From: Junio C Hamano Date: Wed, 2 Jul 2014 19:53:02 +0000 (-0700) Subject: Merge branch 'jk/commit-buffer-length' X-Git-Tag: v2.1.0-rc0~70 X-Git-Url: https://git.lorimer.id.au/gitweb.git/diff_plain/8061ae8b4618d69a67ff6cfb687d7044964d4b04?ds=inline;hp=-c Merge branch 'jk/commit-buffer-length' Move "commit->buffer" out of the in-core commit object and keep track of their lengths. Use this to optimize the code paths to validate GPG signatures in commit objects. * jk/commit-buffer-length: reuse cached commit buffer when parsing signatures commit: record buffer length in cache commit: convert commit->buffer to a slab commit-slab: provide a static initializer use get_commit_buffer everywhere convert logmsg_reencode to get_commit_buffer use get_commit_buffer to avoid duplicate code use get_cached_commit_buffer where appropriate provide helpers to access the commit buffer provide a helper to set the commit buffer provide a helper to free commit buffer sequencer: use logmsg_reencode in get_message logmsg_reencode: return const buffer do not create "struct commit" with xcalloc commit: push commit_index update into alloc_commit_node alloc: include any-object allocations in alloc_report replace dangerous uses of strbuf_attach commit_tree: take a pointer/len pair rather than a const strbuf --- 8061ae8b4618d69a67ff6cfb687d7044964d4b04 diff --combined builtin/blame.c index 662e3fec44,b84e375b5c..d3b256e545 --- a/builtin/blame.c +++ b/builtin/blame.c @@@ -1,8 -1,7 +1,8 @@@ /* * Blame * - * Copyright (c) 2006, Junio C Hamano + * Copyright (c) 2006, 2014 by its authors + * See COPYING for licensing conditions */ #include "cache.h" @@@ -19,9 -18,7 +19,9 @@@ #include "cache-tree.h" #include "string-list.h" #include "mailmap.h" +#include "mergesort.h" #include "parse-options.h" +#include "prio-queue.h" #include "utf8.h" #include "userdiff.h" #include "line-range.h" @@@ -86,42 -83,11 +86,42 @@@ static unsigned blame_copy_score */ struct origin { int refcnt; + /* Record preceding blame record for this blob */ struct origin *previous; + /* origins are put in a list linked via `next' hanging off the + * corresponding commit's util field in order to make finding + * them fast. The presence in this chain does not count + * towards the origin's reference count. It is tempting to + * let it count as long as the commit is pending examination, + * but even under circumstances where the commit will be + * present multiple times in the priority queue of unexamined + * commits, processing the first instance will not leave any + * work requiring the origin data for the second instance. An + * interspersed commit changing that would have to be + * preexisting with a different ancestry and with the same + * commit date in order to wedge itself between two instances + * of the same commit in the priority queue _and_ produce + * blame entries relevant for it. While we don't want to let + * us get tripped up by this case, it certainly does not seem + * worth optimizing for. + */ + struct origin *next; struct commit *commit; + /* `suspects' contains blame entries that may be attributed to + * this origin's commit or to parent commits. When a commit + * is being processed, all suspects will be moved, either by + * assigning them to an origin in a different commit, or by + * shipping them to the scoreboard's ent list because they + * cannot be attributed to a different commit. + */ + struct blame_entry *suspects; mmfile_t file; unsigned char blob_sha1[20]; unsigned mode; + /* guilty gets set when shipping any suspects to the final + * blame list instead of other commits + */ + char guilty; char path[FLEX_ARRAY]; }; @@@ -210,22 -176,10 +210,22 @@@ static inline struct origin *origin_inc static void origin_decref(struct origin *o) { if (o && --o->refcnt <= 0) { + struct origin *p, *l = NULL; if (o->previous) origin_decref(o->previous); free(o->file.ptr); - free(o); + /* Should be present exactly once in commit chain */ + for (p = o->commit->util; p; l = p, p = p->next) { + if (p == o) { + if (l) + l->next = p->next; + else + o->commit->util = p->next; + free(o); + return; + } + } + die("internal error in blame::origin_decref"); } } @@@ -239,12 -193,8 +239,12 @@@ static void drop_origin_blob(struct ori /* * Each group of lines is described by a blame_entry; it can be split - * as we pass blame to the parents. They form a linked list in the - * scoreboard structure, sorted by the target line number. + * as we pass blame to the parents. They are arranged in linked lists + * kept as `suspects' of some unprocessed origin, or entered (when the + * blame origin has been finalized) into the scoreboard structure. + * While the scoreboard structure is only sorted at the end of + * processing (according to final image line number), the lists + * attached to an origin are sorted by the target line number. */ struct blame_entry { struct blame_entry *next; @@@ -260,6 -210,15 +260,6 @@@ /* the commit that introduced this group into the final image */ struct origin *suspect; - /* true if the suspect is truly guilty; false while we have not - * checked if the group came from one of its parents. - */ - char guilty; - - /* true if the entry has been scanned for copies in the current parent - */ - char scanned; - /* the line number of the first line of this group in the * suspect's file; internally all line numbers are 0 based. */ @@@ -271,113 -230,12 +271,113 @@@ unsigned score; }; +/* + * Any merge of blames happens on lists of blames that arrived via + * different parents in a single suspect. In this case, we want to + * sort according to the suspect line numbers as opposed to the final + * image line numbers. The function body is somewhat longish because + * it avoids unnecessary writes. + */ + +static struct blame_entry *blame_merge(struct blame_entry *list1, + struct blame_entry *list2) +{ + struct blame_entry *p1 = list1, *p2 = list2, + **tail = &list1; + + if (!p1) + return p2; + if (!p2) + return p1; + + if (p1->s_lno <= p2->s_lno) { + do { + tail = &p1->next; + if ((p1 = *tail) == NULL) { + *tail = p2; + return list1; + } + } while (p1->s_lno <= p2->s_lno); + } + for (;;) { + *tail = p2; + do { + tail = &p2->next; + if ((p2 = *tail) == NULL) { + *tail = p1; + return list1; + } + } while (p1->s_lno > p2->s_lno); + *tail = p1; + do { + tail = &p1->next; + if ((p1 = *tail) == NULL) { + *tail = p2; + return list1; + } + } while (p1->s_lno <= p2->s_lno); + } +} + +static void *get_next_blame(const void *p) +{ + return ((struct blame_entry *)p)->next; +} + +static void set_next_blame(void *p1, void *p2) +{ + ((struct blame_entry *)p1)->next = p2; +} + +/* + * Final image line numbers are all different, so we don't need a + * three-way comparison here. + */ + +static int compare_blame_final(const void *p1, const void *p2) +{ + return ((struct blame_entry *)p1)->lno > ((struct blame_entry *)p2)->lno + ? 