diff options
author | obrien <obrien@FreeBSD.org> | 2009-05-02 08:22:49 +0000 |
---|---|---|
committer | obrien <obrien@FreeBSD.org> | 2009-05-02 08:22:49 +0000 |
commit | 3d89d1fbe920995209cda59ff0fd5096b6471133 (patch) | |
tree | 7c9b7aee2424612b6dc946ca927b316448f0e54a | |
parent | ad11b985bff1a9b13ab1ee82cf1061fefce20569 (diff) | |
download | FreeBSD-src-3d89d1fbe920995209cda59ff0fd5096b6471133.zip FreeBSD-src-3d89d1fbe920995209cda59ff0fd5096b6471133.tar.gz |
Virgin import of Christos Zoulas's FILE 5.00.
-rw-r--r-- | Magdir/wireless | 5 | ||||
-rw-r--r-- | Makefile.am-src | 18 | ||||
-rw-r--r-- | cdf.c | 1105 | ||||
-rw-r--r-- | cdf.h | 298 | ||||
-rw-r--r-- | cdf_time.c | 182 | ||||
-rw-r--r-- | encoding.c | 484 | ||||
-rw-r--r-- | readcdf.c | 256 |
7 files changed, 2348 insertions, 0 deletions
diff --git a/Magdir/wireless b/Magdir/wireless new file mode 100644 index 0000000..aaae5a5 --- /dev/null +++ b/Magdir/wireless @@ -0,0 +1,5 @@ +#------------------------------------------------------------------------------ +# wireless-regdb: file(1) magic for CRDA wireless-regdb file format +# +0 string RGDB CRDA wireless regulatory database file +>4 belong 19 (Version 1) diff --git a/Makefile.am-src b/Makefile.am-src new file mode 100644 index 0000000..db78d96 --- /dev/null +++ b/Makefile.am-src @@ -0,0 +1,18 @@ +MAGIC = $(pkgdatadir)/magic +lib_LTLIBRARIES = libmagic.la +include_HEADERS = magic.h + +bin_PROGRAMS = file + +AM_CPPFLAGS = -DMAGIC='"$(MAGIC)"' +AM_CFLAGS = @WARNINGS@ + +libmagic_la_SOURCES = magic.c apprentice.c softmagic.c ascmagic.c \ + encoding.c compress.c is_tar.c readelf.c print.c fsmagic.c \ + funcs.c file.h names.h patchlevel.h readelf.h tar.h apptype.c \ + file_opts.h elfclass.h mygetopt.h cdf.c cdf_time.c readcdf.c cdf.h +libmagic_la_LDFLAGS = -no-undefined -version-info 1:0:0 +libmagic_la_LIBADD = $(LTLIBOBJS) + +file_SOURCES = file.c +file_LDADD = libmagic.la @@ -0,0 +1,1105 @@ +/*- + * Copyright (c) 2008 Christos Zoulas + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS + * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED + * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS + * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ +/* + * Parse composite document files, the format used in Microsoft Office + * document files before they switched to zipped xml. + * Info from: http://sc.openoffice.org/compdocfileformat.pdf + */ + +#include "file.h" + +#ifndef lint +FILE_RCSID("@(#)$File: cdf.c,v 1.17 2009/02/03 20:27:51 christos Exp $") +#endif + +#include <assert.h> +#ifdef CDF_DEBUG +#include <err.h> +#endif +#include <stdlib.h> +#include <unistd.h> +#include <string.h> +#include <time.h> +#include <ctype.h> + +#ifndef EFTYPE +#define EFTYPE EINVAL +#endif + +#include "cdf.h" + +#ifndef __arraycount +#define __arraycount(a) (sizeof(a) / sizeof(a[0])) +#endif + +#ifdef CDF_DEBUG +#define DPRINTF(a) printf a +#else +#define DPRINTF(a) +#endif + +static union { + char s[4]; + uint32_t u; +} cdf_bo; + +#define NEED_SWAP (cdf_bo.u == (uint32_t)0x01020304) + +#define CDF_TOLE8(x) (NEED_SWAP ? cdf_tole8(x) : (uint64_t)(x)) +#define CDF_TOLE4(x) (NEED_SWAP ? cdf_tole4(x) : (uint32_t)(x)) +#define CDF_TOLE2(x) (NEED_SWAP ? cdf_tole2(x) : (uint16_t)(x)) + +/* + * swap a short + */ +uint16_t +cdf_tole2(uint16_t sv) +{ + uint16_t rv; + uint8_t *s = (uint8_t *)(void *)&sv; + uint8_t *d = (uint8_t *)(void *)&rv; + d[0] = s[1]; + d[1] = s[0]; + return rv; +} + +/* + * swap an int + */ +uint32_t +cdf_tole4(uint32_t sv) +{ + uint32_t rv; + uint8_t *s = (uint8_t *)(void *)&sv; + uint8_t *d = (uint8_t *)(void *)&rv; + d[0] = s[3]; + d[1] = s[2]; + d[2] = s[1]; + d[3] = s[0]; + return rv; +} + +/* + * swap a quad + */ +uint64_t +cdf_tole8(uint64_t sv) +{ + uint64_t rv; + uint8_t *s = (uint8_t *)(void *)&sv; + uint8_t *d = (uint8_t *)(void *)&rv; + d[0] = s[7]; + d[1] = s[6]; + d[2] = s[5]; + d[3] = s[4]; + d[4] = s[3]; + d[5] = s[2]; + d[6] = s[1]; + d[7] = s[0]; + return rv; +} + +#define CDF_UNPACK(a) \ + (void)memcpy(&(a), &buf[len], sizeof(a)), len += sizeof(a) +#define CDF_UNPACKA(a) \ + (void)memcpy((a), &buf[len], sizeof(a)), len += sizeof(a) + +void +cdf_swap_header(cdf_header_t *h) +{ + size_t i; + + h->h_magic = CDF_TOLE8(h->h_magic); + h->h_uuid[0] = CDF_TOLE8(h->h_uuid[0]); + h->h_uuid[1] = CDF_TOLE8(h->h_uuid[1]); + h->h_revision = CDF_TOLE2(h->h_revision); + h->h_version = CDF_TOLE2(h->h_version); + h->h_byte_order = CDF_TOLE2(h->h_byte_order); + h->h_sec_size_p2 = CDF_TOLE2(h->h_sec_size_p2); + h->h_short_sec_size_p2 = CDF_TOLE2(h->h_short_sec_size_p2); + h->h_num_sectors_in_sat = CDF_TOLE4(h->h_num_sectors_in_sat); + h->h_secid_first_directory = CDF_TOLE4(h->h_secid_first_directory); + h->h_min_size_standard_stream = + CDF_TOLE4(h->h_min_size_standard_stream); + h->h_secid_first_sector_in_short_sat = + CDF_TOLE4(h->h_secid_first_sector_in_short_sat); + h->h_num_sectors_in_short_sat = + CDF_TOLE4(h->h_num_sectors_in_short_sat); + h->h_secid_first_sector_in_master_sat = + CDF_TOLE4(h->h_secid_first_sector_in_master_sat); + h->h_num_sectors_in_master_sat = + CDF_TOLE4(h->h_num_sectors_in_master_sat); + for (i = 0; i < __arraycount(h->h_master_sat); i++) + h->h_master_sat[i] = CDF_TOLE4(h->h_master_sat[i]); +} + +void +cdf_unpack_header(cdf_header_t *h, char *buf) +{ + size_t i; + size_t len = 0; + + CDF_UNPACK(h->h_magic); + CDF_UNPACKA(h->h_uuid); + CDF_UNPACK(h->h_revision); + CDF_UNPACK(h->h_version); + CDF_UNPACK(h->h_byte_order); + CDF_UNPACK(h->h_sec_size_p2); + CDF_UNPACK(h->h_short_sec_size_p2); + CDF_UNPACKA(h->h_unused0); + CDF_UNPACK(h->h_num_sectors_in_sat); + CDF_UNPACK(h->h_secid_first_directory); + CDF_UNPACKA(h->h_unused1); + CDF_UNPACK(h->h_min_size_standard_stream); + CDF_UNPACK(h->h_secid_first_sector_in_short_sat); + CDF_UNPACK(h->h_num_sectors_in_short_sat); + CDF_UNPACK(h->h_secid_first_sector_in_master_sat); + CDF_UNPACK(h->h_num_sectors_in_master_sat); + for (i = 0; i < __arraycount(h->h_master_sat); i++) + CDF_UNPACK(h->h_master_sat[i]); +} + +void +cdf_swap_dir(cdf_directory_t *d) +{ + d->d_namelen = CDF_TOLE2(d->d_namelen); + d->d_left_child = CDF_TOLE4(d->d_left_child); + d->d_right_child = CDF_TOLE4(d->d_right_child); + d->d_storage = CDF_TOLE4(d->d_storage); + d->d_storage_uuid[0] = CDF_TOLE8(d->d_storage_uuid[0]); + d->d_storage_uuid[1] = CDF_TOLE8(d->d_storage_uuid[1]); + d->d_flags = CDF_TOLE4(d->d_flags); + d->d_created = CDF_TOLE8(d->d_created); + d->d_modified = CDF_TOLE8(d->d_modified); + d->d_stream_first_sector = CDF_TOLE4(d->d_stream_first_sector); + d->d_size = CDF_TOLE4(d->d_size); +} + +void +cdf_swap_class(cdf_classid_t *d) +{ + d->cl_dword = CDF_TOLE4(d->cl_dword); + d->cl_word[0] = CDF_TOLE2(d->cl_word[0]); + d->cl_word[1] = CDF_TOLE2(d->cl_word[1]); +} + +void +cdf_unpack_dir(cdf_directory_t *d, char *buf) +{ + size_t len = 0; + + CDF_UNPACKA(d->d_name); + CDF_UNPACK(d->d_namelen); + CDF_UNPACK(d->d_type); + CDF_UNPACK(d->d_color); + CDF_UNPACK(d->d_left_child); + CDF_UNPACK(d->d_right_child); + CDF_UNPACK(d->d_storage); + CDF_UNPACKA(d->d_storage_uuid); + CDF_UNPACK(d->d_flags); + CDF_UNPACK(d->d_created); + CDF_UNPACK(d->d_modified); + CDF_UNPACK(d->d_stream_first_sector); + CDF_UNPACK(d->d_size); + CDF_UNPACK(d->d_unused0); +} + +int +cdf_read_header(int fd, cdf_header_t *h) +{ + (void)memcpy(cdf_bo.s, "\01\02\03\04", 4); + char buf[512]; + if (lseek(fd, (off_t)0, SEEK_SET) == (off_t)-1) + return -1; + if (read(fd, buf, sizeof(buf)) != sizeof(buf)) + return -1; + cdf_unpack_header(h, buf); + cdf_swap_header(h); + if (h->h_magic != CDF_MAGIC) { + DPRINTF(("Bad magic 0x%x != 0x$x\n", h->h_magic, CDF_MAGIC)); + errno = EFTYPE; + return -1; + } + return 0; +} + + +ssize_t +cdf_read_sector(int fd, void *buf, size_t offs, size_t len, + const cdf_header_t *h, cdf_secid_t id) +{ + assert((size_t)CDF_SEC_SIZE(h) == len); + if (lseek(fd, (off_t)CDF_SEC_POS(h, id), SEEK_SET) == (off_t)-1) + return -1; + return read(fd, ((char *)buf) + offs, len); +} + +ssize_t +cdf_read_short_sector(const cdf_stream_t *sst, void *buf, size_t offs, + size_t len, const cdf_header_t *h, cdf_secid_t id) +{ + assert((size_t)CDF_SHORT_SEC_SIZE(h) == len); + (void)memcpy(((char *)buf) + offs, + ((const char *)sst->sst_tab) + CDF_SHORT_SEC_POS(h, id), len); + return len; +} + +/* + * Read the sector allocation table. + */ +int +cdf_read_sat(int fd, cdf_header_t *h, cdf_sat_t *sat) +{ + size_t i, j, k; + size_t ss = CDF_SEC_SIZE(h); + cdf_secid_t *msa, mid; + + for (i = 0; i < __arraycount(h->h_master_sat); i++) + if (h->h_master_sat[i] == CDF_SECID_FREE) + break; + + sat->sat_len = (h->h_num_sectors_in_master_sat + i); + if ((sat->sat_tab = calloc(sat->sat_len, ss)) == NULL) + return -1; + + for (i = 0; i < __arraycount(h->h_master_sat); i++) { + if (h->h_master_sat[i] < 0) + break; + if (cdf_read_sector(fd, sat->sat_tab, ss * i, ss, h, + h->h_master_sat[i]) != (ssize_t)ss) { + DPRINTF(("Reading sector %d", h->h_master_sat[i])); + goto out1; + } + } + + if ((msa = calloc(1, ss)) == NULL) + goto out1; + + mid = h->h_secid_first_sector_in_master_sat; + for (j = 0; j < h->h_num_sectors_in_master_sat; j++) { + if (j >= CDF_LOOP_LIMIT) { + DPRINTF(("Reading master sector loop limit")); + errno = EFTYPE; + goto out2; + } + if (cdf_read_sector(fd, msa, 0, ss, h, mid) != (ssize_t)ss) { + DPRINTF(("Reading master sector %d", mid)); + goto out2; + } + for (k = 0; k < (ss / sizeof(mid)) - 1; k++, i++) + if (cdf_read_sector(fd, sat->sat_tab, ss * i, ss, h, + CDF_TOLE4(msa[k])) != (ssize_t)ss) { + DPRINTF(("Reading sector %d", + CDF_TOLE4(msa[k]))); + goto out2; + } + mid = CDF_TOLE4(msa[(ss / sizeof(mid)) - 1]); + } + free(msa); + return 0; +out2: + free(msa); +out1: + free(sat->sat_tab); + return -1; +} + +size_t +cdf_count_chain(const cdf_header_t *h, const cdf_sat_t *sat, + cdf_secid_t sid) +{ + size_t i, j, s = CDF_SEC_SIZE(h) / sizeof(cdf_secid_t); + cdf_secid_t maxsector = (cdf_secid_t)(sat->sat_len * s); + + DPRINTF(("Chain:")); + for (j = i = 0; sid >= 0; i++, j++) { + DPRINTF((" %d", sid)); + if (j >= CDF_LOOP_LIMIT) { + DPRINTF(("Counting chain loop limit")); + errno = EFTYPE; + return (size_t)-1; + } + if (sid > maxsector) { + DPRINTF(("Sector %d > %d\n", sid, maxsector)); + errno = EFTYPE; + return (size_t)-1; + } + sid = CDF_TOLE4(sat->sat_tab[sid]); + } + DPRINTF(("\n")); + return i; +} + +int +cdf_read_long_sector_chain(int fd, const cdf_header_t *h, const cdf_sat_t *sat, + cdf_secid_t sid, size_t len, cdf_stream_t *scn) +{ + size_t ss = CDF_SEC_SIZE(h), i, j; + ssize_t nr; + scn->sst_len = cdf_count_chain(h, sat, sid); + scn->sst_dirlen = len; + + if (scn->sst_len == (size_t)-1) + return -1; + + scn->sst_tab = calloc(scn->sst_len, ss); + if (scn->sst_tab == NULL) + return -1; + + for (j = i = 0; sid >= 0; i++, j++) { + if ((nr = cdf_read_sector(fd, scn->sst_tab, i * ss, ss, h, + sid)) != (ssize_t)ss) { + if (i == scn->sst_len - 1 && nr > 0) { + /* Last sector might be truncated */ + return 0; + } + DPRINTF(("Reading long sector chain %d", sid)); + goto out; + } + sid = CDF_TOLE4(sat->sat_tab[sid]); + if (j >= CDF_LOOP_LIMIT) { + DPRINTF(("Read long sector chain loop limit")); + errno = EFTYPE; + goto out; + } + } + return 0; +out: + free(scn->sst_tab); + return (size_t)-1; +} + +int +cdf_read_short_sector_chain(const cdf_header_t *h, + const cdf_sat_t *ssat, const cdf_stream_t *sst, + cdf_secid_t sid, size_t len, cdf_stream_t *scn) +{ + size_t ss = CDF_SHORT_SEC_SIZE(h), i, j; + scn->sst_len = cdf_count_chain(h, ssat, sid); + scn->sst_dirlen = len; + + if (scn->sst_len == (size_t)-1) + return -1; + + scn->sst_tab = calloc(scn->sst_len, ss); + if (scn->sst_tab == NULL) + return -1; + + for (j = i = 0; sid >= 0; i++, j++) { + if (j >= CDF_LOOP_LIMIT) { + DPRINTF(("Read short sector chain loop limit")); + errno = EFTYPE; + goto out; + } + if (cdf_read_short_sector(sst, scn->sst_tab, i * ss, ss, h, + sid) != (ssize_t)ss) { + DPRINTF(("Reading short sector chain %d", sid)); + goto out; + } + sid = CDF_TOLE4(ssat->sat_tab[sid]); + } + return 0; +out: + free(scn->sst_tab); + return (size_t)-1; +} + +int +cdf_read_sector_chain(int fd, const cdf_header_t *h, const cdf_sat_t *sat, + const cdf_sat_t *ssat, const cdf_stream_t *sst, + cdf_secid_t sid, size_t len, cdf_stream_t *scn) +{ + + if (len < h->h_min_size_standard_stream) + return cdf_read_short_sector_chain(h, ssat, sst, sid, len, + scn); + else + return cdf_read_long_sector_chain(fd, h, sat, sid, len, scn); +} + +int +cdf_read_dir(int fd, const cdf_header_t *h, const cdf_sat_t *sat, + cdf_dir_t *dir) +{ + size_t i, j; + size_t ss = CDF_SEC_SIZE(h), ns, nd; + char *buf; + cdf_secid_t sid = h->h_secid_first_directory; + + ns = cdf_count_chain(h, sat, sid); + if (ns == (size_t)-1) + return -1; + + nd = ss / CDF_DIRECTORY_SIZE; + + dir->dir_len = ns * nd; + dir->dir_tab = calloc(dir->dir_len, sizeof(dir->dir_tab[0])); + if (dir->dir_tab == NULL) + return -1; + + if ((buf = malloc(ss)) == NULL) { + free(dir->dir_tab); + return -1; + } + + for (j = i = 0; i < ns; i++, j++) { + if (j >= CDF_LOOP_LIMIT) { + DPRINTF(("Read dir loop limit")); + errno = EFTYPE; + goto out; + } + if (cdf_read_sector(fd, buf, 0, ss, h, sid) != (ssize_t)ss) { + DPRINTF(("Reading directory sector %d", sid)); + goto out; + } + for (j = 0; j < nd; j++) { + cdf_unpack_dir(&dir->dir_tab[i * nd + j], + &buf[j * CDF_DIRECTORY_SIZE]); + } + sid = CDF_TOLE4(sat->sat_tab[sid]); + } + if (NEED_SWAP) + for (i = 0; i < dir->dir_len; i++) + cdf_swap_dir(&dir->dir_tab[i]); + free(buf); + return 0; +out: + free(dir->dir_tab); + free(buf); + return -1; +} + + +int +cdf_read_ssat(int fd, const cdf_header_t *h, const cdf_sat_t *sat, + cdf_sat_t *ssat) +{ + size_t i, j; + size_t ss = CDF_SEC_SIZE(h); + cdf_secid_t sid = h->h_secid_first_sector_in_short_sat; + + ssat->sat_len = cdf_count_chain(h, sat, sid); + if (ssat->sat_len == (size_t)-1) + return -1; + + ssat->sat_tab = calloc(ssat->sat_len, ss); + if (ssat->sat_tab == NULL) + return -1; + + for (j = i = 0; sid >= 0; i++, j++) { + if (j >= CDF_LOOP_LIMIT) { + DPRINTF(("Read short sat sector loop limit")); + errno = EFTYPE; + goto out; + } + if (cdf_read_sector(fd, ssat->sat_tab, i * ss, ss, h, sid) != + (ssize_t)ss) { + DPRINTF(("Reading short sat sector %d", sid)); + goto out; + } + sid = CDF_TOLE4(sat->sat_tab[sid]); + } + return 0; +out: + free(ssat->sat_tab); + return -1; +} + +int +cdf_read_short_stream(int fd, const cdf_header_t *h, const cdf_sat_t *sat, + const cdf_dir_t *dir, cdf_stream_t *scn) +{ + size_t i; + const cdf_directory_t *d; + + for (i = 0; i < dir->dir_len; i++) + if (dir->dir_tab[i].d_type == CDF_DIR_TYPE_ROOT_STORAGE) + break; + + if (i == dir->dir_len) { + DPRINTF(("Cannot find root storage node\n")); + errno = EFTYPE; + return -1; + } + d = &dir->dir_tab[i]; + + /* If the it is not there, just fake it; some docs don't have it */ + if (d->d_stream_first_sector < 0) { + scn->sst_tab = NULL; + scn->sst_len = 0; + return 0; + } + + return cdf_read_long_sector_chain(fd, h, sat, + d->d_stream_first_sector, d->d_size, scn); +} + +static int +cdf_namecmp(const char *d, const uint16_t *s, size_t l) +{ + for (; l--; d++, s++) + if (*d != CDF_TOLE2(*s)) + return (unsigned char)*d - CDF_TOLE2(*s); + return 0; +} + +int +cdf_read_summary_info(int fd, const cdf_header_t *h, + const cdf_sat_t *sat, const cdf_sat_t *ssat, const cdf_stream_t *sst, + const cdf_dir_t *dir, cdf_stream_t *scn) +{ + size_t i; + const cdf_directory_t *d; + static const char name[] = "\05SummaryInformation"; + + for (i = 0; i < dir->dir_len; i++) + if (dir->dir_tab[i].d_type == CDF_DIR_TYPE_USER_STREAM && + cdf_namecmp(name, dir->dir_tab[i].d_name, sizeof(name)) + == 0) + break; + + if (i == dir->dir_len) { + DPRINTF(("Cannot find summary information section\n")); + errno = EFTYPE; + return -1; + } + d = &dir->dir_tab[i]; + return cdf_read_sector_chain(fd, h, sat, ssat, sst, + d->d_stream_first_sector, d->d_size, scn); +} + +int +cdf_read_property_info(const cdf_stream_t *sst, uint32_t offs, + cdf_property_info_t **info, size_t *count, size_t *maxcount) +{ + const cdf_section_header_t *shp; + cdf_section_header_t sh; + const uint32_t *p, *q, *e; + int16_t s16; + int32_t s32; + uint32_t u32; + int64_t s64; + uint64_t u64; + cdf_timestamp_t tp; + size_t i, o, nelements, j; + cdf_property_info_t *inp; + + shp = (const void *)((const char *)sst->sst_tab + offs); + sh.sh_len = CDF_TOLE4(shp->sh_len); + sh.sh_properties = CDF_TOLE4(shp->sh_properties); + DPRINTF(("section len: %d properties %d\n", sh.sh_len, + sh.sh_properties)); + if (*maxcount) { + *maxcount += sh.sh_properties; + inp = realloc(*info, *maxcount * sizeof(*inp)); + } else { + *maxcount = sh.sh_properties; + inp = malloc(*maxcount * sizeof(*inp)); + } + if (inp == NULL) + goto out; + *info = inp; + inp += *count; + *count += sh.sh_properties; + p = (const void *)((const char *)sst->sst_tab + offs + sizeof(sh)); + e = (const void *)(((const char *)shp) + sh.sh_len); + for (i = 0; i < sh.sh_properties; i++) { + q = (const uint32_t *)((const char *)p + + CDF_TOLE4(p[(i << 1) + 1])) - 2; + if (q > e) { + DPRINTF(("Ran of the end %p > %p\n", q, e)); + goto out; + } + inp[i].pi_id = CDF_TOLE4(p[i << 1]); + inp[i].pi_type = CDF_TOLE4(q[0]); + DPRINTF(("%d) id=%x type=%x offs=%x\n", i, inp[i].pi_id, + inp[i].pi_type, (const char *)q - (const char *)p)); + if (inp[i].pi_type & CDF_VECTOR) { + nelements = CDF_TOLE4(q[1]); + o = 2; + } else { + nelements = 1; + o = 1; + } + if (inp[i].pi_type & (CDF_ARRAY|CDF_BYREF|CDF_RESERVED)) + goto unknown; + switch (inp[i].pi_type & CDF_TYPEMASK) { + case CDF_EMPTY: + break; + case CDF_SIGNED16: + if (inp[i].pi_type & CDF_VECTOR) + goto unknown; + (void)memcpy(&s16, &q[o], sizeof(s16)); + inp[i].pi_s16 = CDF_TOLE2(s16); + break; + case CDF_SIGNED32: + if (inp[i].pi_type & CDF_VECTOR) + goto unknown; + (void)memcpy(&s32, &q[o], sizeof(s32)); + inp[i].pi_s32 = CDF_TOLE4(s32); + break; + case CDF_BOOL: + case CDF_UNSIGNED32: + if (inp[i].pi_type & CDF_VECTOR) + goto unknown; + (void)memcpy(&u32, &q[o], sizeof(u32)); + inp[i].pi_u32 = CDF_TOLE4(u32); + break; + case CDF_SIGNED64: + if (inp[i].pi_type & CDF_VECTOR) + goto unknown; + (void)memcpy(&s64, &q[o], sizeof(s64)); + inp[i].pi_s64 = CDF_TOLE4(s64); + break; + case CDF_UNSIGNED64: + if (inp[i].pi_type & CDF_VECTOR) + goto unknown; + (void)memcpy(&u64, &q[o], sizeof(u64)); + inp[i].pi_u64 = CDF_TOLE4(u64); + break; + case CDF_LENGTH32_STRING: + if (nelements > 1) { + size_t nelem = inp - *info; + *maxcount += nelements; + inp = realloc(*info, *maxcount * sizeof(*inp)); + if (inp == NULL) + goto out; + *info = inp; + inp = *info + nelem; + } + DPRINTF(("nelements = %d\n", nelements)); + for (j = 0; j < nelements; j++, i++) { + uint32_t l = CDF_TOLE4(q[o]); + inp[i].pi_str.s_len = l; + inp[i].pi_str.s_buf = (const char *)(&q[o+1]); + DPRINTF(("l = %d, r = %d, s = %s\n", l, + CDF_ROUND(l, sizeof(l)), + inp[i].pi_str.s_buf)); + l = 4 + CDF_ROUND(l, sizeof(l)); + o += l >> 2; + } + i--; + break; + case CDF_FILETIME: + if (inp[i].pi_type & CDF_VECTOR) + goto unknown; + (void)memcpy(&tp, &q[o], sizeof(tp)); + inp[i].pi_tp = CDF_TOLE8(tp); + break; + case CDF_CLIPBOARD: + if (inp[i].pi_type & CDF_VECTOR) + goto unknown; + break; + default: + unknown: + DPRINTF(("Don't know how to deal with %x\n", + inp[i].pi_type)); + goto out; + } + } + return 0; +out: + free(*info); + return -1; +} + +int +cdf_unpack_summary_info(const cdf_stream_t *sst, cdf_summary_info_header_t *ssi, + cdf_property_info_t **info, size_t *count) +{ + size_t i, maxcount; + const cdf_summary_info_header_t *si = sst->sst_tab; + const cdf_section_declaration_t *sd = (const void *) + ((const char *)sst->sst_tab + CDF_SECTION_DECLARATION_OFFSET); + + ssi->si_byte_order = CDF_TOLE2(si->si_byte_order); + ssi->si_os_version = CDF_TOLE2(si->si_os_version); + ssi->si_os = CDF_TOLE2(si->si_os); + ssi->si_class = si->si_class; + cdf_swap_class(&ssi->si_class); + ssi->si_count = CDF_TOLE2(si->si_count); + *count = 0; + maxcount = 0; + *info = NULL; + for (i = 0; i < CDF_TOLE4(si->si_count); i++) { + if (i >= CDF_LOOP_LIMIT) { + DPRINTF(("Unpack summary info loop limit")); + errno = EFTYPE; + return -1; + } + if (cdf_read_property_info(sst, CDF_TOLE4(sd->sd_offset), + info, count, &maxcount) == -1) + return -1; + } + return 0; +} + + + +int +cdf_print_classid(char *buf, size_t buflen, const cdf_classid_t *id) +{ + return snprintf(buf, buflen, "%.8x-%.4x-%.4x-%.2x%.2x-" + "%.2x%.2x%.2x%.2x%.2x%.2x", id->cl_dword, id->cl_word[0], + id->cl_word[1], id->cl_two[0], id->cl_two[1], id->cl_six[0], + id->cl_six[1], id->cl_six[2], id->cl_six[3], id->cl_six[4], + id->cl_six[5]); +} + +static const struct { + uint32_t v; + const char *n; +} vn[] = { + { CDF_PROPERTY_CODE_PAGE, "Code page" }, + { CDF_PROPERTY_TITLE, "Title" }, + { CDF_PROPERTY_SUBJECT, "Subject" }, + { CDF_PROPERTY_AUTHOR, "Author" }, + { CDF_PROPERTY_KEYWORDS, "Keywords" }, + { CDF_PROPERTY_COMMENTS, "Comments" }, + { CDF_PROPERTY_TEMPLATE, "Template" }, + { CDF_PROPERTY_LAST_SAVED_BY, "Last Saved By" }, + { CDF_PROPERTY_REVISION_NUMBER, "Revision Number" }, + { CDF_PROPERTY_TOTAL_EDITING_TIME, "Total Editing Time" }, + { CDF_PROPERTY_LAST_PRINTED, "Last Printed" }, + { CDF_PROPERTY_CREATE_TIME, "Create Time/Date" }, + { CDF_PROPERTY_LAST_SAVED_TIME, "Last Saved Time/Date" }, + { CDF_PROPERTY_NUMBER_OF_PAGES, "Number of Pages" }, + { CDF_PROPERTY_NUMBER_OF_WORDS, "Number of Words" }, + { CDF_PROPERTY_NUMBER_OF_CHARACTERS, "Number of Characters" }, + { CDF_PROPERTY_THUMBNAIL, "Thumbnail" }, + { CDF_PROPERTY_NAME_OF_APPLICATION, "Name of Creating Application" }, + { CDF_PROPERTY_SECURITY, "Security" }, + { CDF_PROPERTY_LOCALE_ID, "Locale ID" }, +}; + +int +cdf_print_property_name(char *buf, size_t bufsiz, uint32_t p) +{ + size_t i; + + for (i = 0; i < __arraycount(vn); i++) + if (vn[i].v == p) + return snprintf(buf, bufsiz, "%s", vn[i].n); + return snprintf(buf, bufsiz, "0x%x", p); +} + +int +cdf_print_elapsed_time(char *buf, size_t bufsiz, cdf_timestamp_t ts) +{ + size_t len = 0; + int days, hours, mins, secs; + + ts /= CDF_TIME_PREC; + secs = ts % 60; + ts /= 60; + mins = ts % 60; + ts /= 60; + hours = ts % 24; + ts /= 24; + days = ts; + + if (days) { + len += snprintf(buf + len, bufsiz - len, "%dd+", days); + if (len >= bufsiz) + return len; + } + + if (days || hours) { + len += snprintf(buf + len, bufsiz - len, "%.2d:", hours); + if (len >= bufsiz) + return len; + } + + len += snprintf(buf + len, bufsiz - len, "%.2d:", mins); + if (len >= bufsiz) + return len; + + len += snprintf(buf + len, bufsiz - len, "%.2d", secs); + return len; +} + + +#ifdef CDF_DEBUG +void +cdf_dump_header(const cdf_header_t *h) +{ + size_t i; + +#define DUMP(a, b) printf("%40.40s = " a "\n", # b, h->h_ ## b) + DUMP("%d", revision); + DUMP("%d", version); + DUMP("0x%x", byte_order); + DUMP("%d", sec_size_p2); + DUMP("%d", short_sec_size_p2); + DUMP("%d", num_sectors_in_sat); + DUMP("%d", secid_first_directory); + DUMP("%d", min_size_standard_stream); + DUMP("%d", secid_first_sector_in_short_sat); + DUMP("%d", num_sectors_in_short_sat); + DUMP("%d", secid_first_sector_in_master_sat); + DUMP("%d", num_sectors_in_master_sat); + for (i = 0; i < __arraycount(h->h_master_sat); i++) { + if (h->h_master_sat[i] == CDF_SECID_FREE) + break; + printf("%35.35s[%.3zu] = %d\n", + "master_sat", i, h->h_master_sat[i]); + } +} + +void +cdf_dump_sat(const char *prefix, const cdf_header_t *h, const cdf_sat_t *sat) +{ + size_t i, j, s = CDF_SEC_SIZE(h) / sizeof(cdf_secid_t); + + for (i = 0; i < sat->sat_len; i++) { + printf("%s[%zu]:\n", prefix, i); + for (j = 0; j < s; j++) { + printf("%5d, ", CDF_TOLE4(sat->sat_tab[s * i + j])); + if ((j + 1) % 10 == 0) + printf("\n"); + } + printf("\n"); + } +} + +void +cdf_dump(void *v, size_t len) +{ + size_t i, j; + unsigned char *p = v; + char abuf[16]; + printf("%.4x: ", 0); + for (i = 0, j = 0; i < len; i++, p++) { + printf("%.2x ", *p); + abuf[j++] = isprint(*p) ? *p : '.'; + if (j == 16) { + j = 0; + abuf[15] = '\0'; + printf("%s\n%.4x: ", abuf, i + 1); + } + } + printf("\n"); +} + +void +cdf_dump_stream(const cdf_header_t *h, const cdf_stream_t *sst) +{ + size_t ss = sst->sst_dirlen < h->h_min_size_standard_stream ? + CDF_SHORT_SEC_SIZE(h) : CDF_SEC_SIZE(h); + cdf_dump(sst->sst_tab, ss * sst->sst_len); +} + +void +cdf_dump_dir(int fd, const cdf_header_t *h, const cdf_sat_t *sat, + const cdf_sat_t *ssat, const cdf_stream_t *sst, + const cdf_dir_t *dir) +{ + size_t i, j; + cdf_directory_t *d; + char name[__arraycount(d->d_name)]; + cdf_stream_t scn; + struct timespec ts; + + static const char *types[] = { "empty", "user storage", + "user stream", "lockbytes", "property", "root storage" }; + + for (i = 0; i < dir->dir_len; i++) { + d = &dir->dir_tab[i]; + for (j = 0; j < sizeof(name); j++) + name[j] = (char)CDF_TOLE2(d->d_name[j]); + printf("Directory %zu: %s\n", i, name); + if (d->d_type < __arraycount(types)) + printf("Type: %s\n", types[d->d_type]); + else + printf("Type: %d\n", d->d_type); + printf("Color: %s\n", d->d_color ? "black" : "red"); + printf("Left child: %d\n", d->d_left_child); + printf("Right child: %d\n", d->d_right_child); + printf("Flags: 0x%x\n", d->d_flags); + cdf_timestamp_to_timespec(&ts, d->d_created); + printf("Created %s", ctime(&ts.tv_sec)); + cdf_timestamp_to_timespec(&ts, d->d_modified); + printf("Modified %s", ctime(&ts.tv_sec)); + printf("Stream %d\n", d->d_stream_first_sector); + printf("Size %d\n", d->d_size); + switch (d->d_type) { + case CDF_DIR_TYPE_USER_STORAGE: + printf("Storage: %d\n", d->d_storage); + break; + case CDF_DIR_TYPE_USER_STREAM: + if (sst == NULL) + break; + if (cdf_read_sector_chain(fd, h, sat, ssat, sst, + d->d_stream_first_sector, d->d_size, &scn) == -1) { + warn("Can't read stream for %s at %d len %d", + name, d->d_stream_first_sector, d->d_size); + break; + } + cdf_dump_stream(h, &scn); + free(scn.sst_tab); + break; + default: + break; + } + + } +} + +void +cdf_dump_property_info(const cdf_property_info_t *info, size_t count) +{ + cdf_timestamp_t tp; + struct timespec ts; + char buf[64]; + size_t i; + + for (i = 0; i < count; i++) { + cdf_print_property_name(buf, sizeof(buf), info[i].pi_id); + printf("%zu) %s: ", i, buf); + switch (info[i].pi_type) { + case CDF_SIGNED16: + printf("signed 16 [%hd]\n", info[i].pi_s16); + break; + case CDF_SIGNED32: + printf("signed 32 [%d]\n", info[i].pi_s32); + break; + case CDF_UNSIGNED32: + printf("unsigned 32 [%u]\n", info[i].pi_u32); + break; + case CDF_LENGTH32_STRING: + printf("string %u [%.*s]\n", info[i].pi_str.s_len, + info[i].pi_str.s_len, info[i].pi_str.s_buf); + break; + case CDF_FILETIME: + tp = info[i].pi_tp; + if (tp < 1000000000000000LL) { + cdf_print_elapsed_time(buf, sizeof(buf), tp); + printf("timestamp %s\n", buf); + } else { + cdf_timestamp_to_timespec(&ts, tp); + printf("timestamp %s", ctime(&ts.tv_sec)); + } + break; + case CDF_CLIPBOARD: + printf("CLIPBOARD %u\n", info[i].pi_u32); + break; + default: + DPRINTF(("Don't know how to deal with %x\n", + info[i].pi_type)); + break; + } + } +} + + +void +cdf_dump_summary_info(const cdf_header_t *h, const cdf_stream_t *sst) +{ + char buf[128]; + cdf_summary_info_header_t ssi; + cdf_property_info_t *info; + size_t count; + + (void)&h; + if (cdf_unpack_summary_info(sst, &ssi, &info, &count) == -1) + return; + printf("Endian: %x\n", ssi.si_byte_order); + printf("Os Version %d.%d\n", ssi.si_os_version & 0xff, + ssi.si_os_version >> 8); + printf("Os %d\n", ssi.si_os); + cdf_print_classid(buf, sizeof(buf), &ssi.si_class); + printf("Class %s\n", buf); + printf("Count %d\n", ssi.si_count); + cdf_dump_property_info(info, count); + free(info); +} + +#endif + +#ifdef TEST +int +main(int argc, char *argv[]) +{ + int fd, i; + cdf_header_t h; + cdf_sat_t sat, ssat; + cdf_stream_t sst, scn; + cdf_dir_t dir; + + if (argc < 2) { + (void)fprintf(stderr, "Usage: %s <filename>\n", getprogname()); + return -1; + } + + for (i = 1; i < argc; i++) { + if ((fd = open(argv[1], O_RDONLY)) == -1) + err(1, "Cannot open `%s'", argv[1]); + + if (cdf_read_header(fd, &h) == -1) + err(1, "Cannot read header"); +#ifdef CDF_DEBUG + cdf_dump_header(&h); +#endif + + if (cdf_read_sat(fd, &h, &sat) == -1) + err(1, "Cannot read sat"); +#ifdef CDF_DEBUG + cdf_dump_sat("SAT", &h, &sat); +#endif + + if (cdf_read_ssat(fd, &h, &sat, &ssat) == -1) + err(1, "Cannot read ssat"); +#ifdef CDF_DEBUG + cdf_dump_sat("SSAT", &h, &ssat); +#endif + + if (cdf_read_dir(fd, &h, &sat, &dir) == -1) + err(1, "Cannot read dir"); + + if (cdf_read_short_stream(fd, &h, &sat, &dir, &sst) == -1) + err(1, "Cannot read short stream"); +#ifdef CDF_DEBUG + cdf_dump_stream(&h, &sst); +#endif + +#ifdef CDF_DEBUG + cdf_dump_dir(fd, &h, &sat, &ssat, &sst, &dir); +#endif + + + if (cdf_read_summary_info(fd, &h, &sat, &ssat, &sst, &dir, + &scn) == -1) + err(1, "Cannot read summary info"); +#ifdef CDF_DEBUG + cdf_dump_summary_info(&h, &scn); +#endif + + (void)close(fd); + } + + return 0; +} +#endif @@ -0,0 +1,298 @@ +/*- + * Copyright (c) 2008 Christos Zoulas + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. 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 NETBSD FOUNDATION, INC. 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 FOUNDATION 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. + */ +/* + * Info from: http://sc.openoffice.org/compdocfileformat.pdf + */ + +#ifndef _H_CDF_ +#define _H_CDF_ + +typedef int32_t cdf_secid_t; + +#define CDF_LOOP_LIMIT 10000 + +#define CDF_SECID_NULL 0 +#define CDF_SECID_FREE -1 +#define CDF_SECID_END_OF_CHAIN -2 +#define CDF_SECID_SECTOR_ALLOCATION_TABLE -3 +#define CDF_SECID_MASTER_SECTOR_ALLOCATION_TABLE -4 + +typedef struct { + uint64_t h_magic; +#define CDF_MAGIC 0xE11AB1A1E011CFD0LL + uint64_t h_uuid[2]; + uint16_t h_revision; + uint16_t h_version; + uint16_t h_byte_order; + uint16_t h_sec_size_p2; + uint16_t h_short_sec_size_p2; + uint8_t h_unused0[10]; + uint32_t h_num_sectors_in_sat; + uint32_t h_secid_first_directory; + uint8_t h_unused1[4]; + uint32_t h_min_size_standard_stream; + cdf_secid_t h_secid_first_sector_in_short_sat; + uint32_t h_num_sectors_in_short_sat; + cdf_secid_t h_secid_first_sector_in_master_sat; + uint32_t h_num_sectors_in_master_sat; + cdf_secid_t h_master_sat[436/4]; +} cdf_header_t; + +#define CDF_SEC_SIZE(h) (1 << (h)->h_sec_size_p2) +#define CDF_SEC_POS(h, secid) (CDF_SEC_SIZE(h) + (secid) * CDF_SEC_SIZE(h)) +#define CDF_SHORT_SEC_SIZE(h) (1 << (h)->h_short_sec_size_p2) +#define CDF_SHORT_SEC_POS(h, secid) ((secid) * CDF_SHORT_SEC_SIZE(h)) + +typedef int32_t cdf_dirid_t; +#define CDF_DIRID_NULL -1 + +typedef int64_t cdf_timestamp_t; +#define CDF_BASE_YEAR 1601 +#define CDF_TIME_PREC 10000000 + +typedef struct { + uint16_t d_name[32]; + uint16_t d_namelen; + uint8_t d_type; +#define CDF_DIR_TYPE_EMPTY 0 +#define CDF_DIR_TYPE_USER_STORAGE 1 +#define CDF_DIR_TYPE_USER_STREAM 2 +#define CDF_DIR_TYPE_LOCKBYTES 3 +#define CDF_DIR_TYPE_PROPERTY 4 +#define CDF_DIR_TYPE_ROOT_STORAGE 5 + uint8_t d_color; +#define CDF_DIR_COLOR_READ 0 +#define CDF_DIR_COLOR_BLACK 1 + cdf_dirid_t d_left_child; + cdf_dirid_t d_right_child; + cdf_dirid_t d_storage; + uint64_t d_storage_uuid[2]; + uint32_t d_flags; + cdf_timestamp_t d_created; + cdf_timestamp_t d_modified; + cdf_secid_t d_stream_first_sector; + uint32_t d_size; + uint32_t d_unused0; +} cdf_directory_t; + +#define