : 0); \ + tbl->nonideal_items = 0; \ + for(_he_bkt_i = 0; _he_bkt_i < tbl->num_buckets; _he_bkt_i++) \ + { \ + _he_thh = tbl->buckets[ _he_bkt_i ].hh_head; \ + while (_he_thh) { \ + _he_hh_nxt = _he_thh->hh_next; \ + HASH_TO_BKT( _he_thh->hashv, tbl->num_buckets*2, _he_bkt); \ + _he_newbkt = &(_he_new_buckets[ _he_bkt ]); \ + if (++(_he_newbkt->count) > tbl->ideal_chain_maxlen) { \ + tbl->nonideal_items++; \ + _he_newbkt->expand_mult = _he_newbkt->count / \ + tbl->ideal_chain_maxlen; \ + } \ + _he_thh->hh_prev = NULL; \ + _he_thh->hh_next = _he_newbkt->hh_head; \ + if (_he_newbkt->hh_head) _he_newbkt->hh_head->hh_prev = \ + _he_thh; \ + _he_newbkt->hh_head = _he_thh; \ + _he_thh = _he_hh_nxt; \ + } \ + } \ + uthash_free( tbl->buckets, tbl->num_buckets*sizeof(struct UT_hash_bucket) ); \ + tbl->num_buckets *= 2; \ + tbl->log2_num_buckets++; \ + tbl->buckets = _he_new_buckets; \ + tbl->ineff_expands = (tbl->nonideal_items > (tbl->num_items >> 1)) ? \ + (tbl->ineff_expands+1) : 0; \ + if (tbl->ineff_expands > 1) { \ + tbl->noexpand=1; \ + uthash_noexpand_fyi(tbl); \ + } \ + uthash_expand_fyi(tbl); \ +} while(0) + + +/* This is an adaptation of Simon Tatham's O(n log(n)) mergesort */ +/* Note that HASH_SORT assumes the hash handle name to be hh. + * HASH_SRT was added to allow the hash handle name to be passed in. */ +#define HASH_SORT(head,cmpfcn) HASH_SRT(hh,head,cmpfcn) +#define HASH_SRT(hh,head,cmpfcn) \ +do { \ + unsigned _hs_i; \ + unsigned _hs_looping,_hs_nmerges,_hs_insize,_hs_psize,_hs_qsize; \ + struct UT_hash_handle *_hs_p, *_hs_q, *_hs_e, *_hs_list, *_hs_tail; \ + if (head) { \ + _hs_insize = 1; \ + _hs_looping = 1; \ + _hs_list = &((head)->hh); \ + while (_hs_looping) { \ + _hs_p = _hs_list; \ + _hs_list = NULL; \ + _hs_tail = NULL; \ + _hs_nmerges = 0; \ + while (_hs_p) { \ + _hs_nmerges++; \ + _hs_q = _hs_p; \ + _hs_psize = 0; \ + for ( _hs_i = 0; _hs_i < _hs_insize; _hs_i++ ) { \ + _hs_psize++; \ + _hs_q = (UT_hash_handle*)((_hs_q->next) ? \ + ((void*)((char*)(_hs_q->next) + \ + (head)->hh.tbl->hho)) : NULL); \ + if (! (_hs_q) ) break; \ + } \ + _hs_qsize = _hs_insize; \ + while ((_hs_psize > 0) || ((_hs_qsize > 0) && _hs_q )) { \ + if (_hs_psize == 0) { \ + _hs_e = _hs_q; \ + _hs_q = (UT_hash_handle*)((_hs_q->next) ? \ + ((void*)((char*)(_hs_q->next) + \ + (head)->hh.tbl->hho)) : NULL); \ + _hs_qsize--; \ + } else if ( (_hs_qsize == 0) || !(_hs_q) ) { \ + _hs_e = _hs_p; \ + if (_hs_p){ \ + _hs_p = (UT_hash_handle*)((_hs_p->next) ? \ + ((void*)((char*)(_hs_p->next) + \ + (head)->hh.tbl->hho)) : NULL); \ + } \ + _hs_psize--; \ + } else if (( \ + cmpfcn(DECLTYPE(head)(ELMT_FROM_HH((head)->hh.tbl,_hs_p)), \ + DECLTYPE(head)(ELMT_FROM_HH((head)->hh.tbl,_hs_q))) \ + ) <= 0) { \ + _hs_e = _hs_p; \ + if (_hs_p){ \ + _hs_p = (UT_hash_handle*)((_hs_p->next) ? \ + ((void*)((char*)(_hs_p->next) + \ + (head)->hh.tbl->hho)) : NULL); \ + } \ + _hs_psize--; \ + } else { \ + _hs_e = _hs_q; \ + _hs_q = (UT_hash_handle*)((_hs_q->next) ? \ + ((void*)((char*)(_hs_q->next) + \ + (head)->hh.tbl->hho)) : NULL); \ + _hs_qsize--; \ + } \ + if ( _hs_tail ) { \ + _hs_tail->next = ((_hs_e) ? \ + ELMT_FROM_HH((head)->hh.tbl,_hs_e) : NULL); \ + } else { \ + _hs_list = _hs_e; \ + } \ + if (_hs_e) { \ + _hs_e->prev = ((_hs_tail) ? \ + ELMT_FROM_HH((head)->hh.tbl,_hs_tail) : NULL); \ + } \ + _hs_tail = _hs_e; \ + } \ + _hs_p = _hs_q; \ + } \ + if (_hs_tail){ \ + _hs_tail->next = NULL; \ + } \ + if ( _hs_nmerges <= 1 ) { \ + _hs_looping=0; \ + (head)->hh.tbl->tail = _hs_tail; \ + DECLTYPE_ASSIGN(head,ELMT_FROM_HH((head)->hh.tbl, _hs_list)); \ + } \ + _hs_insize *= 2; \ + } \ + HASH_FSCK(hh,head); \ + } \ +} while (0) + +/* This function selects items from one hash into another hash. + * The end result is that the selected items have dual presence + * in both hashes. There is no copy of the items made; rather + * they are added into the new hash through a secondary hash + * hash handle that must be present in the structure. */ +#define HASH_SELECT(hh_dst, dst, hh_src, src, cond) \ +do { \ + unsigned _src_bkt, _dst_bkt; \ + void *_last_elt=NULL, *_elt; \ + UT_hash_handle *_src_hh, *_dst_hh, *_last_elt_hh=NULL; \ + ptrdiff_t _dst_hho = ((char*)(&(dst)->hh_dst) - (char*)(dst)); \ + if (src) { \ + for(_src_bkt=0; _src_bkt < (src)->hh_src.tbl->num_buckets; _src_bkt++) { \ + for(_src_hh = (src)->hh_src.tbl->buckets[_src_bkt].hh_head; \ + _src_hh; \ + _src_hh = _src_hh->hh_next) { \ + _elt = ELMT_FROM_HH((src)->hh_src.tbl, _src_hh); \ + if (cond(_elt)) { \ + _dst_hh = (UT_hash_handle*)(((char*)_elt) + _dst_hho); \ + _dst_hh->key = _src_hh->key; \ + _dst_hh->keylen = _src_hh->keylen; \ + _dst_hh->hashv = _src_hh->hashv; \ + _dst_hh->prev = _last_elt; \ + _dst_hh->next = NULL; \ + if (_last_elt_hh) { _last_elt_hh->next = _elt; } \ + if (!dst) { \ + DECLTYPE_ASSIGN(dst,_elt); \ + HASH_MAKE_TABLE(hh_dst,dst); \ + } else { \ + _dst_hh->tbl = (dst)->hh_dst.tbl; \ + } \ + HASH_TO_BKT(_dst_hh->hashv, _dst_hh->tbl->num_buckets, _dst_bkt); \ + HASH_ADD_TO_BKT(_dst_hh->tbl->buckets[_dst_bkt],_dst_hh); \ + (dst)->hh_dst.tbl->num_items++; \ + _last_elt = _elt; \ + _last_elt_hh = _dst_hh; \ + } \ + } \ + } \ + } \ + HASH_FSCK(hh_dst,dst); \ +} while (0) + +#define HASH_CLEAR(hh,head) \ +do { \ + if (head) { \ + uthash_free((head)->hh.tbl->buckets, \ + (head)->hh.tbl->num_buckets*sizeof(struct UT_hash_bucket)); \ + HASH_BLOOM_FREE((head)->hh.tbl); \ + uthash_free((head)->hh.tbl, sizeof(UT_hash_table)); \ + (head)=NULL; \ + } \ +} while(0) + +#define HASH_OVERHEAD(hh,head) \ + (size_t)((((head)->hh.tbl->num_items * sizeof(UT_hash_handle)) + \ + ((head)->hh.tbl->num_buckets * sizeof(UT_hash_bucket)) + \ + (sizeof(UT_hash_table)) + \ + (HASH_BLOOM_BYTELEN))) + +#ifdef NO_DECLTYPE +#define HASH_ITER(hh,head,el,tmp) \ +for((el)=(head), (*(char**)(&(tmp)))=(char*)((head)?(head)->hh.next:NULL); \ + el; (el)=(tmp),(*(char**)(&(tmp)))=(char*)((tmp)?(tmp)->hh.next:NULL)) +#else +#define HASH_ITER(hh,head,el,tmp) \ +for((el)=(head),(tmp)=DECLTYPE(el)((head)?(head)->hh.next:NULL); \ + el; (el)=(tmp),(tmp)=DECLTYPE(el)((tmp)?(tmp)->hh.next:NULL)) +#endif + +/* obtain a count of items in the hash */ +#define HASH_COUNT(head) HASH_CNT(hh,head) +#define HASH_CNT(hh,head) ((head)?