diff options
| -rw-r--r-- | .gitignore | 2 | ||||
| -rw-r--r-- | Documentation/git-clone.txt | 21 | ||||
| -rw-r--r-- | Makefile | 9 | ||||
| -rw-r--r-- | daemon.c | 2 | ||||
| -rw-r--r-- | diff-lib.c (renamed from diff.c) | 7 | ||||
| -rwxr-xr-x | git-merge.sh | 2 | ||||
| -rwxr-xr-x | git-repack.sh | 8 | ||||
| -rw-r--r-- | git.c | 2 | ||||
| -rw-r--r-- | gsimm.c | 94 | ||||
| -rw-r--r-- | gsimm.h | 28 | ||||
| -rw-r--r-- | rabinpoly.c | 154 | ||||
| -rw-r--r-- | rabinpoly.h | 31 | ||||
| -rw-r--r-- | revision.c | 1 | ||||
| -rw-r--r-- | sha1_name.c | 16 | ||||
| -rw-r--r-- | templates/hooks--pre-commit | 3 | ||||
| -rw-r--r-- | test-gsimm.c | 209 | ||||
| -rw-r--r-- | tree-walk.c | 50 | ||||
| -rw-r--r-- | tree-walk.h | 2 | ||||
| -rw-r--r-- | unresolve.c | 146 |
19 files changed, 773 insertions, 14 deletions
diff --git a/.gitignore b/.gitignore index b5959d6311..a536f32497 100644 --- a/.gitignore +++ b/.gitignore @@ -111,6 +111,7 @@ git-tag git-tar-tree git-unpack-file git-unpack-objects +git-unresolve git-update-index git-update-ref git-update-server-info @@ -123,6 +124,7 @@ git-write-tree git-core-*/?* test-date test-delta +test-gsimm common-cmds.h *.tar.gz *.dsc diff --git a/Documentation/git-clone.txt b/Documentation/git-clone.txt index 9ac54c282c..131e445747 100644 --- a/Documentation/git-clone.txt +++ b/Documentation/git-clone.txt @@ -10,6 +10,7 @@ SYNOPSIS -------- [verse] 'git-clone' [-l [-s]] [-q] [-n] [--bare] [-o <name>] [-u <upload-pack>] + [--reference <repository>] <repository> [<directory>] DESCRIPTION @@ -46,10 +47,18 @@ OPTIONS -s:: When the repository to clone is on the local machine, instead of using hard links, automatically setup - .git/objects/info/alternatives to share the objects + .git/objects/info/alternates to share the objects with the source repository. The resulting repository starts out without any object of its own. +--reference <repository>:: + If the reference repository is on the local machine + automatically setup .git/objects/info/alternates to + obtain objects from the reference repository. Using + an already existing repository as an alternate will + require less objects to be copied from the repository + being cloned, reducing network and local storage costs. + --quiet:: -q:: Operate quietly. This flag is passed to "rsync" and @@ -112,6 +121,16 @@ $ git show-branch ------------ +Clone from upstream while borrowing from an existing local directory:: ++ +------------ +$ git clone --reference my2.6 \ + git://git.kernel.org/pub/scm/.../linux-2.7 \ + my2.7 +$ cd my2.7 +------------ + + Create a bare repository to publish your changes to the public:: + ------------ @@ -165,7 +165,8 @@ PROGRAMS = \ git-upload-pack$X git-verify-pack$X git-write-tree$X \ git-update-ref$X git-symbolic-ref$X git-check-ref-format$X \ git-name-rev$X git-pack-redundant$X git-repo-config$X git-var$X \ - git-describe$X git-merge-tree$X git-blame$X git-imap-send$X + git-describe$X git-merge-tree$X git-blame$X git-imap-send$X \ + git-unresolve$X BUILT_INS = git-log$X @@ -199,7 +200,7 @@ LIB_H = \ tree-walk.h log-tree.h DIFF_OBJS = \ - diff.o diffcore-break.o diffcore-order.o \ + diff-lib.o diffcore-break.o diffcore-order.o \ diffcore-pickaxe.o diffcore-rename.o tree-diff.o combine-diff.o \ diffcore-delta.o log-tree.o @@ -606,6 +607,9 @@ test-date$X: test-date.c date.o ctype.o test-delta$X: test-delta.c diff-delta.o patch-delta.o $(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) $^ -lz +test-gsimm$X: test-gsimm.c gsimm.o rabinpoly.o + $(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) $^ + check: for i in *.c; do sparse $(ALL_CFLAGS) $(SPARSE_FLAGS) $$i || exit; done @@ -654,6 +658,7 @@ clean: rm -f *.o mozilla-sha1/*.o arm/*.o ppc/*.o compat/*.o xdiff/*.o \ $(LIB_FILE) $(XDIFF_LIB) rm -f $(ALL_PROGRAMS) $(BUILT_INS) git$X + rm -f test-date$X test-delta$X test-gsimm$X rm -f *.spec *.pyc *.pyo */*.pyc */*.pyo common-cmds.h TAGS tags rm -rf $(GIT_TARNAME) rm -f $(GIT_TARNAME).tar.gz git-core_$(GIT_VERSION)-*.tar.gz @@ -535,7 +535,7 @@ static int socksetup(int port, int **socklist_p) if (set_reuse_addr(sockfd)) { close(sockfd); - return 0; /* not fatal */ + continue; } if (bind(sockfd, ai->ai_addr, ai->ai_addrlen) < 0) { @@ -250,13 +250,14 @@ static void show_stats(struct diffstat_t* data) for (i = 0; i < data->nr; i++) { struct diffstat_file *file = data->files[i]; + len = strlen(file->name); + if (max_len < len) + max_len = len; + if (file->is_binary || file->is_unmerged) continue; if (max_change < file->added + file->deleted) max_change = file->added + file->deleted; - len = strlen(file->name); - if (max_len < len) - max_len = len; } for (i = 0; i < data->nr; i++) { diff --git a/git-merge.sh b/git-merge.sh index 78ab422e4e..b834e79c98 100755 --- a/git-merge.sh +++ b/git-merge.sh @@ -335,5 +335,5 @@ Conflicts: then git-rerere fi - die "Automatic merge failed; fix up by hand" + die "Automatic merge failed; fix conflicts and then commit the result." fi diff --git a/git-repack.sh b/git-repack.sh index a5d349fd09..e0c9f323c3 100755 --- a/git-repack.sh +++ b/git-repack.sh @@ -5,9 +5,9 @@ USAGE='[-a] [-d] [-f] [-l] [-n] [-q]' . git-sh-setup - + no_update_info= all_into_one= remove_redundant= -local= quiet= no_reuse_delta= +local= quiet= no_reuse_delta= extra= while case "$#" in 0) break ;; esac do case "$1" in @@ -17,6 +17,8 @@ do -q) quiet=-q ;; -f) no_reuse_delta=--no-reuse-delta ;; -l) local=--local ;; + --window=*) extra="$extra $1" ;; + --depth=*) extra="$extra $1" ;; *) usage ;; esac shift @@ -40,7 +42,7 @@ case ",$all_into_one," in find . -type f \( -name '*.pack' -o -name '*.idx' \) -print` ;; esac -pack_objects="$pack_objects $local $quiet $no_reuse_delta" +pack_objects="$pack_objects $local $quiet $no_reuse_delta$extra" name=$(git-rev-list --objects --all $rev_list 2>&1 | git-pack-objects --non-empty $pack_objects .tmp-pack) || exit 1 @@ -368,8 +368,6 @@ static int cmd_log(int argc, const char **argv, char **envp) init_revisions(&rev); rev.always_show_header = 1; rev.diffopt.recursive = 1; - rev.combine_merges = 1; - rev.ignore_merges = 0; return cmd_log_wc(argc, argv, envp, &rev); } diff --git a/gsimm.c b/gsimm.c new file mode 100644 index 0000000000..7024bf8f58 --- /dev/null +++ b/gsimm.c @@ -0,0 +1,94 @@ +#include "rabinpoly.h" +#include "gsimm.h" + +/* Has to be power of two. Since the Rabin hash only has 63 + usable bits, the number of hashes is limited to 32. + Lower powers of two could be used for speeding up processing + of very large files. */ +#define NUM_HASHES_PER_CHAR 