CDF_DIRECTORY_SIZE 128 + +typedef struct { + cdf_secid_t *sat_tab; + size_t sat_len; +} cdf_sat_t; + +typedef struct { + cdf_directory_t *dir_tab; + size_t dir_len; +} cdf_dir_t; + +typedef struct { + void *sst_tab; + size_t sst_len; + size_t sst_dirlen; +} cdf_stream_t; + +typedef struct { + uint32_t cl_dword; + uint16_t cl_word[2]; + uint8_t cl_two[2]; + uint8_t cl_six[6]; +} cdf_classid_t; + +typedef struct { + uint16_t si_byte_order; + uint16_t si_zero; + uint16_t si_os_version; + uint16_t si_os; + cdf_classid_t si_class; + uint32_t si_count; +} cdf_summary_info_header_t; + +#define CDF_SECTION_DECLARATION_OFFSET 0x1c + +typedef struct { + cdf_classid_t sd_class; + uint32_t sd_offset; +} cdf_section_declaration_t; + +typedef struct { + uint32_t sh_len; + uint32_t sh_properties; +} cdf_section_header_t; + +typedef struct { + uint32_t pi_id; + uint32_t pi_type; + union { + uint16_t _pi_u16; + int16_t _pi_s16; + uint32_t _pi_u32; + int32_t _pi_s32; + uint64_t _pi_u64; + int64_t _pi_s64; + cdf_timestamp_t _pi_tp; + struct { + uint32_t s_len; + const char *s_buf; + } _pi_str; + } pi_val; +#define pi_u64 pi_val._pi_u64 +#define pi_s64 pi_val._pi_s64 +#define pi_u32 pi_val._pi_u32 +#define pi_s32 pi_val._pi_s32 +#define pi_u16 pi_val._pi_u16 +#define pi_s16 pi_val._pi_s16 +#define pi_tp pi_val._pi_tp +#define pi_str pi_val._pi_str +} cdf_property_info_t; + +#define CDF_ROUND(val, by) (((val) + (by) - 1) & ~((by) - 1)) + +/* Variant type definitions */ +#define CDF_EMPTY 0x00000000 +#define CDF_NULL 0x00000001 +#define CDF_SIGNED16 0x00000002 +#define CDF_SIGNED32 0x00000003 +#define CDF_FLOAT 0x00000004 +#define CDF_DOUBLE 0x00000005 +#define CDF_CY 0x00000006 +#define CDF_DATE 0x00000007 +#define CDF_BSTR 0x00000008 +#define CDF_DISPATCH 0x00000009 +#define CDF_ERROR 0x0000000a +#define CDF_BOOL 0x0000000b +#define CDF_VARIANT 0x0000000c +#define CDF_UNKNOWN 0x0000000d +#define CDF_DECIMAL 0x0000000e +#define CDF_SIGNED8 0x00000010 +#define CDF_UNSIGNED8 0x00000011 +#define CDF_UNSIGNED16 0x00000012 +#define CDF_UNSIGNED32 0x00000013 +#define CDF_SIGNED64 0x00000014 +#define CDF_UNSIGNED64 0x00000015 +#define CDF_INT 0x00000016 +#define CDF_UINT 0x00000017 +#define CDF_VOID 0x00000018 +#define CDF_HRESULT 0x00000019 +#define CDF_PTR 0x0000001a +#define CDF_SAFEARRAY 0x0000001b +#define CDF_CARRAY 0x0000001c +#define CDF_USERDEFINED 0x0000001d +#define CDF_LENGTH32_STRING 0x0000001e +#define CDF_LENGTH32_WSTRING 0x0000001f +#define CDF_FILETIME 0x00000040 +#define CDF_BLOB 0x00000041 +#define CDF_STREAM 0x00000042 +#define CDF_STORAGE 0x00000043 +#define CDF_STREAMED_OBJECT 0x00000044 +#define CDF_STORED_OBJECT 0x00000045 +#define CDF_BLOB_OBJECT 0x00000046 +#define CDF_CLIPBOARD 0x00000047 +#define CDF_CLSID 0x00000048 +#define CDF_VECTOR 0x00001000 +#define CDF_ARRAY 0x00002000 +#define CDF_BYREF 0x00004000 +#define CDF_RESERVED 0x00008000 +#define CDF_ILLEGAL 0x0000ffff +#define CDF_ILLEGALMASKED 0x00000fff +#define CDF_TYPEMASK 0x00000fff + +#define CDF_PROPERTY_CODE_PAGE 0x00000001 +#define CDF_PROPERTY_TITLE 0x00000002 +#define CDF_PROPERTY_SUBJECT 0x00000003 +#define CDF_PROPERTY_AUTHOR 0x00000004 +#define CDF_PROPERTY_KEYWORDS 0x00000005 +#define CDF_PROPERTY_COMMENTS 0x00000006 +#define CDF_PROPERTY_TEMPLATE 0x00000007 +#define CDF_PROPERTY_LAST_SAVED_BY 0x00000008 +#define CDF_PROPERTY_REVISION_NUMBER 0x00000009 +#define CDF_PROPERTY_TOTAL_EDITING_TIME 0x0000000a +#define CDF_PROPERTY_LAST_PRINTED 0X0000000b +#define CDF_PROPERTY_CREATE_TIME 0x0000000c +#define CDF_PROPERTY_LAST_SAVED_TIME 0x0000000d +#define CDF_PROPERTY_NUMBER_OF_PAGES 0x0000000e +#define CDF_PROPERTY_NUMBER_OF_WORDS 0x0000000f +#define CDF_PROPERTY_NUMBER_OF_CHARACTERS 0x00000010 +#define CDF_PROPERTY_THUMBNAIL 0x00000011 +#define CDF_PROPERTY_NAME_OF_APPLICATION 0x00000012 +#define CDF_PROPERTY_SECURITY 0x00000013 +#define CDF_PROPERTY_LOCALE_ID 0x80000000 + +struct timespec; +int cdf_timestamp_to_timespec(struct timespec *, cdf_timestamp_t); +int cdf_timespec_to_timestamp(cdf_timestamp_t *, const struct timespec *); +int cdf_read_header(int, cdf_header_t *); +void cdf_swap_header(cdf_header_t *); +void cdf_unpack_header(cdf_header_t *, char *); +void cdf_swap_dir(cdf_directory_t *); +void cdf_unpack_dir(cdf_directory_t *, char *); +void cdf_swap_class(cdf_classid_t *); +ssize_t cdf_read_sector(int, void *, size_t, size_t, const cdf_header_t *, + cdf_secid_t); +ssize_t cdf_read_short_sector(const cdf_stream_t *, void *, size_t, size_t, + const cdf_header_t *, cdf_secid_t); +int cdf_read_sat(int, cdf_header_t *, cdf_sat_t *); +size_t cdf_count_chain(const cdf_header_t *, const cdf_sat_t *, + cdf_secid_t); +int cdf_read_long_sector_chain(int, const cdf_header_t *, + const cdf_sat_t *, cdf_secid_t, size_t, cdf_stream_t *); +int cdf_read_short_sector_chain(const cdf_header_t *, const cdf_sat_t *, + const cdf_stream_t *, cdf_secid_t, size_t, cdf_stream_t *); +int cdf_read_sector_chain(int, const cdf_header_t *, + const cdf_sat_t *, const cdf_sat_t *, const cdf_stream_t *, cdf_secid_t, + size_t, cdf_stream_t *); +int cdf_read_dir(int, const cdf_header_t *, const cdf_sat_t *, cdf_dir_t *); +int cdf_read_ssat(int, const cdf_header_t *, const cdf_sat_t *, cdf_sat_t *); +int cdf_read_short_stream(int, const cdf_header_t *, const cdf_sat_t *, + const cdf_dir_t *, cdf_stream_t *); +int cdf_read_property_info(const cdf_stream_t *, uint32_t, + cdf_property_info_t **, size_t *, size_t *); +int cdf_read_summary_info(int, const cdf_header_t *, const cdf_sat_t *, + const cdf_sat_t *, const cdf_stream_t *, const cdf_dir_t *, + cdf_stream_t *); +int cdf_unpack_summary_info(const cdf_stream_t *, cdf_summary_info_header_t *, + cdf_property_info_t **, size_t *); +int cdf_print_classid(char *, size_t, const cdf_classid_t *); +int cdf_print_property_name(char *, size_t, uint32_t); +int cdf_print_elapsed_time(char *, size_t, cdf_timestamp_t); +uint16_t cdf_tole2(uint16_t); +uint32_t cdf_tole4(uint32_t); +uint64_t cdf_tole8(uint64_t); + +#ifdef CDF_DEBUG +void cdf_dump_header(const cdf_header_t *); +void cdf_dump_sat(const char *, const cdf_header_t *, const cdf_sat_t *); +void cdf_dump(void *, size_t); +void cdf_dump_stream(const cdf_header_t *, const cdf_stream_t *); +void cdf_dump_dir(int, const cdf_header_t *, const cdf_sat_t *, + const cdf_sat_t *, const cdf_stream_t *, const cdf_dir_t *); +void cdf_dump_property_info(const cdf_property_info_t *, size_t); +void cdf_dump_summary_info(const cdf_header_t *, const cdf_stream_t *); +#endif + + +#endif /* _H_CDF_ */ diff --git a/cdf_time.c b/cdf_time.c new file mode 100644 index 0000000..e531b2d --- /dev/null +++ b/cdf_time.c @@ -0,0 +1,182 @@ +/*- + * Copyright (c) 2008 Christos Zoulas + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. 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 NETBSD FOUNDATION, INC. 