mirror/git
1
0
Fork 0
mirror of https://github.com/git/git.git synced 2024-11-18 20:53:55 +01:00
Code
0cb050abc2
git/archive-zip.c

281 lines
7.3 KiB
C
Raw Normal View History

Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
/*
* Copyright (c) 2006 Rene Scharfe
*/
#include "cache.h"
git-archive: wire up ZIP format. Again, this is based on Rene Scharfe's earlier patch, but uses the archiver support introduced by the previous patch. Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com> Acked-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-07 15:12:04 +02:00
#include "archive.h"
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
static int zip_date;
static int zip_time;
static unsigned char *zip_dir;
static unsigned int zip_dir_size;
static unsigned int zip_offset;
static unsigned int zip_dir_offset;
static unsigned int zip_dir_entries;
#define ZIP_DIRECTORY_MIN_SIZE (1024 * 1024)
struct zip_local_header {
unsigned char magic[4];
unsigned char version[2];
unsigned char flags[2];
unsigned char compression_method[2];
unsigned char mtime[2];
unsigned char mdate[2];
unsigned char crc32[4];
unsigned char compressed_size[4];
unsigned char size[4];
unsigned char filename_length[2];
unsigned char extra_length[2];
archive-zip: don't use sizeof(struct ...) We can't rely on sizeof(struct zip_*) returning the sum of all struct members. At least on ARM padding is added at the end, as Gerrit Pape reported. This fixes the problem but still lets the compiler do the summing by introducing explicit padding at the end of the structs and then taking its offset as the combined size of the preceding members. As Junio correctly notes, the _end[] marker array's size must be greater than zero for compatibility with compilers other than gcc. The space wasted by the markers can safely be neglected because we only have one instance of each struct, i.e. in sum 3 wasted bytes on i386, and 0 on ARM. :) We still rely on the compiler to not add padding between the struct members, but that's reasonable given that all of them are unsigned char arrays. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-11-23 23:02:37 +01:00
unsigned char _end[1];
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
};
struct zip_dir_header {
unsigned char magic[4];
unsigned char creator_version[2];
unsigned char version[2];
unsigned char flags[2];
unsigned char compression_method[2];
unsigned char mtime[2];
unsigned char mdate[2];
unsigned char crc32[4];
unsigned char compressed_size[4];
unsigned char size[4];
unsigned char filename_length[2];
unsigned char extra_length[2];
unsigned char comment_length[2];
unsigned char disk[2];
unsigned char attr1[2];
unsigned char attr2[4];
unsigned char offset[4];
archive-zip: don't use sizeof(struct ...) We can't rely on sizeof(struct zip_*) returning the sum of all struct members. At least on ARM padding is added at the end, as Gerrit Pape reported. This fixes the problem but still lets the compiler do the summing by introducing explicit padding at the end of the structs and then taking its offset as the combined size of the preceding members. As Junio correctly notes, the _end[] marker array's size must be greater than zero for compatibility with compilers other than gcc. The space wasted by the markers can safely be neglected because we only have one instance of each struct, i.e. in sum 3 wasted bytes on i386, and 0 on ARM. :) We still rely on the compiler to not add padding between the struct members, but that's reasonable given that all of them are unsigned char arrays. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-11-23 23:02:37 +01:00
unsigned char _end[1];
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
};
struct zip_dir_trailer {
unsigned char magic[4];
unsigned char disk[2];
unsigned char directory_start_disk[2];
unsigned char entries_on_this_disk[2];
unsigned char entries[2];
unsigned char size[4];
unsigned char offset[4];
unsigned char comment_length[2];
archive-zip: don't use sizeof(struct ...) We can't rely on sizeof(struct zip_*) returning the sum of all struct members. At least on ARM padding is added at the end, as Gerrit Pape reported. This fixes the problem but still lets the compiler do the summing by introducing explicit padding at the end of the structs and then taking its offset as the combined size of the preceding members. As Junio correctly notes, the _end[] marker array's size must be greater than zero for compatibility with compilers other than gcc. The space wasted by the markers can safely be neglected because we only have one instance of each struct, i.e. in sum 3 wasted bytes on i386, and 0 on ARM. :) We still rely on the compiler to not add padding between the struct members, but that's reasonable given that all of them are unsigned char arrays. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-11-23 23:02:37 +01:00
unsigned char _end[1];
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
};
archive-zip: don't use sizeof(struct ...) We can't rely on sizeof(struct zip_*) returning the sum of all struct members. At least on ARM padding is added at the end, as Gerrit Pape reported. This fixes the problem but still lets the compiler do the summing by introducing explicit padding at the end of the structs and then taking its offset as the combined size of the preceding members. As Junio correctly notes, the _end[] marker array's size must be greater than zero for compatibility with compilers other than gcc. The space wasted by the markers can safely be neglected because we only have one instance of each struct, i.e. in sum 3 wasted bytes on i386, and 0 on ARM. :) We still rely on the compiler to not add padding between the struct members, but that's reasonable given that all of them are unsigned char arrays. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-11-23 23:02:37 +01:00
/*
* On ARM, padding is added at the end of the struct, so a simple
* sizeof(struct ...) reports two bytes more than the payload size
* we're interested in.