32 + +/* Size of cache used to eliminate duplicate substrings. + Make small enough to comfortably fit in L1 cache. */ +#define DUP_CACHE_SIZE 256 + +/* For the final counting, do not count each bit individually, but + group them. Must be power of two, at most NUM_HASHES_PER_CHAR. + However, larger sizes result in higher cache usage. Use 8 bits + per group for efficient processing of large files on fast machines + with decent caches, or 4 bits for faster processing of small files + and for machines with small caches. */ +#define GROUP_BITS 4 +#define GROUP_COUNTERS (1<<GROUP_BITS) + +static void freq_to_md(u_char *md, int *freq) +{ int j, k; + + for (j = 0; j < MD_LENGTH; j++) + { u_char ch = 0; + + for (k = 0; k < 8; k++) ch = 2*ch + (freq[8*j+k] > 0); + md[j] = ch; + } + bzero (freq, sizeof(freq[0]) * MD_BITS); +} + +void gb_simm_process(u_char *data, unsigned len, u_char *md) +{ size_t j = 0; + u_int32_t ofs; + u_int32_t dup_cache[DUP_CACHE_SIZE]; + u_int32_t count [MD_BITS * (GROUP_COUNTERS/GROUP_BITS)]; + int freq[MD_BITS]; + + bzero (freq, sizeof(freq[0]) * MD_BITS); + bzero (dup_cache, DUP_CACHE_SIZE * sizeof (u_int32_t)); + bzero (count, (MD_BITS * (GROUP_COUNTERS/GROUP_BITS) * sizeof (u_int32_t))); + + /* Ignore incomplete substrings */ + while (j < len && j < RABIN_WINDOW_SIZE) rabin_slide8 (data[j++]); + + while (j < len) + { u_int64_t hash; + u_int32_t ofs, sum; + u_char idx; + int k; + + hash = rabin_slide8 (data[j++]); + + /* In order to update a much larger frequency table + with only 32 bits of checksum, randomly select a + part of the table to update. The selection should + only depend on the content of the represented data, + and be independent of the bits used for the update. + + Instead of updating 32 individual counters, process + the checksum in MD_BITS / GROUP_BITS groups of + GROUP_BITS bits, and count the frequency of each bit pattern. + */ + + idx = (hash >> 32); + sum = (u_int32_t) hash; + ofs = idx % (MD_BITS / NUM_HASHES_PER_CHAR) * NUM_HASHES_PER_CHAR; + idx %= DUP_CACHE_SIZE; + if (dup_cache[idx] != sum) + { dup_cache[idx] = sum; + for (k = 0; k < NUM_HASHES_PER_CHAR / GROUP_BITS; k++) + { count[ofs * GROUP_COUNTERS / GROUP_BITS + (sum % GROUP_COUNTERS)]++; + ofs += GROUP_BITS; + sum >>= GROUP_BITS; + } } } + + /* Distribute the occurrences of each bit group over the frequency table. */ + for (ofs = 0; ofs < MD_BITS; ofs += GROUP_BITS) + { int j; + for (j = 0; j < GROUP_COUNTERS; j++) + { int k; + for (k = 0; k < GROUP_BITS; k++) + { freq[ofs + k] += ((1<<k) & j) + ? count[ofs * GROUP_COUNTERS / GROUP_BITS + j] + : -count[ofs * GROUP_COUNTERS / GROUP_BITS + j]; + } } } + + if (md) + { rabin_reset(); + freq_to_md (md, freq); +} } diff --git a/gsimm.h b/gsimm.h new file mode 100644 index 0000000000..4b023b91a9 --- /dev/null +++ b/gsimm.h @@ -0,0 +1,28 @@ +#ifndef GSIMM_H +#define GSIMM_H + +/* Length of file message digest (MD) in bytes. Longer MD's are + better, but increase processing time for diminishing returns. + Must be multiple of NUM_HASHES_PER_CHAR / 8, and at least 24 + for good results +*/ +#define MD_LENGTH 32 +#define MD_BITS (MD_LENGTH * 8) + +/* The MIN_FILE_SIZE indicates the absolute minimal file size that + can be processed. As indicated above, the