summaryrefslogtreecommitdiffstats
path: root/usr.bin
diff options
context:
space:
mode:
authorrdivacky <rdivacky@FreeBSD.org>2009-09-08 15:55:13 +0000
committerrdivacky <rdivacky@FreeBSD.org>2009-09-08 15:55:13 +0000
commitc96da7a997a165a83d8eb6b9d0cbbdeb2f72a648 (patch)
tree194af88aae2180f7bb945ffb8e0cfffe9e928522 /usr.bin
parent57ba4d8ec0188487a846982bfab3bd067fec199e (diff)
downloadFreeBSD-src-c96da7a997a165a83d8eb6b9d0cbbdeb2f72a648.zip
FreeBSD-src-c96da7a997a165a83d8eb6b9d0cbbdeb2f72a648.tar.gz
Add C/c/f/p/v switches plus a bunch of minor fixes and cleanups.
Obtained from: NetBSD Approved by: des (maintainer) Approved by: ed (mentor, implicit)
Diffstat (limited to 'usr.bin')
-rw-r--r--usr.bin/unzip/unzip.141
-rw-r--r--usr.bin/unzip/unzip.c266
2 files changed, 265 insertions, 42 deletions
diff --git a/usr.bin/unzip/unzip.1 b/usr.bin/unzip/unzip.1
index 031cf3d..b6ee87a 100644
--- a/usr.bin/unzip/unzip.1
+++ b/usr.bin/unzip/unzip.1
@@ -25,7 +25,7 @@
.\"
.\" $FreeBSD$
.\"
-.Dd June 30, 2008
+.Dd September 7, 2009
.Dt UNZIP 1
.Os
.Sh NAME
@@ -33,7 +33,7 @@
.Nd extract files from a ZIP archive
.Sh SYNOPSIS
.Nm
-.Op Fl ajLlnoqtu
+.Op Fl aCcfjLlnopqtuv
.Op Fl d Ar dir
.Ar zipfile
.Sh DESCRIPTION
@@ -44,9 +44,22 @@ The following options are available:
.It Fl a
When extracting a text file, convert DOS-style line endings to
Unix-style line endings.
+.It Fl C
+Match file names case-insensitively.
+.It Fl c
+Extract to stdout/screen.
+When extracting files from the zipfile, they are written to stdout.
+This is similar to
+.Fl p ,
+but doesn't suppress normal output.
.It Fl d Ar dir
Extract files into the specified directory rather than the current
directory.
+.It Fl f
+Update existing.
+Extract only files from the zipfile if a file with the same name
+already exists on disk and is older than the former.
+Otherwise, the file is silently skipped.
.It Fl j
Ignore directories stored in the zipfile; instead, extract all files
directly into the extraction directory.
@@ -56,13 +69,19 @@ Convert the names of the extracted files and directories to lowercase.
List, rather than extract, the contents of the zipfile.
.It Fl n
No overwrite.
-When extacting a file from the zipfile, if a file with the same name
+When extracting a file from the zipfile, if a file with the same name
already exists on disk, the file is silently skipped.
.It Fl o
Overwrite.
-When extacting a file from the zipfile, if a file with the same name
+When extracting a file from the zipfile, if a file with the same name
already exists on disk, the existing file is replaced with the file
from the zipfile.
+.It Fl p
+Extract to stdout.
+When extracting files from the zipfile, they are written to stdout.
+The normal output is suppressed as if
+.Fl q
+was specified.
.It Fl q
Quiet: print less information while extracting.
.It Fl t
@@ -70,15 +89,25 @@ Test: do not extract anything, but verify the checksum of every file
in the archive.
.It Fl u
Update.
-When extacting a file from the zipfile, if a file with the same name
+When extracting a file from the zipfile, if a file with the same name
already exists on disk, the existing file is replaced with the file
from the zipfile if and only if the latter is newer than the former.
Otherwise, the file is silently skipped.
+.It Fl v
+List verbosely, rather than extract, the contents of the zipfile.
+This differs from
+.Fl l
+by using the long listing.
+Note that most of the data is currently fake and does not reflect the
+content of the archive.
+.It Fl x Ar pattern
+Exclude files matching the pattern
+.Ar pattern .
.El
.Pp
Note that only one of
.Fl n ,
-.Fl o
+.Fl o ,
and
.Fl u
may be specified.
diff --git a/usr.bin/unzip/unzip.c b/usr.bin/unzip/unzip.c
index 16373f5..f0856a4 100644
--- a/usr.bin/unzip/unzip.c
+++ b/usr.bin/unzip/unzip.c
@@ -1,4 +1,5 @@
/*-
+ * Copyright (c) 2009 Joerg Sonnenberger <joerg@NetBSD.org>
* Copyright (c) 2007-2008 Dag-Erling Coïdan Smørgrav
* All rights reserved.