((head)->hh.tbl->num_items):0) + +typedef struct UT_hash_bucket { + struct UT_hash_handle *hh_head; + unsigned count; + + /* expand_mult is normally set to 0. In this situation, the max chain length + * threshold is enforced at its default value, HASH_BKT_CAPACITY_THRESH. (If + * the bucket's chain exceeds this length, bucket expansion is triggered). + * However, setting expand_mult to a non-zero value delays bucket expansion + * (that would be triggered by additions to this particular bucket) + * until its chain length reaches a *multiple* of HASH_BKT_CAPACITY_THRESH. + * (The multiplier is simply expand_mult+1). The whole idea of this + * multiplier is to reduce bucket expansions, since they are expensive, in + * situations where we know that a particular bucket tends to be overused. + * It is better to let its chain length grow to a longer yet-still-bounded + * value, than to do an O(n) bucket expansion too often. + */ + unsigned expand_mult; + +} UT_hash_bucket; + +/* random signature used only to find hash tables in external analysis */ +#define HASH_SIGNATURE 0xa0111fe1 +#define HASH_BLOOM_SIGNATURE 0xb12220f2 + +typedef struct UT_hash_table { + UT_hash_bucket *buckets; + unsigned num_buckets, log2_num_buckets; + unsigned num_items; + struct UT_hash_handle *tail; /* tail hh in app order, for fast append */ + ptrdiff_t hho; /* hash handle offset (byte pos of hash handle in element */ + + /* in an ideal situation (all buckets used equally), no bucket would have + * more than ceil(#items/#buckets) items. that's the ideal chain length. */ + unsigned ideal_chain_maxlen; + + /* nonideal_items is the number of items in the hash whose chain position + * exceeds the ideal chain maxlen. these items pay the penalty for an uneven + * hash distribution; reaching them in a chain traversal takes >ideal steps */ + unsigned nonideal_items; + + /* ineffective expands occur when a bucket doubling was performed, but + * afterward, more than half the items in the hash had nonideal chain + * positions. If this happens on two consecutive expansions we inhibit any + * further expansion, as it's not helping; this happens when the hash + * function isn't a good fit for the key domain. When expansion is inhibited + * the hash will still work, albeit no longer in constant time. */ + unsigned ineff_expands, noexpand; + + uint32_t signature; /* used only to find hash tables in external analysis */ +#ifdef HASH_BLOOM + uint32_t bloom_sig; /* used only to test bloom exists in external analysis */ + uint8_t *bloom_bv; + char bloom_nbits; +#endif + +} UT_hash_table; + +typedef struct UT_hash_handle { + struct UT_hash_table *tbl; + void *prev; /* prev element in app order */ + void *next; /* next element in app order */ + struct UT_hash_handle *hh_prev; /* previous hh in bucket order */ + struct UT_hash_handle *hh_next; /* next hh in bucket order */ + const void *key; /* ptr to enclosing struct's key */ + unsigned keylen; /* enclosing struct's key len */ + unsigned hashv; /* result of hash-fcn(key) */ +} UT_hash_handle; + +#endif /* UTHASH_H */ diff --git a/contrib/libucl/uthash/utlist.h b/contrib/libucl/uthash/utlist.h new file mode 100644 index 0000000..c82dd91 --- /dev/null +++ b/contrib/libucl/uthash/utlist.h @@ -0,0 +1,757 @@ +/* +Copyright (c) 2007-2013, Troy D. Hanson http://troydhanson.github.com/uthash/ +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS +IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED +TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A +PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER +OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, +EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, +PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR +PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF +LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING +NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +*/ + +#ifndef UTLIST_H +#define UTLIST_H + +#define UTLIST_VERSION 1.9.8 + +#include <assert.h> + +/* + * This file contains macros to manipulate singly and doubly-linked lists. + * + * 1. LL_ macros: singly-linked lists. + * 2. DL_ macros: doubly-linked lists. + * 3. CDL_ macros: circular doubly-linked lists. + * + * To use singly-linked lists, your structure must have a "next" pointer. + * To use doubly-linked lists, your structure must "prev" and "next" pointers. + * Either way, the pointer to the head of the list must be initialized to NULL. + * + * ----------------.EXAMPLE ------------------------- + * struct item { + * int id; + * struct item *prev, *next; + * } + * + * struct item *list = NULL: + * + * int main() { + * struct item *item; + * ... allocate and populate item ... + * DL_APPEND(list, item); + * } + * -------------------------------------------------- + * + * For doubly-linked lists, the append and delete macros are O(1) + * For singly-linked lists, append and delete are O(n) but prepend is O(1) + * The sort macro is O(n log(n)) for all types of single/double/circular lists. + */ + +/* These macros use decltype or the earlier __typeof GNU extension. + As decltype is only available in newer compilers (VS2010 or gcc 4.3+ + when compiling c++ code), this code uses whatever method is needed + or, for VS2008 where neither is available, uses casting workarounds. */ +#ifdef _MSC_VER /* MS compiler */ +#if _MSC_VER >= 1600 && defined(__cplusplus) /* VS2010 or newer in C++ mode */ +#define LDECLTYPE(x) decltype(x) +#else /* VS2008 or older (or VS2010 in C mode) */ +#define NO_DECLTYPE +#define LDECLTYPE(x) char* +#endif +#elif defined(__ICCARM__) +#define NO_DECLTYPE +#define LDECLTYPE(x) char* +#else /* GNU, Sun and other compilers */ +#define LDECLTYPE(x) __typeof(x) +#endif + +/* for VS2008 we use some workarounds to get around the lack of decltype, + * namely, we always reassign our tmp variable to the list head if we need + * to dereference its prev/next pointers, and save/restore the real head.