diff options
-rw-r--r-- | Documentation/git-fsck-objects.txt | 15 | ||||
-rw-r--r-- | Makefile | 4 | ||||
-rw-r--r-- | blame.c | 239 | ||||
-rw-r--r-- | commit.c | 36 | ||||
-rw-r--r-- | commit.h | 20 | ||||
-rw-r--r-- | count-delta.c | 72 | ||||
-rw-r--r-- | count-delta.h | 10 | ||||
-rw-r--r-- | diff.c | 2 | ||||
-rw-r--r-- | diffcore-break.c | 46 | ||||
-rw-r--r-- | diffcore-delta.c | 215 | ||||
-rw-r--r-- | diffcore-rename.c | 23 | ||||
-rw-r--r-- | diffcore.h | 7 | ||||
-rw-r--r-- | fsck-objects.c | 20 | ||||
-rwxr-xr-x | git-fetch.sh | 10 | ||||
-rwxr-xr-x | git-fmt-merge-msg.perl | 8 | ||||
-rwxr-xr-x | git-parse-remote.sh | 8 | ||||
-rw-r--r-- | http-push.c | 1051 | ||||
-rw-r--r-- | http.c | 8 | ||||
-rw-r--r-- | http.h | 1 | ||||
-rw-r--r-- | pack-check.c | 20 | ||||
-rw-r--r-- | pack-objects.c | 38 | ||||
-rw-r--r-- | rev-list.c | 6 | ||||
-rw-r--r-- | revision.c | 97 | ||||
-rw-r--r-- | revision.h | 18 |
24 files changed, 1406 insertions, 568 deletions
diff --git a/Documentation/git-fsck-objects.txt b/Documentation/git-fsck-objects.txt index 387b435484..93ce9dcc92 100644 --- a/Documentation/git-fsck-objects.txt +++ b/Documentation/git-fsck-objects.txt @@ -10,7 +10,7 @@ SYNOPSIS -------- [verse] 'git-fsck-objects' [--tags] [--root] [--unreachable] [--cache] - [--standalone | --full] [--strict] [<object>*] + [--full] [--strict] [<object>*] DESCRIPTION ----------- @@ -38,21 +38,14 @@ index file and all SHA1 references in .git/refs/* as heads. Consider any object recorded in the index also as a head node for an unreachability trace. ---standalone:: - Limit checks to the contents of GIT_OBJECT_DIRECTORY - ($GIT_DIR/objects), making sure that it is consistent and - complete without referring to objects found in alternate - object pools listed in GIT_ALTERNATE_OBJECT_DIRECTORIES, - nor packed git archives found in $GIT_DIR/objects/pack; - cannot be used with --full. - --full:: Check not just objects in GIT_OBJECT_DIRECTORY ($GIT_DIR/objects), but also the ones found in alternate - object pools listed in GIT_ALTERNATE_OBJECT_DIRECTORIES, + object pools listed in GIT_ALTERNATE_OBJECT_DIRECTORIES + or $GIT_DIR/objects/info/alternates, and in packed git archives found in $GIT_DIR/objects/pack and corresponding pack subdirectories in alternate - object pools; cannot be used with --standalone. + object pools. --strict:: Enable more strict checking, namely to catch a file mode @@ -190,7 +190,7 @@ PYMODULES = \ LIB_FILE=libgit.a LIB_H = \ - blob.h cache.h commit.h count-delta.h csum-file.h delta.h \ + blob.h cache.h commit.h csum-file.h delta.h \ diff.h object.h pack.h pkt-line.h quote.h refs.h \ run-command.h strbuf.h tag.h tree.h git-compat-util.h revision.h @@ -200,7 +200,7 @@ DIFF_OBJS = \ diffcore-delta.o LIB_OBJS = \ - blob.o commit.o connect.o count-delta.o csum-file.o \ + blob.o commit.o connect.o csum-file.o \ date.o diff-delta.o entry.o exec_cmd.o ident.o index.o \ object.o pack-check.o patch-delta.o path.o pkt-line.o \ quote.o read-cache.o refs.o run-command.o \ @@ -14,6 +14,7 @@ #include "tree.h" #include "blob.h" #include "diff.h" +#include "diffcore.h" #include "revision.h" #define DEBUG 0 @@ -34,7 +35,9 @@ struct util_info { char *buf; unsigned long size; int num_lines; -// const char* path; + const char* pathname; + + void* topo_data; }; struct chunk { @@ -342,25 +345,34 @@ static int map_line(struct commit *commit, int line) return info->line_map[line]; } -static int fill_util_info(struct commit *commit, const char *path) +static struct util_info* get_util(struct commit *commit) { - struct util_info *util; - if (commit->object.util) - return 0; + struct util_info *util = commit->object.util; + + if (util) + return util; util = xmalloc(sizeof(struct util_info)); + util->buf = NULL; + util->size = 0; + util->line_map = NULL; + util->num_lines = -1; + util->pathname = NULL; + commit->object.util = util; + return util; +} + +static int fill_util_info(struct commit *commit) +{ + struct util_info *util = commit->object.util; + + assert(util); + assert(util->pathname); - if (get_blob_sha1(commit->tree, path, util->sha1)) { - free(util); + if (get_blob_sha1(commit->tree, util->pathname, util->sha1)) return 1; - } else { - util->buf = NULL; - util->size = 0; - util->line_map = NULL; - util->num_lines = -1; - commit->object.util = util; + else return 0; - } } static void alloc_line_map(struct commit *commit) @@ -389,10 +401,11 @@ static void alloc_line_map(struct commit *commit) static void init_first_commit(struct commit* commit, const char* filename) { - struct util_info* util; + struct util_info* util = commit->object.util; int i; - if (fill_util_info(commit, filename)) + util->pathname = filename; + if (fill_util_info(commit)) die("fill_util_info failed"); alloc_line_map(commit); @@ -453,7 +466,7 @@ static void process_commits(struct rev_info *rev, const char *path, if(num_parents == 0) *initial = commit; - if(fill_util_info(commit, path)) + if (fill_util_info(commit)) continue; alloc_line_map(commit); @@ -471,7 +484,7 @@ static void process_commits(struct rev_info *rev, const char *path, printf("parent: %s\n", sha1_to_hex(parent->object.sha1)); - if(fill_util_info(parent, path)) { + if (fill_util_info(parent)) { num_parents--; continue; } @@ -511,6 +524,135 @@ static void process_commits(struct rev_info *rev, const char *path, } while ((commit = get_revision(rev)) != NULL); } + +static int compare_tree_path(struct rev_info* revs, + struct commit* c1, struct commit* c2) +{ + const char* paths[2]; + struct util_info* util = c2->object.util; + paths[0] = util->pathname; + paths[1] = NULL; + + diff_tree_setup_paths(get_pathspec(revs->prefix, paths)); + return rev_compare_tree(c1->tree, c2->tree); +} + + +static int same_tree_as_empty_path(struct rev_info *revs, struct tree* t1, + const char* path) +{ + const char* paths[2]; + paths[0] = path; + paths[1] = NULL; + + diff_tree_setup_paths(get_pathspec(revs->prefix, paths)); + return rev_same_tree_as_empty(t1); +} + +static const char* find_rename(struct commit* commit, struct commit* parent) +{ + struct util_info* cutil = commit->object.util; + struct diff_options diff_opts; + const char *paths[1]; + int i; + + if (DEBUG) { + printf("find_rename commit: %s ", + sha1_to_hex(commit->object.sha1)); + puts(sha1_to_hex(parent->object.sha1)); + } + + diff_setup(&diff_opts); + diff_opts.recursive = 1; + diff_opts.detect_rename = DIFF_DETECT_RENAME; + paths[0] = NULL; + diff_tree_setup_paths(paths); + if (diff_setup_done(&diff_opts) < 0) + die("diff_setup_done failed"); + + diff_tree_sha1(commit->tree->object.sha1, parent->tree->object.sha1, + "", &diff_opts); + diffcore_std(&diff_opts); + + for (i = 0; i < diff_queued_diff.nr; i++) { + struct diff_filepair *p = diff_queued_diff.queue[i]; + + if (p->status == 'R' && !strcmp(p->one->path, cutil->pathname)) { + if (DEBUG) + printf("rename %s -> %s\n", p->one->path, p->two->path); + return p->two->path; + } + } + + return 0; +} + +static void simplify_commit(struct rev_info *revs, struct commit *commit) +{ + struct commit_list **pp, *parent; + + if (!commit->tree) + return; + + if (!commit->parents) { + struct util_info* util = commit->object.util; + if (!same_tree_as_empty_path(revs, commit->tree, + util->pathname)) + commit->object.flags |= TREECHANGE; + return; + } + + pp = &commit->parents; + while ((parent = *pp) != NULL) { + struct commit *p = parent->item; + + if (p->object.flags & UNINTERESTING) { + pp = &parent->next; + continue; + } + + parse_commit(p); + switch (compare_tree_path(revs, p, commit)) { + case REV_TREE_SAME: + parent->next = NULL; + commit->parents = parent; + get_util(p)->pathname = get_util(commit)->pathname; + return; + + case REV_TREE_NEW: + { + + struct util_info* util = commit->object.util; + if (revs->remove_empty_trees && + same_tree_as_empty_path(revs, p->tree, + util->pathname)) { + const char* new_name = find_rename(commit, p); + if (new_name) { + struct util_info* putil = get_util(p); + if (!putil->pathname) + putil->pathname = strdup(new_name); + } else { + *pp = parent->next; + continue; + } + } + } + + /* fallthrough */ + case REV_TREE_DIFFERENT: + pp = &parent->next; + if (!get_util(p)->pathname) + get_util(p)->pathname = + get_util(commit)->pathname; + continue; + } + die("bad tree compare for commit %s", + sha1_to_hex(commit->object.sha1)); + } + commit->object.flags |= TREECHANGE; +} + + struct commit_info { char* author; @@ -569,6 +711,18 @@ static const char* format_time(unsigned long time, const char* tz_str) return time_buf; } +static void topo_setter(struct commit* c, void* data) +{ + struct util_info* util = c->object.util; + util->topo_data = data; +} + +static void* topo_getter(struct commit* c) +{ + struct util_info* util = c->object.util; + return util->topo_data; +} + int main(int argc, const char **argv) { int i; @@ -580,8 +734,8 @@ int main(int argc, const char **argv) int sha1_len = 8; int compability = 0; int options = 1; + struct commit* start_commit; - int num_args; const char* args[10]; struct rev_info rev; @@ -634,28 +788,29 @@ int main(int argc, const char **argv) strcpy(filename_buf, filename); filename = filename_buf; - { - struct commit* c; - if (get_sha1(commit, sha1)) - die("get_sha1 failed, commit '%s' not found", commit); - c = lookup_commit_reference(sha1); - - if (fill_util_info(c, filename)) { - printf("%s not found in %s\n", filename, commit); - return 1; - } + if (get_sha1(commit, sha1)) + die("get_sha1 failed, commit '%s' not found", commit); + start_commit = lookup_commit_reference(sha1); + get_util(start_commit)->pathname = filename; + if (fill_util_info(start_commit)) { + printf("%s not found in %s\n", filename, commit); + return 1; } - num_args = 0; - args[num_args++] = NULL; - args[num_args++] = "--topo-order"; - args[num_args++] = "--remove-empty"; - args[num_args++] = commit; - args[num_args++] = "--"; - args[num_args++] = filename; - args[num_args] = NULL; - setup_revisions(num_args, args, &rev, "HEAD"); + init_revisions(&rev); + rev.remove_empty_trees = 1; + rev.topo_order = 1; + rev.prune_fn = simplify_commit; + rev.topo_setter = topo_setter; + rev.topo_getter = topo_getter; + rev.limited = 1; + + commit_list_insert(start_commit, &rev.commits); + + args[0] = filename; + args[1] = NULL; + diff_tree_setup_paths(args); prepare_revision_walk(&rev); process_commits(&rev, filename, &initial); @@ -665,17 +820,21 @@ int main(int argc, const char **argv) for (i = 0; i < num_blame_lines; i++) { struct commit *c = blame_lines[i]; + struct util_info* u; + if (!c) c = initial; + u = c->object.util; get_commit_info(c, &ci); fwrite(sha1_to_hex(c->object.sha1), sha1_len, 1, stdout); if(compability) printf("\t(%10s\t%10s\t%d)", ci.author, format_time(ci.author_time, ci.author_tz), i+1); else - printf(" (%-15.15s %10s %*d) ", ci.author, - format_time(ci.author_time, ci.author_tz), + printf(" %s (%-15.15s %10s %*d) ", u->pathname, + ci.author, format_time(ci.author_time, + ci.author_tz), max_digits, i+1); if(i == num_blame_lines - 1) { @@ -569,11 +569,29 @@ int count_parents(struct commit * commit) return count; } +void topo_sort_default_setter(struct commit *c, void *data) +{ + c->object.util = data; +} + +void *topo_sort_default_getter(struct commit *c) +{ + return c->object.util; +} + /* * Performs an in-place topological sort on the list supplied. */ void sort_in_topological_order(struct commit_list ** list, int lifo) { + sort_in_topological_order_fn(list, lifo, topo_sort_default_setter, + topo_sort_default_getter); +} + +void sort_in_topological_order_fn(struct commit_list ** list, int lifo, + topo_sort_set_fn_t setter, + topo_sort_get_fn_t getter) +{ struct commit_list * next = *list; struct