*
@@ -51,15 +52,19 @@
/* command-line options */
static int a_opt; /* convert EOL */
+static int C_opt; /* match case-insensitively */
+static int c_opt; /* extract to stdout */
static const char *d_arg; /* directory */
+static int f_opt; /* update existing files only */
static int j_opt; /* junk directories */
static int L_opt; /* lowercase names */
-static int l_opt; /* list */
static int n_opt; /* never overwrite */
static int o_opt; /* always overwrite */
+static int p_opt; /* extract to stdout, quiet */
static int q_opt; /* quiet */
static int t_opt; /* test */
static int u_opt; /* update */
+static int v_opt; /* verbose/list */
/* time when unzip started */
static time_t now;
@@ -70,9 +75,6 @@ static int unzip_debug;
/* running on tty? */
static int tty;
-/* error flag for -t */
-static int test_failed;
-
/* convenience macro */
/* XXX should differentiate between ARCHIVE_{WARN,FAIL,RETRY} */
#define ac(call) \
@@ -162,7 +164,6 @@ static void
info(const char *fmt, ...)
{
va_list ap;
- int i;
if (q_opt && !unzip_debug)
return;
@@ -171,9 +172,10 @@ info(const char *fmt, ...)
va_end(ap);
fflush(stdout);
- for (i = 0; fmt[i] != '\0'; ++i)
- /* nothing */ ;
- noeol = !(i && fmt[i - 1] == '\n');
+ if (*fmt == '\0')
+ noeol = 1;
+ else
+ noeol = fmt[strlen(fmt) - 1] != '\n';
}
/* debug message (if unzip_debug) */
@@ -181,7 +183,6 @@ static void
debug(const char *fmt, ...)
{
va_list ap;
- int i;
if (!unzip_debug)
return;
@@ -190,9 +191,10 @@ debug(const char *fmt, ...)
va_end(ap);
fflush(stderr);
- for (i = 0; fmt[i] != '\0'; ++i)
- /* nothing */ ;
- noeol = !(i && fmt[i - 1] == '\n');
+ if (*fmt == '\0')
+ noeol = 1;
+ else
+ noeol = fmt[strlen(fmt) - 1] != '\n';
}
/* duplicate a path name, possibly converting to lower case */
@@ -200,7 +202,7 @@ static char *
pathdup(const char *path)
{
char *str;
- int len;
+ size_t i, len;
len = strlen(path);
while (len && path[len - 1] == '/')
@@ -209,8 +211,12 @@ pathdup(const char *path)
errno = ENOMEM;
error("malloc()");
}
- for (int i = 0; i < len; ++i)
- str[i] = L_opt ? tolower(path[i]) : path[i];
+ if (L_opt) {
+ for (i = 0; i < len; ++i)
+ str[i] = tolower((unsigned char)path[i]);
+ } else {
+ memcpy(str, path, len);
+ }
str[len] = '\0';
return (str);
@@ -221,7 +227,7 @@ static char *
pathcat(const char *prefix, const char *path)
{
char *str;
- int prelen, len;
+ size_t prelen, len;
prelen = prefix ? strlen(prefix) + 1 : 0;
len = strlen(path) + 1;
@@ -257,7 +263,7 @@ static void
add_pattern(struct pattern_list *list, const char *pattern)
{
struct pattern *entry;
- int len;
+ size_t len;
debug("adding pattern '%s'\n", pattern);
len = strlen(pattern);
@@ -265,7 +271,6 @@ add_pattern(struct pattern_list *list, const char *pattern)
errno = ENOMEM;
error("malloc()");
}
- memset(&entry->link, 0, sizeof entry->link);
memcpy(entry->pattern, pattern, len + 1);
STAILQ_INSERT_TAIL(list, entry, link);
}
@@ -279,7 +284,7 @@ match_pattern(struct pattern_list *list, const char *str)
struct pattern *entry;
STAILQ_FOREACH(entry, list, link) {
- if (fnmatch(entry->pattern, str, 0) == 0)
+ if (fnmatch(entry->pattern, str, C_opt ? FNM_CASEFOLD : 0) == 0)
return (1);
}
return (0);
@@ -427,9 +432,9 @@ extract_file(struct archive *a, struct archive_entry *e, const char *path)