*/
#define ZIP_LOCAL_HEADER_SIZE offsetof(struct zip_local_header, _end)
#define ZIP_DIR_HEADER_SIZE offsetof(struct zip_dir_header, _end)
#define ZIP_DIR_TRAILER_SIZE offsetof(struct zip_dir_trailer, _end)
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
static void copy_le16(unsigned char *dest, unsigned int n)
{
dest[0] = 0xff & n;
dest[1] = 0xff & (n >> 010);
}
static void copy_le32(unsigned char *dest, unsigned int n)
{
dest[0] = 0xff & n;
dest[1] = 0xff & (n >> 010);
dest[2] = 0xff & (n >> 020);
dest[3] = 0xff & (n >> 030);
}
static void *zlib_deflate(void *data, unsigned long size,
archive: make zip compression level independent from core git zlib_compression_level is the compression level used for git's object store. It's 1 by default, which is the fastest setting. This variable is also used as the default compression level for ZIP archives created by git archive. For archives, however, zlib's own default of 6 is more appropriate, as it's favouring small size over speed -- archive creation is not that performance critical most of the time. This patch makes git archive independent from git's internal compression level setting. It affects invocations of git archive without explicitly specified compression level option, only. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-18 16:30:32 +02:00
int compression_level, unsigned long *compressed_size)
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
{
z_stream stream;
unsigned long maxsize;
void *buffer;
int result;
memset(&stream, 0, sizeof(stream));
archive: make zip compression level independent from core git zlib_compression_level is the compression level used for git's object store. It's 1 by default, which is the fastest setting. This variable is also used as the default compression level for ZIP archives created by git archive. For archives, however, zlib's own default of 6 is more appropriate, as it's favouring small size over speed -- archive creation is not that performance critical most of the time. This patch makes git archive independent from git's internal compression level setting. It affects invocations of git archive without explicitly specified compression level option, only. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-18 16:30:32 +02:00
deflateInit(&stream, compression_level);
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
maxsize = deflateBound(&stream, size);
buffer = xmalloc(maxsize);
stream.next_in = data;
stream.avail_in = size;
stream.next_out = buffer;
stream.avail_out = maxsize;
do {
result = deflate(&stream, Z_FINISH);
} while (result == Z_OK);
if (result != Z_STREAM_END) {
free(buffer);
return NULL;
}
deflateEnd(&stream);
*compressed_size = stream.total_out;
return buffer;
}
archive: centralize archive entry writing Add the exported function write_archive_entries() to archive.c, which uses the new ability of read_tree_recursive() to pass a context pointer to its callback in order to centralize previously duplicated code. The new callback function write_archive_entry() does the work that every archiver backend needs to do: loading file contents, entering subdirectories, handling file attributes, constructing the full path of the entry. All that done, it calls the backend specific write_archive_entry_fn_t function. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-14 21:22:24 +02:00
static int write_zip_entry(struct archiver_args *args,
const unsigned char *sha1, const char *path, size_t pathlen,
unsigned int mode, void *buffer, unsigned long size)
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
{
struct zip_local_header header;
struct zip_dir_header dirent;
git-archive --format=zip: add symlink support Add symlink support to ZIP file creation, and a few tests. This implementation sets the "version made by" field (creator_version) to Unix for symlinks, only; regular files and directories are still marked as originating from FAT/VFAT/NTFS. Also set "external file attributes" (attr2) to 0 for regular files and 16 for directories (FAT attribute), and to the file mode for symlinks. We could always set the creator_version to Unix and include the mode, but then Info-ZIP unzip would set the mode of the extracted files to *exactly* the value stored in attr2. The FAT trick makes it apply the umask instead. Note: FAT has no executable bit, so this information is not stored in the ZIP file. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-10-07 01:47:35 +02:00
unsigned long attr2;
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
unsigned long compressed_size;
unsigned long uncompressed_size;
unsigned long crc;
unsigned long direntsize;
int method;
unsigned char *out;
void *deflated = NULL;
sparse fix: Using plain integer as NULL pointer Z_NULL is defined as 0, use a proper NULL pointer in its stead. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-11-18 13:06:56 +01:00
crc = crc32(0, NULL, 0);