first and last + RABIN_WINDOW_SIZE - 1 bytes are skipped. + In order to get at least an average of 12 samples + per bit in the final message digest, require at least 3 * MD_LENGTH + complete windows in the file. */ +#define MIN_FILE_SIZE (3 * MD_LENGTH + 2 * (RABIN_WINDOW_SIZE - 1)) + +/* Limit matching algorithm to files less than 256 MB, so we can use + 32 bit integers everywhere without fear of overflow. For larger + files we should add logic to mmap the file by piece and accumulate + the frequency counts. */ +#define MAX_FILE_SIZE (256*1024*1024 - 1) + +void gb_simm_process(u_char *data, unsigned len, u_char *md); + +#endif diff --git a/rabinpoly.c b/rabinpoly.c new file mode 100644 index 0000000000..c26f3829c3 --- /dev/null +++ b/rabinpoly.c @@ -0,0 +1,154 @@ +/* + * + * Copyright (C) 1999 David Mazieres (dm@uun.org) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; either version 2, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 + * USA + * + */ + + /* Faster generic_fls */ + /* (c) 2002, D.Phillips and Sistina Software */ + +#include "rabinpoly.h" +#define MSB64 0x8000000000000000ULL + +static inline unsigned fls8(unsigned n) +{ + return n & 0xf0? + n & 0xc0? (n >> 7) + 7: (n >> 5) + 5: + n & 0x0c? (n >> 3) + 3: n - ((n + 1) >> 2); +} + +static inline unsigned fls16(unsigned n) +{ + return n & 0xff00? fls8(n >> 8) + 8: fls8(n); +} + +static inline unsigned fls32(unsigned n) +{ + return n & 0xffff0000? fls16(n >> 16) + 16: fls16(n); +} + +static inline unsigned fls64(unsigned long long n) /* should be u64 */ +{ + return n & 0xffffffff00000000ULL? fls32(n >> 32) + 32: fls32(n); +} + + +static u_int64_t polymod (u_int64_t nh, u_int64_t nl, u_int64_t d); +static void polymult (u_int64_t *php, u_int64_t *plp, + u_int64_t x, u_int64_t y); +static u_int64_t polymmult (u_int64_t x, u_int64_t y, u_int64_t d); + +static u_int64_t poly = 0xb15e234bd3792f63ull; // Actual polynomial +static u_int64_t T[256]; // Lookup table for mod +static int shift; + +u_int64_t append8 (u_int64_t p, u_char m) +{ return ((p << 8) | m) ^ T[p >> shift]; +} + +static u_int64_t +polymod (u_int64_t nh, u_int64_t nl, u_int64_t d) +{ int i, k; + assert (d); + k = fls64 (d) - 1; + d <<= 63 - k; + + if (nh) { + if (nh & MSB64) + nh ^= d; + for (i = 62; i >= 0; i--) + if (nh & 1ULL << i) { + nh ^= d >> (63 - i); + nl ^= d << (i + 1); + } + } + for (i = 63; i >= k; i--) + if (nl & 1ULL << i) + nl ^= d >> (63 - i); + return nl; +} + +static void +polymult (u_int64_t *php, u_int64_t *plp, u_int64_t x, u_int64_t y) +{ int i; + u_int64_t ph = 0, pl = 0; + if (x & 1) + pl = y; + for (i = 1; i < 64; i++) + if (x & (1ULL << i)) { + ph ^= y >> (64 - i); + pl ^= y << i; + } + if (php) + *php = ph; + if (plp) + *plp = pl; +} + +static u_int64_t +polymmult (u_int64_t x, u_int64_t y, u_int64_t d) +{ + u_int64_t h, l; + polymult (&h, &l, x, y); + return polymod (h, l, d); +} + +static int size = RABIN_WINDOW_SIZE; +static u_int64_t fingerprint = 0; +static int bufpos = -1; +static u_int64_t U[256]; +static u_char buf[RABIN_WINDOW_SIZE]; + +void rabin_init () +{ u_int64_t sizeshift = 1; + u_int64_t T1; + int xshift; + int i, j; + assert (poly >= 0x100); + xshift = fls64 (poly) - 1; + shift = xshift - 8; + T1 = polymod (0, 1ULL << xshift, poly); + for (j = 0; j < 256; j++) + T[j] = polymmult (j, T1, poly) | ((u_int64_t) j << xshift); + + for (i = 1; i < size; i++) + sizeshift = append8 (sizeshift, 0); + for (i = 0; i < 256; i++) + U[i] = polymmult (i, sizeshift, poly); + bzero (buf, sizeof (buf)); +} + +void +rabin_reset () +{ rabin_init(); + fingerprint = 0; + bzero (buf, sizeof (buf)); +} + +u_int64_t +rabin_slide8 (u_char m) +{ u_char om; + if (++bufpos >= size) bufpos = 0; + + om = buf[bufpos]; + buf[bufpos] = m; + fingerprint = append8 (fingerprint ^ U[om], m); + + return fingerprint; +} + diff --git a/rabinpoly.h b/rabinpoly.h new file mode 100644 index 0000000000..a19a097459 --- /dev/null +++ b/rabinpoly.h @@ -0,0 +1,31 @@ +/* + * + * Copyright (C) 2000 David Mazieres (dm@uun.org) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; either version 2, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 + * USA + * + * Translated to C and simplified by Geert Bosch (bosch@gnat.com) + */ + +#include <assert.h> +#include <strings.h> +#include <sys/types.h> + +#ifndef RABIN_WINDOW_SIZE +#define RABIN_WINDOW_SIZE 48 +#endif +void rabin_reset(); +u_int64_t rabin_slide8(u_char c); diff --git a/revision.c b/revision.c index dbd54da5ba..113dd5a89f 100644 --- a/revision.c +++ b/revision.c @@ -152,6 +152,7 @@ static struct commit *handle_commit(struct rev_info *revs, struct object *object if (parse_commit(commit) < 0) die("unable to parse commit %s", name); if (flags & UNINTERESTING) { + commit->object.flags |= UNINTERESTING; mark_parents_uninteresting(commit); revs->limited = 1; } diff --git a/sha1_name.c b/sha1_name.c index 4f92e12a8d..345935bb2b 100644 --- a/sha1_name.c +++ b/sha1_name.c @@ -3,6 +3,7 @@ #include "commit.h" #include "tree.h" #include "blob.h" +#include "tree-walk.h" static int find_short_object_filename(int len, const char *name, unsigned char *sha1) { @@ -455,6 +456,19 @@ static int get_sha1_1(const char *name, int len, unsigned char *sha1) */ int get_sha1(const char *name, unsigned char *sha1) { + int ret; + unsigned unused; + prepare_alt_odb(); - return get_sha1_1(name, strlen(name), sha1); + ret = get_sha1_1(name, strlen(name), sha1); + if (ret < 0) { + const char *cp = strchr(name, ':'); + if (cp) { + unsigned char tree_sha1[20]; + if (!get_sha1_1(name, cp-name, tree_sha1)) + return get_tree_entry(tree_sha1, cp+1, sha1, + &unused); + } + } + return ret; } diff --git a/templates/hooks--pre-commit b/templates/hooks--pre-commit index 43d3b6ef4a..723a9ef210 100644 --- a/templates/hooks--pre-commit +++ b/templates/hooks--pre-commit @@ -61,6 +61,9 @@ perl -e ' if (/^\s* /) { bad_line("indent SP followed by a TAB", $_); } + if (/^(?:[<>=]){7}/) { + bad_line("unresolved merge conflict", $_); + } } } exit($found_bad); diff --git a/test-gsimm.c b/test-gsimm.c new file mode 100644 index 0000000000..bd28b7da28 --- /dev/null +++ b/test-gsimm.c @@ -0,0 +1,209 @@ +#include <unistd.h> +#include <stdlib.h> +#include <fcntl.h> +#include <libgen.h> +#include <stdio.h> +#include <assert.h> +#include <math.h> +#include <string.