/* look for existing file of same name */
if (lstat(path, &sb) == 0) {
- if (u_opt) {
+ if (u_opt || f_opt) {
/* check if up-to-date */
- if (S_ISREG(sb.st_mode) && sb.st_mtime > mtime)
+ if (S_ISREG(sb.st_mode) && sb.st_mtime >= mtime)
return;
(void)unlink(path);
} else if (o_opt) {
@@ -442,13 +447,16 @@ extract_file(struct archive *a, struct archive_entry *e, const char *path)
/* XXX ask user */
errorx("not implemented");
}
+ } else {
+ if (f_opt)
+ return;
}
if ((fd = open(path, O_RDWR|O_CREAT|O_TRUNC, mode)) < 0)
error("open('%s')", path);
/* loop over file contents and write to disk */
- info("x %s", path);
+ info(" extracting: %s", path);
text = a_opt;
warn = 0;
cr = 0;
@@ -618,40 +626,172 @@ extract(struct archive *a, struct archive_entry *e)
free(pathname);
}
+static void
+extract_stdout(struct archive *a, struct archive_entry *e)
+{
+ char *pathname;
+ mode_t filetype;
+ int cr, text, warn;
+ ssize_t len;
+ unsigned char *p, *q, *end;
+
+ pathname = pathdup(archive_entry_pathname(e));
+ filetype = archive_entry_filetype(e);
+
+ /* I don't think this can happen in a zipfile.. */
+ if (!S_ISDIR(filetype) && !S_ISREG(filetype)) {
+ warningx("skipping non-regular entry '%s'", pathname);
+ ac(archive_read_data_skip(a));
+ free(pathname);
+ return;
+ }
+
+ /* skip directories in -j case */
+ if (S_ISDIR(filetype)) {
+ ac(archive_read_data_skip(a));
+ free(pathname);
+ return;
+ }
+
+ /* apply include / exclude patterns */
+ if (!accept_pathname(pathname)) {
+ ac(archive_read_data_skip(a));
+ free(pathname);
+ return;
+ }
+
+ if (c_opt)
+ info("x %s\n", pathname);
+
+ text = a_opt;
+ warn = 0;
+ cr = 0;
+ for (int n = 0; ; n++) {
+ len = archive_read_data(a, buffer, sizeof buffer);
+
+ if (len < 0)
+ ac(len);
+
+ /* left over CR from previous buffer */
+ if (a_opt && cr) {
+ if (len == 0 || buffer[0] != '\n') {
+ if (fwrite("\r", 1, 1, stderr) != 1)
+ error("write('%s')", pathname);
+ }
+ cr = 0;
+ }
+
+ /* EOF */
+ if (len == 0)
+ break;
+ end = buffer + len;
+
+ /*
+ * Detect whether this is a text file. The correct way to
+ * do this is to check the least significant bit of the
+ * "internal file attributes" field of the corresponding
+ * file header in the central directory, but libarchive
+ * does not read the central directory, so we have to
+ * guess by looking for non-ASCII characters in the
+ * buffer. Hopefully we won't guess wrong. If we do
+ * guess wrong, we print a warning message later.
+ */
+ if (a_opt && n == 0) {
+ for (p = buffer; p < end; ++p) {
+ if (!isascii((unsigned char)*p)) {
+ text = 0;
+ break;
+ }
+ }
+ }
+
+ /* simple case */
+ if (!a_opt || !text) {
+ if (fwrite(buffer, 1, len, stdout) != (size_t)len)
+ error("write('%s')", pathname);
+ continue;
+ }
+
+ /* hard case: convert \r\n to \n (sigh...) */
+ for (p = buffer; p < end; p = q + 1) {
+ for (q = p; q < end; q++) {
+ if (!warn && !isascii(*q)) {
+ warningx("%s may be corrupted due"
+ " to weak text file detection"
+ " heuristic", pathname);
+ warn = 1;
+ }
+ if (q[0] != '\r')
+ continue;
+ if (&q[1] == end) {
+ cr = 1;
+ break;
+ }
+ if (q[1] == '\n')
+ break;
+ }
+ if (fwrite(p, 1, q - p, stdout) != (size_t)(q - p))
+ error("write('%s')", pathname);
+ }
+ }
+
+ free(pathname);
+}
+
/*
* Print the name of an entry to stdout.