*/ +#ifdef NO_DECLTYPE +#define _SV(elt,list) _tmp = (char*)(list); {char **_alias = (char**)&(list); *_alias = (elt); } +#define _NEXT(elt,list,next) ((char*)((list)->next)) +#define _NEXTASGN(elt,list,to,next) { char **_alias = (char**)&((list)->next); *_alias=(char*)(to); } +/* #define _PREV(elt,list,prev) ((char*)((list)->prev)) */ +#define _PREVASGN(elt,list,to,prev) { char **_alias = (char**)&((list)->prev); *_alias=(char*)(to); } +#define _RS(list) { char **_alias = (char**)&(list); *_alias=_tmp; } +#define _CASTASGN(a,b) { char **_alias = (char**)&(a); *_alias=(char*)(b); } +#else +#define _SV(elt,list) +#define _NEXT(elt,list,next) ((elt)->next) +#define _NEXTASGN(elt,list,to,next) ((elt)->next)=(to) +/* #define _PREV(elt,list,prev) ((elt)->prev) */ +#define _PREVASGN(elt,list,to,prev) ((elt)->prev)=(to) +#define _RS(list) +#define _CASTASGN(a,b) (a)=(b) +#endif + +/****************************************************************************** + * The sort macro is an adaptation of Simon Tatham's O(n log(n)) mergesort * + * Unwieldy variable names used here to avoid shadowing passed-in variables. * + *****************************************************************************/ +#define LL_SORT(list, cmp) \ + LL_SORT2(list, cmp, next) + +#define LL_SORT2(list, cmp, next) \ +do { \ + LDECLTYPE(list) _ls_p; \ + LDECLTYPE(list) _ls_q; \ + LDECLTYPE(list) _ls_e; \ + LDECLTYPE(list) _ls_tail; \ + int _ls_insize, _ls_nmerges, _ls_psize, _ls_qsize, _ls_i, _ls_looping; \ + if (list) { \ + _ls_insize = 1; \ + _ls_looping = 1; \ + while (_ls_looping) { \ + _CASTASGN(_ls_p,list); \ + list = NULL; \ + _ls_tail = NULL; \ + _ls_nmerges = 0; \ + while (_ls_p) { \ + _ls_nmerges++; \ + _ls_q = _ls_p; \ + _ls_psize = 0; \ + for (_ls_i = 0; _ls_i < _ls_insize; _ls_i++) { \ + _ls_psize++; \ + _SV(_ls_q,list); _ls_q = _NEXT(_ls_q,list,next); _RS(list); \ + if (!_ls_q) break; \ + } \ + _ls_qsize = _ls_insize; \ + while (_ls_psize > 0 || (_ls_qsize > 0 && _ls_q)) { \ + if (_ls_psize == 0) { \ + _ls_e = _ls_q; _SV(_ls_q,list); _ls_q = \ + _NEXT(_ls_q,list,next); _RS(list); _ls_qsize--; \ + } else if (_ls_qsize == 0 || !_ls_q) { \ + _ls_e = _ls_p; _SV(_ls_p,list); _ls_p = \ + _NEXT(_ls_p,list,next); _RS(list); _ls_psize--; \ + } else if (cmp(_ls_p,_ls_q) <= 0) { \ + _ls_e = _ls_p; _SV(_ls_p,list); _ls_p = \ + _NEXT(_ls_p,list,next); _RS(list); _ls_psize--; \ + } else { \ + _ls_e = _ls_q; _SV(_ls_q,list); _ls_q = \ + _NEXT(_ls_q,list,next); _RS(list); _ls_qsize--; \ + } \ + if (_ls_tail) { \ + _SV(_ls_tail,list); _NEXTASGN(_ls_tail,list,_ls_e,next); _RS(list); \ + } else { \ + _CASTASGN(list,_ls_e); \ + } \ + _ls_tail = _ls_e; \ + } \ + _ls_p = _ls_q; \ + } \ + if (_ls_tail) { \ + _SV(_ls_tail,list); _NEXTASGN(_ls_tail,list,NULL,next); _RS(list); \ + } \ + if (_ls_nmerges <= 1) { \ + _ls_looping=0; \ + } \ + _ls_insize *= 2; \ + } \ + } \ +} while (0) + + +#define DL_SORT(list, cmp) \ + DL_SORT2(list, cmp, prev, next) + +#define DL_SORT2(list, cmp, prev, next) \ +do { \ + LDECLTYPE(list) _ls_p; \ + LDECLTYPE(list) _ls_q; \ + LDECLTYPE(list) _ls_e; \ + LDECLTYPE(list) _ls_tail; \ + int _ls_insize, _ls_nmerges, _ls_psize, _ls_qsize, _ls_i, _ls_looping; \ + if (list) { \ + _ls_insize = 1; \ + _ls_looping = 1; \ + while (_ls_looping) { \ + _CASTASGN(_ls_p,list); \ + list = NULL; \ + _ls_tail = NULL; \ + _ls_nmerges = 0; \ + while (_ls_p) { \ + _ls_nmerges++; \ + _ls_q = _ls_p; \ + _ls_psize = 0; \ + for (_ls_i = 0; _ls_i < _ls_insize; _ls_i++) { \ + _ls_psize++; \ + _SV(_ls_q,list); _ls_q = _NEXT(_ls_q,list,next); _RS(list); \ + if (!_ls_q) break; \ + } \ + _ls_qsize = _ls_insize; \ + while (_ls_psize > 0 || (_ls_qsize > 0 && _ls_q)) { \ + if (_ls_psize == 0) { \ + _ls_e = _ls_q; _SV(_ls_q,list); _ls_q = \ + _NEXT(_ls_q,list,next); _RS(list); _ls_qsize--; \ + } else if (_ls_qsize == 0 || !_ls_q) { \ + _ls_e = _ls_p; _SV(_ls_p,list); _ls_p = \ + _NEXT(_ls_p,list,next); _RS(list); _ls_psize--; \ + } else if (cmp(_ls_p,_ls_q) <= 0) { \ + _ls_e = _ls_p; _SV(_ls_p,list); _ls_p = \ + _NEXT(_ls_p,list,next); _RS(list); _ls_psize--; \ + } else { \ + _ls_e = _ls_q; _SV(_ls_q,list); _ls_q = \ + _NEXT(_ls_q,list,next); _RS(list); _ls_qsize--; \ + } \ + if (_ls_tail) { \ + _SV(_ls_tail,list); _NEXTASGN(_ls_tail,list,_ls_e,next); _RS(list); \ + } else { \ + _CASTASGN(list,_ls_e); \ + } \ + _SV(_ls_e,list); _PREVASGN(_ls_e,list,_ls_tail,prev); _RS(list); \ + _ls_tail = _ls_e; \ + } \ + _ls_p = _ls_q; \ + } \ + _CASTASGN(list->prev, _ls_tail); \ + _SV(_ls_tail,list); _NEXTASGN(_ls_tail,list,NULL,next); _RS(list); \ + if (_ls_nmerges <= 1) { \ + _ls_looping=0; \ + } \ + _ls_insize *= 2; \ + } \ + } \ +} while (0) + +#define CDL_SORT(list, cmp) \ + CDL_SORT2(list, cmp, prev, next) + +#define CDL_SORT2(list, cmp, prev, next) \ +do { \ + LDECLTYPE(list) _ls_p; \ + LDECLTYPE(list) _ls_q; \ + LDECLTYPE(list) _ls_e; \ + LDECLTYPE(list) _ls_tail; \ + LDECLTYPE(list) _ls_oldhead; \ + LDECLTYPE(list) _tmp; \ + int _ls_insize, _ls_nmerges, _ls_psize, _ls_qsize, _ls_i, _ls_looping; \ + if (list) { \ + _ls_insize = 1; \ + _ls_looping = 1; \ + while (_ls_looping) { \ + _CASTASGN(_ls_p,list); \ + _CASTASGN(_ls_oldhead,list); \ + list = NULL; \ + _ls_tail = NULL; \ + _ls_nmerges = 0; \ + while (_ls_p) { \ + _ls_nmerges++; \ + _ls_q = _ls_p; \ + _ls_psize = 0; \ + for (_ls_i = 0; _ls_i < _ls_insize; _ls_i++) { \ + _ls_psize++; \ + _SV(_ls_q,list); \ + if (_NEXT(_ls_q,list,next) == _ls_oldhead) { \ + _ls_q = NULL; \ + } else { \ + _ls_q = _NEXT(_ls_q,list,next); \ + } \ + _RS(list); \ + if (!_ls_q) break; \ + } \ + _ls_qsize = _ls_insize; \ + while (_ls_psize > 0 || (_ls_qsize > 0 && _ls_q)) { \ + if (_ls_psize == 0) { \ + _ls_e = _ls_q; _SV(_ls_q,list); _ls_q = \ + _NEXT(_ls_q,list,next); _RS(list); _ls_qsize--; \ + if (_ls_q == _ls_oldhead) { _ls_q = NULL; } \ + } else if (_ls_qsize == 0 || !