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
if (pathlen > 0xffff) {
archive: centralize archive entry writing Add the exported function write_archive_entries() to archive.c, which uses the new ability of read_tree_recursive() to pass a context pointer to its callback in order to centralize previously duplicated code. The new callback function write_archive_entry() does the work that every archiver backend needs to do: loading file contents, entering subdirectories, handling file attributes, constructing the full path of the entry. All that done, it calls the backend specific write_archive_entry_fn_t function. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-14 21:22:24 +02:00
return error("path too long (%d chars, SHA1: %s): %s",
(int)pathlen, sha1_to_hex(sha1), path);
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
}
rename dirlink to gitlink. Unify naming of plumbing dirlink/gitlink concept: git ls-files -z '*.[ch]' | xargs -0 perl -pi -e 's/dirlink/gitlink/g;' -e 's/DIRLNK/GITLINK/g;' Signed-off-by: Junio C Hamano <junkio@cox.net>
2007-05-21 22:08:28 +02:00
if (S_ISDIR(mode) || S_ISGITLINK(mode)) {
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
method = 0;
git-archive --format=zip: add symlink support Add symlink support to ZIP file creation, and a few tests. This implementation sets the "version made by" field (creator_version) to Unix for symlinks, only; regular files and directories are still marked as originating from FAT/VFAT/NTFS. Also set "external file attributes" (attr2) to 0 for regular files and 16 for directories (FAT attribute), and to the file mode for symlinks. We could always set the creator_version to Unix and include the mode, but then Info-ZIP unzip would set the mode of the extracted files to *exactly* the value stored in attr2. The FAT trick makes it apply the umask instead. Note: FAT has no executable bit, so this information is not stored in the ZIP file. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-10-07 01:47:35 +02:00
attr2 = 16;
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
out = NULL;
uncompressed_size = 0;
compressed_size = 0;
git-archive --format=zip: add symlink support Add symlink support to ZIP file creation, and a few tests. This implementation sets the "version made by" field (creator_version) to Unix for symlinks, only; regular files and directories are still marked as originating from FAT/VFAT/NTFS. Also set "external file attributes" (attr2) to 0 for regular files and 16 for directories (FAT attribute), and to the file mode for symlinks. We could always set the creator_version to Unix and include the mode, but then Info-ZIP unzip would set the mode of the extracted files to *exactly* the value stored in attr2. The FAT trick makes it apply the umask instead. Note: FAT has no executable bit, so this information is not stored in the ZIP file. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-10-07 01:47:35 +02:00
} else if (S_ISREG(mode) || S_ISLNK(mode)) {
method = 0;
preserve executable bits in zip archives Correct `git-archive --format=zip' command to preserve executable bits in zip archives. Signed-off-by: Junio C Hamano <gitster@pobox.com>
2007-09-16 19:07:38 +02:00
attr2 = S_ISLNK(mode) ? ((mode | 0777) << 16) :
(mode & 0111) ? ((mode) << 16) : 0;
archive: make zip compression level independent from core git zlib_compression_level is the compression level used for git's object store. It's 1 by default, which is the fastest setting. This variable is also used as the default compression level for ZIP archives created by git archive. For archives, however, zlib's own default of 6 is more appropriate, as it's favouring small size over speed -- archive creation is not that performance critical most of the time. This patch makes git archive independent from git's internal compression level setting. It affects invocations of git archive without explicitly specified compression level option, only. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-18 16:30:32 +02:00
if (S_ISREG(mode) && args->compression_level != 0)
git-archive --format=zip: add symlink support Add symlink support to ZIP file creation, and a few tests. This implementation sets the "version made by" field (creator_version) to Unix for symlinks, only; regular files and directories are still marked as originating from FAT/VFAT/NTFS. Also set "external file attributes" (attr2) to 0 for regular files and 16 for directories (FAT attribute), and to the file mode for symlinks. We could always set the creator_version to Unix and include the mode, but then Info-ZIP unzip would set the mode of the extracted files to *exactly* the value stored in attr2. The FAT trick makes it apply the umask instead. Note: FAT has no executable bit, so this information is not stored in the ZIP file. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-10-07 01:47:35 +02:00
method = 8;
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