commit_list * work = NULL, **insert; struct commit_list ** pptr = list; @@ -596,7 +614,7 @@ void sort_in_topological_order(struct commit_list ** list, int lifo) next=*list; while (next) { next_nodes->list_item = next; - next->item->object.util = next_nodes; + setter(next->item, next_nodes); next_nodes++; next = next->next; } @@ -606,8 +624,8 @@ void sort_in_topological_order(struct commit_list ** list, int lifo) struct commit_list * parents = next->item->parents; while (parents) { struct commit * parent=parents->item; - struct sort_node * pn = (struct sort_node *)parent->object.util; - + struct sort_node * pn = (struct sort_node *) getter(parent); + if (pn) pn->indegree++; parents=parents->next; @@ -624,7 +642,7 @@ void sort_in_topological_order(struct commit_list ** list, int lifo) next=*list; insert = &work; while (next) { - struct sort_node * node = (struct sort_node *)next->item->object.util; + struct sort_node * node = (struct sort_node *) getter(next->item); if (node->indegree == 0) { insert = &commit_list_insert(next->item, insert)->next; @@ -637,15 +655,15 @@ void sort_in_topological_order(struct commit_list ** list, int lifo) sort_by_date(&work); while (work) { struct commit * work_item = pop_commit(&work); - struct sort_node * work_node = (struct sort_node *)work_item->object.util; + struct sort_node * work_node = (struct sort_node *) getter(work_item); struct commit_list * parents = work_item->parents; while (parents) { struct commit * parent=parents->item; - struct sort_node * pn = (struct sort_node *)parent->object.util; - + struct sort_node * pn = (struct sort_node *) getter(parent); + if (pn) { - /* + /* * parents are only enqueued for emission * when all their children have been emitted thereby * guaranteeing topological order. @@ -667,7 +685,7 @@ void sort_in_topological_order(struct commit_list ** list, int lifo) *pptr = work_node->list_item; pptr = &(*pptr)->next; *pptr = NULL; - work_item->object.util = NULL; + setter(work_item, NULL); } free(nodes); } @@ -65,15 +65,29 @@ int count_parents(struct commit * commit); /* * Performs an in-place topological sort of list supplied. * - * Pre-conditions: + * Pre-conditions for sort_in_topological_order: * all commits in input list and all parents of those * commits must have object.util == NULL - * - * Post-conditions: + * + * Pre-conditions for sort_in_topological_order_fn: + * all commits in input list and all parents of those + * commits must have getter(commit) == NULL + * + * Post-conditions: * invariant of resulting list is: * a reachable from b => ord(b) < ord(a) * in addition, when lifo == 0, commits on parallel tracks are * sorted in the dates order. */ + +typedef void (*topo_sort_set_fn_t)(struct commit*, void *data); +typedef void* (*topo_sort_get_fn_t)(struct commit*); + +void topo_sort_default_setter(struct commit *c, void *data); +void *topo_sort_default_getter(struct commit *c); + void sort_in_topological_order(struct commit_list ** list, int lifo); +void sort_in_topological_order_fn(struct commit_list ** list, int lifo, + topo_sort_set_fn_t setter, + topo_sort_get_fn_t getter); #endif /* COMMIT_H */ diff --git a/count-delta.c b/count-delta.c deleted file mode 100644 index 058a2aadb1..0000000000 --- a/count-delta.c +++ /dev/null @@ -1,72 +0,0 @@ -/* - * Copyright (C) 2005 Junio C Hamano - * The delta-parsing part is almost straight copy of patch-delta.c - * which is (C) 2005 Nicolas Pitre <nico@cam.org>. - */ -#include <stdlib.h> -#include <string.h> -#include <limits.h> -#include "delta.h" -#include "count-delta.h" - -/* - * NOTE. We do not _interpret_ delta fully. As an approximation, we - * just count the number of bytes that are copied from the source, and - * the number of literal data bytes that are inserted. - * - * Number of bytes that are _not_ copied from the source is deletion, - * and number of inserted literal bytes are addition, so sum of them - * is the extent of damage. - */ -int count_delta(void *delta_buf, unsigned long delta_size, - unsigned long *src_copied, unsigned long *literal_added) -{ - unsigned long copied_from_source, added_literal; - const unsigned char *data, *top; - unsigned char cmd; - unsigned long src_size, dst_size, out; - - if (delta_size < DELTA_SIZE_MIN) - return -1; - - data = delta_buf; - top = delta_buf + delta_size; - - src_size = get_delta_hdr_size(&data); - dst_size = get_delta_hdr_size(&data); - - added_literal = copied_from_source = out = 0; - while (data < top) { - cmd = *data++; - if (cmd & 0x80) { - unsigned long cp_off = 0, cp_size = 0; - if (cmd & 0x01) cp_off = *data++; - if (cmd & 0x02) cp_off |= (*data++ << 8); - if (cmd & 0x04) cp_off |= (*data++ << 16); - if (cmd & 0x08) cp_off |= (*data++ << 24); - if (cmd & 0x10) cp_size = *data++; - if (cmd & 0x20) cp_size |= (*data++ << 8); - if (cmd & 0x40) cp_size |= (*data++ << 16); - if (cp_size == 0) cp_size = 0x10000; - - copied_from_source += cp_size; - out += cp_size; - } else { - /* write literal into dst */ - added_literal += cmd; - out += cmd; - data += cmd; - } - } - - /* sanity check */ - if (data != top || out != dst_size) - return -1; - - /* delete size is what was _not_ copied from source. - * edit size is that and literal additions. - */ - *src_copied = copied_from_source; - *literal_added = added_literal; - return 0; -} diff --git a/count-delta.h b/count-delta.h deleted file mode 100644 index 7359629827..0000000000 --- a/count-delta.h +++ /dev/null @@ -1,10 +0,0 @@ -/* - * Copyright (C) 2005 Junio C Hamano - */ -#ifndef COUNT_DELTA_H -#define COUNT_DELTA_H - -int count_delta(void *, unsigned long, - unsigned long *src_copied, unsigned long *literal_added); - -#endif @@ -463,6 +463,8 @@ void diff_free_filespec_data(struct diff_filespec *s) munmap(s->data, s->size); s->should_free = s->should_munmap = 0; s->data = NULL; + free(s->cnt_data); + s->cnt_data = NULL; } static void prep_temp_blob(struct diff_tempfile *temp, diff --git a/diffcore-break.c b/diffcore-break.c index 0fc2b860be..ed0e14c6d8 100644 --- a/diffcore-break.c +++ b/diffcore-break.c @@ -45,8 +45,8 @@ static int should_break(struct diff_filespec *src, * The value we return is 1 if we want the pair to be broken, * or 0 if we do not. */ - unsigned long delta_size, base_size, src_copied, literal_added; - int to_break = 0; + unsigned long delta_size, base_size, src_copied, literal_added, + src_removed; *merge_score_p = 0; /* assume no deletion --- "do not break" * is the default. @@ -68,37 +68,45 @@ static int should_break(struct diff_filespec *src, if (diffcore_count_changes(src->data, src->size, dst->data, dst->size, + NULL, NULL, 0, &src_copied, &literal_added)) return 0; + /* sanity */ + if (src->size < src_copied) + src_copied = src->size; + if (dst->size < literal_added + src_copied) { + if (src_copied < dst->size) + literal_added = dst->size - src_copied; + else + literal_added = 0; + } + src_removed = src->size - src_copied; + /* Compute merge-score, which is "how much is removed * from the source material". The clean-up stage will * merge the surviving pair together if the score is * less than the minimum, after rename/copy runs. */ - if (src->size <= src_copied) - ; /* all copied, nothing removed */ - else { - delta_size = src->size - src_copied; - *merge_score_p = delta_size * MAX_SCORE / src->size; - } - + *merge_score_p = src_removed * MAX_SCORE / src->size; + /* Extent of damage, which counts both inserts and * deletes. */ - if (src->size + literal_added <= src_copied) - delta_size = 0; /* avoid wrapping around */ - else - delta_size = (src->size - src_copied) + literal_added; - - /* We break if the edit exceeds the minimum. - * i.e. (break_score / MAX_SCORE < delta_size / base_size) + delta_size = src_removed + literal_added; + if (delta_size * MAX_SCORE / base_size < break_score) + return 0; + + /* If you removed a lot without adding new material, that is + * not really a rewrite. */ - if (break_score * base_size < delta_size * MAX_SCORE) - to_break = 1; + if ((src->size * break_score < src_removed * MAX_SCORE) && + (literal_added * 20 < src_removed) && + (literal_added * 20 < src_copied)) + return 0; - return to_break; + return 1; } void diffcore_break(int break_score) diff --git a/diffcore-delta.c b/diffcore-delta.c index 1e6a6911ec..471b98f05d 100644 --- a/diffcore-delta.c +++ b/diffcore-delta.c @@ -1,43 +1,202 @@ #include "cache.h" #include "diff.h" #include "diffcore.h" -#include "delta.h" -#include "count-delta.h" - -static int diffcore_count_changes_1(void *src, unsigned long src_size, - void *dst, unsigned long dst_size, - unsigned long delta_limit, - unsigned long *src_copied, - unsigned long *literal_added) + +/* + * Idea here is very simple. + * + * We have total of (sz-N+1) N-byte overlapping sequences in buf whose + * size is sz. If the same N-byte sequence appears in both source and + * destination, we say the byte that starts that sequence is shared + * between them (i.e. copied from source to destination). + * + * For each possible N-byte sequence, if the source buffer has more + * instances of it than the destination buffer, that means the + * difference are the number of bytes not copied from source to + * destination. If the counts are the same, everything was copied + * from source to destination. If the destination has more, + * everything was copied, and destination added more. + * + * We are doing an approximation so we do not really have to waste + * memory by actually storing the sequence. We just hash them into + * somewhere around 2^16 hashbuckets and count the occurrences. + * + * The length of the sequence is arbitrarily set to 8 for now. + */ + +/* Wild guess at the initial hash size */ +#define INITIAL_HASH_SIZE 10 +#define HASHBASE 65537 /* next_prime(2^16) */ + +struct spanhash { + unsigned long hashval; + unsigned long cnt; +}; +struct spanhash_top { + int alloc_log2; + int free; + struct spanhash data[FLEX_ARRAY]; +}; + +static struct spanhash *spanhash_find(struct spanhash_top *top, unsigned long hashval) { - void *delta; - unsigned long delta_size; - - delta = diff_delta(src, src_size, - dst, dst_size, - &delta_size, delta_limit); - if (!delta) - /* If delta_limit is exceeded, we have too much differences */ - return -1; + int sz = 1 << top->alloc_log2; + int bucket = hashval & (sz - 1); + while (1) { + struct spanhash *h = &(top->data[bucket++]); + if (!h->cnt) + return NULL; + if (h->hashval == hashval) + return h; + if (sz <= bucket) + bucket = 0; + } +} - /* Estimate the edit size by interpreting delta. */ - if (count_delta(delta, delta_size, src_copied, literal_added)) { - free(delta); - return -1; +static struct spanhash_top *spanhash_rehash(struct spanhash_top *orig) +{ + struct spanhash_top *new; + int i; + int osz = 1 << orig->alloc_log2; + int sz = osz << 1; + + new = xmalloc(sizeof(*orig) + sizeof(struct spanhash) * sz); + new->alloc_log2 = orig->alloc_log2 + 1; + new->free = osz; + memset(new->data, 0, sizeof(struct spanhash) * sz); + for (i = 0; i < osz; i++) { + struct spanhash *o = &(orig->data[i]); + int bucket; + if (!o->cnt) + continue; + bucket = o->hashval & (sz - 1); + while (1) { + struct spanhash *h = &(new->data[bucket++]); + if (!h->cnt) { + h->hashval = o->hashval; + h->cnt = o->cnt; + new->free--; + break; + } + if (sz <= bucket) + bucket = 0; + } + } + free(orig); + return new; +} + +static struct spanhash_top *add_spanhash(struct spanhash_top *top, + unsigned long hashval) +{ + int bucket, lim; + struct spanhash *h; + + lim = (1 << top->alloc_log2); + bucket = hashval & (lim - 1); + while (1) { + h = &(top->data[bucket++]); + if (!h->cnt) { + h->hashval = hashval; + h->cnt = 1; + top->free--; + if (top->free < 0) + return