*/
static void
list(struct archive *a, struct archive_entry *e)
{
+ char buf[20];
+ time_t mtime;
- printf("%s\n", archive_entry_pathname(e));
+ mtime = archive_entry_mtime(e);
+ strftime(buf, sizeof(buf), "%m-%d-%g %R", localtime(&mtime));
+
+ if (v_opt == 1) {
+ printf(" %8ju %s %s\n",
+ (uintmax_t)archive_entry_size(e),
+ buf, archive_entry_pathname(e));
+ } else if (v_opt == 2) {
+ printf("%8ju Stored %7ju 0%% %s %08x %s\n",
+ (uintmax_t)archive_entry_size(e),
+ (uintmax_t)archive_entry_size(e),
+ buf,
+ 0U,
+ archive_entry_pathname(e));
+ }
ac(archive_read_data_skip(a));
}
/*
* Extract to memory to check CRC
*/
-static void
+static int
test(struct archive *a, struct archive_entry *e)
{
ssize_t len;
+ int error_count;
+ error_count = 0;
if (S_ISDIR(archive_entry_filetype(e)))
- return;
+ return 0;
- info("%s ", archive_entry_pathname(e));
+ info(" testing: %s\t", archive_entry_pathname(e));
while ((len = archive_read_data(a, buffer, sizeof buffer)) > 0)
/* nothing */;
if (len < 0) {
- info("%s\n", archive_error_string(a));
- ++test_failed;
+ info(" %s\n", archive_error_string(a));
+ ++error_count;
} else {
- info("OK\n");
+ info(" OK\n");
}
/* shouldn't be necessary, but it doesn't hurt */
ac(archive_read_data_skip(a));
+
+ return error_count;
}
@@ -665,6 +805,7 @@ unzip(const char *fn)
struct archive *a;
struct archive_entry *e;
int fd, ret;
+ uintmax_t total_size, file_count, error_count;
if ((fd = open(fn, O_RDONLY)) < 0)
error("%s", fn);
@@ -673,33 +814,69 @@ unzip(const char *fn)
ac(archive_read_support_format_zip(a));
ac(archive_read_open_fd(a, fd, 8192));
+ printf("Archive: %s\n", fn);
+ if (v_opt == 1) {
+ printf(" Length Date Time Name\n");
+ printf(" -------- ---- ---- ----\n");
+ } else if (v_opt == 2) {
+ printf(" Length Method Size Ratio Date Time CRC-32 Name\n");
+ printf("-------- ------ ------- ----- ---- ---- ------ ----\n");
+ }
+
+ total_size = 0;
+ file_count = 0;
+ error_count = 0;
for (;;) {
ret = archive_read_next_header(a, &e);
if (ret == ARCHIVE_EOF)
break;
ac(ret);
if (t_opt)
- test(a, e);
- else if (l_opt)
+ error_count += test(a, e);
+ else if (v_opt)
list(a, e);
+ else if (p_opt || c_opt)
+ extract_stdout(a, e);
else
extract(a, e);
+
+ total_size += archive_entry_size(e);
+ ++file_count;
+ }
+
+ if (v_opt == 1) {
+ printf(" -------- -------\n");
+ printf(" %8ju %ju file%s\n",
+ total_size, file_count, file_count != 1 ? "s" : "");
+ } else if (v_opt == 2) {
+ printf("-------- ------- --- -------\n");
+ printf("%8ju %7ju 0%% %ju file%s\n",
+ total_size, total_size, file_count,
+ file_count != 1 ? "s" : "");
}
ac(archive_read_close(a));
(void)archive_read_finish(a);
+
if (close(fd) != 0)
error("%s", fn);
- if (t_opt && test_failed)
- errorx("%d checksum error(s) found.", test_failed);
+ if (t_opt) {
+ if (error_count > 0) {
+ errorx("%d checksum error(s) found.", error_count);
+ }
+ else {
+ printf("No errors detected in compressed data of %s.\n",
+ fn);
+ }
+ }
}
static void
usage(void)
{
- fprintf(stderr, "usage: unzip [-ajLlnoqtu] [-d dir] zipfile\n");
+ fprintf(stderr, "usage: unzip [-aCcfjLlnopqtuv] [-d dir] [-x pattern] zipfile\n");
exit(1);
}
@@ -709,14 +886,23 @@ getopts(int argc, char *argv[])
int opt;
optreset = optind = 1;
- while ((opt = getopt(argc, argv, "ad:jLlnoqtux:")) != -1)
+ while ((opt = getopt(argc, argv, "aCcd:fjLlnopqtuvx:")) != -1)
switch (opt) {
case 'a':
a_opt = 1;
break;
+ case 'C':
+ C_opt = 1;
+ break;
+ case 'c':
+ c_opt = 1;
+ break;
case 'd':
d_arg = optarg;
break;
+ case 'f':
+ f_opt = 1;
+ break;
case 'j':
j_opt = 1;
break;
@@ -724,13 +910,18 @@ getopts(int argc, char *argv[])
L_opt = 1;
break;
case 'l':
- l_opt = 1;
+ if (v_opt == 0)
+ v_opt = 1;
break;
case 'n':
n_opt = 1;
break;
case 'o':
o_opt = 1;
+ q_opt = 1;
+ break;
+ case 'p':
+ p_opt = 1;
break;
case 'q':
q_opt = 1;
@@ -741,6 +932,9 @@ getopts(int argc, char *argv[])
case 'u':
u_opt = 1;
break;
+ case 'v':
+ v_opt = 2;
+ break;
case 'x':
add_pattern(&exclude, optarg);
break;
OpenPOWER on IntegriCloud