crc = crc32(crc, buffer, size);
out = buffer;
uncompressed_size = size;
compressed_size = size;
} else {
archive: centralize archive entry writing Add the exported function write_archive_entries() to archive.c, which uses the new ability of read_tree_recursive() to pass a context pointer to its callback in order to centralize previously duplicated code. The new callback function write_archive_entry() does the work that every archiver backend needs to do: loading file contents, entering subdirectories, handling file attributes, constructing the full path of the entry. All that done, it calls the backend specific write_archive_entry_fn_t function. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-14 21:22:24 +02:00
return error("unsupported file mode: 0%o (SHA1: %s)", mode,
sha1_to_hex(sha1));
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
}
if (method == 8) {
archive: make zip compression level independent from core git zlib_compression_level is the compression level used for git's object store. It's 1 by default, which is the fastest setting. This variable is also used as the default compression level for ZIP archives created by git archive. For archives, however, zlib's own default of 6 is more appropriate, as it's favouring small size over speed -- archive creation is not that performance critical most of the time. This patch makes git archive independent from git's internal compression level setting. It affects invocations of git archive without explicitly specified compression level option, only. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-18 16:30:32 +02:00
deflated = zlib_deflate(buffer, size, args->compression_level,
&compressed_size);
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
if (deflated && compressed_size - 6 < size) {
/* ZLIB --> raw compressed data (see RFC 1950) */
/* CMF and FLG ... */
out = (unsigned char *)deflated + 2;
compressed_size -= 6; /* ... and ADLER32 */
} else {
method = 0;
compressed_size = size;
}
}
/* make sure we have enough free space in the dictionary */
archive-zip: don't use sizeof(struct ...) We can't rely on sizeof(struct zip_*) returning the sum of all struct members. At least on ARM padding is added at the end, as Gerrit Pape reported. This fixes the problem but still lets the compiler do the summing by introducing explicit padding at the end of the structs and then taking its offset as the combined size of the preceding members. As Junio correctly notes, the _end[] marker array's size must be greater than zero for compatibility with compilers other than gcc. The space wasted by the markers can safely be neglected because we only have one instance of each struct, i.e. in sum 3 wasted bytes on i386, and 0 on ARM. :) We still rely on the compiler to not add padding between the struct members, but that's reasonable given that all of them are unsigned char arrays. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-11-23 23:02:37 +01:00
direntsize = ZIP_DIR_HEADER_SIZE + pathlen;
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
while (zip_dir_size < zip_dir_offset + direntsize) {
zip_dir_size += ZIP_DIRECTORY_MIN_SIZE;
zip_dir = xrealloc(zip_dir, zip_dir_size);
}
copy_le32(dirent.magic, 0x02014b50);
preserve executable bits in zip archives Correct `git-archive --format=zip' command to preserve executable bits in zip archives. Signed-off-by: Junio C Hamano <gitster@pobox.com>
2007-09-16 19:07:38 +02:00
copy_le16(dirent.creator_version,
S_ISLNK(mode) || (S_ISREG(mode) && (mode & 0111)) ? 0x0317 : 0);
git-archive --format=zip: use default version ID Use 10 for the "version needed to extract" field. This is the default value, and we want to use it because we don't do anything special. Info-ZIP's zip uses it, too. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-10-07 01:47:24 +02:00
copy_le16(dirent.version, 10);
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
copy_le16(dirent.flags, 0);
copy_le16(dirent.compression_method, method);
copy_le16(dirent.mtime, zip_time);
copy_le16(dirent.mdate, zip_date);
copy_le32(dirent.crc32, crc);
copy_le32(dirent.compressed_size, compressed_size);
copy_le32(dirent.size, uncompressed_size);
copy_le16(dirent.filename_length, pathlen);
copy_le16(dirent.extra_length, 0);
copy_le16(dirent.comment_length, 0);
copy_le16(dirent.disk, 0);
copy_le16(dirent.attr1, 0);
git-archive --format=zip: add symlink support Add symlink support to ZIP file creation, and a few tests. This implementation sets the "version made by" field (creator_version) to Unix for symlinks, only; regular files and directories are still marked as originating from FAT/VFAT/NTFS. Also set "external file attributes" (attr2) to 0 for regular files and 16 for directories (FAT attribute), and to the file mode for symlinks. We could always set the creator_version to Unix and include the mode, but then Info-ZIP unzip would set the mode of the extracted files to *exactly* the value stored in attr2. The FAT trick makes it apply the umask instead. Note: FAT has no executable bit, so this information is not stored in the ZIP file. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-10-07 01:47:35 +02:00