spanhash_rehash(top); + return top; + } + if (h->hashval == hashval) { + h->cnt++; + return top; + } + if (lim <= bucket) + bucket = 0; } - free(delta); - return 0; +} + +static struct spanhash_top *hash_chars(unsigned char *buf, unsigned long sz) +{ + int i; + unsigned long accum1, accum2, hashval; + struct spanhash_top *hash; + + i = INITIAL_HASH_SIZE; + hash = xmalloc(sizeof(*hash) + sizeof(struct spanhash) * (1<<i)); + hash->alloc_log2 = i; + hash->free = (1<<i)/2; + memset(hash->data, 0, sizeof(struct spanhash) * (1<<i)); + + /* an 8-byte shift register made of accum1 and accum2. New + * bytes come at LSB of accum2, and shifted up to accum1 + */ + for (i = accum1 = accum2 = 0; i < 7; i++, sz--) { + accum1 = (accum1 << 8) | (accum2 >> 24); + accum2 = (accum2 << 8) | *buf++; + } + while (sz) { + accum1 = (accum1 << 8) | (accum2 >> 24); + accum2 = (accum2 << 8) | *buf++; + hashval = (accum1 + accum2 * 0x61) % HASHBASE; + hash = add_spanhash(hash, hashval); + sz--; + } + return hash; } int diffcore_count_changes(void *src, unsigned long src_size, void *dst, unsigned long dst_size, + void **src_count_p, + void **dst_count_p, unsigned long delta_limit, unsigned long *src_copied, unsigned long *literal_added) { - return diffcore_count_changes_1(src, src_size, - dst, dst_size, - delta_limit, - src_copied, - literal_added); + int i, ssz; + struct spanhash_top *src_count, *dst_count; + unsigned long sc, la; + + if (src_size < 8 || dst_size < 8) + return -1; + + src_count = dst_count = NULL; + if (src_count_p) + src_count = *src_count_p; + if (!src_count) { + src_count = hash_chars(src, src_size); + if (src_count_p) + *src_count_p = src_count; + } + if (dst_count_p) + dst_count = *dst_count_p; + if (!dst_count) { + dst_count = hash_chars(dst, dst_size); + if (dst_count_p) + *dst_count_p = dst_count; + } + sc = la = 0; + + ssz = 1 << src_count->alloc_log2; + for (i = 0; i < ssz; i++) { + struct spanhash *s = &(src_count->data[i]); + struct spanhash *d; + unsigned dst_cnt, src_cnt; + if (!s->cnt) + continue; + src_cnt = s->cnt; + d = spanhash_find(dst_count, s->hashval); + dst_cnt = d ? d->cnt : 0; + if (src_cnt < dst_cnt) { + la += dst_cnt - src_cnt; + sc += src_cnt; + } + else + sc += dst_cnt; + } + + if (!src_count_p) + free(src_count); + if (!dst_count_p) + free(dst_count); + *src_copied = sc; + *literal_added = la; + return 0; } diff --git a/diffcore-rename.c b/diffcore-rename.c index 55cf1c37f3..b80b4320ff 100644 --- a/diffcore-rename.c +++ b/diffcore-rename.c @@ -166,23 +166,20 @@ static int estimate_similarity(struct diff_filespec *src, delta_limit = base_size * (MAX_SCORE-minimum_score) / MAX_SCORE; if (diffcore_count_changes(src->data, src->size, dst->data, dst->size, + &src->cnt_data, &dst->cnt_data, delta_limit, &src_copied, &literal_added)) return 0; - /* Extent of damage */ - if (src->size + literal_added < src_copied) - delta_size = 0; - else - delta_size = (src->size - src_copied) + literal_added; - - /* - * Now we will give some score to it. 100% edit gets 0 points - * and 0% edit gets MAX_SCORE points. + /* How similar are they? + * what percentage of material in dst are from source? */ - score = MAX_SCORE - (MAX_SCORE * delta_size / base_size); - if (score < 0) return 0; - if (MAX_SCORE < score) return MAX_SCORE; + if (dst->size < src_copied) + score = MAX_SCORE; + else if (!dst->size) + score = 0; /* should not happen */ + else + score = src_copied * MAX_SCORE / dst->size; return score; } @@ -310,6 +307,8 @@ void diffcore_rename(struct diff_options *options) m->score = estimate_similarity(one, two, minimum_score); } + free(two->cnt_data); + two->cnt_data = NULL; dst_cnt++; } /* cost matrix sorted by most to least similar pair */ diff --git a/diffcore.h b/diffcore.h index dba4f17658..73c7842cc7 100644 --- a/diffcore.h +++ b/diffcore.h @@ -17,8 +17,8 @@ */ #define MAX_SCORE 60000.0 #define DEFAULT_RENAME_SCORE 30000 /* rename/copy similarity minimum (50%) */ -#define DEFAULT_BREAK_SCORE 30000 /* minimum for break to happen (50%)*/ -#define DEFAULT_MERGE_SCORE 48000 /* maximum for break-merge to happen (80%)*/ +#define DEFAULT_BREAK_SCORE 30000 /* minimum for break to happen (50%) */ +#define DEFAULT_MERGE_SCORE 36000 /* maximum for break-merge to happen 60%) */ #define MINIMUM_BREAK_SIZE 400 /* do not break a file smaller than this */ @@ -26,6 +26,7 @@ struct diff_filespec { unsigned char sha1[20]; char *path; void *data; + void *cnt_data; unsigned long size; int xfrm_flags; /* for use by the xfrm */ unsigned short mode; /* file mode */ @@ -103,6 +104,8 @@ void diff_debug_queue(const char *, struct diff_queue_struct *); extern int diffcore_count_changes(void *src, unsigned long src_size, void *dst, unsigned long dst_size, + void **src_count_p, + void **dst_count_p, unsigned long delta_limit, unsigned long *src_copied, unsigned long *literal_added); diff --git a/fsck-objects.c b/fsck-objects.c index 4ddd67699c..59b25904cb 100644 --- a/fsck-objects.c +++ b/fsck-objects.c @@ -14,10 +14,9 @@ static int show_root = 0; static int show_tags = 0; static int show_unreachable = 0; -static int standalone = 0; static int check_full = 0; static int check_strict = 0; -static int keep_cache_objects = 0; +static int keep_cache_objects = 0; static unsigned char head_sha1[20]; #ifdef NO_D_INO_IN_DIRENT @@ -68,7 +67,7 @@ static void check_connectivity(void) continue; if (!obj->parsed) { - if (!standalone && has_sha1_file(obj->sha1)) + if (has_sha1_file(obj->sha1)) ; /* it is in pack */ else printf("missing %s %s\n", @@ -82,7 +81,7 @@ static void check_connectivity(void) for (j = 0; j < refs->count; j++) { struct object *ref = refs->ref[j]; if (ref->parsed || - (!standalone && has_sha1_file(ref->sha1))) + (has_sha1_file(ref->sha1))) continue; printf("broken link from %7s %s\n", obj->type, sha1_to_hex(obj->sha1)); @@ -390,7 +389,7 @@ static int fsck_handle_ref(const char *refname, const unsigned char *sha1) obj = lookup_object(sha1); if (!obj) { - if (!standalone && has_sha1_file(sha1)) { + if (has_sha1_file(sha1)) { default_refs++; return 0; /* it is in a pack */ } @@ -464,10 +463,6 @@ int main(int argc, char **argv) keep_cache_objects = 1; continue; } - if (!strcmp(arg, "--standalone")) { - standalone = 1; - continue; - } if (!strcmp(arg, "--full")) { check_full = 1; continue; @@ -477,14 +472,9 @@ int main(int argc, char **argv) continue; } if (*arg == '-') - usage("git-fsck-objects [--tags] [--root] [[--unreachable] [--cache] [--standalone | --full] [--strict] <head-sha1>*]"); + usage("git-fsck-objects [--tags] [--root] [[--unreachable] [--cache] [--full] [--strict] <head-sha1>*]"); } - if (standalone && check_full) - die("Only one of --standalone or --full can be used."); - if (standalone) - putenv("GIT_ALTERNATE_OBJECT_DIRECTORIES="); - fsck_head_link(); fsck_object_dir(get_object_directory()); if (check_full) { diff --git a/git-fetch.sh b/git-fetch.sh index 0346d4a45c..c0eb96752e 100755 --- a/git-fetch.sh +++ b/git-fetch.sh @@ -94,6 +94,9 @@ append_fetch_head () { # remote-nick is the URL given on the command line (or a shorthand) # remote-name is the $GIT_DIR relative refs/ path we computed # for this refspec. + + # the $note_ variable will be fed to git-fmt-merge-msg for further + # processing. case "$remote_name_" in HEAD) note_= ;; @@ -103,6 +106,9 @@ append_fetch_head () { refs/tags/*) note_="$(expr "$remote_name_" : 'refs/tags/\(.*\)')" note_="tag '$note_' of " ;; + refs/remotes/*) + note_="$(expr "$remote_name_" : 'refs/remotes/\(.*\)')" + note_="remote branch '$note_' of " ;; *) note_="$remote_name of " ;; esac @@ -147,10 +153,10 @@ fast_forward_local () { else echo >&2 "* $1: storing $3" fi - git-update-ref "$1" "$2" + git-update-ref "$1" "$2" ;; - refs/heads/*) + refs/heads/* | refs/remotes/*) # $1 is the ref being updated. # $2 is the new value for the ref. local=$(git-rev-parse --verify "$1^0" 2>/dev/null) diff --git a/git-fmt-merge-msg.perl b/git-fmt-merge-msg.perl index afe80e6321..5986e5414a 100755 --- a/git-fmt-merge-msg.perl +++ b/git-fmt-merge-msg.perl @@ -75,6 +75,7 @@ while (<>) { $src{$src} = { BRANCH => [], TAG => [], + R_BRANCH => [], GENERIC => [], # &1 == has HEAD. # &2 == has others. @@ -91,6 +92,11 @@ while (<>) { push @{$src{$src}{TAG}}, $1; $src{$src}{HEAD_STATUS} |= 2; } + elsif (/^remote branch (.*)$/) { + $origin = $1; + push @{$src{$src}{R_BRANCH}}, $1; + $src{$src}{HEAD_STATUS} |= 2; + } elsif (/^HEAD$/) { $origin = $src; $src{$src}{HEAD_STATUS} |= 1; @@ -123,6 +129,8 @@ for my $src (@src) { } push @this, andjoin("branch ", "branches ", $src{$src}{BRANCH}); + push @this, andjoin("remote branch ", "remote branches ", + $src{$src}{R_BRANCH}); push @this, andjoin("tag ", "tags ", $src{$src}{TAG}); push @this, andjoin("commit ", "commits ", diff --git a/git-parse-remote.sh b/git-parse-remote.sh index 5f158c613f..63f22818e6 100755 --- a/git-parse-remote.sh +++ b/git-parse-remote.sh @@ -86,14 +86,14 @@ canon_refs_list_for_fetch () { local=$(expr "$ref" : '[^:]*:\(.*\)') case "$remote" in '') remote=HEAD ;; - refs/heads/* | refs/tags/*) ;; - heads/* | tags/* ) remote="refs/$remote" ;; + refs/heads/* | refs/tags/* | refs/remotes/*) ;; + heads/* | tags/* | remotes/* ) remote="refs/$remote" ;; *) remote="refs/heads/$remote" ;; esac case "$local" in '') local= ;; - refs/heads/* | refs/tags/*) ;; - heads/* | tags/* ) local="refs/$local" ;; + refs/heads/* | refs/tags/* | refs/remotes/*) ;; + heads/* | tags/* | remotes/* ) local="refs/$local" ;; *) local="refs/heads/$local" ;; esac diff --git a/http-push.c b/http-push.c index 226d71966d..42b0d59e8c 100644 --- a/http-push.c +++ b/http-push.c @@ -11,7 +11,7 @@ #include <expat.h> static const char http_push_usage[] = -"git-http-push [--complete] [--force] [--verbose] <url> <ref> [<ref>...]\n"; +"git-http-push [--all] [--force] [--verbose] <remote> [<head>...]\n"; #ifndef XML_STATUS_OK enum XML_Status { @@ -22,6 +22,7 @@ enum XML_Status { #define XML_STATUS_ERROR 0 #endif +#define PREV_BUF_SIZE 4096 #define RANGE_HEADER_SIZE 30 /* DAV methods */ @@ -58,9 +59,10 @@ enum XML_Status { /* bits #0-4 in revision.h */ -#define LOCAL (1u << 5) -#define REMOTE (1u << 6) -#define PUSHING (1u << 7) +#define LOCAL (1u << 5) +#define REMOTE (1u << 6) +#define FETCHING (1u << 7) +#define PUSHING (1u << 8) static int pushing = 0; static int aborted = 0; @@ -79,13 +81,19 @@ struct repo { char *url; int path_len; + int has_info_refs; + int can_update_info_refs; + int has_info_packs; struct packed_git *packs; + struct remote_lock *locks; }; static struct repo *remote = NULL; -static struct remote_lock *remote_locks = NULL; enum transfer_state { + NEED_FETCH, + RUN_FETCH_LOOSE, + RUN_FETCH_PACKED, NEED_PUSH, RUN_MKCOL, RUN_PUT, @@ -104,6 +112,8 @@ struct transfer_request struct buffer buffer; char filename[PATH_MAX]; char tmpfile[PATH_MAX]; + int local_fileno; + FILE *local_stream; enum transfer_state state; CURLcode curl_result; char errorstr[CURL_ERROR_SIZE]; @@ -113,6 +123,7 @@ struct transfer_request z_stream stream; int zret; int rename; + void *userData; struct active_request_slot *slot; struct transfer_request *next; }; @@ -135,19 +146,31 @@ struct remote_lock char *token; time_t start_time; long timeout; - int active; int refreshing; struct remote_lock *next; }; -struct remote_dentry +/* Flags that control remote_ls processing */ +#define PROCESS_FILES (1u << 0) +#define PROCESS_DIRS (1u << 1) +#define RECURSIVE (1u << 2) + +/* Flags that remote_ls passes to callback functions */ +#define IS_DIR (1u << 0) + +struct remote_ls_ctx { - char *base; - char *name; - int is_dir; + char *path; + void (*userFunc)(struct remote_ls_ctx *ls); + void *userData; + int flags; + char *dentry_name; + int dentry_flags; + struct