_ls_q) { \ + _ls_e = _ls_p; _SV(_ls_p,list); _ls_p = \ + _NEXT(_ls_p,list,next); _RS(list); _ls_psize--; \ + if (_ls_p == _ls_oldhead) { _ls_p = NULL; } \ + } else if (cmp(_ls_p,_ls_q) <= 0) { \ + _ls_e = _ls_p; _SV(_ls_p,list); _ls_p = \ + _NEXT(_ls_p,list,next); _RS(list); _ls_psize--; \ + if (_ls_p == _ls_oldhead) { _ls_p = NULL; } \ + } else { \ + _ls_e = _ls_q; _SV(_ls_q,list); _ls_q = \ + _NEXT(_ls_q,list,next); _RS(list); _ls_qsize--; \ + if (_ls_q == _ls_oldhead) { _ls_q = NULL; } \ + } \ + if (_ls_tail) { \ + _SV(_ls_tail,list); _NEXTASGN(_ls_tail,list,_ls_e,next); _RS(list); \ + } else { \ + _CASTASGN(list,_ls_e); \ + } \ + _SV(_ls_e,list); _PREVASGN(_ls_e,list,_ls_tail,prev); _RS(list); \ + _ls_tail = _ls_e; \ + } \ + _ls_p = _ls_q; \ + } \ + _CASTASGN(list->prev,_ls_tail); \ + _CASTASGN(_tmp,list); \ + _SV(_ls_tail,list); _NEXTASGN(_ls_tail,list,_tmp,next); _RS(list); \ + if (_ls_nmerges <= 1) { \ + _ls_looping=0; \ + } \ + _ls_insize *= 2; \ + } \ + } \ +} while (0) + +/****************************************************************************** + * singly linked list macros (non-circular) * + *****************************************************************************/ +#define LL_PREPEND(head,add) \ + LL_PREPEND2(head,add,next) + +#define LL_PREPEND2(head,add,next) \ +do { \ + (add)->next = head; \ + head = add; \ +} while (0) + +#define LL_CONCAT(head1,head2) \ + LL_CONCAT2(head1,head2,next) + +#define LL_CONCAT2(head1,head2,next) \ +do { \ + LDECLTYPE(head1) _tmp; \ + if (head1) { \ + _tmp = head1; \ + while (_tmp->next) { _tmp = _tmp->next; } \ + _tmp->next=(head2); \ + } else { \ + (head1)=(head2); \ + } \ +} while (0) + +#define LL_APPEND(head,add) \ + LL_APPEND2(head,add,next) + +#define LL_APPEND2(head,add,next) \ +do { \ + LDECLTYPE(head) _tmp; \ + (add)->next=NULL; \ + if (head) { \ + _tmp = head; \ + while (_tmp->next) { _tmp = _tmp->next; } \ + _tmp->next=(add); \ + } else { \ + (head)=(add); \ + } \ +} while (0) + +#define LL_DELETE(head,del) \ + LL_DELETE2(head,del,next) + +#define LL_DELETE2(head,del,next) \ +do { \ + LDECLTYPE(head) _tmp; \ + if ((head) == (del)) { \ + (head)=(head)->next; \ + } else { \ + _tmp = head; \ + while (_tmp->next && (_tmp->next != (del))) { \ + _tmp = _tmp->next; \ + } \ + if (_tmp->next) { \ + _tmp->next = ((del)->next); \ + } \ + } \ +} while (0) + +/* Here are VS2008 replacements for LL_APPEND and LL_DELETE */ +#define LL_APPEND_VS2008(head,add) \ + LL_APPEND2_VS2008(head,add,next) + +#define LL_APPEND2_VS2008(head,add,next) \ +do { \ + if (head) { \ + (add)->next = head; /* use add->next as a temp variable */ \ + while ((add)->next->next) { (add)->next = (add)->next->next; } \ + (add)->next->next=(add); \ + } else { \ + (head)=(add); \ + } \ + (add)->next=NULL; \ +} while (0) + +#define LL_DELETE_VS2008(head,del) \ + LL_DELETE2_VS2008(head,del,next) + +#define LL_DELETE2_VS2008(head,del,next) \ +do { \ + if ((head) == (del)) { \ + (head)=(head)->next; \ + } else { \ + char *_tmp = (char*)(head); \ + while ((head)->next && ((head)->next != (del))) { \ + head = (head)->next; \ + } \ + if ((head)->next) { \ + (head)->next = ((del)->next); \ + } \ + { \ + char **_head_alias = (char**)&(head); \ + *_head_alias = _tmp; \ + } \ + } \ +} while (0) +#ifdef NO_DECLTYPE +#undef LL_APPEND +#define LL_APPEND LL_APPEND_VS2008 +#undef LL_DELETE +#define LL_DELETE LL_DELETE_VS2008 +#undef LL_DELETE2 +#define LL_DELETE2 LL_DELETE2_VS2008 +#undef LL_APPEND2 +#define LL_APPEND2 LL_APPEND2_VS2008 +#undef LL_CONCAT /* no LL_CONCAT_VS2008 */ +#undef DL_CONCAT /* no DL_CONCAT_VS2008 */ +#endif +/* end VS2008 replacements */ + +#define LL_COUNT(head,el,counter) \ + LL_COUNT2(head,el,counter,next) \ + +#define LL_COUNT2(head,el,counter,next) \ +{ \ + counter = 0; \ + LL_FOREACH2(head,el,next){ ++counter; } \ +} + +#define LL_FOREACH(head,el) \ + LL_FOREACH2(head,el,next) + +#define LL_FOREACH2(head,el,next) \ + for(el=head;el;el=(el)->next) + +#define LL_FOREACH_SAFE(head,el,tmp) \ + LL_FOREACH_SAFE2(head,el,tmp,next) + +#define LL_FOREACH_SAFE2(head,el,tmp,next) \ + for((el)=(head);(el) && (tmp = (el)->next, 1); (el) = tmp) + +#define LL_SEARCH_SCALAR(head,out,field,val) \ + LL_SEARCH_SCALAR2(head,out,field,val,next) + +#define LL_SEARCH_SCALAR2(head,out,field,val,next) \ +do { \ + LL_FOREACH2(head,out,next) { \ + if ((out)->field == (val)) break; \ + } \ +} while(0) + +#define LL_SEARCH(head,out,elt,cmp) \ + LL_SEARCH2(head,out,elt,cmp,next) + +#define LL_SEARCH2(head,out,elt,cmp,next) \ +do { \ + LL_FOREACH2(head,out,next) { \ + if ((cmp(out,elt))==0) break; \ + } \ +} while(0) + +#define LL_REPLACE_ELEM(head, el, add) \ +do { \ + LDECLTYPE(head) _tmp; \ + assert(head != NULL); \ + assert(el != NULL); \ + assert(add != NULL); \ + (add)->next = (el)->next; \ + if ((head) == (el)) { \ + (head) = (add); \ + } else { \ + _tmp = head; \ + while (_tmp->next && (_tmp->next != (el))) { \ + _tmp = _tmp->next; \ + } \ + if (_tmp->next) { \ + _tmp->next = (add); \ + } \ + } \ +} while (0) + +#define LL_PREPEND_ELEM(head, el, add) \ +do { \ + LDECLTYPE(head) _tmp; \ + assert(head != NULL); \ + assert(el != NULL); \ + assert(add != NULL); \ + (add)->next = (el); \ + if ((head) == (el)) { \ + (head) = (add); \ + } else { \ + _tmp = head; \ + while (_tmp->next && (_tmp->next != (el))) { \ + _tmp = _tmp->next; \ + } \ + if (_tmp->next) { \ + _tmp->next = (add); \ + } \ + } \ +} while (0) \ + + +/****************************************************************************** + * doubly linked list macros (non-circular) * + *****************************************************************************/ +#define DL_PREPEND(head,add) \ + DL_PREPEND2(head,add,prev,next) + +#define DL_PREPEND2(head,add,prev,next) \ +do { \ + (add)->next = head; \ + if (head) { \ + (add)->prev = (head)->prev; \ + (head)->prev = (add); \ + } else { \ + (add)->prev = (add); \ + } \ + (head) = (add); \ +} while (0) + +#define DL_APPEND(head,add) \ + DL_APPEND2(head,add,prev,next) + +#define DL_APPEND2(head,add,prev,next) \ +do { \ + if (head) { \ + (add)->prev = (head)->prev; \ + (head)->prev->next = (add); \ + (head)->prev = (add); \ + (add)->next = NULL; \ + } else { \ + (head)=(add); \ + (head)->prev = (head); \ + (head)->next = NULL; \ + } \ +} while (0) + +#define DL_CONCAT(head1,head2) \ + DL_CONCAT2(head1,head2,prev,next) + +#define DL_CONCAT2(head1,head2,prev,next) \ +do { \ + LDECLTYPE(head1) _tmp; \ + if (head2) { \ + if (head1) { \ + _tmp = (head2)->prev; \ + (head2)->prev = (head1)->prev; \ + (head1)->prev->next = (head2); \ + (head1)->prev = _tmp; \ + } else { \ + (head1)=(head2); \ + } \ + } \ +} while (0) + +#define DL_DELETE(head,del) \ + DL_DELETE2(head,del,prev,next) + +#define DL_DELETE2(head,del,prev,next) \ +do { \ + assert((del)->prev != NULL); \ + if ((del)->prev == (del)) { \ + (head)=NULL; \ + } else if ((del)==(head)) { \ + (del)->next->prev = (del)->prev; \ + (head) = (del)->next; \ + } else { \ + (del)->prev->next = (del)->next; \ + if ((del)->next) { \ + (del)->next->prev = (del)->prev; \ + } else { \ + (head)->prev = (del)->prev; \ + } \ + } \ +} while (0) + +#define DL_COUNT(head,el,counter) \ + DL_COUNT2(head,el,counter,next) \ + +#define DL_COUNT2(head,el,counter,next) \ +{ \ + counter = 0; \ + DL_FOREACH2(head,el,next){ ++counter; } \ +} + +#define DL_FOREACH(head,el) \ + DL_FOREACH2(head,el,next) + +#define DL_FOREACH2(head,el,next) \ + for(el=head;el;el=(el)->next) + +/* this version is safe for deleting the elements during iteration */ +#define DL_FOREACH_SAFE(head,el,tmp) \ + DL_FOREACH_SAFE2(head,el,tmp,next) + +#define DL_FOREACH_SAFE2(head,el,tmp,next) \ + for((el)=(head);(el) && (tmp = (el)->next, 1); (el) = tmp) + +/* these are identical to their singly-linked list counterparts */ +#define DL_SEARCH_SCALAR LL_SEARCH_SCALAR +#define DL_SEARCH LL_SEARCH +#define DL_SEARCH_SCALAR2 LL_SEARCH_SCALAR2 +#define DL_SEARCH2 LL_SEARCH2 + +#define DL_REPLACE_ELEM(head, el, add) \ +do { \ + assert(head != NULL); \ + assert(el != NULL); \ + assert(add != NULL); \ + if ((head) == (el)) { \ + (head) = (add); \ + (add)->next = (el)->next; \ + if ((el)->next == NULL) { \ + (add)->prev = (add); \ + } else { \ + (add)->prev = (el)->prev; \ + (add)->next->prev = (add); \ + } \ + } else { \ + (add)->next = (el)->next; \ + (add)->prev = (el)->prev; \ + (add)->prev->next = (add); \ + if ((el)->next == NULL) { \ + (head)->prev = (add); \ + } else { \ + (add)->next->prev = (add); \ + } \ + } \ +} while (0) + +#define DL_PREPEND_ELEM(head, el, add) \ +do { \ + assert(head != NULL); \ + assert(el != NULL); \ + assert(add != NULL); \ + (add)->next = (el); \ + (add)->prev = (el)->prev; \ + (el)->prev = (add); \ + if ((head) == (el)) { \ + (head) = (add); \ + } else { \ + (add)->prev->next = (add); \ + } \ +} while (0) \ + + +/****************************************************************************** + * circular doubly linked list macros * + *****************************************************************************/ +#define CDL_PREPEND(head,add) \ + CDL_PREPEND2(head,add,prev,next) + +#define CDL_PREPEND2(head,add,prev,next) \ +do { \ + if (head) { \ + (add)->prev = (head)->prev; \ + (add)->next = (head); \ + (head)->prev = (add); \ + (add)->prev->next = (add); \ + } else { \ + (add)->prev = (add); \ + (add)->next = (add); \ + } \ +(head)=(add); \ +} while (0) + +#define CDL_DELETE(head,del) \ + CDL_DELETE2(head,del,prev,next) + +#define CDL_DELETE2(head,del,prev,next) \ +do { \ + if ( ((head)==(del)) && ((head)->next == (head))) { \ + (head) = 0L; \ + } else { \ + (del)->next->prev = (del)->prev; \ + (del)->prev->next = (del)->next; \ + if ((del) == (head)) (head)=(del)->next; \ + } \ +} while (0) + +#define CDL_COUNT(head,el,counter) \ + CDL_COUNT2(head,el,counter,next) \ + +#define CDL_COUNT2(head, el, counter,next) \ +{ \ + counter = 0; \ + CDL_FOREACH2(head,el,next){ ++counter; } \ +} + +#define CDL_FOREACH(head,el) \ + CDL_FOREACH2(head,el,next) + +#define CDL_FOREACH2(head,el,next) \ + for(el=head;el;el=((el)->next==head ? 0L : (el)->next)) + +#define CDL_FOREACH_SAFE(head,el,tmp1,tmp2) \ + CDL_FOREACH_SAFE2(head,el,tmp1,tmp2,prev,next) + +#define CDL_FOREACH_SAFE2(head,el,tmp1,tmp2,prev,next) \ + for((el)=(head), ((tmp1)=(head)?((head)->prev):NULL); \ + (el) && ((tmp2)=(el)->next, 1); \ + ((el) = (((el)==(tmp1)) ? 0L : (tmp2)))) + +#define CDL_SEARCH_SCALAR(head,out,field,val) \ + CDL_SEARCH_SCALAR2(head,out,field,val,next) + +#define CDL_SEARCH_SCALAR2(head,out,field,val,next) \ +do { \ + CDL_FOREACH2(head,out,next) { \ + if ((out)->field == (val)) break; \ + } \ +} while(0) + +#define CDL_SEARCH(head,out,elt,cmp) \ + CDL_SEARCH2(head,out,elt,cmp,next) + +#define CDL_SEARCH2(head,out,elt,cmp,next) \ +do { \ + CDL_FOREACH2(head,out,next) { \ + if ((cmp(out,elt))==0) break; \ + } \ +} while(0) + +#define CDL_REPLACE_ELEM(head, el, add) \ +do { \ + assert(head != NULL); \ + assert(el != NULL); \ + assert(add != NULL); \ + if ((el)->next == (el)) { \ + (add)->next = (add); \ + (add)->prev = (add); \ + (head) = (add); \ + } else { \ + (add)->next = (el)->next; \ + (add)->prev = (el)->prev; \ + (add)->next->prev = (add); \ + (add)->prev->next = (add); \ + if ((head) == (el)) { \ + (head) = (add); \ + } \ + } \ +} while (0) + +#define CDL_PREPEND_ELEM(head, el, add) \ +do { \ + assert(head != NULL); \ + assert(el != NULL); \ + assert(add != NULL); \ + (add)->next = (el); \ + (add)->prev = (el)->prev; \ + (el)->prev = (add); \ + (add)->prev->next = (add); \ + if ((head) == (el)) { \ + (head) = (add); \ + } \ +} while (0) \ + +#endif /* UTLIST_H */ + diff --git a/contrib/libucl/uthash/utstring.h b/contrib/libucl/uthash/utstring.h new file mode 100644 index 0000000..b439948 --- /dev/null +++ b/contrib/libucl/uthash/utstring.h @@ -0,0 +1,412 @@ +/* +Copyright (c) 2008-2013, Troy D. Hanson http://troydhanson.github.com/uthash/ +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS +IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED +TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A +PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER +OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, +EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, +PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR +PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF +LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING +NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +*/ + +/* a dynamic string implementation using macros + */ +#ifndef UTSTRING_H +#define UTSTRING_H + +#define UTSTRING_VERSION 1.9.8 + +#ifdef __GNUC__ +#define _UNUSED_ __attribute__ ((__unused__)) +#else +#define _UNUSED_ +#endif + +#include <stdlib.h> +#include <string.h> +#include <stdarg.h> + +#ifndef oom +#define oom abort +#endif + +typedef struct { + char *d; + void **pd; + size_t n; /* allocd size */ + size_t i; /* index of first unused byte */ +} UT_string; + +#define utstring_reserve(s,amt) \ +do { \ + if (((s)->n - (s)->i) < (size_t)(amt)) { \ + (s)->d = (char*)realloc((s)->d, (s)->n + amt); \ + if ((s)->d == NULL) oom(); \ + else {(s)->n += amt; \ + if ((s)->pd) *((s)->pd) = (s)->d;} \ + } \ +} while(0) + +#define utstring_init(s) \ +do { \ + (s)->n = 0; (s)->i = 0; (s)->d = NULL; \ + utstring_reserve(s,128); \ + (s)->d[0] = '\0'; \ +} while(0) + +#define utstring_done(s) \ +do { \ + if ((s)->d != NULL) free((s)->d); \ + (s)->n = 0; \ +} while(0) + +#define utstring_free(s) \ +do { \ + utstring_done(s); \ + free(s); \ +} while(0) + +#define utstring_new(s) \ +do { \ + s = (UT_string*)calloc(1, sizeof(UT_string)); \ + if (!s) oom(); \ + else utstring_init(s); \ +} while(0) + +#define utstring_renew(s) \ +do { \ + if (s) { \ + utstring_clear(s); \ + } else { \ + utstring_new(s); \ + } \ +} while(0) + +#define utstring_clear(s) \ +do { \ + (s)->i = 0; \ + (s)->d[0] = '\0'; \ +} while(0) + +#define utstring_bincpy(s,b,l) \ +do { \ + utstring_reserve((s),(l)+1); \ + if (l) memcpy(&(s)->d[(s)->i], b, l); \ + (s)->i += l; \ + (s)->d[(s)->i]='\0'; \ +} while(0) + +#define utstring_concat(dst,src) \ +do { \ + utstring_reserve((dst),((src)->i)+1); \ + if ((src)->i) memcpy(&(dst)->d[(dst)->i], (src)->d, (src)->i); \ + (dst)->i += (src)->i; \ + (dst)->d[(dst)->i]='\0'; \ +} while(0) + +#define utstring_len(s) ((unsigned)((s)->i)) + +#define utstring_body(s) ((s)->d) + +_UNUSED_ static void utstring_printf_va(UT_string *s, const char *fmt, va_list ap) { + int n; + va_list cp; + while (1) { +#ifdef _WIN32 + cp = ap; +#else + va_copy(cp, ap); +#endif + n = vsnprintf (&s->d[s->i], s->n-s->i, fmt, cp); + va_end(cp); + + if ((n > -1) && (n < (int)(s->n-s->i))) { + s->i += n; + return; + } + + /* Else try again with more space. */ + if (n > -1) utstring_reserve(s,n+1); /* exact */ + else utstring_reserve(s,(s->n)*2); /* 2x */ + } +} +#ifdef __GNUC__ +/* support printf format checking (2=the format string, 3=start of varargs) */ +static void utstring_printf(UT_string *s, const char *fmt, ...) + __attribute__ (( format( printf, 2, 3) )); +#endif +_UNUSED_ static void utstring_printf(UT_string *s, const char *fmt, ...) { + va_list ap; + va_start(ap,fmt); + utstring_printf_va(s,fmt,ap); + va_end(ap); +} + +#define utstring_append_len(dst, src, len) \ +do { \ + while ((dst)->n-(dst)->i <= (len)) utstring_reserve((dst),((dst)->n)*2); \ + memcpy(&(dst)->d[(dst)->i], (src), (len)); \ + (dst)->i+=(len); \ + (dst)->d[(dst)->i]='\0'; \ +} while(0) + +#define utstring_append_c(dst, c) \ +do { \ + if ((dst)->n-(dst)->i < 2) utstring_reserve((dst),((dst)->n)*2); \ + (dst)->d[(dst)->i++] = (c); \ + (dst)->d[(dst)->i]='\0'; \ +} while(0) + +/******************************************************************************* + * begin substring search functions * + ******************************************************************************/ +/* Build KMP table from left to right. */ +_UNUSED_ static void _utstring_BuildTable( + const char *P_Needle, + ssize_t P_NeedleLen, + long *P_KMP_Table) +{ + long i, j; + + i = 0; + j = i - 1; + P_KMP_Table[i] = j; + while (i < P_NeedleLen) + { + while ( (j > -1) && (P_Needle[i] != P_Needle[j]) ) + { + j = P_KMP_Table[j]; + } + i++; + j++; + if (i < P_NeedleLen) + { + if (P_Needle[i] == P_Needle[j]) + { + P_KMP_Table[i] = P_KMP_Table[j]; + } + else + { + P_KMP_Table[i] = j; + } + } + else + { + P_KMP_Table[i] = j; + } + } + + return; +} + + +/* Build KMP table from right to left. */ +_UNUSED_ static void _utstring_BuildTableR( + const char *P_Needle, + ssize_t P_NeedleLen, + long *P_KMP_Table) +{ + long i, j; + + i = P_NeedleLen - 1; + j = i + 1; + P_KMP_Table[i + 1] = j; + while (i >= 0) + { + while ( (j < P_NeedleLen) && (P_Needle[i] != P_Needle[j]) ) + { + j = P_KMP_Table[j + 1]; + } + i--; + j--; + if (i >= 0) + { + if (P_Needle[i] == P_Needle[j]) + { + P_KMP_Table[i + 1] = P_KMP_Table[j + 1]; + } + else + { + P_KMP_Table[i + 1] = j; + } + } + else + { + P_KMP_Table[i + 1] = j; + } + } + + return; +} + + +/* Search data from left to right. ( Multiple search mode. ) */ +_UNUSED_ static long _utstring_find( + const char *P_Haystack, + size_t P_HaystackLen, + const char *P_Needle, + size_t P_NeedleLen, + long *P_KMP_Table) +{ + long i, j; + long V_FindPosition = -1; + + /* Search from left to right. */ + i = j = 0; + while ( (j < (int)P_HaystackLen) && (((P_HaystackLen - j) + i) >= P_NeedleLen) ) + { + while ( (i > -1) && (P_Needle[i] != P_Haystack[j]) ) + { + i = P_KMP_Table[i]; + } + i++; + j++; + if (i >= (int)P_NeedleLen) + { + /* Found. */ + V_FindPosition = j - i; + break; + } + } + + return V_FindPosition; +} + + +/* Search data from right to left. ( Multiple search mode. ) */ +_UNUSED_ static long _utstring_findR( + const char *P_Haystack, + size_t P_HaystackLen, + const char *P_Needle, + size_t P_NeedleLen, + long *P_KMP_Table) +{ + long i, j; + long V_FindPosition = -1; + + /* Search from right to left. */ + j = (P_HaystackLen - 1); + i = (P_NeedleLen - 1); + while ( (j >= 0) && (j >= i) ) + { + while ( (i < (int)P_NeedleLen) && (P_Needle[i] != P_Haystack[j]) ) + { + i = P_KMP_Table[i + 1]; + } + i--; + j--; + if (i < 0) + { + /* Found. */ + V_FindPosition = j + 1; + break; + } + } + + return V_FindPosition; +} + + +/* Search data from left to right. ( One time search mode. ) */ +_UNUSED_ static long utstring_find( + UT_string *s, + long P_StartPosition, /* Start from 0. -1 means last position. */ + const char *P_Needle, + ssize_t P_NeedleLen) +{ + long V_StartPosition; + long V_HaystackLen; + long *V_KMP_Table; + long V_FindPosition = -1; + + if (P_StartPosition < 0) + { + V_StartPosition = s->i + P_StartPosition; + } + else + { + V_StartPosition = P_StartPosition; + } + V_HaystackLen = s->i - V_StartPosition; + if ( (V_HaystackLen >= P_NeedleLen) && (P_NeedleLen > 0) ) + { + V_KMP_Table = (long *)malloc(sizeof(long) * (P_NeedleLen + 1)); + if (V_KMP_Table != NULL) + { + _utstring_BuildTable(P_Needle, P_NeedleLen, V_KMP_Table); + + V_FindPosition = _utstring_find(s->d + V_StartPosition, + V_HaystackLen, + P_Needle, + P_NeedleLen, + V_KMP_Table); + if (V_FindPosition >= 0) + { + V_FindPosition += V_StartPosition; + } + + free(V_KMP_Table); + } + } + + return V_FindPosition; +} + + +/* Search data from right to left. ( One time search mode. ) */ +_UNUSED_ static long utstring_findR( + UT_string *s, + long P_StartPosition, /* Start from 0. -1 means last position. */ + const char *P_Needle, + ssize_t P_NeedleLen) +{ + long V_StartPosition; + long V_HaystackLen; + long *V_KMP_Table; + long V_FindPosition = -1; + + if (P_StartPosition < 0) + { + V_StartPosition = s->i + P_StartPosition; + } + else + { + V_StartPosition = P_StartPosition; + } + V_HaystackLen = V_StartPosition + 1; + if ( (V_HaystackLen >= P_NeedleLen) && (P_NeedleLen > 0) ) + { + V_KMP_Table = (long *)malloc(sizeof(long) * (P_NeedleLen + 1)); + if (V_KMP_Table != NULL) + { + _utstring_BuildTableR(P_Needle, P_NeedleLen, V_KMP_Table); + + V_FindPosition = _utstring_findR(s->d, + V_HaystackLen, + P_Needle, + P_NeedleLen, + V_KMP_Table); + + free(V_KMP_Table); + } + } + + return V_FindPosition; +} +/******************************************************************************* + * end substring search functions * + ******************************************************************************/ + +#endif /* UTSTRING_H */ diff --git a/contrib/libucl/utils/Makefile.am b/contrib/libucl/utils/Makefile.am new file mode 100644 index 0000000..ec85aaa --- /dev/null +++ b/contrib/libucl/utils/Makefile.am @@ -0,0 +1,23 @@ +common_utils_cflags = -I$(top_srcdir)/include \ + -I$(top_srcdir)/src \ + -I$(top_srcdir)/uthash +common_utils_ldadd = $(top_builddir)/src/libucl.la + +ucl_chargen_SOURCES = chargen.c +ucl_chargen_LDADD = $(common_utils_ldadd) +ucl_chargen_CFLAGS = $(common_utils_cflags) + +ucl_objdump_SOURCES = objdump.c +ucl_objdump_LDADD = $(common_utils_ldadd) +ucl_objdump_CFLAGS = $(common_utils_cflags) + +ucl_tool_SOURCES = ucl-tool.c +ucl_tool_LDADD = $(common_utils_ldadd) +ucl_tool_CFLAGS = $(common_utils_cflags) + +if UTILS +UTL = ucl_chargen ucl_objdump ucl_tool +else +UTL = +endif +bin_PROGRAMS = $(UTL) diff --git a/contrib/libucl/utils/chargen.c b/contrib/libucl/utils/chargen.c new file mode 100644 index 0000000..3981340 --- /dev/null +++ b/contrib/libucl/utils/chargen.c @@ -0,0 +1,128 @@ +/* Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +/** + * @file this utility generates character table for ucl + */ + +#include <stdio.h> +#include <ctype.h> +#include <stdbool.h> + +static inline int +print_flag (const char *flag, bool *need_or, char *val) +{ + int res; + res = sprintf (val, "%s%s", *need_or ? "|" : "", flag); + + *need_or |= true; + + return res; +} + +int +main (int argc, char **argv) +{ + int i, col, r; + const char *name = "ucl_chartable"; + bool need_or; + char valbuf[2048]; + + col = 0; + + if (argc > 1) { + name = argv[1]; + } + + printf ("static const unsigned int %s[256] = {\n", name); + + for (i = 0; i < 256; i ++) { + need_or = false; + r = 0; + /* UCL_CHARACTER_VALUE_END */ + + if (i == ' ' || i == '\t') { + r += print_flag ("UCL_CHARACTER_WHITESPACE", &need_or, valbuf + r); + } + if (isspace (i)) { + r += print_flag ("UCL_CHARACTER_WHITESPACE_UNSAFE", &need_or, valbuf + r); + } + if (isalnum (i) || i >= 0x80 || i == '/' || i == '_') { + r += print_flag ("UCL_CHARACTER_KEY_START", &need_or, valbuf + r); + } + if (isalnum (i) || i == '-' || i == '_' || i == '/' || i == '.' || i >= 0x80) { + r += print_flag ("UCL_CHARACTER_KEY", &need_or, valbuf + r); + } + if (i == 0 || i == '\r' || i == '\n' || i == ']' || i == '}' || i == ';' || i == ',' || i == '#') { + r += print_flag ("UCL_CHARACTER_VALUE_END", &need_or, valbuf + r); + } + else { + if (isprint (i) || i >= 0x80) { + r += print_flag ("UCL_CHARACTER_VALUE_STR", &need_or, valbuf + r); + } + if (isdigit (i) || i == '-') { + r += print_flag ("UCL_CHARACTER_VALUE_DIGIT_START", &need_or, valbuf + r); + } + if (isalnum (i) || i == '.' || i == '-' || i == '+') { + r += print_flag ("UCL_CHARACTER_VALUE_DIGIT", &need_or, valbuf + r); + } + } + if (i == '"' || i == '\\' || i == '/' || i == 'b' || + i == 'f' || i == 'n' || i == 'r' || i == 't' || i == 'u') { + r += print_flag ("UCL_CHARACTER_ESCAPE", &need_or, valbuf + r); + } + if (i == ' ' || i == '\t' || i == ':' || i == '=') { + r += print_flag ("UCL_CHARACTER_KEY_SEP", &need_or, valbuf + r); + } + if (i == '\n' || i == '\r' || i == '\\' || i == '\b' || i == '\t' || + i == '"' || i == '\f') { + r += print_flag ("UCL_CHARACTER_JSON_UNSAFE", &need_or, valbuf + r); + } + if (i == '\n' || i == '\r' || i == '\\' || i == '\b' || i == '\t' || + i == '"' || i == '\f' || i == '=' || i == ':' || i == '{' || i == '[' || i == ' ') { + r += print_flag ("UCL_CHARACTER_UCL_UNSAFE", &need_or, valbuf + r); + } + + if (!need_or) { + r += print_flag ("UCL_CHARACTER_DENIED", &need_or, valbuf + r); + } + + if (isprint (i)) { + r += sprintf (valbuf + r, " /* %c */", i); + } + if (i != 255) { + r += sprintf (valbuf + r, ", "); + } + col += r; + if (col > 80) { + printf ("\n%s", valbuf); + col = r; + } + else { + printf ("%s", valbuf); + } + } + printf ("\n}\n"); + + return 0; +} diff --git a/contrib/libucl/utils/objdump.c b/contrib/libucl/utils/objdump.c new file mode 100644 index 0000000..6fde2f4 --- /dev/null +++ b/contrib/libucl/utils/objdump.c @@ -0,0 +1,182 @@ +/* Copyright (c) 2013, Dmitriy V. Reshetnikov + * Copyright (c) 2013, Vsevolod Stakhov + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <stdio.h> +#include <errno.h> + +#include "ucl.h" + +void +ucl_obj_dump (const ucl_object_t *obj, unsigned int shift) +{ + int num = shift * 4 + 5; + char *pre = (char *) malloc (num * sizeof(char)); + const ucl_object_t *cur, *tmp; + ucl_object_iter_t it = NULL, it_obj = NULL; + + pre[--num] = 0x00; + while (num--) + pre[num] = 0x20; + + tmp = obj; + + while ((obj = ucl_iterate_object (tmp, &it, false))) { + printf ("%sucl object address: %p\n", pre + 4, obj); + if (obj->key != NULL) { + printf ("%skey: \"%s\"\n", pre, ucl_object_key (obj)); + } + printf ("%sref: %u\n", pre, obj->ref); + printf ("%slen: %u\n", pre, obj->len); + printf ("%sprev: %p\n", pre, obj->prev); + printf ("%snext: %p\n", pre, obj->next); + if (obj->type == UCL_OBJECT) { + printf ("%stype: UCL_OBJECT\n", pre); + printf ("%svalue: %p\n", pre, obj->value.ov); + it_obj = NULL; + while ((cur = ucl_iterate_object (obj, &it_obj, true))) { + ucl_obj_dump (cur, shift + 2); + } + } + else if (obj->type == UCL_ARRAY) { + printf ("%stype: UCL_ARRAY\n", pre); + printf ("%svalue: %p\n", pre, obj->value.av); + it_obj = NULL; + while ((cur = ucl_iterate_object (obj, &it_obj, true))) { + ucl_obj_dump (cur, shift + 2); + } + } + else if (obj->type == UCL_INT) { + printf ("%stype: UCL_INT\n", pre); + printf ("%svalue: %jd\n", pre, (intmax_t)ucl_object_toint (obj)); + } + else if (obj->type == UCL_FLOAT) { + printf ("%stype: UCL_FLOAT\n", pre); + printf ("%svalue: %f\n", pre, ucl_object_todouble (obj)); + } + else