copy_le32(dirent.attr2, attr2);
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
copy_le32(dirent.offset, zip_offset);
archive-zip: don't use sizeof(struct ...) We can't rely on sizeof(struct zip_*) returning the sum of all struct members. At least on ARM padding is added at the end, as Gerrit Pape reported. This fixes the problem but still lets the compiler do the summing by introducing explicit padding at the end of the structs and then taking its offset as the combined size of the preceding members. As Junio correctly notes, the _end[] marker array's size must be greater than zero for compatibility with compilers other than gcc. The space wasted by the markers can safely be neglected because we only have one instance of each struct, i.e. in sum 3 wasted bytes on i386, and 0 on ARM. :) We still rely on the compiler to not add padding between the struct members, but that's reasonable given that all of them are unsigned char arrays. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-11-23 23:02:37 +01:00
memcpy(zip_dir + zip_dir_offset, &dirent, ZIP_DIR_HEADER_SIZE);
zip_dir_offset += ZIP_DIR_HEADER_SIZE;
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
memcpy(zip_dir + zip_dir_offset, path, pathlen);
zip_dir_offset += pathlen;
zip_dir_entries++;
copy_le32(header.magic, 0x04034b50);
git-archive --format=zip: use default version ID Use 10 for the "version needed to extract" field. This is the default value, and we want to use it because we don't do anything special. Info-ZIP's zip uses it, too. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-10-07 01:47:24 +02:00
copy_le16(header.version, 10);
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
copy_le16(header.flags, 0);
copy_le16(header.compression_method, method);
copy_le16(header.mtime, zip_time);
copy_le16(header.mdate, zip_date);
copy_le32(header.crc32, crc);
copy_le32(header.compressed_size, compressed_size);
copy_le32(header.size, uncompressed_size);
copy_le16(header.filename_length, pathlen);
copy_le16(header.extra_length, 0);
archive-zip: don't use sizeof(struct ...) We can't rely on sizeof(struct zip_*) returning the sum of all struct members. At least on ARM padding is added at the end, as Gerrit Pape reported. This fixes the problem but still lets the compiler do the summing by introducing explicit padding at the end of the structs and then taking its offset as the combined size of the preceding members. As Junio correctly notes, the _end[] marker array's size must be greater than zero for compatibility with compilers other than gcc. The space wasted by the markers can safely be neglected because we only have one instance of each struct, i.e. in sum 3 wasted bytes on i386, and 0 on ARM. :) We still rely on the compiler to not add padding between the struct members, but that's reasonable given that all of them are unsigned char arrays. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-11-23 23:02:37 +01:00
write_or_die(1, &header, ZIP_LOCAL_HEADER_SIZE);
zip_offset += ZIP_LOCAL_HEADER_SIZE;
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
write_or_die(1, path, pathlen);
zip_offset += pathlen;
if (compressed_size > 0) {
write_or_die(1, out, compressed_size);
zip_offset += compressed_size;
}
free(deflated);
archive: centralize archive entry writing Add the exported function write_archive_entries() to archive.c, which uses the new ability of read_tree_recursive() to pass a context pointer to its callback in order to centralize previously duplicated code. The new callback function write_archive_entry() does the work that every archiver backend needs to do: loading file contents, entering subdirectories, handling file attributes, constructing the full path of the entry. All that done, it calls the backend specific write_archive_entry_fn_t function. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-14 21:22:24 +02:00
return 0;
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
}
static void write_zip_trailer(const unsigned char *sha1)
{
struct zip_dir_trailer trailer;
copy_le32(trailer.magic, 0x06054b50);
copy_le16(trailer.disk, 0);
copy_le16(trailer.directory_start_disk, 0);
copy_le16(trailer.entries_on_this_disk, zip_dir_entries);
copy_le16(trailer.entries, zip_dir_entries);
copy_le32(trailer.size, zip_dir_offset);
copy_le32(trailer.offset, zip_offset);
copy_le16(trailer.comment_length, sha1 ? 40 : 0);
write_or_die(1, zip_dir, zip_dir_offset);