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <sys/mman.h> + +#include "rabinpoly.h" +#include "gsimm.h" + +#define MIN(x,y) ((y)<(x) ? (y) : (x)) +#define MAX(x,y) ((y)>(x) ? (y) : (x)) + +/* The RABIN_WINDOW_SIZE is the size of fingerprint window used by + Rabin algorithm. This is not a modifiable parameter. + + The first RABIN_WINDOW_SIZE - 1 bytes are skipped, in order to ensure + fingerprints are good hashes. This does somewhat reduce the + influence of the first few bytes in the file (they're part of + fewer windows, like the last few bytes), but that actually isn't + so bad as files often start with fixed content that may bias comparisons. +*/ + +typedef struct fileinfo +{ char *name; + size_t length; + u_char md[MD_LENGTH]; + int match; +} File; + +int flag_verbose = 0; +int flag_debug = 0; +char *flag_relative = 0; + +char cmd[12] = " ..."; +char md_strbuf[MD_LENGTH * 2 + 1]; +u_char relative_md [MD_LENGTH]; + +File *file; +int file_count; +size_t file_bytes; + +char hex[17] = "0123456789abcdef"; + +void usage() +{ fprintf (stderr, "usage: %s [-dhvw] [-r fingerprint] file ...\n", cmd); + fprintf (stderr, " -d\tdebug output, repeate for more verbosity\n"); + fprintf (stderr, " -h\tshow this usage information\n"); + fprintf (stderr, " -r\tshow distance relative to fingerprint " + "(%u hex digits)\n", MD_LENGTH * 2); + fprintf (stderr, " -v\tverbose output, repeat for even more verbosity\n"); + fprintf (stderr, " -w\tenable warnings for suspect statistics\n"); + exit (1); +} + +int dist (u_char *l, u_char *r) +{ int j, k; + int d = 0; + + for (j = 0; j < MD_LENGTH; j++) + { u_char ch = l[j] ^ r[j]; + + for (k = 0; k < 8; k++) d += ((ch & (1<<k)) > 0); + } + + return d; +} + +char *md_to_str(u_char *md) +{ int j; + + for (j = 0; j < MD_LENGTH; j++) + { u_char ch = md[j]; + + md_strbuf[j*2] = hex[ch >> 4]; + md_strbuf[j*2+1] = hex[ch & 0xF]; + } + + md_strbuf[j*2] = 0; + return md_strbuf; +} + +void process_file (char *name) +{ int fd; + struct stat fs; + u_char *data; + File *fi = file+file_count;; + + fd = open (name, O_RDONLY, 0); + if (fd < 0) + { perror (name); + exit (2); + } + + if (fstat (fd, &fs)) + { perror (name); + exit (2); + } + + if (fs.st_size >= MIN_FILE_SIZE + && fs.st_size <= MAX_FILE_SIZE) + { fi->length = fs.st_size; + fi->name = name; + + data = (u_char *) mmap (0, fs.st_size, PROT_READ, MAP_PRIVATE, fd, 0); + + if (data == (u_char *) -1) + { perror (name); + exit (2); + } + + gb_simm_process (data, fs.st_size, fi->md); + if (flag_relative) + { int d = dist (fi->md, relative_md); + double sim = 1.0 - MIN (1.0, (double) (d) / (MD_LENGTH * 4 - 1)); + fprintf (stdout, "%s %llu %u %s %u %3.1f\n", + md_to_str (fi->md), (long long unsigned) 0, + (unsigned) fs.st_size, name, + d, 100.0 * sim); + } + else + { + fprintf (stdout, "%s %llu %u %s\n", + md_to_str (fi->md), (long long unsigned) 0, + (unsigned) fs.st_size, name); + } + munmap (data, fs.st_size); + file_bytes += fs.st_size; + file_count++; + } else if (flag_verbose) + { fprintf (stdout, "skipping %s (size %llu)\n", name, (long long unsigned) fs.st_size); } + + close (fd); +} + +u_char *str_to_md(char *str, u_char *md) +{ int j; + + if (!md || !str) return 0; + + bzero (md, MD_LENGTH); + + for (j = 0; j < MD_LENGTH * 2; j++) + { char ch = str[j]; + + if (ch >= '0' && ch <= '9') + { md [j/2] = (md [j/2] << 4) + (ch - '0'); + } + else + { ch |= 32; + + if (ch < 'a' || ch > 'f') break; + md [j/2] = (md[j/2] << 4) + (ch - 'a' + 10); + } } + + return (j != MD_LENGTH * 2 || str[j] != 0) ? 