1 : -1; +} + +static int compare_blame_suspect(const void *p1, const void *p2) +{ + const struct blame_entry *s1 = p1, *s2 = p2; + /* + * to allow for collating suspects, we sort according to the + * respective pointer value as the primary sorting criterion. + * The actual relation is pretty unimportant as long as it + * establishes a total order. Comparing as integers gives us + * that. + */ + if (s1->suspect != s2->suspect) + return (intptr_t)s1->suspect > (intptr_t)s2->suspect ? 1 : -1; + if (s1->s_lno == s2->s_lno) + return 0; + return s1->s_lno > s2->s_lno ? 1 : -1; +} + +static struct blame_entry *blame_sort(struct blame_entry *head, + int (*compare_fn)(const void *, const void *)) +{ + return llist_mergesort (head, get_next_blame, set_next_blame, compare_fn); +} + +static int compare_commits_by_reverse_commit_date(const void *a, + const void *b, + void *c) +{ + return -compare_commits_by_commit_date(a, b, c); +} + /* * The current state of the blame assignment. */ struct scoreboard { /* the final commit (i.e. where we started digging from) */ struct commit *final; + /* Priority queue for commits with unassigned blame records */ + struct prio_queue commits; struct rev_info *revs; const char *path; @@@ -410,6 -268,7 +410,6 @@@ static void coalesce(struct scoreboard for (ent = sb->ent; ent && (next = ent->next); ent = next) { if (ent->suspect == next->suspect && - ent->guilty == next->guilty && ent->s_lno + ent->num_lines == next->s_lno) { ent->num_lines += next->num_lines; ent->next = next->next; @@@ -424,30 -283,6 +424,30 @@@ sanity_check_refcnt(sb); } +/* + * Merge the given sorted list of blames into a preexisting origin. + * If there were no previous blames to that commit, it is entered into + * the commit priority queue of the score board. + */ + +static void queue_blames(struct scoreboard *sb, struct origin *porigin, + struct blame_entry *sorted) +{ + if (porigin->suspects) + porigin->suspects = blame_merge(porigin->suspects, sorted); + else { + struct origin *o; + for (o = porigin->commit->util; o; o = o->next) { + if (o->suspects) { + porigin->suspects = sorted; + return; + } + } + porigin->suspects = sorted; + prio_queue_put(&sb->commits, porigin->commit); + } +} + /* * Given a commit and a path in it, create a new origin structure. * The callers that add blame to the scoreboard should use @@@ -460,32 -295,23 +460,32 @@@ static struct origin *make_origin(struc o = xcalloc(1, sizeof(*o) + strlen(path) + 1); o->commit = commit; o->refcnt = 1; + o->next = commit->util; + commit->util = o; strcpy(o->path, path); return o; } /* * Locate an existing origin or create a new one. + * This moves the origin to front position in the commit util list. */ static struct origin *get_origin(struct scoreboard *sb, struct commit *commit, const char *path) { - struct blame_entry *e; + struct origin *o, *l; - for (e = sb->ent; e; e = e->next) { - if (e->suspect->commit == commit && - !strcmp(e->suspect->path, path)) - return origin_incref(e->suspect); + for (o = commit->util, l = NULL; o; l = o, o = o->next) { + if (!strcmp(o->path, path)) { + /* bump to front */ + if (l) { + l->next = o->next; + o->next = commit->util; + commit->util = o; + } + return origin_incref(o); + } } return make_origin(commit, path); } @@@ -524,19 -350,41 +524,19 @@@ static struct origin *find_origin(struc struct commit *parent, struct origin *origin) { - struct origin *porigin = NULL; + struct origin *porigin; struct diff_options diff_opts; const char *paths[2]; - if (parent->util) { - /* - * Each commit object can cache one origin in that - * commit. This is a freestanding copy of origin and - * not refcounted. - */ - struct origin *cached = parent->util; - if (!strcmp(cached->path, origin->path)) { + /* First check any existing origins */ + for (porigin = parent->util; porigin; porigin = porigin->next) + if (!strcmp(porigin->path, origin->path)) { /* * The same path between origin and its parent * without renaming -- the most common case. */ - porigin = get_origin(sb, parent, cached->path); - - /* - * If the origin was newly created (i.e. get_origin - * would call make_origin if none is found in the - * scoreboard), it does not know the blob_sha1/mode, - * so copy it. Otherwise porigin was in the - * scoreboard and already knows blob_sha1/mode. - */ - if (porigin->refcnt == 1) { - hashcpy(porigin->blob_sha1, cached->blob_sha1); - porigin->mode = cached->mode; - } - return porigin; + return origin_incref (porigin); } - /* otherwise it was not very useful; free it */ - free(parent->util); - parent->util = NULL; - } /* See if the origin->path is different between parent * and origin first. Most of the time they are the @@@ -602,6 -450,19 +602,6 @@@ } diff_flush(&diff_opts); free_pathspec(&diff_opts.pathspec); - if (porigin) { - /* - * Create a freestanding copy that is not part of - * the refcounted origin found in the scoreboard, and - * cache it in the commit. - */ - struct origin *cached; - - cached = make_origin(porigin->commit, porigin->path); - hashcpy(cached->blob_sha1, porigin->blob_sha1); - cached->mode = porigin->mode; - parent->util = cached; - } return porigin; } @@@ -648,31 -509,46 +648,31 @@@ static struct origin *find_rename(struc } /* - * Link in a new blame entry to the scoreboard. Entries that cover the - * same line range have been removed from the scoreboard previously. + * Append a new blame entry to a given output queue. */ -static void add_blame_entry(struct scoreboard *sb, struct blame_entry *e) +static