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 FOUNDATION 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. + */ + +#include "file.h" + +#ifndef lint +FILE_RCSID("@(#)$File: cdf_time.c,v 1.5 2009/02/03 20:27:51 christos Exp $") +#endif + +#include <time.h> +#ifdef TEST +#include <err.h> +#endif +#include <string.h> + +#include "cdf.h" + +#define isleap(y) ((((y) % 4) == 0) && \ + ((((y) % 100) != 0) || (((y) % 400) == 0))) + +static const int mdays[] = { + 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31 +}; + +/* + * Return the number of days between jan 01 1601 and jan 01 of year. + */ +static int +cdf_getdays(int year) +{ + int days = 0; + int y; + + for (y = CDF_BASE_YEAR; y < year; y++) + days += isleap(y) + 365; + + return days; +} + +/* + * Return the day within the month + */ +static int +cdf_getday(int year, int days) +{ + size_t m; + + for (m = 0; m < sizeof(mdays) / sizeof(mdays[0]); m++) { + int sub = mdays[m] + (m == 1 && isleap(year)); + if (days < sub) + return days; + days -= sub; + } + return days; +} + +/* + * Return the 0...11 month number. + */ +static int +cdf_getmonth(int year, int days) +{ + size_t m; + + for (m = 0; m < sizeof(mdays) / sizeof(mdays[0]); m++) { + days -= mdays[m]; + if (m == 1 && isleap(year)) + days--; + if (days <= 0) + return (int)m; + } + return (int)m; +} + +int +cdf_timestamp_to_timespec(struct timespec *ts, cdf_timestamp_t t) +{ + struct tm tm; +#ifdef HAVE_STRUCT_TM_TM_ZONE + static char UTC[] = "UTC"; +#endif + + /* Unit is 100's of nanoseconds */ + ts->tv_nsec = (t % CDF_TIME_PREC) * 100; + + t /= CDF_TIME_PREC; + tm.tm_sec = t % 60; + t /= 60; + + tm.tm_min = t % 60; + t /= 60; + + tm.tm_hour = t % 24; + t /= 24; + + // XXX: Approx + tm.tm_year = CDF_BASE_YEAR + (t / 365); + + int rdays = cdf_getdays(tm.tm_year); + t -= rdays; + tm.tm_mday = cdf_getday(tm.tm_year, t); + tm.tm_mon = cdf_getmonth(tm.tm_year, t); + tm.tm_wday = 0; + tm.tm_yday = 0; + tm.tm_isdst = 0; +#ifdef HAVE_STRUCT_TM_TM_GMTOFF + tm.tm_gmtoff = 0; +#endif +#ifdef HAVE_STRUCT_TM_TM_ZONE + tm.tm_zone = UTC; +#endif + tm.tm_year -= 1900; + ts->tv_sec = mktime(&tm); + if (ts->tv_sec == -1) { + errno = EINVAL; + return -1; + } + return 0; +} + +int +cdf_timespec_to_timestamp(cdf_timestamp_t *t, const struct timespec *ts) +{ + (void)&t; + (void)&ts; +#ifdef notyet + struct tm tm; + if (gmtime_r(&ts->ts_sec, &tm) == NULL) { + errno = EINVAL; + return -1; + } + *t = (ts->ts_nsec / 100) * CDF_TIME_PREC; + *t = tm.tm_sec; + *t += tm.tm_min * 60; + *t += tm.tm_hour * 60 * 60; + *t += tm.tm_mday * 60 * 60 * 24; +#endif + return 0; +} + + +#ifdef TEST +int +main(int argc, char *argv[]) +{ + struct timespec ts; + static const cdf_timestamp_t tst = 0x01A5E403C2D59C00ULL; + static const char *ref = "Sat Apr 23 01:30:00 1977"; + char *p, *q; + + cdf_timestamp_to_timespec(&ts, tst); + p = ctime(&ts.tv_sec); + if ((q = strchr(p, '\n')) != NULL) + *q = '\0'; + if (strcmp(ref, p) != 0) + errx(1, "Error date %s != %s\n", ref, p); + return 0; +} +#endif diff --git a/encoding.c b/encoding.c new file mode 100644 index 0000000..4e94f9b --- /dev/null +++ b/encoding.c @@ -0,0 +1,484 @@ +/* + * Copyright (c) Ian F. Darwin 1986-1995. + * Software written by Ian F. Darwin and others; + * maintained 1995-present by Christos Zoulas and others. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice immediately at the beginning of the file, without modification, + * this list of conditions, and the following disclaimer. + * 2. 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 AUTHOR 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 AUTHOR 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. + */ +/* + * Encoding -- determine the character encoding of a text file. + * + * Joerg Wunsch <joerg@freebsd.org> wrote the original support for 8-bit + * international characters. + */ + +#include "file.h" + +#ifndef lint +FILE_RCSID("@(#)$File: encoding.c,v 1.3 2009/02/03 20:27:51 christos Exp $") +#endif /* lint */ + +#include "magic.h" +#include <string.h> +#include <memory.h> +#include <stdlib.h> + + +private int looks_ascii(const unsigned char *, size_t, unichar *, size_t *); +private int looks_utf8_with_BOM(const unsigned char *, size_t, unichar *, + size_t *); +private int looks_ucs16(const unsigned char *, size_t, unichar *, size_t *); +private int looks_latin1(const unsigned char *, size_t, unichar *, size_t *); +private int looks_extended(const unsigned char *, size_t, unichar *, size_t *); +private void from_ebcdic(const unsigned char *, size_t, unsigned char *); + +/* + * Try to determine whether text is in some character code we can + * identify. Each of these tests, if it succeeds, will leave + * the text converted into one-unichar-per-character Unicode in + * ubuf, and the number of characters converted in ulen. + */ +protected int +file_encoding(struct magic_set *ms, const unsigned char *buf, size_t nbytes, unichar **ubuf, size_t *ulen, const char **code, const char **code_mime, const char **type) +{ + size_t mlen; + int rv = 1, ucs_type; + unsigned char *nbuf = NULL; + + mlen = (nbytes + 1) * sizeof(nbuf[0]); + if ((nbuf = CAST(unsigned char *, calloc((size_t)1, mlen))) == NULL) { + file_oomem(ms, mlen); + goto done; + } + mlen = (nbytes + 1) * sizeof((*ubuf)[0]); + if ((*ubuf = CAST(unichar *, calloc((size_t)1, mlen))) == NULL) { + file_oomem(ms, mlen); + goto done; + } + + *type = "text"; + if (looks_ascii(buf, nbytes, *ubuf, ulen)) { + *code = "ASCII"; + *code_mime = "us-ascii"; + } else if (looks_utf8_with_BOM(buf, nbytes, *ubuf, ulen) > 0) { + *code = "UTF-8 Unicode (with BOM)"; + *code_mime = "utf-8"; + } else if (file_looks_utf8(buf, nbytes, *ubuf, ulen) > 1) { + *code = "UTF-8 Unicode"; + *code_mime = "utf-8"; + } else if ((ucs_type = looks_ucs16(buf, nbytes, *ubuf, ulen)) != 0) { + if (ucs_type == 1) { + *code = "Little-endian UTF-16 Unicode"; + *code_mime = "utf-16le"; + } else { + *code = "Big-endian UTF-16 Unicode"; + *code_mime = "utf-16be"; + } + } else if (looks_latin1(buf, nbytes, *ubuf, ulen)) { + *code = "ISO-8859"; + *code_mime = "iso-8859-1"; + } else if (looks_extended(buf, nbytes, *ubuf, ulen)) { + *code = "Non-ISO extended-ASCII"; + *code_mime = "unknown-8bit"; + } else { + from_ebcdic(buf, nbytes, nbuf); + + if (looks_ascii(nbuf, nbytes, *ubuf, ulen)) { + *code = "EBCDIC"; + *code_mime = "ebcdic"; + } else if (looks_latin1(nbuf, nbytes, *ubuf, ulen)) { + *code = "International EBCDIC"; + *code_mime = "ebcdic"; + } else { /* Doesn't look like text at all */ + rv = 0; + *type = "binary"; + } + } + + done: + if (nbuf) + free(nbuf); + + return rv; +} + +/* + * This table reflects a particular philosophy about what constitutes + * "text," and there is room for disagreement about it. + * + * Version 3.31 of the file command considered a file to be ASCII if + * each of its characters was approved by either the isascii() or + * isalpha() function. On most systems, this would mean that any + * file consisting only of characters in the range 0x00 ... 0x7F + * would be called ASCII text, but many systems might reasonably + * consider some characters outside this range to be alphabetic, + * so the file command would call such characters ASCII. It might + * have been more accurate to call this "considered textual on the + * local system" than "ASCII." + * + * It considered a file to be "International language text" if each + * of its characters was either an ASCII printing character (according + * to the real ASCII standard, not the above test), a character in + * the range 0x80 ... 0xFF, or one of the following control characters: + * backspace, tab, line feed, vertical tab, form feed, carriage return, + * escape. No attempt was made to determine the language in which files + * of this type were written. + * + * + * The table below considers a file to be ASCII if all of its characters + * are either ASCII printing characters (again, according to the X3.4 + * standard, not isascii()) or any of the following controls: bell, + * backspace, tab, line feed, form feed, carriage return, esc, nextline. + * + * I include bell because some programs (particularly shell scripts) + * use it literally, even though it is rare in normal text. I exclude + * vertical tab because it never seems to be used in real text. I also + * include, with hesitation, the X3.64/ECMA-43 control nextline (0x85), + * because that's what the dd EBCDIC->ASCII table maps the EBCDIC newline + * character to. It might be more appropriate to include it in the 8859 + * set instead of the ASCII set, but it's got to be included in *something* + * we recognize or EBCDIC files aren't going to be considered textual. + * Some old Unix source files use SO/SI (^N/^O) to shift between Greek + * and Latin characters, so these should possibly be allowed. But they + * make a real mess on VT100-style displays if they're not paired properly, + * so we are probably better off not calling them text. + * + * A file is considered to be ISO-8859 text if its characters are all + * either ASCII, according to the above definition, or printing characters + * from the ISO-8859 8-bit extension, characters 0xA0 ... 