0 : md; +} + +int main (int argc, char *argv[]) +{ int ch, j; + + strncpy (cmd, basename (argv[0]), 8); + + while ((ch = getopt(argc, argv, "dhr:vw")) != -1) + { switch (ch) + { case 'd': flag_debug++; + break; + case 'r': if (!optarg) + { fprintf (stderr, "%s: missing argument for -r\n", cmd); + return 1; + } + if (str_to_md (optarg, relative_md)) flag_relative = optarg; + else + { fprintf (stderr, "%s: not a valid fingerprint\n", optarg); + return 1; + } + break; + case 'v': flag_verbose++; + break; + case 'w': break; + default : usage(); + return (ch != 'h'); + } } + + argc -= optind; + argv += optind; + + if (argc == 0) usage(); + + rabin_reset (); + if (flag_verbose && flag_relative) + { fprintf (stdout, "distances are relative to %s\n", flag_relative); + } + + file = (File *) calloc (argc, sizeof (File)); + + for (j = 0; j < argc; j++) process_file (argv[j]); + + if (flag_verbose) + { fprintf (stdout, "%li bytes in %i files\n", (long) file_bytes, file_count); + } + + return 0; +} diff --git a/tree-walk.c b/tree-walk.c index bf8bfdfdf8..9f7abb7cb3 100644 --- a/tree-walk.c +++ b/tree-walk.c @@ -115,3 +115,53 @@ void traverse_trees(int n, struct tree_desc *t, const char *base, traverse_callb free(entry); } +static int find_tree_entry(struct tree_desc *t, const char *name, unsigned char *result, unsigned *mode) +{ + int namelen = strlen(name); + while (t->size) { + const char *entry; + const unsigned char *sha1; + int entrylen, cmp; + + sha1 = tree_entry_extract(t, &entry, mode); + update_tree_entry(t); + entrylen = strlen(entry); + if (entrylen > namelen) + continue; + cmp = memcmp(name, entry, entrylen); + if (cmp > 0) + continue; + if (cmp < 0) + break; + if (entrylen == namelen) { + memcpy(result, sha1, 20); + return 0; + } + if (name[entrylen] != '/') + continue; + if (!S_ISDIR(*mode)) + break; + if (++entrylen == namelen) { + memcpy(result, sha1, 20); + return 0; + } + return get_tree_entry(sha1, name + entrylen, result, mode); + } + return -1; +} + +int get_tree_entry(const unsigned char *tree_sha1, const char *name, unsigned char *sha1, unsigned *mode) +{ + int retval; + void *tree; + struct tree_desc t; + + tree = read_object_with_reference(tree_sha1, tree_type, &t.size, NULL); + if (!tree) + return -1; + t.buf = tree; + retval = find_tree_entry(&t, name, sha1, mode); + free(tree); + return retval; +} + diff --git a/tree-walk.h b/tree-walk.h index 76893e36c3..47438fe1c0 100644 --- a/tree-walk.h +++ b/tree-walk.h @@ -22,4 +22,6 @@ typedef void (*traverse_callback_t)(int n, unsigned long mask, struct name_entry void traverse_trees(int n, struct tree_desc *t, const char *base, traverse_callback_t callback); +int get_tree_entry(const unsigned char *, const char *, unsigned char *, unsigned *); + #endif diff --git a/unresolve.c b/unresolve.c new file mode 100644 index 0000000000..0b23b9bc75 --- /dev/null +++ b/unresolve.c @@ -0,0 +1,146 @@ +#include "cache.h" +#include "tree-walk.h" + +static const char unresolve_usage[] = +"git-unresolve <paths>..."; + +static struct cache_file cache_file; +static unsigned char head_sha1[20]; +static unsigned char merge_head_sha1[20]; + +static struct cache_entry *read_one_ent(const char *which, + unsigned