void add_blame_entry(struct blame_entry ***queue, struct blame_entry *e) { - struct blame_entry *ent, *prev = NULL; - origin_incref(e->suspect); - for (ent = sb->ent; ent && ent->lno < e->lno; ent = ent->next) - prev = ent; - - /* prev, if not NULL, is the last one that is below e */ - - if (prev) { - e->next = prev->next; - prev->next = e; - } - else { - e->next = sb->ent; - sb->ent = e; - } + e->next = **queue; + **queue = e; + *queue = &e->next; } /* * src typically is on-stack; we want to copy the information in it to - * a malloced blame_entry that is already on the linked list of the - * scoreboard. The origin of dst loses a refcnt while the origin of src - * gains one. + * a malloced blame_entry that gets added to the given queue. The + * origin of dst loses a refcnt. */ -static void dup_entry(struct blame_entry *dst, struct blame_entry *src) +static void dup_entry(struct blame_entry ***queue, + struct blame_entry *dst, struct blame_entry *src) { - struct blame_entry *n; - - n = dst->next; origin_incref(src->suspect); origin_decref(dst->suspect); memcpy(dst, src, sizeof(*src)); - dst->next = n; - dst->score = 0; + dst->next = **queue; + **queue = dst; + *queue = &dst->next; } static const char *nth_line(struct scoreboard *sb, long lno) @@@ -744,11 -620,10 +744,11 @@@ static void split_overlap(struct blame_ /* * split_overlap() divided an existing blame e into up to three parts - * in split. Adjust the linked list of blames in the scoreboard to + * in split. Any assigned blame is moved to queue to * reflect the split. */ -static void split_blame(struct scoreboard *sb, +static void split_blame(struct blame_entry ***blamed, + struct blame_entry ***unblamed, struct blame_entry *split, struct blame_entry *e) { @@@ -756,39 -631,61 +756,39 @@@ if (split[0].suspect && split[2].suspect) { /* The first part (reuse storage for the existing entry e) */ - dup_entry(e, &split[0]); + dup_entry(unblamed, e, &split[0]); /* The last part -- me */ new_entry = xmalloc(sizeof(*new_entry)); memcpy(new_entry, &(split[2]), sizeof(struct blame_entry)); - add_blame_entry(sb, new_entry); + add_blame_entry(unblamed, new_entry); /* ... and the middle part -- parent */ new_entry = xmalloc(sizeof(*new_entry)); memcpy(new_entry, &(split[1]), sizeof(struct blame_entry)); - add_blame_entry(sb, new_entry); + add_blame_entry(blamed, new_entry); } else if (!split[0].suspect && !split[2].suspect) /* * The parent covers the entire area; reuse storage for * e and replace it with the parent. */ - dup_entry(e, &split[1]); + dup_entry(blamed, e, &split[1]); else if (split[0].suspect) { /* me and then parent */ - dup_entry(e, &split[0]); + dup_entry(unblamed, e, &split[0]); new_entry = xmalloc(sizeof(*new_entry)); memcpy(new_entry, &(split[1]), sizeof(struct blame_entry)); - add_blame_entry(sb, new_entry); + add_blame_entry(blamed, new_entry); } else { /* parent and then me */ - dup_entry(e, &split[1]); + dup_entry(blamed, e, &split[1]); new_entry = xmalloc(sizeof(*new_entry)); memcpy(new_entry, &(split[2]), sizeof(struct blame_entry)); - add_blame_entry(sb, new_entry); - } - - if (DEBUG) { /* sanity */ - struct blame_entry *ent; - int lno = sb->ent->lno, corrupt = 0; - - for (ent = sb->ent; ent; ent = ent->next) { - if (lno != ent->lno) - corrupt = 1; - if (ent->s_lno < 0) - corrupt = 1; - lno += ent->num_lines; - } - if (corrupt) { - lno = sb->ent->lno; - for (ent = sb->ent; ent; ent = ent->next) { - printf("L %8d l %8d n %8d\n", - lno, ent->lno, ent->num_lines); - lno = ent->lno + ent->num_lines; - } - die("oops"); - } + add_blame_entry(unblamed, new_entry); } } @@@ -805,146 -702,74 +805,146 @@@ static void decref_split(struct blame_e } /* - * Helper for blame_chunk(). blame_entry e is known to overlap with - * the patch hunk; split it and pass blame to the parent. + * reverse_blame reverses the list given in head, appending tail. + * That allows us to build lists in reverse order, then reverse them + * afterwards. This can be faster than building the list in proper + * order right away. The reason is that building in proper order + * requires writing a link in the _previous_ element, while building + * in reverse order just requires placing the list head into the + * _current_ element. */ -static void blame_overlap(struct scoreboard *sb, struct blame_entry *e, - int tlno, int plno, int same, - struct origin *parent) -{ - struct blame_entry split[3]; - split_overlap(split, e, tlno, plno, same, parent); - if (split[1].suspect) - split_blame(sb, split, e); - decref_split(split); -} - -/* - * Find the line number of the last line the target is suspected for. - */ -static int find_last_in_target(struct scoreboard *sb, struct origin *target) +static struct blame_entry *reverse_blame(struct blame_entry *head, + struct blame_entry *tail) { - struct blame_entry *e; - int last_in_target = -1; - - for (e = sb->ent; e; e = e->next) { - if (e->guilty || e->suspect != target) - continue; - if (last_in_target < e->s_lno + e->num_lines) - last_in_target = e->s_lno + e->num_lines; + while (head) { + struct blame_entry *next = head->next; + head->next = tail; + tail = head; + head = next; } - return last_in_target; + return tail; } /* * Process one hunk from the patch between the current suspect for - * blame_entry e and its parent. Find and split the overlap, and - * pass blame to the overlapping part to the parent. + * blame_entry e and its parent. This first blames any unfinished + * entries before the chunk (which is where target and parent start + * differing) on the parent, and then splits blame entries at the + * start and at the end of the difference region. Since use of -M and + * -C options may lead to overlapping/duplicate source line number + * ranges, all we can rely on from sorting/merging is the order of the + * first suspect line number. */ -static void blame_chunk(struct scoreboard *sb, - int tlno, int plno, int same, - struct origin *target, struct origin *parent) +static void blame_chunk(struct blame_entry ***dstq, struct blame_entry ***srcq, + int tlno, int offset, int same, + struct origin *parent) { - struct blame_entry *e; + struct blame_entry *e = **srcq; + struct blame_entry *samep = NULL, *diffp = NULL; - for (e = sb->ent; e; e = e->next) { - if (e->guilty || e->suspect != target) - continue; - if (same <= e->s_lno) - continue; - if (tlno < e->s_lno + e->num_lines) - blame_overlap(sb, e, tlno, plno, same, parent); + while (e && e->s_lno < tlno) { + struct blame_entry *next = e->next; + /* + * current record starts before differing portion. If + * it reaches into it, we need to split it up and + * examine the second part separately. + */ + if (e->s_lno + e->num_lines > tlno) { + /* Move second half to a new record */ + int len = tlno - e->s_lno; + struct blame_entry *n = xcalloc(1, sizeof (struct blame_entry)); + n->suspect = e->suspect; + n->lno = e->lno + len; + n->s_lno = e->s_lno + len; + n->num_lines = e->num_lines - len; + e->num_lines = len; + e->score = 0; + /* Push new record to diffp */ + n->next = diffp; + diffp = n; + } else + origin_decref(e->suspect); + /* Pass blame for everything before the differing + * chunk to the parent */ + e->suspect = origin_incref(parent); + e->s_lno += offset; + e->next = samep; + samep = e; + e = next; + } + /* + * As we don't know how much of a common stretch after this + * diff will occur, the currently blamed parts are all that we + * can assign to the parent for now. + */ + + if (samep) { + **dstq = reverse_blame(samep, **dstq); + *dstq = &samep->next; } + /* + * Prepend the split off portions: everything after e starts + * after the blameable portion. + */ + e = reverse_blame(diffp, e); + + /* + * Now retain records on the target while parts are different + * from the parent. + */ + samep = NULL; + diffp = NULL; + while (e && e->s_lno < same) { + struct blame_entry *next = e->next; + + /* + * If current record extends into sameness, need to split. + */ + if (e->s_lno + e->num_lines > same) { + /* + * Move second half to a new record to be + * processed by later chunks + */ + int len = same - e->s_lno; + struct blame_entry *n = xcalloc(1, sizeof (struct blame_entry)); + n->suspect = origin_incref(e->suspect); + n->lno = e->lno + len; + n->s_lno = e->s_lno + len; + n->num_lines = e->num_lines - len; + e->num_lines = len; + e->score = 0; + /* Push new record to samep */ + n->next = samep; + samep = n; + } + e->next = diffp; + diffp = e; + e = next; + } + **srcq = reverse_blame(diffp, reverse_blame(samep, e)); + /* Move across elements that are in the unblamable portion */ + if (diffp) + *srcq = &diffp->next; } struct blame_chunk_cb_data { - struct scoreboard *sb; - struct origin *target; struct origin *parent; - long plno; - long tlno; + long offset; + struct blame_entry **dstq; + struct blame_entry **srcq; }; +/* diff chunks are from parent to target */ static int blame_chunk_cb(long start_a, long count_a, long start_b, long count_b, void *data) { struct blame_chunk_cb_data *d = data; - blame_chunk(d->sb, d->tlno, d->plno, start_b, d->target, d->parent); - d->plno = start_a + count_a; - d->tlno = start_b + count_b; + if (start_a - start_b != d->offset) + die("internal error in blame::blame_chunk_cb"); + blame_chunk(&d->dstq, &d->srcq, start_b, start_a - start_b, + start_b + count_b, d->parent); + d->offset = start_a + count_a - (start_b + count_b); return 0; } @@@ -953,32 -778,29 +953,32 @@@ * for the lines it is suspected to its parent. Run diff to find * which lines came from parent and pass blame for them. */ -static int pass_blame_to_parent(struct scoreboard *sb, - struct origin *target, - struct origin *parent) +static void pass_blame_to_parent(struct scoreboard *sb, + struct origin *target, + struct origin *parent) { - int last_in_target; mmfile_t file_p, file_o; struct blame_chunk_cb_data d; + struct blame_entry *newdest = NULL; - memset(&d, 0, sizeof(d)); - d.sb = sb; d.target = target; d.parent = parent; - last_in_target = find_last_in_target(sb, target); - if (last_in_target < 0) - return 1; /* nothing remains for this target */ + if (!target->suspects) + return; /* nothing remains for this target */ + + d.parent = parent; + d.offset = 0; + d.dstq = &newdest; d.srcq = &target->suspects; fill_origin_blob(&sb->revs->diffopt, parent, &file_p); fill_origin_blob(&sb->revs->diffopt, target, &file_o); num_get_patch++; diff_hunks(&file_p, &file_o, 0, blame_chunk_cb, &d); - /* The rest (i.e. anything after tlno) are the same as the parent */ - blame_chunk(sb, d.tlno, d.plno, last_in_target, target, parent); + /* The rest are the same as the parent */ + blame_chunk(&d.dstq, &d.srcq, INT_MAX, d.offset, INT_MAX, parent); + *d.dstq = NULL; + queue_blames(sb, parent, newdest); - return 0; + return; } /* @@@ -1123,80 -945,43 +1123,80 @@@ static void find_copy_in_blob(struct sc handle_split(sb, ent, d.tlno, d.plno, ent->num_lines, parent, split); } +/* Move all blame entries from list *source that have a score smaller + * than score_min to the front of list *small. + * Returns a pointer to the link pointing to the old head of the small list. + */ + +static struct blame_entry **filter_small(struct scoreboard *sb, + struct blame_entry **small, + struct blame_entry **source, + unsigned score_min) +{ + struct blame_entry *p = *source; + struct blame_entry *oldsmall = *small; + while (p) { + if (ent_score(sb, p) <= score_min) { + *small = p; + small = &p->next; + p = *small; + } else { + *source = p; + source = &p->next; + p = *source; + } + } + *small = oldsmall; + *source = NULL; + return small; +} + /* * See if lines currently target is suspected for can be attributed to * parent. */ -static int find_move_in_parent(struct scoreboard *sb, - struct origin *target, - struct origin *parent) +static void find_move_in_parent(struct scoreboard *sb, + struct blame_entry ***blamed, + struct blame_entry **toosmall, + struct origin *target, + struct origin *parent) { - int last_in_target, made_progress; struct blame_entry *e, split[3]; + struct blame_entry *unblamed = target->suspects; + struct blame_entry *leftover = NULL; mmfile_t file_p; - last_in_target = find_last_in_target(sb, target); - if (last_in_target < 0) - return 1; /* nothing remains for this target */ + if (!unblamed) + return; /* nothing remains for this target */ fill_origin_blob(&sb->revs->diffopt, parent, &file_p); if (!file_p.ptr) - return 