if (obj->type == UCL_STRING) { + printf ("%stype: UCL_STRING\n", pre); + printf ("%svalue: \"%s\"\n", pre, ucl_object_tostring (obj)); + } + else if (obj->type == UCL_BOOLEAN) { + printf ("%stype: UCL_BOOLEAN\n", pre); + printf ("%svalue: %s\n", pre, ucl_object_tostring_forced (obj)); + } + else if (obj->type == UCL_TIME) { + printf ("%stype: UCL_TIME\n", pre); + printf ("%svalue: %f\n", pre, ucl_object_todouble (obj)); + } + else if (obj->type == UCL_USERDATA) { + printf ("%stype: UCL_USERDATA\n", pre); + printf ("%svalue: %p\n", pre, obj->value.ud); + } + } + + free (pre); +} + +int +main(int argc, char **argv) +{ + const char *fn = NULL; + unsigned char *inbuf; + struct ucl_parser *parser; + int k, ret = 0, r = 0; + ssize_t bufsize; + ucl_object_t *obj = NULL; + const ucl_object_t *par; + FILE *in; + + if (argc > 1) { + fn = argv[1]; + } + + if (fn != NULL) { + in = fopen (fn, "r"); + if (in == NULL) { + exit (-errno); + } + } + else { + in = stdin; + } + + parser = ucl_parser_new (0); + inbuf = malloc (BUFSIZ); + bufsize = BUFSIZ; + r = 0; + + while (!feof (in) && !ferror (in)) { + if (r == bufsize) { + inbuf = realloc (inbuf, bufsize * 2); + bufsize *= 2; + if (inbuf == NULL) { + perror ("realloc"); + exit (EXIT_FAILURE); + } + } + r += fread (inbuf + r, 1, bufsize - r, in); + } + + if (ferror (in)) { + fprintf (stderr, "Failed to read the input file.\n"); + exit (EXIT_FAILURE); + } + + ucl_parser_add_chunk (parser, inbuf, r); + fclose (in); + if (ucl_parser_get_error(parser)) { + printf ("Error occured: %s\n", ucl_parser_get_error(parser)); + ret = 1; + goto end; + } + + obj = ucl_parser_get_object (parser); + if (ucl_parser_get_error (parser)) { + printf ("Error occured: %s\n", ucl_parser_get_error(parser)); + ret = 1; + goto end; + } + + if (argc > 2) { + for (k = 2; k < argc; k++) { + printf ("search for \"%s\"... ", argv[k]); + par = ucl_object_find_key (obj, argv[k]); + printf ("%sfound\n", (par == NULL )?"not ":""); + ucl_obj_dump (par, 0); + } + } + else { + ucl_obj_dump (obj, 0); + } + +end: + if (parser != NULL) { + ucl_parser_free (parser); + } + if (obj != NULL) { + ucl_object_unref (obj); + } + + return ret; +} diff --git a/contrib/libucl/utils/ucl-tool.c b/contrib/libucl/utils/ucl-tool.c new file mode 100644 index 0000000..feea9c2 --- /dev/null +++ b/contrib/libucl/utils/ucl-tool.c @@ -0,0 +1,168 @@ +/* Copyright (c) 2015, Cesanta Software + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <stdio.h> +#include <getopt.h> +#include <stdlib.h> + +#include "ucl.h" + +static struct option opts[] = { + {"help", no_argument, NULL, 'h'}, + {"in", required_argument, NULL, 'i' }, + {"out", required_argument, NULL, 'o' }, + {"schema", required_argument, NULL, 's'}, + {"format", required_argument, NULL, 'f'}, + {0, 0, 0, 0} +}; + +void usage(const char *name, FILE *out) { + fprintf(out, "Usage: %s [--help] [-i|--in file] [-o|--out file]\n", name); + fprintf(out, " [-s|--schema file] [-f|--format format]\n\n"); + fprintf(out, " --help - print this message and exit\n"); + fprintf(out, " --in - specify input filename " + "(default: standard input)\n"); + fprintf(out, " --out - specify output filename " + "(default: standard output)\n"); + fprintf(out, " --schema - specify schema file for validation\n"); + fprintf(out, " --format - output format. Options: ucl (default), " + "json, compact_json, yaml, msgpack\n"); +} + +int main(int argc, char **argv) { + char ch; + FILE *in = stdin, *out = stdout; + const char *schema = NULL; + unsigned char *buf = NULL; + size_t size = 0, r = 0; + struct ucl_parser *parser = NULL; + ucl_object_t *obj = NULL; + ucl_emitter_t emitter = UCL_EMIT_CONFIG; + + while((ch = getopt_long(argc, argv, "hi:o:s:f:", opts, NULL)) != -1) { + switch (ch) { + case 'i': + in = fopen(optarg, "r"); + if (in == NULL) { + perror("fopen on input file"); + exit(EXIT_FAILURE); + } + break; + case 'o': + out = fopen(optarg, "w"); + if (out == NULL) { + perror("fopen on output file"); + exit(EXIT_FAILURE); + } + break; + case 's': + schema = optarg; + break; + case 'f': + if (strcmp(optarg, "ucl") == 0) { + emitter = UCL_EMIT_CONFIG; + } else if (strcmp(optarg, "json") == 0) { + emitter = UCL_EMIT_JSON; + } else if (strcmp(optarg, "yaml") == 0) { + emitter = UCL_EMIT_YAML; + } else if (strcmp(optarg, "compact_json") == 0) { + emitter = UCL_EMIT_JSON_COMPACT; + } else if (strcmp(optarg, "msgpack") == 0) { + emitter = UCL_EMIT_MSGPACK; + } else { + fprintf(stderr, "Unknown output format: %s\n", optarg); + exit(EXIT_FAILURE); + } + break; + case 'h': + usage(argv[0], stdout); + exit(0); + default: + usage(argv[0], stderr); + exit(EXIT_FAILURE); + break; + } + } + + parser = ucl_parser_new(0); + buf = malloc(BUFSIZ); + size = BUFSIZ; + while(!feof(in) && !ferror(in)) { + if (r == size) { + buf = realloc(buf, size*2); + size *= 2; + if (buf == NULL) { + perror("realloc"); + exit(EXIT_FAILURE); + } + } + r += fread(buf + r, 1, size - r, in); + } + if (ferror(in)) { + fprintf(stderr, "Failed to read the input file.\n"); + exit(EXIT_FAILURE); + } + fclose(in); + if (!ucl_parser_add_chunk(parser, buf, r)) { + fprintf(stderr, "Failed to parse input file: %s\n", + ucl_parser_get_error(parser)); + exit(EXIT_FAILURE); + } + if ((obj = ucl_parser_get_object(parser)) == NULL) { + fprintf(stderr, "Failed to get root object: %s\n", + ucl_parser_get_error(parser)); + exit(EXIT_FAILURE); + } + if (schema != NULL) { + struct ucl_parser *schema_parser = ucl_parser_new(0); + ucl_object_t *schema_obj = NULL; + struct ucl_schema_error error; + + if (!ucl_parser_add_file(schema_parser, schema)) { + fprintf(stderr, "Failed to parse schema file: %s\n", + ucl_parser_get_error(schema_parser)); + exit(EXIT_FAILURE); + } + if ((schema_obj = ucl_parser_get_object(schema_parser)) == NULL) { + fprintf(stderr, "Failed to get root object: %s\n", + ucl_parser_get_error(schema_parser)); + exit(EXIT_FAILURE); + } + if (!ucl_object_validate(schema_obj, obj, &error)) { + fprintf(stderr, "Validation failed: %s\n", error.msg); + exit(EXIT_FAILURE); + } + } + + if (emitter != UCL_EMIT_MSGPACK) { + fprintf(out, "%s\n", ucl_object_emit(obj, emitter)); + } + else { + size_t len; + unsigned char *res; + + res = ucl_object_emit_len(obj, emitter, &len); + fwrite(res, 1, len, out); + } + + return 0; +} |