char *ent, const char *path, + int namelen, int stage) +{ + unsigned mode; + unsigned char sha1[20]; + int size; + struct cache_entry *ce; + + if (get_tree_entry(ent, path, sha1, &mode)) { + error("%s: not in %s branch.", path, which); + return NULL; + } + if (mode == S_IFDIR) { + error("%s: not a blob in %s branch.", path, which); + return NULL; + } + size = cache_entry_size(namelen); + ce = xcalloc(1, size); + + memcpy(ce->sha1, sha1, 20); + memcpy(ce->name, path, namelen); + ce->ce_flags = create_ce_flags(namelen, stage); + ce->ce_mode = create_ce_mode(mode); + return ce; +} + +static int unresolve_one(const char *path) +{ + int namelen = strlen(path); + int pos; + int ret = 0; + struct cache_entry *ce_2 = NULL, *ce_3 = NULL; + + /* See if there is such entry in the index. */ + pos = cache_name_pos(path, namelen); + if (pos < 0) { + /* If there isn't, either it is unmerged, or + * resolved as "removed" by mistake. We do not + * want to do anything in the former case. + */ + pos = -pos-1; + if (pos < active_nr) { + struct cache_entry *ce = active_cache[pos]; + if (ce_namelen(ce) == namelen && + !memcmp(ce->name, path, namelen)) { + fprintf(stderr, + "%s: skipping still unmerged path.\n", + path); + goto free_return; + } + } + } + + /* Grab blobs from given path from HEAD and MERGE_HEAD, + * stuff HEAD version in stage #2, + * stuff MERGE_HEAD version in stage #3. + */ + ce_2 = read_one_ent("our", head_sha1, path, namelen, 2); + ce_3 = read_one_ent("their", merge_head_sha1, path, namelen, 3); + + if (!ce_2 || !ce_3) { + ret = -1; + goto free_return; + } + if (!memcmp(ce_2->sha1, ce_3->sha1, 20) && + ce_2->ce_mode == ce_3->ce_mode) { + fprintf(stderr, "%s: identical in both, skipping.\n", + path); + goto free_return; + } + + remove_file_from_cache(path); + if (add_cache_entry(ce_2, ADD_CACHE_OK_TO_ADD)) { + error("%s: cannot add our version to the index.", path); + ret = -1; + goto free_return; + } + if (!add_cache_entry(ce_3, ADD_CACHE_OK_TO_ADD)) + return 0; + error("%s: cannot add their version to the index.", path); + ret = -1; + free_return: + free(ce_2); + free(ce_3); + return ret; +} + +static void read_head_pointers(void) +{ + if (read_ref(git_path("HEAD"), head_sha1)) + die("Cannot read HEAD -- no initial commit yet?"); + if (read_ref(git_path("MERGE_HEAD"), merge_head_sha1)) { + fprintf(stderr, "Not in the middle of a merge.\n"); + exit(0); + } +} + +int main(int ac, char **av) +{ + int i; + int err = 0; + int newfd; + + if (ac < 2) + usage(unresolve_usage); + + git_config(git_default_config); + + /* Read HEAD and MERGE_HEAD; if MERGE_HEAD does not exist, we + * are not doing a merge, so exit with success status. + */ + read_head_pointers(); + + /* Otherwise we would need to update the cache. */ + newfd= hold_index_file_for_update(&cache_file, get_index_file()); + if (newfd < 0) + die("unable to create new cachefile"); + + if (read_cache() < 0) + die("cache corrupted"); + + for (i = 1; i < ac; i++) { + char *arg = av[i]; + err |= unresolve_one(arg); + } + if (err) + die("Error encountered; index not updated."); + + if (active_cache_changed) { + if (write_cache(newfd, active_cache, active_nr) || + commit_index_file(&cache_file)) + die("Unable to write new cachefile"); + } + return 0; +} |