0; + return; - made_progress = 1; - while (made_progress) { - made_progress = 0; - for (e = sb->ent; e; e = e->next) { - if (e->guilty || e->suspect != target || - ent_score(sb, e) < blame_move_score) - continue; + /* At each iteration, unblamed has a NULL-terminated list of + * entries that have not yet been tested for blame. leftover + * contains the reversed list of entries that have been tested + * without being assignable to the parent. + */ + do { + struct blame_entry **unblamedtail = &unblamed; + struct blame_entry *next; + for (e = unblamed; e; e = next) { + next = e->next; find_copy_in_blob(sb, e, parent, split, &file_p); if (split[1].suspect && blame_move_score < ent_score(sb, &split[1])) { - split_blame(sb, split, e); - made_progress = 1; + split_blame(blamed, &unblamedtail, split, e); + } else { + e->next = leftover; + leftover = e; } decref_split(split); } - } - return 0; + *unblamedtail = NULL; + toosmall = filter_small(sb, toosmall, &unblamed, blame_move_score); + } while (unblamed); + target->suspects = reverse_blame(leftover, NULL); } struct blame_list { @@@ -1208,46 -993,62 +1208,46 @@@ * Count the number of entries the target is suspected for, * and prepare a list of entry and the best split. */ -static struct blame_list *setup_blame_list(struct scoreboard *sb, - struct origin *target, - int min_score, +static struct blame_list *setup_blame_list(struct blame_entry *unblamed, int *num_ents_p) { struct blame_entry *e; int num_ents, i; struct blame_list *blame_list = NULL; - for (e = sb->ent, num_ents = 0; e; e = e->next) - if (!e->scanned && !e->guilty && - e->suspect == target && - min_score < ent_score(sb, e)) - num_ents++; + for (e = unblamed, num_ents = 0; e; e = e->next) + num_ents++; if (num_ents) { blame_list = xcalloc(num_ents, sizeof(struct blame_list)); - for (e = sb->ent, i = 0; e; e = e->next) - if (!e->scanned && !e->guilty && - e->suspect == target && - min_score < ent_score(sb, e)) - blame_list[i++].ent = e; + for (e = unblamed, i = 0; e; e = e->next) + blame_list[i++].ent = e; } *num_ents_p = num_ents; return blame_list; } -/* - * Reset the scanned status on all entries. - */ -static void reset_scanned_flag(struct scoreboard *sb) -{ - struct blame_entry *e; - for (e = sb->ent; e; e = e->next) - e->scanned = 0; -} - /* * For lines target is suspected for, see if we can find code movement * across file boundary from the parent commit. porigin is the path * in the parent we already tried. */ -static int find_copy_in_parent(struct scoreboard *sb, - struct origin *target, - struct commit *parent, - struct origin *porigin, - int opt) +static void find_copy_in_parent(struct scoreboard *sb, + struct blame_entry ***blamed, + struct blame_entry **toosmall, + struct origin *target, + struct commit *parent, + struct origin *porigin, + int opt) { struct diff_options diff_opts; int i, j; - int retval; struct blame_list *blame_list; int num_ents; + struct blame_entry *unblamed = target->suspects; + struct blame_entry *leftover = NULL; - blame_list = setup_blame_list(sb, target, blame_copy_score, &num_ents); - if (!blame_list) - return 1; /* nothing remains for this target */ + if (!unblamed) + return; /* nothing remains for this target */ diff_setup(&diff_opts); DIFF_OPT_SET(&diff_opts, RECURSIVE); @@@ -1277,9 -1078,9 +1277,9 @@@ if (!DIFF_OPT_TST(&diff_opts, FIND_COPIES_HARDER)) diffcore_std(&diff_opts); - retval = 0; - while (1) { - int made_progress = 0; + do { + struct blame_entry **unblamedtail = &unblamed; + blame_list = setup_blame_list(unblamed, &num_ents); for (i = 0; i < diff_queued_diff.nr; i++) { struct diff_filepair *p = diff_queued_diff.queue[i]; @@@ -1316,21 -1117,27 +1316,21 @@@ struct blame_entry *split = blame_list[j].split; if (split[1].suspect && blame_copy_score < ent_score(sb, &split[1])) { - split_blame(sb, split, blame_list[j].ent); - made_progress = 1; + split_blame(blamed, &unblamedtail, split, + blame_list[j].ent); + } else { + blame_list[j].ent->next = leftover; + leftover = blame_list[j].ent; } - else - blame_list[j].ent->scanned = 1; decref_split(split); } free(blame_list); - - if (!made_progress) - break; - blame_list = setup_blame_list(sb, target, blame_copy_score, &num_ents); - if (!blame_list) { - retval = 1; - break; - } - } - reset_scanned_flag(sb); + *unblamedtail = NULL; + toosmall = filter_small(sb, toosmall, &unblamed, blame_copy_score); + } while (unblamed); + target->suspects = reverse_blame(leftover, NULL); diff_flush(&diff_opts); free_pathspec(&diff_opts.pathspec); - return retval; } /* @@@ -1340,21 -1147,20 +1340,21 @@@ static void pass_whole_blame(struct scoreboard *sb, struct origin *origin, struct origin *porigin) { - struct blame_entry *e; + struct blame_entry *e, *suspects; if (!porigin->file.ptr && origin->file.ptr) { /* Steal its file */ porigin->file = origin->file; origin->file.ptr = NULL; } - for (e = sb->ent; e; e = e->next) { - if (e->suspect != origin) - continue; + suspects = origin->suspects; + origin->suspects = NULL; + for (e = suspects; e; e = e->next) { origin_incref(porigin); origin_decref(e->suspect); e->suspect = porigin; } + queue_blames(sb, porigin, suspects); } /* @@@ -1378,27 -1184,6 +1378,27 @@@ static int num_scapegoats(struct rev_in return cnt; } +/* Distribute collected unsorted blames to the respected sorted lists + * in the various origins. + */ +static void distribute_blame(struct scoreboard *sb, struct blame_entry *blamed) +{ + blamed = blame_sort(blamed, compare_blame_suspect); + while (blamed) + { + struct origin *porigin = blamed->suspect; + struct blame_entry *suspects = NULL; + do { + struct blame_entry *next = blamed->next; + blamed->next = suspects; + suspects = blamed; + blamed = next; + } while (blamed && blamed->suspect == porigin); + suspects = reverse_blame(suspects, NULL); + queue_blames(sb, porigin, suspects); + } +} + #define MAXSG 16 static void pass_blame(struct scoreboard *sb, struct origin *origin, int opt) @@@ -1409,8 -1194,6 +1409,8 @@@ struct commit_list *sg; struct origin *sg_buf[MAXSG]; struct origin *porigin, **sg_origin = sg_buf; + struct blame_entry *toosmall = NULL; + struct blame_entry *blames, **blametail = &blames; num_sg = num_scapegoats(revs, commit); if (!num_sg) @@@ -1472,71 -1255,38 +1472,71 @@@ origin_incref(porigin); origin->previous = porigin; } - if (pass_blame_to_parent(sb, origin, porigin)) + pass_blame_to_parent(sb, origin, porigin); + if (!origin->suspects) goto finish; } /* * Optionally find moves in parents' files. */ - if (opt & PICKAXE_BLAME_MOVE) - for (i = 0, sg = first_scapegoat(revs, commit); - i < num_sg && sg; - sg = sg->next, i++) { - struct origin *porigin = sg_origin[i]; - if (!porigin) - continue; - if (find_move_in_parent(sb, origin, porigin)) - goto finish; + if (opt & PICKAXE_BLAME_MOVE) { + filter_small(sb, &toosmall, &origin->suspects, blame_move_score); + if (origin->suspects) { + for (i = 0, sg = first_scapegoat(revs, commit); + i < num_sg && sg; + sg = sg->next, i++) { + struct origin *porigin = sg_origin[i]; + if (!porigin) + continue; + find_move_in_parent(sb, &blametail, &toosmall, origin, porigin); + if (!origin->suspects) + break; + } } + } /* * Optionally find copies from parents' files. */ - if (opt & PICKAXE_BLAME_COPY) + if (opt & PICKAXE_BLAME_COPY) { + if (blame_copy_score > blame_move_score) + filter_small(sb, &toosmall, &origin->suspects, blame_copy_score); + else if (blame_copy_score < blame_move_score) { + origin->suspects = blame_merge(origin->suspects, toosmall); + toosmall = NULL; + filter_small(sb, &toosmall, &origin->suspects, blame_copy_score); + } + if (!origin->suspects) + goto finish; + for (i = 0, sg = first_scapegoat(revs, commit); i < num_sg && sg; sg = sg->next, i++) { struct origin *porigin = sg_origin[i]; - if (find_copy_in_parent(sb, origin, sg->item, - porigin, opt)) + find_copy_in_parent(sb, &blametail, &toosmall, + origin, sg->item, porigin, opt); + if (!origin->suspects) goto finish; } + } - finish: +finish: + *blametail = NULL; + distribute_blame(sb, blames); + /* + * prepend toosmall to origin->suspects + * + * There is no point in sorting: this ends up on a big + * unsorted list in the caller anyway. + */ + if (toosmall) { + struct blame_entry **tail = &toosmall; + while (*tail) + tail = &(*tail)->next; + *tail = origin->suspects; + origin->suspects = toosmall; + } for (i = 0; i < num_sg; i++) { if (sg_origin[i]) { drop_origin_blob(sg_origin[i]); @@@ -1655,7 -1405,7 +1655,7 @@@ static void get_commit_info(struct comm { int len; const char *subject, *encoding; - char *message; + const char *message; commit_info_init(ret); @@@ -1666,7 -1416,7 +1666,7 @@@ &ret->author_time, &ret->author_tz); if (!detailed) { - logmsg_free(message, commit); + unuse_commit_buffer(commit, message); return; } @@@ -1680,7 -1430,7 +1680,7 @@@ else strbuf_addf(&ret->summary, "(%s)", sha1_to_hex(commit->object.sha1)); - logmsg_free(message, commit); + unuse_commit_buffer(commit, message); } /* @@@ -1731,11 -1481,14 +1731,11 @@@ static int emit_one_suspect_detail(stru } /* - * The blame_entry is found to be guilty for the range. Mark it - * as such, and show it in incremental output. + * The blame_entry is found to be guilty for the range. + * Show it in incremental output. */ static void found_guilty_entry(struct blame_entry *ent) { - if (ent->guilty) - return; - ent->guilty = 1; if (incremental) { struct origin *suspect = ent->suspect; @@@ -1749,34 -1502,32 +1749,34 @@@ } /* - * The main loop -- while the scoreboard has lines whose true origin - * is still unknown, pick one blame_entry, and allow its current - * suspect to pass blames to its parents. - */ + * The main loop -- while we have blobs with lines whose true origin + * is still unknown, pick one blob, and allow its lines to pass blames + * to its parents. */ static void assign_blame(struct scoreboard *sb, int opt) { struct rev_info *revs = sb->revs; + struct commit *commit = prio_queue_get(&sb->commits); - while (1) { + while (commit) { struct blame_entry *ent; - struct commit *commit; - struct origin *suspect = NULL; + struct origin *suspect = commit->util; /* find one suspect to break down */ - for (ent = sb->ent; !suspect && ent; ent = ent->next) - if (!ent->guilty) - suspect = ent->suspect; - if (!suspect) - return; /* all done */ + while (suspect && !suspect->suspects) + suspect = suspect->next; + + if (!suspect) { + commit = prio_queue_get(&sb->commits); + continue; + } + + assert(commit == suspect->commit); /* * We will use this suspect later in the loop, * so hold onto it in the meantime. */ origin_incref(suspect); - commit = suspect->commit; parse_commit(commit); if (reverse || (!(commit->object.flags & UNINTERESTING) && @@@ -1792,22 -1543,9 +1792,22 @@@ commit->object.flags |= UNINTERESTING; /* Take responsibility for the remaining entries */ - for (ent = sb->ent; ent; ent = ent->next) - if (ent->suspect == suspect) + ent = suspect->suspects; + if (ent) { + suspect->guilty = 1; + for (;;) { + struct blame_entry *next = ent->next; found_guilty_entry(ent); + if (next) { + ent = next; + continue; + } + ent->next = sb->ent; + sb->ent = suspect->suspects; + suspect->suspects = NULL; + break; + } + } origin_decref(suspect); if (DEBUG) /* sanity */ @@@ -1818,29 -1556,22 +1818,29 @@@ static const char *format_time(unsigned long time, const char *tz_str, int show_raw_time) { - static char time_buf[128]; + static struct strbuf time_buf = STRBUF_INIT; + strbuf_reset(&time_buf); if (show_raw_time) { - snprintf(time_buf, sizeof(time_buf), "%lu %s", time, tz_str); + strbuf_addf(&time_buf, "%lu %s", time, tz_str); } else { const char *time_str; - int time_len; + size_t time_width; int tz; tz = atoi(tz_str); time_str = show_date(time, tz, blame_date_mode); - time_len = strlen(time_str); - memcpy(time_buf, time_str, time_len); - memset(time_buf + time_len, ' ', blame_date_width - time_len); + strbuf_addstr(&time_buf, time_str); + /* + * Add space paddings to time_buf to display a fixed width + * string, and use time_width for display width calibration. + */ + for (time_width = utf8_strwidth(time_str); + time_width < blame_date_width; + time_width++) + strbuf_addch(&time_buf, ' '); } - return time_buf; + return time_buf.buf; } #define OUTPUT_ANNOTATE_COMPAT 001 @@@ -1871,8 -1602,9 +1871,8 @@@ static void emit_porcelain(struct score char hex[41]; strcpy(hex, sha1_to_hex(suspect->commit->object.sha1)); - printf("%s%c%d %d %d\n", + printf("%s %d %d %d\n", hex, - ent->guilty ? ' ' : '*', /* purely for debugging */ ent->s_lno + 1, ent->lno + 1, ent->num_lines); @@@ -1985,16 -1717,17 +1985,16 @@@ static void output(struct scoreboard *s if (option & OUTPUT_PORCELAIN) { for (ent = sb->ent; ent; ent = ent->next) { - struct blame_entry *oth; - struct origin *suspect = ent->suspect; - struct commit *commit = suspect->commit; + int count = 0; + struct origin *suspect; + struct commit *commit = ent->suspect->commit; if (commit->object.flags & MORE_THAN_ONE_PATH) continue; - for (oth = ent->next; oth; oth = oth->next) { - if ((oth->suspect->commit != commit) || - !strcmp(oth->suspect->path, suspect->path)) - continue; - commit->object.flags |= MORE_THAN_ONE_PATH; - break; + for (suspect = commit->util; suspect; suspect = suspect->next) { + if (suspect->guilty && count++) { + commit->object.flags |= MORE_THAN_ONE_PATH; + break; + } } } } @@@ -2008,12 -1741,6 +2008,12 @@@ } } +static const char *get_next_line(const char *start, const char *end) +{ + const char *nl = memchr(start, '\n', end - start); + return nl ? nl + 1 : end; +} + /* * To allow quick access to the contents of nth line in the * final image, prepare an index in the scoreboard. @@@ -2025,19 -1752,39 +2025,19 @@@ static int prepare_lines(struct scorebo const char *end = buf + len; const char *p; int *lineno; - int num = 0, incomplete = 0; + int num = 0; - for (p = buf;;) { - p = memchr(p, '\n', end - p); - if (p) { - p++; - num++; - continue; - } - break; - } + for (p = buf; p < end; p = get_next_line(p, end)) + num++; - if (len && end[-1] != '\n') - incomplete++; /* incomplete line at the end */ + sb->lineno = lineno = xmalloc(sizeof(*sb->lineno) * (num + 1)); - sb->lineno = xmalloc(sizeof(*sb->lineno) * (num + incomplete + 1)); - lineno = sb->lineno; + for (p = buf; p < end; p = get_next_line(p, end)) + *lineno++ = p - buf; - *lineno++ = 0; - for (p = buf;;) { - p = memchr(p, '\n', end - p); - if (p) { - p++; - *lineno++ = p - buf; - continue; - } - break; - } + *lineno = len; - if (incomplete) - *lineno++ = len; - - sb->num_lines = num + incomplete; + sb->num_lines = num; return sb->num_lines; } @@@ -2251,6 -1998,18 +2251,18 @@@ static void append_merge_parents(struc strbuf_release(&line); } + /* + * This isn't as simple as passing sb->buf and sb->len, because we + * want to transfer ownership of the buffer to the commit (so we + * must use detach). + */ + static void set_commit_buffer_from_strbuf(struct commit *c, struct strbuf *sb) + { + size_t len; + void *buf = strbuf_detach(sb, &len); + set_commit_buffer(c, buf, len); + } + /* * Prepare a dummy commit that represents the work tree (or staged) item. * Note that annotating work tree item never works in the reverse. @@@ -2272,7 -2031,7 +2284,7 @@@ static struct commit *fake_working_tree struct strbuf msg = STRBUF_INIT; time(&now); - commit = xcalloc(1, sizeof(*commit)); + commit = alloc_commit_node(); commit->object.parsed = 1; commit->date = now; commit->object.type = OBJ_COMMIT; @@@ -2299,7 -2058,7 +2311,7 @@@ ident, ident, path, (!contents_from ? path : (!strcmp(contents_from, "-") ? "standard input" : contents_from))); - commit->buffer = strbuf_detach(&msg, NULL); + set_commit_buffer_from_strbuf(commit, &msg); if (!contents_from || strcmp("-", contents_from)) { struct stat st; @@@ -2341,9 -2100,11 +2353,9 @@@ if (strbuf_read(&buf, 0, 0) < 0) die_errno("failed to read from stdin"); } - convert_to_git(path, buf.buf, buf.len, &buf, 0); origin->file.ptr = buf.buf; origin->file.size = buf.len; pretend_sha1_file(buf.buf, buf.len, OBJ_BLOB, origin->blob_sha1); - commit->util = origin; /* * Read the current index, replace the path entry with @@@ -2582,14 -2343,7 +2594,14 @@@ parse_done blame_date_width = sizeof("2006-10-19"); break; case DATE_RELATIVE: - /* "normal" is used as the fallback for "relative" */ + /* TRANSLATORS: This string is used to tell us the maximum + display width for a relative timestamp in "git blame" + output. For C locale, "4 years, 11 months ago", which + takes 22 places, is the longest among various forms of + relative timestamps, but your language may need more or + fewer display columns. */ + blame_date_width = utf8_strwidth(_("4 years, 11 months ago")) + 1; /* add the null */ + break; case DATE_LOCAL: case DATE_NORMAL: blame_date_width = sizeof("Thu Oct 19 16:00:04 2006 -0700"); @@@ -2661,16 -2415,12 +2673,16 @@@ memset(&sb, 0, sizeof(sb)); sb.revs = &revs; - if (!reverse) + if (!reverse) { final_commit_name = prepare_final(&sb); + sb.commits.compare = compare_commits_by_commit_date; + } else if (contents_from) die("--contents and --children do not blend well."); - else + else { final_commit_name = prepare_initial(&sb); + sb.commits.compare = compare_commits_by_reverse_commit_date; + } if (!sb.final) { /* @@@ -2759,16 -2509,12 +2771,16 @@@ ent->next = next; origin_incref(o); } + + o->suspects = ent; + prio_queue_put(&sb.commits, o->commit); + origin_decref(o); range_set_release(&ranges); string_list_clear(&range_list, 0); - sb.ent = ent; + sb.ent = NULL; sb.path = path; read_mailmap(&mailmap, NULL); @@@ -2781,8 -2527,6 +2793,8 @@@ if (incremental) return 0; + sb.ent = blame_sort(sb.ent, compare_blame_final); + coalesce(&sb); if (!(output_option & OUTPUT_PORCELAIN)) diff --combined builtin/commit.c index 5e2221c8e8,8b85df475b..84cec9a715 --- a/builtin/commit.c +++ b/builtin/commit.c @@@ -526,29 -526,10 +526,29 @@@ static int sane_ident_split(struct iden return 1; } +static int parse_force_date(const char *in, char *out, int len) +{ + if (len < 1) + return -1; + *out++ = '@'; + len--; + + if (parse_date(in, out, len) < 0) { + int errors = 0; + unsigned long t = approxidate_careful(in, &errors); + if (errors) + return -1; + snprintf(out, len, "%lu", t); + } + + return 0; +} + static void determine_author_info(struct strbuf *author_ident) { char *name, *email, *date; struct ident_split author; + char date_buf[64]; name = getenv("GIT_AUTHOR_NAME"); email = getenv("GIT_AUTHOR_EMAIL"); @@@ -593,12 -574,8 +593,12 @@@ email = xstrndup(lb + 2, rb - (lb + 2)); } - if (force_date) - date = force_date; + if (force_date) { + if (parse_force_date(force_date, date_buf, sizeof(date_buf))) + die(_("invalid date format: %s"), force_date); + date = date_buf; + } + strbuf_addstr(author_ident, fmt_ident(name, email, date, IDENT_STRICT)); if (!split_ident_line(&author, author_ident->buf, author_ident->len) && sane_ident_split(&author)) { @@@ -608,46 -585,13 +608,46 @@@ } } -static char *cut_ident_timestamp_part(char *string) +static void split_ident_or_die(struct ident_split *id, const struct strbuf *buf) +{ + if (split_ident_line(id, buf->buf, buf->len) || + !sane_ident_split(id)) + die(_("Malformed ident string: '%s'"), buf->buf); +} + +static int author_date_is_interesting(void) { - char *ket = strrchr(string, '>'); - if (!ket || ket[1] != ' ') - die(_("Malformed ident string: '%s'"), string); - *++ket = '\0'; - return ket; + return author_message || force_date; +} + +static void adjust_comment_line_char(const struct strbuf *sb) +{ + char candidates[] = "#;@!