0xFF. + * + * Finally, a file is considered to be international text from some other + * character code if its characters are all either ISO-8859 (according to + * the above definition) or characters in the range 0x80 ... 0x9F, which + * ISO-8859 considers to be control characters but the IBM PC and Macintosh + * consider to be printing characters. + */ + +#define F 0 /* character never appears in text */ +#define T 1 /* character appears in plain ASCII text */ +#define I 2 /* character appears in ISO-8859 text */ +#define X 3 /* character appears in non-ISO extended ASCII (Mac, IBM PC) */ + +private char text_chars[256] = { + /* BEL BS HT LF FF CR */ + F, F, F, F, F, F, F, T, T, T, T, F, T, T, F, F, /* 0x0X */ + /* ESC */ + F, F, F, F, F, F, F, F, F, F, F, T, F, F, F, F, /* 0x1X */ + T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x2X */ + T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x3X */ + T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x4X */ + T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x5X */ + T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, /* 0x6X */ + T, T, T, T, T, T, T, T, T, T, T, T, T, T, T, F, /* 0x7X */ + /* NEL */ + X, X, X, X, X, T, X, X, X, X, X, X, X, X, X, X, /* 0x8X */ + X, X, X, X, X, X, X, X, X, X, X, X, X, X, X, X, /* 0x9X */ + I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xaX */ + I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xbX */ + I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xcX */ + I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xdX */ + I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, /* 0xeX */ + I, I, I, I, I, I, I, I, I, I, I, I, I, I, I, I /* 0xfX */ +}; + +private int +looks_ascii(const unsigned char *buf, size_t nbytes, unichar *ubuf, + size_t *ulen) +{ + size_t i; + + *ulen = 0; + + for (i = 0; i < nbytes; i++) { + int t = text_chars[buf[i]]; + + if (t != T) + return 0; + + ubuf[(*ulen)++] = buf[i]; + } + + return 1; +} + +private int +looks_latin1(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen) +{ + size_t i; + + *ulen = 0; + + for (i = 0; i < nbytes; i++) { + int t = text_chars[buf[i]]; + + if (t != T && t != I) + return 0; + + ubuf[(*ulen)++] = buf[i]; + } + + return 1; +} + +private int +looks_extended(const unsigned char *buf, size_t nbytes, unichar *ubuf, + size_t *ulen) +{ + size_t i; + + *ulen = 0; + + for (i = 0; i < nbytes; i++) { + int t = text_chars[buf[i]]; + + if (t != T && t != I && t != X) + return 0; + + ubuf[(*ulen)++] = buf[i]; + } + + return 1; +} + +/* + * Decide whether some text looks like UTF-8. Returns: + * + * -1: invalid UTF-8 + * 0: uses odd control characters, so doesn't look like text + * 1: 7-bit text + * 2: definitely UTF-8 text (valid high-bit set bytes) + * + * If ubuf is non-NULL on entry, text is decoded into ubuf, *ulen; + * ubuf must be big enough! + */ +protected int +file_looks_utf8(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen) +{ + size_t i; + int n; + unichar c; + int gotone = 0, ctrl = 0; + + if (ubuf) + *ulen = 0; + + for (i = 0; i < nbytes; i++) { + if ((buf[i] & 0x80) == 0) { /* 0xxxxxxx is plain ASCII */ + /* + * Even if the whole file is valid UTF-8 sequences, + * still reject it if it uses weird control characters. + */ + + if (text_chars[buf[i]] != T) + ctrl = 1; + + if (ubuf) + ubuf[(*ulen)++] = buf[i]; + } else if ((buf[i] & 0x40) == 0) { /* 10xxxxxx never 1st byte */ + return -1; + } else { /* 11xxxxxx begins UTF-8 */ + int following; + + if ((buf[i] & 0x20) == 0) { /* 110xxxxx */ + c = buf[i] & 0x1f; + following = 1; + } else if ((buf[i] & 0x10) == 0) { /* 1110xxxx */ + c = buf[i] & 0x0f; + following = 2; + } else if ((buf[i] & 0x08) == 0) { /* 11110xxx */ + c = buf[i] & 0x07; + following = 3; + } else if ((buf[i] & 0x04) == 0) { /* 111110xx */ + c = buf[i] & 0x03; + following = 4; + } else if ((buf[i] & 0x02) == 0) { /* 1111110x */ + c = buf[i] & 0x01; + following = 5; + } else + return -1; + + for (n = 0; n < following; n++) { + i++; + if (i >= nbytes) + goto done; + + if ((buf[i] & 0x80) == 0 || (buf[i] & 0x40)) + return -1; + + c = (c << 6) + (buf[i] & 0x3f); + } + + if (ubuf) + ubuf[(*ulen)++] = c; + gotone = 1; + } + } +done: + return ctrl ? 0 : (gotone ? 2 : 1); +} + +/* + * Decide whether some text looks like UTF-8 with BOM. If there is no + * BOM, return -1; otherwise return the result of looks_utf8 on the + * rest of the text. + */ +private int +looks_utf8_with_BOM(const unsigned char *buf, size_t nbytes, unichar *ubuf, + size_t *ulen) +{ + if (nbytes > 3 && buf[0] == 0xef && buf[1] == 0xbb && buf[2] == 0xbf) + return file_looks_utf8(buf + 3, nbytes - 3, ubuf, ulen); + else + return -1; +} + +private int +looks_ucs16(const unsigned char *buf, size_t nbytes, unichar *ubuf, + size_t *ulen) +{ + int bigend; + size_t i; + + if (nbytes < 2) + return 0; + + if (buf[0] == 0xff && buf[1] == 0xfe) + bigend = 0; + else if (buf[0] == 0xfe && buf[1] == 0xff) + bigend = 1; + else + return 0; + + *ulen = 0; + + for (i = 2; i + 1 < nbytes; i += 2) { + /* XXX fix to properly handle chars > 65536 */ + + if (bigend) + ubuf[(*ulen)++] = buf[i + 1] + 256 * buf[i]; + else + ubuf[(*ulen)++] = buf[i] + 256 * buf[i + 1]; + + if (ubuf[*ulen - 1] == 0xfffe) + return 0; + if (ubuf[*ulen - 1] < 128 && + text_chars[(size_t)ubuf[*ulen - 1]] != T) + return 0; + } + + return 1 + bigend; +} + +#undef F +#undef T +#undef I +#undef X + +/* + * This table maps each EBCDIC character to an (8-bit extended) ASCII + * character, as specified in the rationale for the dd(1) command in + * draft 11.2 (September, 1991) of the POSIX P1003.2 standard. + * + * Unfortunately it does not seem to correspond exactly to any of the + * five variants of EBCDIC documented in IBM's _Enterprise Systems + * Architecture/390: Principles of Operation_, SA22-7201-06, Seventh + * Edition, July, 1999, pp. I-1 - I-4. + * + * Fortunately, though, all versions of EBCDIC, including this one, agree + * on most of the printing characters that also appear in (7-bit) ASCII. + * Of these, only '|', '!', '~', '^', '[', and ']' are in question at all. + * + * Fortunately too, there is general agreement that codes 0x00 through + * 0x3F represent control characters, 0x41 a nonbreaking space, and the + * remainder printing characters. + * + * This is sufficient to allow us to identify EBCDIC text and to distinguish + * between old-style and internationalized examples of text. + */ + +private unsigned char ebcdic_to_ascii[] = { + 0, 1, 2, 3, 156, 9, 134, 127, 151, 141, 142, 11, 12, 13, 14, 15, + 16, 17, 18, 19, 157, 133, 8, 135, 24, 25, 146, 143, 28, 29, 30, 31, +128, 129, 130, 131, 132, 10, 23, 27, 136, 137, 138, 139, 140, 5, 6, 7, +144, 145, 22, 147, 148, 149, 150, 4, 152, 153, 154, 155, 20, 21, 158, 26, +' ', 160, 161, 162, 163, 164, 165, 166, 167, 168, 213, '.', '<', '(', '+', '|', +'&', 169, 170, 171, 172, 173, 174, 175, 176, 177, '!', '$', '*', ')', ';', '~', +'-', '/', 178, 179, 180, 181, 182, 183, 184, 185, 203, ',', '%', '_', '>', '?', +186, 187, 188, 189, 190, 191, 192, 193, 194, '`', ':', '#', '@', '\'','=', '"', +195, 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 196, 197, 198, 199, 200, 201, +202, 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', '^', 204, 205, 206, 207, 208, +209, 229, 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', 210, 211, 212, '[', 214, 215, +216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, ']', 230, 231, +'{', 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 