remote_ls_ctx *parent; }; static void finish_request(struct transfer_request *request); +static void release_request(struct transfer_request *request); static void process_response(void *callback_data) { @@ -157,6 +180,258 @@ static void process_response(void *callback_data) finish_request(request); } +static size_t fwrite_sha1_file(void *ptr, size_t eltsize, size_t nmemb, + void *data) +{ + unsigned char expn[4096]; + size_t size = eltsize * nmemb; + int posn = 0; + struct transfer_request *request = (struct transfer_request *)data; + do { + ssize_t retval = write(request->local_fileno, + ptr + posn, size - posn); + if (retval < 0) + return posn; + posn += retval; + } while (posn < size); + + request->stream.avail_in = size; + request->stream.next_in = ptr; + do { + request->stream.next_out = expn; + request->stream.avail_out = sizeof(expn); + request->zret = inflate(&request->stream, Z_SYNC_FLUSH); + SHA1_Update(&request->c, expn, + sizeof(expn) - request->stream.avail_out); + } while (request->stream.avail_in && request->zret == Z_OK); + data_received++; + return size; +} + +static void start_fetch_loose(struct transfer_request *request) +{ + char *hex = sha1_to_hex(request->obj->sha1); + char *filename; + char prevfile[PATH_MAX]; + char *url; + char *posn; + int prevlocal; + unsigned char prev_buf[PREV_BUF_SIZE]; + ssize_t prev_read = 0; + long prev_posn = 0; + char range[RANGE_HEADER_SIZE]; + struct curl_slist *range_header = NULL; + struct active_request_slot *slot; + + filename = sha1_file_name(request->obj->sha1); + snprintf(request->filename, sizeof(request->filename), "%s", filename); + snprintf(request->tmpfile, sizeof(request->tmpfile), + "%s.temp", filename); + + snprintf(prevfile, sizeof(prevfile), "%s.prev", request->filename); + unlink(prevfile); + rename(request->tmpfile, prevfile); + unlink(request->tmpfile); + + if (request->local_fileno != -1) + error("fd leakage in start: %d", request->local_fileno); + request->local_fileno = open(request->tmpfile, + O_WRONLY | O_CREAT | O_EXCL, 0666); + /* This could have failed due to the "lazy directory creation"; + * try to mkdir the last path component. + */ + if (request->local_fileno < 0 && errno == ENOENT) { + char *dir = strrchr(request->tmpfile, '/'); + if (dir) { + *dir = 0; + mkdir(request->tmpfile, 0777); + *dir = '/'; + } + request->local_fileno = open(request->tmpfile, + O_WRONLY | O_CREAT | O_EXCL, 0666); + } + + if (request->local_fileno < 0) { + request->state = ABORTED; + error("Couldn't create temporary file %s for %s: %s", + request->tmpfile, request->filename, strerror(errno)); + return; + } + + memset(&request->stream, 0, sizeof(request->stream)); + + inflateInit(&request->stream); + + SHA1_Init(&request->c); + + url = xmalloc(strlen(remote->url) + 50); + request->url = xmalloc(strlen(remote->url) + 50); + strcpy(url, remote->url); + posn = url + strlen(remote->url); + strcpy(posn, "objects/"); + posn += 8; + memcpy(posn, hex, 2); + posn += 2; + *(posn++) = '/'; + strcpy(posn, hex + 2); + strcpy(request->url, url); + + /* If a previous temp file is present, process what was already + fetched. */ + prevlocal = open(prevfile, O_RDONLY); + if (prevlocal != -1) { + do { + prev_read = read(prevlocal, prev_buf, PREV_BUF_SIZE); + if (prev_read>0) { + if (fwrite_sha1_file(prev_buf, + 1, + prev_read, + request) == prev_read) { + prev_posn += prev_read; + } else { + prev_read = -1; + } + } + } while (prev_read > 0); + close(prevlocal); + } + unlink(prevfile); + + /* Reset inflate/SHA1 if there was an error reading the previous temp + file; also rewind to the beginning of the local file. */ + if (prev_read == -1) { + memset(&request->stream, 0, sizeof(request->stream)); + inflateInit(&request->stream); + SHA1_Init(&request->c); + if (prev_posn>0) { + prev_posn = 0; + lseek(request->local_fileno, SEEK_SET, 0); + ftruncate(request->local_fileno, 0); + } + } + + slot = get_active_slot(); + slot->callback_func = process_response; + slot->callback_data = request; + request->slot = slot; + + curl_easy_setopt(slot->curl, CURLOPT_FILE, request); + curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_sha1_file); + curl_easy_setopt(slot->curl, CURLOPT_ERRORBUFFER, request->errorstr); + curl_easy_setopt(slot->curl, CURLOPT_URL, url); + curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, no_pragma_header); + + /* If we have successfully processed data from a previous fetch + attempt, only fetch the data we don't already have. */ + if (prev_posn>0) { + if (push_verbosely) + fprintf(stderr, + "Resuming fetch of object %s at byte %ld\n", + hex, prev_posn); + sprintf(range, "Range: bytes=%ld-", prev_posn); + range_header = curl_slist_append(range_header, range); + curl_easy_setopt(slot->curl, + CURLOPT_HTTPHEADER, range_header); + } + + /* Try to get the request started, abort the request on error */ + request->state = RUN_FETCH_LOOSE; + if (!start_active_slot(slot)) { + fprintf(stderr, "Unable to start GET request\n"); + remote->can_update_info_refs = 0; + release_request(request); + } +} + +static void start_fetch_packed(struct transfer_request *request) +{ + char *url; + struct packed_git *target; + FILE *packfile; + char *filename; + long prev_posn = 0; + char range[RANGE_HEADER_SIZE]; + struct curl_slist *range_header = NULL; + + struct transfer_request *check_request = request_queue_head; + struct active_request_slot *slot; + + target = find_sha1_pack(request->obj->sha1, remote->packs); + if (!target) { + fprintf(stderr, "Unable to fetch %s, will not be able to update server info refs\n", sha1_to_hex(request->obj->sha1)); + remote->can_update_info_refs = 0; + release_request(request); + return; + } + + fprintf(stderr, "Fetching pack %s\n", sha1_to_hex(target->sha1)); + fprintf(stderr, " which contains %s\n", sha1_to_hex(request->obj->sha1)); + + filename = sha1_pack_name(target->sha1); + snprintf(request->filename, sizeof(request->filename), "%s", filename); + snprintf(request->tmpfile, sizeof(request->tmpfile), + "%s.temp", filename); + + url = xmalloc(strlen(remote->url) + 64); + sprintf(url, "%sobjects/pack/pack-%s.pack", + remote->url, sha1_to_hex(target->sha1)); + + /* Make sure there isn't another open request for this pack */ + while (check_request) { + if (check_request->state == RUN_FETCH_PACKED && + !strcmp(check_request->url, url)) { + free(url); + release_request(request); + return; + } + check_request = check_request->next; + } + + packfile = fopen(request->tmpfile, "a"); + if (!packfile) { + fprintf(stderr, "Unable to open local file %s for pack", + filename); + remote->can_update_info_refs = 0; + free(url); + return; + } + + slot = get_active_slot(); + slot->callback_func = process_response; + slot->callback_data = request; + request->slot = slot; + request->local_stream = packfile; + request->userData = target; + + request->url = url; + curl_easy_setopt(slot->curl, CURLOPT_FILE, packfile); + curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite); + curl_easy_setopt(slot->curl, CURLOPT_URL, url); + curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, no_pragma_header); + slot->local = packfile; + + /* If there is data present from a previous transfer attempt, + resume where it left off */ + prev_posn = ftell(packfile); + if (prev_posn>0) { + if (push_verbosely) + fprintf(stderr, + "Resuming fetch of pack %s at byte %ld\n", + sha1_to_hex(target->sha1), prev_posn); + sprintf(range, "Range: bytes=%ld-", prev_posn); + range_header = curl_slist_append(range_header, range); + curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, range_header); + } + + /* Try to get the request started, abort the request on error */ + request->state = RUN_FETCH_PACKED; + if (!start_active_slot(slot)) { + fprintf(stderr, "Unable to start GET request\n"); + remote->can_update_info_refs = 0; + release_request(request); + } +} + static void start_mkcol(struct transfer_request *request) { char *hex = sha1_to_hex(request->obj->sha1); @@ -299,62 +574,69 @@ static void start_move(struct transfer_request *request) } } -static int refresh_lock(struct remote_lock *check_lock) +static int refresh_lock(struct remote_lock *lock) { struct active_request_slot *slot; + struct slot_results results; char *if_header; char timeout_header[25]; struct curl_slist *dav_headers = NULL; - struct remote_lock *lock; - int time_remaining; - time_t current_time; + int rc = 0; - /* Refresh all active locks if they're close to expiring */ - for (lock = remote_locks; lock; lock = lock->next) { - if (!lock->active) - continue; + lock->refreshing = 1; - current_time = time(NULL); - time_remaining = lock->start_time + lock->timeout - - current_time; - if (time_remaining > LOCK_REFRESH) - continue; + if_header = xmalloc(strlen(lock->token) + 25); + sprintf(if_header, "If: (<opaquelocktoken:%s>)", lock->token); + sprintf(timeout_header, "Timeout: Second-%ld", lock->timeout); + dav_headers = curl_slist_append(dav_headers, if_header); + dav_headers = curl_slist_append(dav_headers, timeout_header); - lock->refreshing = 1; + slot = get_active_slot(); + slot->results = &results; + curl_easy_setopt(slot->curl, CURLOPT_HTTPGET, 1); + curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_null); + curl_easy_setopt(slot->curl, CURLOPT_URL, lock->url); + curl_easy_setopt(slot->curl, CURLOPT_CUSTOMREQUEST, DAV_LOCK); + curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, dav_headers); - if_header = xmalloc(strlen(lock->token) + 25); - sprintf(if_header, "If: (<opaquelocktoken:%s>)", lock->token); - sprintf(timeout_header, "Timeout: Second-%ld", lock->timeout); - dav_headers = curl_slist_append(dav_headers, if_header); - dav_headers = curl_slist_append(dav_headers, timeout_header); + if (start_active_slot(slot)) { + run_active_slot(slot); + if (results.curl_result != CURLE_OK) { + fprintf(stderr, "LOCK HTTP error %ld\n", + results.http_code); + } else { + lock->start_time = time(NULL); + rc = 1; + } + } - slot = get_active_slot(); - curl_easy_setopt(slot->curl, CURLOPT_HTTPGET, 1); - curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_null); - curl_easy_setopt(slot->curl, CURLOPT_URL, lock->url); - curl_easy_setopt(slot->curl, CURLOPT_CUSTOMREQUEST, DAV_LOCK); - curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, dav_headers); + lock->refreshing = 0; + curl_slist_free_all(dav_headers); + free(if_header); - if (start_active_slot(slot)) { - run_active_slot(slot); - if (slot->curl_result != CURLE_OK) { - fprintf(stderr, "Got HTTP error %ld\n", slot->http_code); - lock->active = 0; - } else { - lock->active = 1; - lock->start_time = time(NULL); + return rc; +} + +static void check_locks() +{ + struct remote_lock *lock = remote->locks; + time_t current_time = time(NULL); + int time_remaining; + + while (lock) { + time_remaining = lock->start_time + lock->timeout - + current_time; + if (!lock->refreshing && time_remaining < LOCK_REFRESH) { + if (!refresh_lock(lock)) { + fprintf(stderr, + "Unable to refresh lock for %s\n", + lock->url); + aborted = 1; + return; } } - - lock->refreshing = 0; - curl_slist_free_all(dav_headers); - free(if_header); + lock = lock->next; } - - if (check_lock) - return check_lock->active; - else - return 0; } static void release_request(struct transfer_request *request) @@ -370,6 +652,10 @@ static void release_request(struct transfer_request *request) entry->next = entry->next->next; } + if (request->local_fileno != -1) + close(request->local_fileno); + if (request->local_stream) + fclose(request->local_stream); if (request->url != NULL) free(request->url); free(request); @@ -377,12 +663,16 @@ static void release_request(struct transfer_request *request) static void finish_request(struct transfer_request *request) { - request->curl_result = request->slot->curl_result; + struct stat st; + struct packed_git *target; + struct packed_git **lst; + + request->curl_result = request->slot->curl_result; request->http_code = request->slot->http_code; request->slot = NULL; /* Keep locks active */ - refresh_lock(request->lock); + check_locks(); if (request->headers != NULL) curl_slist_free_all(request->headers); @@ -417,9 +707,9 @@ static void finish_request(struct transfer_request *request) } } else if (request->state == RUN_MOVE) { if (request->curl_result == CURLE_OK) { - fprintf(stderr, " sent %s\n", - sha1_to_hex(request->obj->sha1)); - request->state = COMPLETE; + if (push_verbosely) + fprintf(stderr, " sent %s\n", + sha1_to_hex(request->obj->sha1)); request->obj->flags |= REMOTE; release_request(request); } else { @@ -429,12 +719,73 @@ static void finish_request(struct transfer_request *request) request->state = ABORTED; aborted = 1; } + } else if (request->state == RUN_FETCH_LOOSE) { + fchmod(request->local_fileno, 0444); + close(request->local_fileno); request->local_fileno = -1; + + if (request->curl_result != CURLE_OK && + request->http_code != 416) { + if (stat(request->tmpfile, &st) == 0) { + if (st.st_size == 0) + unlink(request->tmpfile); + } + } else { + if (request->http_code == 416) + fprintf(stderr, "Warning: requested range invalid; we may already have all the data.