archive-zip: don't use sizeof(struct ...) We can't rely on sizeof(struct zip_*) returning the sum of all struct members. At least on ARM padding is added at the end, as Gerrit Pape reported. This fixes the problem but still lets the compiler do the summing by introducing explicit padding at the end of the structs and then taking its offset as the combined size of the preceding members. As Junio correctly notes, the _end[] marker array's size must be greater than zero for compatibility with compilers other than gcc. The space wasted by the markers can safely be neglected because we only have one instance of each struct, i.e. in sum 3 wasted bytes on i386, and 0 on ARM. :) We still rely on the compiler to not add padding between the struct members, but that's reasonable given that all of them are unsigned char arrays. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-11-23 23:02:37 +01:00
write_or_die(1, &trailer, ZIP_DIR_TRAILER_SIZE);
Add git-zip-tree In the Windows world ZIP files are better supported than tar files. Windows even includes built-in support for ZIP files nowadays. git-zip-tree is similar to git-tar-tree; it creates ZIP files out of git trees. It stores the commit ID (if available) in a ZIP file comment which can be extracted by unzip. There's still quite some room for improvement: this initial version supports no symlinks, calls write() way too often (three times per file) and there is no unit test. [jc: with a minor typefix to avoid void* arithmetic] Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-08-26 23:19:21 +02:00
if (sha1)
write_or_die(1, sha1_to_hex(sha1), 40);
}
static void dos_time(time_t *time, int *dos_date, int *dos_time)
{
struct tm *t = localtime(time);
*dos_date = t->tm_mday + (t->tm_mon + 1) * 32 +
(t->tm_year + 1900 - 1980) * 512;
*dos_time = t->tm_sec / 2 + t->tm_min * 32 + t->tm_hour * 2048;
}
git-archive: wire up ZIP format. Again, this is based on Rene Scharfe's earlier patch, but uses the archiver support introduced by the previous patch. Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com> Acked-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-07 15:12:04 +02:00
int write_zip_archive(struct archiver_args *args)
{
archive: centralize archive entry writing Add the exported function write_archive_entries() to archive.c, which uses the new ability of read_tree_recursive() to pass a context pointer to its callback in order to centralize previously duplicated code. The new callback function write_archive_entry() does the work that every archiver backend needs to do: loading file contents, entering subdirectories, handling file attributes, constructing the full path of the entry. All that done, it calls the backend specific write_archive_entry_fn_t function. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-14 21:22:24 +02:00
int err;
git-archive: wire up ZIP format. Again, this is based on Rene Scharfe's earlier patch, but uses the archiver support introduced by the previous patch. Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com> Acked-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-07 15:12:04 +02:00
dos_time(&args->time, &zip_date, &zip_time);
zip_dir = xmalloc(ZIP_DIRECTORY_MIN_SIZE);
zip_dir_size = ZIP_DIRECTORY_MIN_SIZE;
archive: centralize archive entry writing Add the exported function write_archive_entries() to archive.c, which uses the new ability of read_tree_recursive() to pass a context pointer to its callback in order to centralize previously duplicated code. The new callback function write_archive_entry() does the work that every archiver backend needs to do: loading file contents, entering subdirectories, handling file attributes, constructing the full path of the entry. All that done, it calls the backend specific write_archive_entry_fn_t function. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-14 21:22:24 +02:00
err = write_archive_entries(args, write_zip_entry);
if (!err)
write_zip_trailer(args->commit_sha1);
git-archive: wire up ZIP format. Again, this is based on Rene Scharfe's earlier patch, but uses the archiver support introduced by the previous patch. Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com> Acked-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-07 15:12:04 +02:00
free(zip_dir);
archive: centralize archive entry writing Add the exported function write_archive_entries() to archive.c, which uses the new ability of read_tree_recursive() to pass a context pointer to its callback in order to centralize previously duplicated code. The new callback function write_archive_entry() does the work that every archiver backend needs to do: loading file contents, entering subdirectories, handling file attributes, constructing the full path of the entry. All that done, it calls the backend specific write_archive_entry_fn_t function. Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-07-14 21:22:24 +02:00
return err;
git-archive: wire up ZIP format. Again, this is based on Rene Scharfe's earlier patch, but uses the archiver support introduced by the previous patch. Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com> Acked-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-09-07 15:12:04 +02:00
}
Copy Permalink