232, 233, 234, 235, 236, 237, +'}', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 238, 239, 240, 241, 242, 243, +'\\',159, 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 244, 245, 246, 247, 248, 249, +'0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 250, 251, 252, 253, 254, 255 +}; + +#ifdef notdef +/* + * The following EBCDIC-to-ASCII table may relate more closely to reality, + * or at least to modern reality. It comes from + * + * http://ftp.s390.ibm.com/products/oe/bpxqp9.html + * + * and maps the characters of EBCDIC code page 1047 (the code used for + * Unix-derived software on IBM's 390 systems) to the corresponding + * characters from ISO 8859-1. + * + * If this table is used instead of the above one, some of the special + * cases for the NEL character can be taken out of the code. + */ + +private unsigned char ebcdic_1047_to_8859[] = { +0x00,0x01,0x02,0x03,0x9C,0x09,0x86,0x7F,0x97,0x8D,0x8E,0x0B,0x0C,0x0D,0x0E,0x0F, +0x10,0x11,0x12,0x13,0x9D,0x0A,0x08,0x87,0x18,0x19,0x92,0x8F,0x1C,0x1D,0x1E,0x1F, +0x80,0x81,0x82,0x83,0x84,0x85,0x17,0x1B,0x88,0x89,0x8A,0x8B,0x8C,0x05,0x06,0x07, +0x90,0x91,0x16,0x93,0x94,0x95,0x96,0x04,0x98,0x99,0x9A,0x9B,0x14,0x15,0x9E,0x1A, +0x20,0xA0,0xE2,0xE4,0xE0,0xE1,0xE3,0xE5,0xE7,0xF1,0xA2,0x2E,0x3C,0x28,0x2B,0x7C, +0x26,0xE9,0xEA,0xEB,0xE8,0xED,0xEE,0xEF,0xEC,0xDF,0x21,0x24,0x2A,0x29,0x3B,0x5E, +0x2D,0x2F,0xC2,0xC4,0xC0,0xC1,0xC3,0xC5,0xC7,0xD1,0xA6,0x2C,0x25,0x5F,0x3E,0x3F, +0xF8,0xC9,0xCA,0xCB,0xC8,0xCD,0xCE,0xCF,0xCC,0x60,0x3A,0x23,0x40,0x27,0x3D,0x22, +0xD8,0x61,0x62,0x63,0x64,0x65,0x66,0x67,0x68,0x69,0xAB,0xBB,0xF0,0xFD,0xFE,0xB1, +0xB0,0x6A,0x6B,0x6C,0x6D,0x6E,0x6F,0x70,0x71,0x72,0xAA,0xBA,0xE6,0xB8,0xC6,0xA4, +0xB5,0x7E,0x73,0x74,0x75,0x76,0x77,0x78,0x79,0x7A,0xA1,0xBF,0xD0,0x5B,0xDE,0xAE, +0xAC,0xA3,0xA5,0xB7,0xA9,0xA7,0xB6,0xBC,0xBD,0xBE,0xDD,0xA8,0xAF,0x5D,0xB4,0xD7, +0x7B,0x41,0x42,0x43,0x44,0x45,0x46,0x47,0x48,0x49,0xAD,0xF4,0xF6,0xF2,0xF3,0xF5, +0x7D,0x4A,0x4B,0x4C,0x4D,0x4E,0x4F,0x50,0x51,0x52,0xB9,0xFB,0xFC,0xF9,0xFA,0xFF, +0x5C,0xF7,0x53,0x54,0x55,0x56,0x57,0x58,0x59,0x5A,0xB2,0xD4,0xD6,0xD2,0xD3,0xD5, +0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37,0x38,0x39,0xB3,0xDB,0xDC,0xD9,0xDA,0x9F +}; +#endif + +/* + * Copy buf[0 ... nbytes-1] into out[], translating EBCDIC to ASCII. + */ +private void +from_ebcdic(const unsigned char *buf, size_t nbytes, unsigned char *out) +{ + size_t i; + + for (i = 0; i < nbytes; i++) { + out[i] = ebcdic_to_ascii[buf[i]]; + } +} diff --git a/readcdf.c b/readcdf.c new file mode 100644 index 0000000..27031db --- /dev/null +++ b/readcdf.c @@ -0,0 +1,256 @@ +/*- + * Copyright (c) 2008 Christos Zoulas + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. 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 NETBSD FOUNDATION, INC. 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 FOUNDATION 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. + */ +#include "file.h" + +#ifndef lint +FILE_RCSID("@(#)$File: readcdf.c,v 1.11 2009/02/03 20:27:51 christos Exp $") +#endif + +#include <stdlib.h> +#include <unistd.h> +#include <string.h> +#include <time.h> +#include <ctype.h> + +#include "cdf.h" +#include "magic.h" + +#define NOTMIME(ms) (((ms)->flags & MAGIC_MIME) == 0) + +private int +cdf_file_property_info(struct magic_set *ms, const cdf_property_info_t *info, + size_t count) +{ + size_t i; + cdf_timestamp_t tp; + struct timespec ts; + char buf[64]; + const char *str = "vnd.ms-office"; + const char *s; + int len; + + for (i = 0; i < count; i++) { + cdf_print_property_name(buf, sizeof(buf), info[i].pi_id); + switch (info[i].pi_type) { + case CDF_SIGNED16: + if (NOTMIME(ms) && file_printf(ms, ", %s: %hd", buf, + info[i].pi_s16) == -1) + return -1; + break; + case CDF_SIGNED32: + if (NOTMIME(ms) && file_printf(ms, ", %s: %d", buf, + info[i].pi_s32) == -1) + return -1; + break; + case CDF_UNSIGNED32: + if (NOTMIME(ms) && file_printf(ms, ", %s: %u", buf, + info[i].pi_u32) == -1) + return -1; + break; + case CDF_LENGTH32_STRING: + len = info[i].pi_str.s_len; + if (len > 1) { + s = info[i].pi_str.s_buf; + if (NOTMIME(ms)) { + if (file_printf(ms, ", %s: %.*s", buf, + len, s) == -1) + return -1; + } else if (info[i].pi_id == + CDF_PROPERTY_NAME_OF_APPLICATION) { + if (strstr(s, "Word")) + str = "msword"; + else if (strstr(s, "Excel")) + str = "vnd.ms-excel"; + else if (strstr(s, "Powerpoint")) + str = "vnd.ms-powerpoint"; + } + } + break; + case CDF_FILETIME: + tp = info[i].pi_tp; + if (tp != 0) { + if (tp < 1000000000000000LL) { + char tbuf[64]; + cdf_print_elapsed_time(tbuf, + sizeof(tbuf), tp); + if (NOTMIME(ms) && file_printf(ms, + ", %s: %s", buf, tbuf) == -1) + return -1; + } else { + char *c, *ec; + cdf_timestamp_to_timespec(&ts, tp); + c = ctime(&ts.tv_sec); + if ((ec = strchr(c, '\n')) != NULL) + *ec = '\0'; + + if (NOTMIME(ms) && file_printf(ms, + ", %s: %s", buf, c) == -1) + return -1; + } + } + break; + case CDF_CLIPBOARD: + break; + default: + file_error(ms, 0, "Internal parsing error"); + return -1; + } + } + if (!NOTMIME(ms)) { + if (file_printf(ms, "application/%s", str) == -1) + return -1; + } + return 1; +} + +private int +cdf_file_summary_info(struct magic_set *ms, const cdf_stream_t *sst) +{ + cdf_summary_info_header_t si; + cdf_property_info_t *info; + size_t count; + int m; + + if (cdf_unpack_summary_info(sst, &si, &info, &count) == -1) { + if (si.si_byte_order != 0xfffe) + return 0; + else + return -1; + } + + if (si.si_byte_order != 0xfffe) + return 0; + + if (NOTMIME(ms)) { + if (file_printf(ms, "CDF V2 Document") == -1) + return -1; + + if (file_printf(ms, ", %s Endian", + si.si_byte_order == 0xfffe ? "Little" : "Big") == -1) + return -1; + switch (si.si_os) { + case 2: + if (file_printf(ms, ", Os: Windows, Version %d.%d", + si.si_os_version & 0xff, si.si_os_version >> 8) + == -1) + return -1; + break; + case 1: + if (file_printf(ms, ", Os: MacOS, Version %d.%d", + si.si_os_version >> 8, si.si_os_version & 0xff) + == -1) + return -1; + break; + default: + if (file_printf(ms, ", Os %d, Version: %d.%d", si.si_os, + si.si_os_version & 0xff, si.si_os_version >> 8) + == -1) + return -1; + break; + } + } + + m = cdf_file_property_info(ms, info, count); + free(info); + + return m; +} + +protected int +file_trycdf(struct magic_set *ms, int fd, const unsigned char *buf, + size_t nbytes) +{ + cdf_header_t h; + cdf_sat_t sat, ssat; + cdf_stream_t sst, scn; + cdf_dir_t dir; + int i; + (void)&nbytes; + (void)&buf; + + if (ms->flags & MAGIC_APPLE) + return 0; + if (cdf_read_header(fd, &h) == -1) + return 0; +#ifdef CDF_DEBUG + cdf_dump_header(&h); +#endif + + if (cdf_read_sat(fd, &h, &sat) == -1) { + file_error(ms, errno, "Can't read SAT"); + return -1; + } +#ifdef CDF_DEBUG + cdf_dump_sat("SAT", &h, &sat); +#endif + + if ((i = cdf_read_ssat(fd, &h, &sat, &ssat)) == -1) { + file_error(ms, errno, "Can't read SAT"); + goto out1; + } +#ifdef CDF_DEBUG + cdf_dump_sat("SSAT", &h, &ssat); +#endif + + if ((i = cdf_read_dir(fd, &h, &sat, &dir)) == -1) { + file_error(ms, errno, "Can't read directory"); + goto out2; + } + + if ((i = cdf_read_short_stream(fd, &h, &sat, &dir, &sst)) == -1) { + file_error(ms, errno, "Cannot read short stream"); + goto out3; + } + +#ifdef CDF_DEBUG + cdf_dump_dir(fd, &h, &sat, &ssat, &sst, &dir); +#endif + if ((i = cdf_read_summary_info(fd, &h, &sat, &ssat, &sst, &dir, &scn)) + == -1) { + /* Some files don't have summary info! */ +#ifdef notyet + file_error(ms, errno, "Can't read summary_info"); +#else + i = 0; +#endif + goto out4; + } +#ifdef CDF_DEBUG + cdf_dump_summary_info(&h, &scn); +#endif + if ((i = cdf_file_summary_info(ms, &scn)) == -1) + file_error(ms, errno, "Can't expand summary_info"); + free(scn.sst_tab); +out4: + free(sst.sst_tab); +out3: + free(dir.dir_tab); +out2: + free(ssat.sat_tab); +out1: + free(sat.sat_tab); + return i; +} |