\n"); + + inflateEnd(&request->stream); + SHA1_Final(request->real_sha1, &request->c); + if (request->zret != Z_STREAM_END) { + unlink(request->tmpfile); + } else if (memcmp(request->obj->sha1, request->real_sha1, 20)) { + unlink(request->tmpfile); + } else { + request->rename = + move_temp_to_file( + request->tmpfile, + request->filename); + if (request->rename == 0) { + request->obj->flags |= (LOCAL | REMOTE); + } + } + } + + /* Try fetching packed if necessary */ + if (request->obj->flags & LOCAL) + release_request(request); + else + start_fetch_packed(request); + + } else if (request->state == RUN_FETCH_PACKED) { + if (request->curl_result != CURLE_OK) { + fprintf(stderr, "Unable to get pack file %s\n%s", + request->url, curl_errorstr); + remote->can_update_info_refs = 0; + } else { + fclose(request->local_stream); + request->local_stream = NULL; + if (!move_temp_to_file(request->tmpfile, + request->filename)) { + target = (struct packed_git *)request->userData; + lst = &remote->packs; + while (*lst != target) + lst = &((*lst)->next); + *lst = (*lst)->next; + + if (!verify_pack(target, 0)) + install_packed_git(target); + else + remote->can_update_info_refs = 0; + } + } + release_request(request); } } void fill_active_slots(void) { struct transfer_request *request = request_queue_head; + struct transfer_request *next; struct active_request_slot *slot = active_queue_head; int num_transfers; @@ -442,7 +793,10 @@ void fill_active_slots(void) return; while (active_requests < max_requests && request != NULL) { - if (pushing && request->state == NEED_PUSH) { + next = request->next; + if (request->state == NEED_FETCH) { + start_fetch_loose(request); + } else if (pushing && request->state == NEED_PUSH) { if (remote_dir_exists[request->obj->sha1[0]] == 1) { start_put(request); } else { @@ -450,7 +804,7 @@ void fill_active_slots(void) } curl_multi_perform(curlm, &num_transfers); } - request = request->next; + request = next; } while (slot != NULL) { @@ -464,11 +818,45 @@ void fill_active_slots(void) static void get_remote_object_list(unsigned char parent); -static void add_request(struct object *obj, struct remote_lock *lock) +static void add_fetch_request(struct object *obj) +{ + struct transfer_request *request; + + check_locks(); + + /* + * Don't fetch the object if it's known to exist locally + * or is already in the request queue + */ + if (remote_dir_exists[obj->sha1[0]] == -1) + get_remote_object_list(obj->sha1[0]); + if (obj->flags & (LOCAL | FETCHING)) + return; + + obj->flags |= FETCHING; + request = xmalloc(sizeof(*request)); + request->obj = obj; + request->url = NULL; + request->lock = NULL; + request->headers = NULL; + request->local_fileno = -1; + request->local_stream = NULL; + request->state = NEED_FETCH; + request->next = request_queue_head; + request_queue_head = request; + + fill_active_slots(); + step_active_slots(); +} + +static int add_send_request(struct object *obj, struct remote_lock *lock) { struct transfer_request *request = request_queue_head; struct packed_git *target; + /* Keep locks active */ + check_locks(); + /* * Don't push the object if it's known to exist on the remote * or is already in the request queue @@ -476,11 +864,11 @@ static void add_request(struct object *obj, struct remote_lock *lock) if (remote_dir_exists[obj->sha1[0]] == -1) get_remote_object_list(obj->sha1[0]); if (obj->flags & (REMOTE | PUSHING)) - return; + return 0; target = find_sha1_pack(obj->sha1, remote->packs); if (target) { obj->flags |= REMOTE; - return; + return 0; } obj->flags |= PUSHING; @@ -489,12 +877,16 @@ static void add_request(struct object *obj, struct remote_lock *lock) request->url = NULL; request->lock = lock; request->headers = NULL; + request->local_fileno = -1; + request->local_stream = NULL; request->state = NEED_PUSH; request->next = request_queue_head; request_queue_head = request; fill_active_slots(); step_active_slots(); + + return 1; } static int fetch_index(unsigned char *sha1) @@ -509,16 +901,18 @@ static int fetch_index(unsigned char *sha1) FILE *indexfile; struct active_request_slot *slot; + struct slot_results results; /* Don't use the index if the pack isn't there */ - url = xmalloc(strlen(remote->url) + 65); - sprintf(url, "%s/objects/pack/pack-%s.pack", remote->url, hex); + url = xmalloc(strlen(remote->url) + 64); + sprintf(url, "%sobjects/pack/pack-%s.pack", remote->url, hex); slot = get_active_slot(); + slot->results = &results; curl_easy_setopt(slot->curl, CURLOPT_URL, url); curl_easy_setopt(slot->curl, CURLOPT_NOBODY, 1); if (start_active_slot(slot)) { run_active_slot(slot); - if (slot->curl_result != CURLE_OK) { + if (results.curl_result != CURLE_OK) { free(url); return error("Unable to verify pack %s is available", hex); @@ -532,9 +926,9 @@ static int fetch_index(unsigned char *sha1) if (push_verbosely) fprintf(stderr, "Getting index for pack %s\n", hex); - - sprintf(url, "%s/objects/pack/pack-%s.idx", remote->url, hex); - + + sprintf(url, "%sobjects/pack/pack-%s.idx", remote->url, hex); + filename = sha1_pack_index_name(sha1); snprintf(tmpfile, sizeof(tmpfile), "%s.temp", filename); indexfile = fopen(tmpfile, "a"); @@ -543,6 +937,7 @@ static int fetch_index(unsigned char *sha1) filename); slot = get_active_slot(); + slot->results = &results; curl_easy_setopt(slot->curl, CURLOPT_NOBODY, 0); curl_easy_setopt(slot->curl, CURLOPT_HTTPGET, 1); curl_easy_setopt(slot->curl, CURLOPT_FILE, indexfile); @@ -566,7 +961,7 @@ static int fetch_index(unsigned char *sha1) if (start_active_slot(slot)) { run_active_slot(slot); - if (slot->curl_result != CURLE_OK) { + if (results.curl_result != CURLE_OK) { free(url); fclose(indexfile); return error("Unable to get pack index %s\n%s", url, @@ -606,6 +1001,7 @@ static int fetch_indices(void) int i = 0; struct active_request_slot *slot; + struct slot_results results; data = xmalloc(4096); memset(data, 0, 4096); @@ -615,21 +1011,22 @@ static int fetch_indices(void) if (push_verbosely) fprintf(stderr, "Getting pack list\n"); - - url = xmalloc(strlen(remote->url) + 21); - sprintf(url, "%s/objects/info/packs", remote->url); + + url = xmalloc(strlen(remote->url) + 20); + sprintf(url, "%sobjects/info/packs", remote->url); slot = get_active_slot(); + slot->results = &results; curl_easy_setopt(slot->curl, CURLOPT_FILE, &buffer); curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_buffer); curl_easy_setopt(slot->curl, CURLOPT_URL, url); curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, NULL); if (start_active_slot(slot)) { run_active_slot(slot); - if (slot->curl_result != CURLE_OK) { + if (results.curl_result != CURLE_OK) { free(buffer.buffer); free(url); - if (slot->http_code == 404) + if (results.http_code == 404) return 0; else return error("%s", curl_errorstr); @@ -716,20 +1113,22 @@ int fetch_ref(char *ref, unsigned char *sha1) struct buffer buffer; char *base = remote->url; struct active_request_slot *slot; + struct slot_results results; buffer.size = 41; buffer.posn = 0; buffer.buffer = hex; hex[41] = '\0'; - + url = quote_ref_url(base, ref); slot = get_active_slot(); + slot->results = &results; curl_easy_setopt(slot->curl, CURLOPT_FILE, &buffer); curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_buffer); curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, NULL); curl_easy_setopt(slot->curl, CURLOPT_URL, url); if (start_active_slot(slot)) { run_active_slot(slot); - if (slot->curl_result != CURLE_OK) + if (results.curl_result != CURLE_OK) return error("Couldn't get %s for %s\n%s", url, ref, curl_errorstr); } else { @@ -803,55 +1202,6 @@ static void handle_new_lock_ctx(struct xml_ctx *ctx, int tag_closed) } static void one_remote_ref(char *refname); -static void crawl_remote_refs(char *path); - -static void handle_crawl_ref_ctx(struct xml_ctx *ctx, int tag_closed) -{ - struct remote_dentry *dentry = (struct remote_dentry *)ctx->userData; - - - if (tag_closed) { - if (!strcmp(ctx->name, DAV_PROPFIND_RESP) && dentry->name) { - if (dentry->is_dir) { - if (strcmp(dentry->name, dentry->base)) { - crawl_remote_refs(dentry->name); - } - } else { - one_remote_ref(dentry->name); - } - } else if (!strcmp(ctx->name, DAV_PROPFIND_NAME) && ctx->cdata) { - dentry->name = xmalloc(strlen(ctx->cdata) - - remote->path_len + 1); - strcpy(dentry->name, - ctx->cdata + remote->path_len); - } else if (!strcmp(ctx->name, DAV_PROPFIND_COLLECTION)) { - dentry->is_dir = 1; - } - } else if (!strcmp(ctx->name, DAV_PROPFIND_RESP)) { - dentry->name = NULL; - dentry->is_dir = 0; - } -} - -static void handle_remote_object_list_ctx(struct xml_ctx *ctx, int tag_closed) -{ - char *path; - char *obj_hex; - - if (tag_closed) { - if (!strcmp(ctx->name, DAV_PROPFIND_NAME) && ctx->cdata) { - path = ctx->cdata + remote->path_len; - if (strlen(path) != 50) - return; - path += 9; - obj_hex = xmalloc(strlen(path)); - strncpy(obj_hex, path, 2); - strcpy(obj_hex + 2, path + 3); - one_remote_object(obj_hex); - free(obj_hex); - } - } -} static void xml_start_tag(void *userData, const char *name, const char **atts) @@ -913,6 +1263,7 @@ xml_cdata(void *userData, const XML_Char *s, int len) static struct remote_lock *lock_remote(char *path, long timeout) { struct active_request_slot *slot; + struct slot_results results; struct buffer out_buffer; struct buffer in_buffer; char *out_data; @@ -920,7 +1271,7 @@ static struct remote_lock *lock_remote(char *path, long timeout) char *url; char *ep; char timeout_header[25]; - struct remote_lock *lock = remote_locks; + struct remote_lock *lock = NULL; XML_Parser parser = XML_ParserCreate(NULL); enum XML_Status result; struct curl_slist *dav_headers = NULL; @@ -929,31 +1280,20 @@ static struct remote_lock *lock_remote(char *path, long timeout) url = xmalloc(strlen(remote->url) + strlen(path) + 1); sprintf(url, "%s%s", remote->url, path); - /* Make sure the url is not already locked */ - while (lock && strcmp(lock->url, url)) { - lock = lock->next; - } - if (lock) { - free(url); - if (refresh_lock(lock)) - return lock; - else - return NULL; - } - /* Make sure leading directories exist for the remote ref */ ep = strchr(url + strlen(remote->url) + 11, '/'); while (ep) { *ep = 0; slot = get_active_slot(); + slot->results = &results; curl_easy_setopt(slot->curl, CURLOPT_HTTPGET, 1); curl_easy_setopt(slot->curl, CURLOPT_URL, url); curl_easy_setopt(slot->curl, CURLOPT_CUSTOMREQUEST, DAV_MKCOL); curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_null); if (start_active_slot(slot)) { run_active_slot(slot); - if (slot->curl_result != CURLE_OK && - slot->http_code != 405) { + if (results.curl_result != CURLE_OK && + results.http_code != 405) { fprintf(stderr, "Unable to create branch path %s\n", url); @@ -961,7 +1301,7 @@ static struct remote_lock *lock_remote(char *path, long timeout) return NULL; } } else { - fprintf(stderr, "Unable to start request\n"); + fprintf(stderr, "Unable to start MKCOL request\n"); free(url); return NULL; } @@ -985,6 +1325,7 @@ static struct remote_lock *lock_remote(char *path, long timeout) dav_headers = curl_slist_append(dav_headers, "Content-Type: text/xml"); slot = get_active_slot(); + slot->results = &results; curl_easy_setopt(slot->curl, CURLOPT_INFILE, &out_buffer); curl_easy_setopt(slot->curl, CURLOPT_INFILESIZE, out_buffer.size); curl_easy_setopt(slot->curl, CURLOPT_READFUNCTION, fread_buffer); @@ -996,14 +1337,11 @@ static struct remote_lock *lock_remote(char *path, long timeout) curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, dav_headers); lock = xcalloc(1, sizeof(*lock)); - lock->owner = NULL; - lock->token = NULL; lock->timeout = -1; - lock->refreshing = 0; if (start_active_slot(slot)) { run_active_slot(slot); - if (slot->curl_result == CURLE_OK) { + if (results.curl_result == CURLE_OK) { ctx.name = xcalloc(10, 1); ctx.len = 0; ctx.cdata = NULL; @@ -1024,7 +1362,7 @@ static struct remote_lock *lock_remote(char *path, long timeout) } } } else { - fprintf(stderr, "Unable to start request\n"); + fprintf(stderr, "Unable to start LOCK request\n"); } curl_slist_free_all(dav_headers); @@ -1041,10 +1379,9 @@ static struct remote_lock *lock_remote(char *path, long timeout) lock = NULL; } else { lock->url = url; - lock->active = 1; lock->start_time = time(NULL); - lock->next = remote_locks; - remote_locks = lock; + lock->next = remote->locks; + remote->locks = lock; } return lock; @@ -1053,6 +1390,8 @@ static struct remote_lock *lock_remote(char *path, long timeout) static int unlock_remote(struct remote_lock *lock) { struct active_request_slot *slot; + struct slot_results results; + struct remote_lock *prev = remote->locks; char *lock_token_header; struct curl_slist *dav_headers = NULL; int rc = 0; @@ -1063,6 +1402,7 @@ static int unlock_remote(struct remote_lock *lock) dav_headers = curl_slist_append(dav_headers, lock_token_header); slot = get_active_slot(); + slot->results = &results; curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_null); curl_easy_setopt(slot->curl, CURLOPT_URL, lock->url); curl_easy_setopt(slot->curl, CURLOPT_CUSTOMREQUEST, DAV_UNLOCK); @@ -1070,107 +1410,115 @@ static int unlock_remote(struct remote_lock *lock) if (start_active_slot(slot)) { run_active_slot(slot); - if (slot->curl_result == CURLE_OK) + if (results.curl_result == CURLE_OK) rc = 1; else - fprintf(stderr, "Got HTTP error %ld\n", - slot->http_code); + fprintf(stderr, "UNLOCK HTTP error %ld\n", + results.http_code); } else { - fprintf(stderr, "Unable to start request\n"); + fprintf(stderr, "Unable to start UNLOCK request\n"); } curl_slist_free_all(dav_headers); free(lock_token_header); - lock->active = 0; + if (remote->locks == lock) { + remote->locks = lock->next; + } else { + while (prev && prev->next != lock) + prev = prev->next; + if (prev) + prev->next = prev->next->next; + } + + if (lock->owner != NULL) + free(lock->owner); + free(lock->url); + free(lock->token); + free(lock); return rc; } -static void crawl_remote_refs(char *path) -{ - char *url; - struct active_request_slot *slot; - struct buffer in_buffer; - struct buffer out_buffer; - char *in_data; - char *out_data; - XML_Parser parser = XML_ParserCreate(NULL); - enum XML_Status result; - struct curl_slist *dav_headers = NULL; - struct xml_ctx ctx; - struct remote_dentry dentry; - - fprintf(stderr, " %s\n", path); - - dentry.base = path; - dentry.name = NULL; - dentry.is_dir = 0; +static void remote_ls(const char *path, int flags, + void (*userFunc)(struct remote_ls_ctx *ls), + void *userData); - url = xmalloc(strlen(remote->url) + strlen(path) + 1); - sprintf(url, "%s%s", remote->url, path); +static void process_ls_object(struct remote_ls_ctx *ls) +{ + unsigned int *parent = (unsigned int *)ls->userData; + char *path = ls->dentry_name; + char *obj_hex; - out_buffer.size = strlen(PROPFIND_ALL_REQUEST); - out_data = xmalloc(out_buffer.size + 1); - snprintf(out_data, out_buffer.size + 1, PROPFIND_ALL_REQUEST); - out_buffer.posn = 0; - out_buffer.buffer = out_data; + if (!strcmp(ls->path, ls->dentry_name) && (ls->flags & IS_DIR)) { + remote_dir_exists[*parent] = 1; + return; + } - in_buffer.size = 4096; - in_data = xmalloc(in_buffer.size); - in_buffer.posn = 0; - in_buffer.buffer = in_data; + if (strlen(path) != 49) + return; + path += 8; + obj_hex = xmalloc(strlen(path)); + strncpy(obj_hex, path, 2); + strcpy(obj_hex + 2, path + 3); + one_remote_object(obj_hex); + free(obj_hex); +} - dav_headers = curl_slist_append(dav_headers, "Depth: 1"); - dav_headers = curl_slist_append(dav_headers, "Content-Type: text/xml"); +static void process_ls_ref(struct remote_ls_ctx *ls) +{ + if (!strcmp(ls->path, ls->dentry_name) && (ls->dentry_flags & IS_DIR)) { + fprintf(stderr, " %s\n", ls->dentry_name); + return; + } - slot = get_active_slot(); - curl_easy_setopt(slot->curl, CURLOPT_INFILE, &out_buffer); - curl_easy_setopt(slot->curl, CURLOPT_INFILESIZE, out_buffer.size); - curl_easy_setopt(slot->curl, CURLOPT_READFUNCTION, fread_buffer); - curl_easy_setopt(slot->curl, CURLOPT_FILE, &in_buffer); - curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_buffer); - curl_easy_setopt(slot->curl, CURLOPT_URL, url); - curl_easy_setopt(slot->curl, CURLOPT_UPLOAD, 1); - curl_easy_setopt(slot->curl, CURLOPT_CUSTOMREQUEST, DAV_PROPFIND); - curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, dav_headers); + if (!(ls->dentry_flags & IS_DIR)) + one_remote_ref(ls->dentry_name); +} - if (start_active_slot(slot)) { - run_active_slot(slot); - if (slot->curl_result == CURLE_OK) { - ctx.name = xcalloc(10, 1); - ctx.len = 0; - ctx.cdata = NULL; - ctx.userFunc = handle_crawl_ref_ctx; - ctx.userData = &dentry; - XML_SetUserData(parser, &ctx); - XML_SetElementHandler(parser, xml_start_tag, - xml_end_tag); - XML_SetCharacterDataHandler(parser, xml_cdata); - result = XML_Parse(parser, in_buffer.buffer, - in_buffer.posn, 1); - free(ctx.name); +static void handle_remote_ls_ctx(struct xml_ctx *ctx, int tag_closed) +{ + struct remote_ls_ctx *ls = (struct remote_ls_ctx *)ctx->userData; - if (result != XML_STATUS_OK) { - fprintf(stderr, "XML error: %s\n", - XML_ErrorString( - XML_GetErrorCode(parser))); + if (tag_closed) { + if (!strcmp(ctx->name, DAV_PROPFIND_RESP) && ls->dentry_name) { + if (ls->dentry_flags & IS_DIR) { + if (ls->flags & PROCESS_DIRS) { + ls->userFunc(ls); + } + if (strcmp(ls->dentry_name, ls->path) && + ls->flags & RECURSIVE) { + remote_ls(ls->dentry_name, + ls->flags, + ls->userFunc, + ls->userData); + } + } else if (ls->flags & PROCESS_FILES) { + ls->userFunc(ls); } + } else if (!strcmp(ctx->name, DAV_PROPFIND_NAME) && ctx->cdata) { + ls->dentry_name = xmalloc(strlen(ctx->cdata) - + remote->path_len + 1); + strcpy(ls->dentry_name, ctx->cdata + remote->path_len); + } else if (!strcmp(ctx->name, DAV_PROPFIND_COLLECTION)) { + ls->dentry_flags |= IS_DIR; } - } else { - fprintf(stderr, "Unable to start request\n"); + } else if (!strcmp(ctx->name, DAV_PROPFIND_RESP)) { + if (ls->dentry_name) { + free(ls->dentry_name); + } + ls->dentry_name = NULL; + ls->dentry_flags = 0; } - - free(url); - free(out_data); - free(in_buffer.buffer); - curl_slist_free_all(dav_headers); } -static void get_remote_object_list(unsigned char parent) +static void remote_ls(const char *path, int flags, + void (*userFunc)(struct remote_ls_ctx *ls), + void *userData) { - char *url; + char *url = xmalloc(strlen(remote->url) + strlen(path) + 1); struct active_request_slot *slot; + struct slot_results results; struct buffer in_buffer; struct buffer out_buffer; char *in_data; @@ -1179,13 +1527,15 @@ static void get_remote_object_list(unsigned char parent) enum XML_Status result; struct curl_slist *dav_headers = NULL; struct xml_ctx ctx; - char path[] = "/objects/XX/"; - static const char hex[] = "0123456789abcdef"; - unsigned int val = parent; + struct remote_ls_ctx ls; + + ls.flags = flags; + ls.path = strdup(path); + ls.dentry_name = NULL; + ls.dentry_flags = 0; + ls.userData = userData; + ls.userFunc = userFunc; - path[9] = hex[val >> 4]; - path[10] = hex[val & 0xf]; - url = xmalloc(strlen(remote->url) + strlen(path) + 1); sprintf(url, "%s%s", remote->url, path); out_buffer.size = strlen(PROPFIND_ALL_REQUEST); @@ -1203,6 +1553,7 @@ static void get_remote_object_list(unsigned char parent) dav_headers = curl_slist_append(dav_headers, "Content-Type: text/xml"); slot = get_active_slot(); + slot->results = &results; curl_easy_setopt(slot->curl, CURLOPT_INFILE, &out_buffer); curl_easy_setopt(slot->curl, CURLOPT_INFILESIZE, out_buffer.size); curl_easy_setopt(slot->curl, CURLOPT_READFUNCTION, fread_buffer); @@ -1215,12 +1566,12 @@ static void get_remote_object_list(unsigned char parent) if (start_active_slot(slot)) { run_active_slot(slot); - if (slot->curl_result == CURLE_OK) { - remote_dir_exists[parent] = 1; + if (results.curl_result == CURLE_OK) { ctx.name = xcalloc(10, 1); ctx.len = 0; ctx.cdata = NULL; - ctx.userFunc = handle_remote_object_list_ctx; + ctx.userFunc = handle_remote_ls_ctx; + ctx.userData = &ls; XML_SetUserData(parser, &ctx); XML_SetElementHandler(parser, xml_start_tag, xml_end_tag); @@ -1234,22 +1585,35 @@ static void get_remote_object_list(unsigned char parent) XML_ErrorString( XML_GetErrorCode(parser))); } - } else { - remote_dir_exists[parent] = 0; } } else { - fprintf(stderr, "Unable to start request\n"); + fprintf(stderr, "Unable to start PROPFIND request\n"); } + free(ls.path); free(url); free(out_data); free(in_buffer.buffer); curl_slist_free_all(dav_headers); } +static void get_remote_object_list(unsigned char parent) +{ + char path[] = "objects/XX/"; + static const char hex[] = "0123456789abcdef"; + unsigned int val = parent; + + path[8] = hex[val >> 4]; + path[9] = hex[val & 0xf]; + remote_dir_exists[val] = 0; + remote_ls(path, (PROCESS_FILES | PROCESS_DIRS), + process_ls_object, &val); +} + static int locking_available(void) { struct active_request_slot *slot; + struct slot_results results; struct buffer in_buffer; struct buffer out_buffer; char *in_data; @@ -1276,8 +1640,9 @@ static int locking_available(void) dav_headers = curl_slist_append(dav_headers, "Depth: 0"); dav_headers = curl_slist_append(dav_headers, "Content-Type: text/xml"); - + slot = get_active_slot(); + slot->results = &results; curl_easy_setopt(slot->curl, CURLOPT_INFILE, &out_buffer); curl_easy_setopt(slot->curl, CURLOPT_INFILESIZE, out_buffer.size); curl_easy_setopt(slot->curl, CURLOPT_READFUNCTION, fread_buffer); @@ -1290,7 +1655,7 @@ static int locking_available(void) if (start_active_slot(slot)) { run_active_slot(slot); - if (slot->curl_result == CURLE_OK) { + if (results.curl_result == CURLE_OK) { ctx.name = xcalloc(10, 1); ctx.len = 0; ctx.cdata = NULL; @@ -1311,7 +1676,7 @@ static int locking_available(void) } } } else { - fprintf(stderr, "Unable to start request\n"); + fprintf(stderr, "Unable to start PROPFIND request\n"); } free(out_data); @@ -1372,16 +1737,17 @@ static struct object_list **process_tree(struct tree *tree, return p; } -static void get_delta(struct rev_info *revs, struct remote_lock *lock) +static int get_delta(struct rev_info *revs, struct remote_lock *lock) { struct commit *commit; struct object_list **p = &objects, *pending; + int count = 0; while ((commit = get_revision(revs)) != NULL) { p = process_tree(commit->tree, p, NULL, ""); commit->object.flags |= LOCAL; if (!(commit->object.flags & UNINTERESTING)) - add_request(&commit->object, lock); + count += add_send_request(&commit->object, lock); } for (pending = revs->pending_objects; pending; pending = pending->next) { @@ -1408,14 +1774,17 @@ static void get_delta(struct rev_info *revs, struct remote_lock *lock) while (objects) { if (!(objects->item->flags & UNINTERESTING)) - add_request(objects->item, lock); + count += add_send_request(objects->item, lock); objects = objects->next; } + + return count; } static int update_remote(unsigned char *sha1, struct remote_lock *lock) { struct active_request_slot *slot; + struct slot_results results; char *out_data; char *if_header; struct buffer out_buffer; @@ -1437,6 +1806,7 @@ static int update_remote(unsigned char *sha1, struct remote_lock *lock) out_buffer.buffer = out_data; slot = get_active_slot(); + slot->results = &results; curl_easy_setopt(slot->curl, CURLOPT_INFILE, &out_buffer); curl_easy_setopt(slot->curl, CURLOPT_INFILESIZE, out_buffer.size); curl_easy_setopt(slot->curl, CURLOPT_READFUNCTION, fread_buffer); @@ -1451,10 +1821,10 @@ static int update_remote(unsigned char *sha1, struct remote_lock *lock) run_active_slot(slot); free(out_data); free(if_header); - if (slot->curl_result != CURLE_OK) { + if (results.curl_result != CURLE_OK) { fprintf(stderr, "PUT error: curl result=%d, HTTP code=%ld\n", - slot->curl_result, slot->http_code); + results.curl_result, results.http_code); /* We should attempt recovery? */ return 0; } @@ -1487,6 +1857,7 @@ static void one_remote_ref(char *refname) { struct ref *ref; unsigned char remote_sha1[20]; + struct object *obj; if (fetch_ref(refname, remote_sha1) != 0) { fprintf(stderr, @@ -1495,6 +1866,19 @@ static void one_remote_ref(char *refname) return; } + /* + * Fetch a copy of the object if it doesn't exist locally - it + * may be required for updating server info later. + */ + if (remote->can_update_info_refs && !has_sha1_file(remote_sha1)) { + obj = lookup_unknown_object(remote_sha1); + if (obj) { + fprintf(stderr, " fetch %s for %s\n", + sha1_to_hex(remote_sha1), refname); + add_fetch_request(obj); + } + } + int len = strlen(refname) + 1; ref = xcalloc(1, sizeof(*ref) + len); memcpy(ref->old_sha1, remote_sha1, 20); @@ -1512,7 +1896,7 @@ static void get_local_heads(void) static void get_dav_remote_heads(void) { remote_tail = &remote_refs; - crawl_remote_refs("refs/"); + remote_ls("refs/", (PROCESS_FILES | PROCESS_DIRS | RECURSIVE), process_ls_ref, NULL); } static int is_zero_sha1(const unsigned char *sha1) @@ -1600,24 +1984,142 @@ static void mark_edges_uninteresting(struct commit_list *list) } } +static void add_remote_info_ref(struct remote_ls_ctx *ls) +{ + struct buffer *buf = (struct buffer *)ls->userData; + unsigned char remote_sha1[20]; + struct object *o; + int len; + char *ref_info; + + if (fetch_ref(ls->dentry_name, remote_sha1) != 0) { + fprintf(stderr, + "Unable to fetch ref %s from %s\n", + ls->dentry_name, remote->url); + aborted = 1; + return; + } + + o = parse_object(remote_sha1); + if (!o) { + fprintf(stderr, + "Unable to parse object %s for remote ref %s\n", + sha1_to_hex(remote_sha1), ls->dentry_name); + aborted = 1; + return; + } + + len = strlen(ls->dentry_name) + 42; + ref_info = xcalloc(len + 1, 1); + sprintf(ref_info, "%s %s\n", + sha1_to_hex(remote_sha1), ls->dentry_name); + fwrite_buffer(ref_info, 1, len, buf); + free(ref_info); + + if (o->type == tag_type) { + o = deref_tag(o, ls->dentry_name, 0); + if (o) { + len = strlen(ls->dentry_name) + 45; + ref_info = xcalloc(len + 1, 1); + sprintf(ref_info, "%s %s^{}\n", + sha1_to_hex(o->sha1), ls->dentry_name); + fwrite_buffer(ref_info, 1, len, buf); + free(ref_info); + } + } +} + +static void update_remote_info_refs(struct remote_lock *lock) +{ + struct buffer buffer; + struct active_request_slot *slot; + struct slot_results results; + char *if_header; + struct curl_slist *dav_headers = NULL; + + buffer.buffer = xmalloc(4096); + memset(buffer.buffer, 0, 4096); + buffer.size = 4096; + buffer.posn = 0; + remote_ls("refs/", (PROCESS_FILES | RECURSIVE), + add_remote_info_ref, &buffer); + if (!aborted) { + if_header = xmalloc(strlen(lock->token) + 25); + sprintf(if_header, "If: (<opaquelocktoken:%s>)", lock->token); + dav_headers = curl_slist_append(dav_headers, if_header); + + slot = get_active_slot(); + slot->results = &results; + curl_easy_setopt(slot->curl, CURLOPT_INFILE, &buffer); + curl_easy_setopt(slot->curl, CURLOPT_INFILESIZE, buffer.posn); + curl_easy_setopt(slot->curl, CURLOPT_READFUNCTION, fread_buffer); + curl_easy_setopt(slot->curl, CURLOPT_WRITEFUNCTION, fwrite_null); + curl_easy_setopt(slot->curl, CURLOPT_CUSTOMREQUEST, DAV_PUT); + curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, dav_headers); + curl_easy_setopt(slot->curl, CURLOPT_UPLOAD, 1); + curl_easy_setopt(slot->curl, CURLOPT_PUT, 1); + curl_easy_setopt(slot->curl, CURLOPT_URL, lock->url); + + buffer.posn = 0; + + if (start_active_slot(slot)) { + run_active_slot(slot); + if (results.curl_result != CURLE_OK) { + fprintf(stderr, + "PUT error: curl result=%d, HTTP code=%ld\n", + results.curl_result, results.http_code); + } + } + free(if_header); + } + free(buffer.buffer); +} + +static int remote_exists(const char *path) +{ + char *url = xmalloc(strlen(remote->url) + strlen(path) + 1); + struct active_request_slot *slot; + struct slot_results results; + + sprintf(url, "%s%s", remote->url, path); + + slot = get_active_slot(); + slot->results = &results; + curl_easy_setopt(slot->curl, CURLOPT_URL, url); + curl_easy_setopt(slot->curl, CURLOPT_NOBODY, 1); + + if (start_active_slot(slot)) { + run_active_slot(slot); + if (results.http_code == 404) + return 0; + else if (results.curl_result == CURLE_OK) + return 1; + else + fprintf(stderr, "HEAD HTTP error %ld\n", results.http_code); + } else { + fprintf(stderr, "Unable to start HEAD request\n"); + } + + return -1; +} + int main(int argc, char **argv) { struct transfer_request *request; struct transfer_request *next_request; int nr_refspec = 0; char **refspec = NULL; - struct remote_lock *ref_lock; + struct remote_lock *ref_lock = NULL; + struct remote_lock *info_ref_lock = NULL; struct rev_info revs; + int objects_to_send; int rc = 0; int i; setup_git_directory(); setup_ident(); - remote = xmalloc(sizeof(*remote)); - remote->url = NULL; - remote->path_len = 0; - remote->packs = NULL; + remote = xcalloc(sizeof(*remote), 1); argv++; for (i = 1; i < argc; i++, argv++) { @@ -1674,6 +2176,18 @@ int main(int argc, char **argv) goto cleanup; } + /* Check whether the remote has server info files */ + remote->can_update_info_refs = 0; + remote->has_info_refs = remote_exists("info/refs"); + remote->has_info_packs = remote_exists("objects/info/packs"); + if (remote->has_info_refs) { + info_ref_lock = lock_remote("info/refs", LOCK_TIME); + if (info_ref_lock) + remote->can_update_info_refs = 1; + } + if (remote->has_info_packs) + fetch_indices(); + /* Get a list of all local and remote heads to validate refspecs */ get_local_heads(); fprintf(stderr, "Fetching remote heads...\n"); @@ -1690,7 +2204,6 @@ int main(int argc, char **argv) return 0; } - int ret = 0; int new_refs = 0; struct ref *ref; for (ref = remote_refs; ref; ref = ref->next) { @@ -1722,14 +2235,14 @@ int main(int argc, char **argv) "need to pull first?", ref->name, ref->peer_ref->name); - ret = -2; + rc = -2; continue; } } memcpy(ref->new_sha1, ref->peer_ref->new_sha1, 20); if (is_zero_sha1(ref->new_sha1)) { error("cannot happen anymore"); - ret = -3; + rc = -3; continue; } new_refs++; @@ -1752,23 +2265,20 @@ int main(int argc, char **argv) } /* Set up revision info for this refspec */ - const char *commit_argv[3]; - int commit_argc = 2; + const char *commit_argv[4]; + int commit_argc = 3; char *new_sha1_hex = strdup(sha1_to_hex(ref->new_sha1)); char *old_sha1_hex = NULL; - commit_argv[1] = new_sha1_hex; + commit_argv[1] = "--objects"; + commit_argv[2] = new_sha1_hex; if (!push_all && !is_zero_sha1(ref->old_sha1)) { old_sha1_hex = xmalloc(42); sprintf(old_sha1_hex, "^%s", sha1_to_hex(ref->old_sha1)); - commit_argv[2] = old_sha1_hex; + commit_argv[3] = old_sha1_hex; commit_argc++; } - revs.commits = NULL; setup_revisions(commit_argc, commit_argv, &revs, NULL); - revs.tag_objects = 1; - revs.tree_objects = 1; - revs.blob_objects = 1; free(new_sha1_hex); if (old_sha1_hex) { free(old_sha1_hex); @@ -1779,13 +2289,15 @@ int main(int argc, char **argv) pushing = 0; prepare_revision_walk(&revs); mark_edges_uninteresting(revs.commits); - fetch_indices(); - get_delta(&revs, ref_lock); + objects_to_send = get_delta(&revs, ref_lock); finish_all_active_slots(); /* Push missing objects to remote, this would be a convenient time to pack them first if appropriate. */ pushing = 1; + if (objects_to_send) + fprintf(stderr, " sending %d objects\n", + objects_to_send); fill_active_slots(); finish_all_active_slots(); @@ -1799,7 +2311,20 @@ int main(int argc, char **argv) if (!rc) fprintf(stderr, " done\n"); unlock_remote(ref_lock); + check_locks(); + } + + /* Update remote server info if appropriate */ + if (remote->has_info_refs && new_refs) { + if (info_ref_lock && remote->can_update_info_refs) { + fprintf(stderr, "Updating remote server info\n"); + update_remote_info_refs(info_ref_lock); + } else { + fprintf(stderr, "Unable to update server info\n"); + } } + if (info_ref_lock) + unlock_remote(info_ref_lock); cleanup: free(remote); @@ -339,6 +339,7 @@ struct active_request_slot *get_active_slot(void) slot->in_use = 1; slot->local = NULL; slot->results = NULL; + slot->finished = NULL; slot->callback_data = NULL; slot->callback_func = NULL; curl_easy_setopt(slot->curl, CURLOPT_HTTPHEADER, pragma_header); @@ -389,8 +390,10 @@ void run_active_slot(struct active_request_slot *slot) fd_set excfds; int max_fd; struct timeval select_timeout; + int finished = 0; - while (slot->in_use) { + slot->finished = &finished; + while (!finished) { data_received = 0; step_active_slots(); @@ -442,6 +445,9 @@ static void finish_active_slot(struct active_request_slot *slot) closedown_active_slot(slot); curl_easy_getinfo(slot->curl, CURLINFO_HTTP_CODE, &slot->http_code); + if (slot->finished != NULL) + (*slot->finished) = 1; + /* Store slot results so they can be read after the slot is reused */ if (slot->results != NULL) { slot->results->curl_result = slot->curl_result; @@ -35,6 +35,7 @@ struct active_request_slot int in_use; CURLcode curl_result; long http_code; + int *finished; struct slot_results *results; void *callback_data; void (*callback_func)(void *data); diff --git a/pack-check.c b/pack-check.c index eca32b6cab..84ed90d369 100644 --- a/pack-check.c +++ b/pack-check.c @@ -70,13 +70,17 @@ static int verify_packfile(struct packed_git *p) } +#define MAX_CHAIN 40 + static void show_pack_info(struct packed_git *p) { struct pack_header *hdr; int nr_objects, i; + unsigned int chain_histogram[MAX_CHAIN]; hdr = p->pack_base; nr_objects = ntohl(hdr->hdr_entries); + memset(chain_histogram, 0, sizeof(chain_histogram)); for (i = 0; i < nr_objects; i++) { unsigned char sha1[20], base_sha1[20]; @@ -97,11 +101,25 @@ static void show_pack_info(struct packed_git *p) printf("%s ", sha1_to_hex(sha1)); if (!delta_chain_length) printf("%-6s %lu %u\n", type, size, e.offset); - else + else { printf("%-6s %lu %u %u %s\n", type, size, e.offset, delta_chain_length, sha1_to_hex(base_sha1)); + if (delta_chain_length < MAX_CHAIN) + chain_histogram[delta_chain_length]++; + else + chain_histogram[0]++; + } } + for (i = 0; i < MAX_CHAIN; i++) { + if (!chain_histogram[i]) + continue; + printf("chain length %s %d: %d object%s\n", + i ? "=" : ">=", + i ? i : MAX_CHAIN, + chain_histogram[i], + 1 < chain_histogram[i] ? "s" : ""); + } } int verify_pack(struct packed_git *p, int verbose) diff --git a/pack-objects.c b/pack-objects.c index 136a7f5aad..49357c6735 100644 --- a/pack-objects.c +++ b/pack-objects.c @@ -32,9 +32,6 @@ struct object_entry { * be used as the base objectto delta huge * objects against. */ - int based_on_preferred; /* current delta candidate is a preferred - * one, or delta against a preferred one. - */ }; /* @@ -824,8 +821,6 @@ static int try_delta(struct unpacked *cur, struct unpacked *old, unsigned max_de { struct object_entry *cur_entry = cur->entry; struct object_entry *old_entry = old->entry; - int old_preferred = (old_entry->preferred_base || - old_entry->based_on_preferred); unsigned long size, oldsize, delta_size, sizediff; long max_size; void *delta_buf; @@ -867,27 +862,8 @@ static int try_delta(struct unpacked *cur, struct unpacked *old, unsigned max_de * delete). */ max_size = size / 2 - 20; - if (cur_entry->delta) { - if (cur_entry->based_on_preferred) { - if (old_preferred) - max_size = cur_entry->delta_size-1; - else - /* trying with non-preferred one when we - * already have a delta based on preferred - * one is pointless. - */ - return -1; - } - else if (!old_preferred) - max_size = cur_entry->delta_size-1; - else - /* otherwise... even if delta with a - * preferred one produces a bigger result than - * what we currently have, which is based on a - * non-preferred one, it is OK. - */ - ; - } + if (cur_entry->delta) + max_size = cur_entry->delta_size-1; if (sizediff >= max_size) return -1; delta_buf = diff_delta(old->data, oldsize, @@ -897,7 +873,6 @@ static int try_delta(struct unpacked *cur, struct unpacked *old, unsigned max_de cur_entry->delta = old_entry; cur_entry->delta_size = delta_size; cur_entry->depth = old_entry->depth + 1; - cur_entry->based_on_preferred = old_preferred; free(delta_buf); return 0; } @@ -966,6 +941,15 @@ static void find_deltas(struct object_entry **list, int window, int depth) if (try_delta(n, m, depth) < 0) break; } +#if 0 + /* if we made n a delta, and if n is already at max + * depth, leaving it in the window is pointless. we + * should evict it first. + * ... in theory only; somehow this makes things worse. + */ + if (entry->delta && depth <= entry->depth) + continue; +#endif idx++; if (idx >= window) idx = 0; diff --git a/rev-list.c b/rev-list.c index 8e4d83efba..812d237f47 100644 --- a/rev-list.c +++ b/rev-list.c @@ -190,7 +190,7 @@ static int count_distance(struct commit_list *entry) if (commit->object.flags & (UNINTERESTING | COUNTED)) break; - if (!revs.paths || (commit->object.flags & TREECHANGE)) + if (!revs.prune_fn || (commit->object.flags & TREECHANGE)) nr++; commit->object.flags |= COUNTED; p = commit->parents; @@ -224,7 +224,7 @@ static struct commit_list *find_bisection(struct commit_list *list) nr = 0; p = list; while (p) { - if (!revs.paths || (p->item->object.flags & TREECHANGE)) + if (!revs.prune_fn || (p->item->object.flags & TREECHANGE)) nr++; p = p->next; } @@ -234,7 +234,7 @@ static struct commit_list *find_bisection(struct commit_list *list) for (p = list; p; p = p->next) { int distance; - if (revs.paths && !(p->item->object.flags & TREECHANGE)) + if (revs.prune_fn && !(p->item->object.flags & TREECHANGE)) continue; distance = count_distance(p); diff --git a/revision.c b/revision.c index 03085ffe9d..38a8d1b937 100644 --- a/revision.c +++ b/revision.c @@ -199,31 +199,27 @@ static int everybody_uninteresting(struct commit_list *orig) return 1; } -#define TREE_SAME 0 -#define TREE_NEW 1 -#define TREE_DIFFERENT 2 -static int tree_difference = TREE_SAME; +static int tree_difference = REV_TREE_SAME; static void file_add_remove(struct diff_options *options, int addremove, unsigned mode, const unsigned char *sha1, const char *base, const char *path) { - int diff = TREE_DIFFERENT; + int diff = REV_TREE_DIFFERENT; /* - * Is it an add of a new file? It means that - * the old tree didn't have it at all, so we - * will turn "TREE_SAME" -> "TREE_NEW", but - * leave any "TREE_DIFFERENT" alone (and if - * it already was "TREE_NEW", we'll keep it - * "TREE_NEW" of course). + * Is it an add of a new file? It means that the old tree + * didn't have it at all, so we will turn "REV_TREE_SAME" -> + * "REV_TREE_NEW", but leave any "REV_TREE_DIFFERENT" alone + * (and if it already was "REV_TREE_NEW", we'll keep it + * "REV_TREE_NEW" of course). */ if (addremove == '+') { diff = tree_difference; - if (diff != TREE_SAME) + if (diff != REV_TREE_SAME) return; - diff = TREE_NEW; + diff = REV_TREE_NEW; } tree_difference = diff; } @@ -234,7 +230,7 @@ static void file_change(struct diff_options *options, const unsigned char *new_sha1, const char *base, const char *path) { - tree_difference = TREE_DIFFERENT; + tree_difference = REV_TREE_DIFFERENT; } static struct diff_options diff_opt = { @@ -243,19 +239,19 @@ static struct diff_options diff_opt = { .change = file_change, }; -static int compare_tree(struct tree *t1, struct tree *t2) +int rev_compare_tree(struct tree *t1, struct tree *t2) { if (!t1) - return TREE_NEW; + return REV_TREE_NEW; if (!t2) - return TREE_DIFFERENT; - tree_difference = TREE_SAME; + return REV_TREE_DIFFERENT; + tree_difference = REV_TREE_SAME; if (diff_tree_sha1(t1->object.sha1, t2->object.sha1, "", &diff_opt) < 0) - return TREE_DIFFERENT; + return REV_TREE_DIFFERENT; return tree_difference; } -static int same_tree_as_empty(struct tree *t1) +int rev_same_tree_as_empty(struct tree *t1) { int retval; void *tree; @@ -288,7 +284,7 @@ static void try_to_simplify_commit(struct rev_info *revs, struct commit *commit) return; if (!commit->parents) { - if (!same_tree_as_empty(commit->tree)) + if (!rev_same_tree_as_empty(commit->tree)) commit->object.flags |= TREECHANGE; return; } @@ -298,8 +294,8 @@ static void try_to_simplify_commit(struct rev_info *revs, struct commit *commit) struct commit *p = parent->item; parse_commit(p); - switch (compare_tree(p->tree, commit->tree)) { - case TREE_SAME: + switch (rev_compare_tree(p->tree, commit->tree)) { + case REV_TREE_SAME: if (p->object.flags & UNINTERESTING) { /* Even if a merge with an uninteresting * side branch brought the entire change @@ -314,9 +310,9 @@ static void try_to_simplify_commit(struct rev_info *revs, struct commit *commit) commit->parents = parent; return; - case TREE_NEW: + case REV_TREE_NEW: if (revs->remove_empty_trees && - same_tree_as_empty(p->tree)) { + rev_same_tree_as_empty(p->tree)) { /* We are adding all the specified paths from * this parent, so the parents of it is * not interesting, but the difference between @@ -325,7 +321,7 @@ static void try_to_simplify_commit(struct rev_info *revs, struct commit *commit) p->object.flags |= UNINTERESTING; } /* fallthrough */ - case TREE_DIFFERENT: + case REV_TREE_DIFFERENT: tree_changed = 1; pp = &parent->next; continue; @@ -373,8 +369,8 @@ static void add_parents_to_list(struct rev_info *revs, struct commit *commit, st * simplify the commit history and find the parent * that has no differences in the path set if one exists. */ - if (revs->paths) - try_to_simplify_commit(revs, commit); + if (revs->prune_fn) + revs->prune_fn(revs, commit); parent = commit->parents; while (parent) { @@ -396,9 +392,6 @@ static void limit_list(struct rev_info *revs) struct commit_list *newlist = NULL; struct commit_list **p = &newlist; - if (revs->paths) - diff_tree_setup_paths(revs->paths); - while (list) { struct commit_list *entry = list; struct commit *commit = list->item; @@ -450,6 +443,23 @@ static void handle_all(struct rev_info *revs, unsigned flags) for_each_ref(handle_one_ref); } +void init_revisions(struct rev_info *revs) +{ + memset(revs, 0, sizeof(*revs)); + revs->lifo = 1; + revs->dense = 1; + revs->prefix = setup_git_directory(); + revs->max_age = -1; + revs->min_age = -1; + revs->max_count = -1; + + revs->prune_fn = NULL; + revs->prune_data = NULL; + + revs->topo_setter = topo_sort_default_setter; + revs->topo_getter = topo_sort_default_getter; +} + /* * Parse revision information, filling in the "rev_info" structure, * and removing the used arguments from the argument list. @@ -463,13 +473,7 @@ int setup_revisions(int argc, const char **argv, struct rev_info *revs, const ch const char **unrecognized = argv + 1; int left = 1; - memset(revs, 0, sizeof(*revs)); - revs->lifo = 1; - revs->dense = 1; - revs->prefix = setup_git_directory(); - revs->max_age = -1; - revs->min_age = -1; - revs->max_count = -1; + init_revisions(revs); /* First, search for "--" */ seen_dashdash = 0; @@ -479,7 +483,7 @@ int setup_revisions(int argc, const char **argv, struct rev_info *revs, const ch continue; argv[i] = NULL; argc = i; - revs->paths = get_pathspec(revs->prefix, argv + i + 1); + revs->prune_data = get_pathspec(revs->prefix, argv + i + 1); seen_dashdash = 1; break; } @@ -643,7 +647,7 @@ int setup_revisions(int argc, const char **argv, struct rev_info *revs, const ch if (lstat(argv[j], &st) < 0) die("'%s': %s", arg, strerror(errno)); } - revs->paths = get_pathspec(revs->prefix, argv + i); + revs->prune_data = get_pathspec(revs->prefix, argv + i); break; } commit = get_commit_reference(revs, arg, sha1, flags ^ local_flags); @@ -657,8 +661,13 @@ int setup_revisions(int argc, const char **argv, struct rev_info *revs, const ch commit = get_commit_reference(revs, def, sha1, 0); add_one_commit(commit, revs); } - if (revs->paths) + + if (revs->prune_data) { + diff_tree_setup_paths(revs->prune_data); + revs->prune_fn = try_to_simplify_commit; revs->limited = 1; + } + return left; } @@ -668,7 +677,9 @@ void prepare_revision_walk(struct rev_info *revs) if (revs->limited) limit_list(revs); if (revs->topo_order) - sort_in_topological_order(&revs->commits, revs->lifo); + sort_in_topological_order_fn(&revs->commits, revs->lifo, + revs->topo_setter, + revs->topo_getter); } static int rewrite_one(struct commit **pp) @@ -724,7 +735,7 @@ struct commit *get_revision(struct rev_info *revs) return NULL; if (revs->no_merges && commit->parents && commit->parents->next) goto next; - if (revs->paths && revs->dense) { + if (revs->prune_fn && revs->dense) { if (!(commit->object.flags & TREECHANGE)) goto next; rewrite_parents(commit); diff --git a/revision.h b/revision.h index 31e8f61567..6c2becad13 100644 --- a/revision.h +++ b/revision.h @@ -7,6 +7,10 @@ #define SHOWN (1u<<3) #define TMP_MARK (1u<<4) /* for isolated cases; clean after use */ +struct rev_info; + +typedef void (prune_fn_t)(struct rev_info *revs, struct commit *commit); + struct rev_info { /* Starting list */ struct commit_list *commits; @@ -14,7 +18,8 @@ struct rev_info { /* Basic information */ const char *prefix; - const char **paths; + void *prune_data; + prune_fn_t *prune_fn; /* Traversal flags */ unsigned int dense:1, @@ -33,9 +38,20 @@ struct rev_info { int max_count; unsigned long max_age; unsigned long min_age; + + topo_sort_set_fn_t topo_setter; + topo_sort_get_fn_t topo_getter; }; +#define REV_TREE_SAME 0 +#define REV_TREE_NEW 1 +#define REV_TREE_DIFFERENT 2 + /* revision.c */ +extern int rev_same_tree_as_empty(struct tree *t1); +extern int rev_compare_tree(struct tree *t1, struct tree *t2); + +extern void init_revisions(struct rev_info *revs); extern int setup_revisions(int argc, const char **argv, struct rev_info *revs, const char *def); extern void prepare_revision_walk(struct rev_info *revs); extern struct commit *get_revision(struct rev_info *revs); |