$%^&|:"; + char *candidate; + const char *p; + + comment_line_char = candidates[0]; + if (!memchr(sb->buf, comment_line_char, sb->len)) + return; + + p = sb->buf; + candidate = strchr(candidates, *p); + if (candidate) + *candidate = ' '; + for (p = sb->buf; *p; p++) { + if ((p[0] == '\n' || p[0] == '\r') && p[1]) { + candidate = strchr(candidates, p[1]); + if (candidate) + *candidate = ' '; + } + } + + for (p = candidates; *p == ' '; p++) + ; + if (!*p) + die(_("unable to select a comment character that is not used\n" + "in the current commit message")); + comment_line_char = *p; } static int prepare_to_commit(const char *index_file, const char *prefix, @@@ -706,8 -650,9 +706,8 @@@ } else if (use_message) { char *buffer; buffer = strstr(use_message_buffer, "\n\n"); - if (!use_editor && (!buffer || buffer[2] == '\0')) - die(_("commit has empty message")); - strbuf_add(&sb, buffer + 2, strlen(buffer + 2)); + if (buffer) + strbuf_add(&sb, buffer + 2, strlen(buffer + 2)); hook_arg1 = "commit"; hook_arg2 = use_message; } else if (fixup_message) { @@@ -803,8 -748,6 +803,8 @@@ if (fwrite(sb.buf, 1, sb.len, s->fp) < sb.len) die_errno(_("could not write commit template")); + if (auto_comment_line_char) + adjust_comment_line_char(&sb); strbuf_release(&sb); /* This checks if committer ident is explicitly given */ @@@ -812,8 -755,7 +812,8 @@@ if (use_editor && include_status) { int ident_shown = 0; int saved_color_setting; - char *ai_tmp, *ci_tmp; + struct ident_split ci, ai; + if (whence != FROM_COMMIT) { if (cleanup_mode == CLEANUP_SCISSORS) wt_status_add_cut_line(s->fp); @@@ -853,39 -795,32 +853,39 @@@ status_printf_ln(s, GIT_COLOR_NORMAL, "%s", only_include_assumed); - ai_tmp = cut_ident_timestamp_part(author_ident->buf); - ci_tmp = cut_ident_timestamp_part(committer_ident.buf); - if (strcmp(author_ident->buf, committer_ident.buf)) + split_ident_or_die(&ai, author_ident); + split_ident_or_die(&ci, &committer_ident); + + if (ident_cmp(&ai, &ci)) + status_printf_ln(s, GIT_COLOR_NORMAL, + _("%s" + "Author: %.*s <%.*s>"), + ident_shown++ ? "" : "\n", + (int)(ai.name_end - ai.name_begin), ai.name_begin, + (int)(ai.mail_end - ai.mail_begin), ai.mail_begin); + + if (author_date_is_interesting()) status_printf_ln(s, GIT_COLOR_NORMAL, _("%s" - "Author: %s"), + "Date: %s"), ident_shown++ ? "" : "\n", - author_ident->buf); + show_ident_date(&ai, DATE_NORMAL)); if (!committer_ident_sufficiently_given()) status_printf_ln(s, GIT_COLOR_NORMAL, _("%s" - "Committer: %s"), + "Committer: %.*s <%.*s>"), ident_shown++ ? "" : "\n", - committer_ident.buf); + (int)(ci.name_end - ci.name_begin), ci.name_begin, + (int)(ci.mail_end - ci.mail_begin), ci.mail_begin); if (ident_shown) - status_printf_ln(s, GIT_COLOR_NORMAL, ""); + status_printf_ln(s, GIT_COLOR_NORMAL, "%s", ""); saved_color_setting = s->use_color; s->use_color = 0; commitable = run_status(s->fp, index_file, prefix, 1, s); s->use_color = saved_color_setting; - - *ai_tmp = ' '; - *ci_tmp = ' '; } else { unsigned char sha1[20]; const char *parent = "HEAD"; @@@ -898,22 -833,8 +898,22 @@@ if (get_sha1(parent, sha1)) commitable = !!active_nr; - else - commitable = index_differs_from(parent, 0); + else { + /* + * Unless the user did explicitly request a submodule + * ignore mode by passing a command line option we do + * not ignore any changed submodule SHA-1s when + * comparing index and parent, no matter what is + * configured. Otherwise we won't commit any + * submodules which were manually staged, which would + * be really confusing. + */ + int diff_flags = DIFF_OPT_OVERRIDE_SUBMODULE_CONFIG; + if (ignore_submodule_arg && + !strcmp(ignore_submodule_arg, "all")) + diff_flags |= DIFF_OPT_IGNORE_SUBMODULES; + commitable = index_differs_from(parent, diff_flags); + } } strbuf_release(&committer_ident); @@@ -1435,13 -1356,6 +1435,13 @@@ static void print_summary(const char *p strbuf_addstr(&format, "\n Author: "); strbuf_addbuf_percentquote(&format, &author_ident); } + if (author_date_is_interesting()) { + struct strbuf date = STRBUF_INIT; + format_commit_message(commit, "%ad", &date, &pctx); + strbuf_addstr(&format, "\n Date: "); + strbuf_addbuf_percentquote(&format, &date); + strbuf_release(&date); + } if (!committer_ident_sufficiently_given()) { strbuf_addstr(&format, "\n Committer: "); strbuf_addbuf_percentquote(&format, &committer_ident); @@@ -1745,8 -1659,8 +1745,8 @@@ int cmd_commit(int argc, const char **a append_merge_tag_headers(parents, &tail); } - if (commit_tree_extended(&sb, active_cache_tree->sha1, parents, sha1, - author_ident.buf, sign_commit, extra)) { + if (commit_tree_extended(sb.buf, sb.len, active_cache_tree->sha1, + parents, sha1, author_ident.buf, sign_commit, extra)) { rollback_index_files(); die(_("failed to write commit object")); } @@@ -1758,10 -1672,6 +1758,10 @@@ ? NULL : current_head->object.sha1, 0, NULL); + if (!ref_lock) { + rollback_index_files(); + die(_("cannot lock HEAD ref")); + } nl = strchr(sb.buf, '\n'); if (nl) @@@ -1771,6 -1681,10 +1771,6 @@@ strbuf_insert(&sb, 0, reflog_msg, strlen(reflog_msg)); strbuf_insert(&sb, strlen(reflog_msg), ": ", 2); - if (!ref_lock) { - rollback_index_files(); - die(_("cannot lock HEAD ref")); - } if (write_ref_sha1(ref_lock, sha1, sb.buf) < 0) { rollback_index_files(); die(_("cannot update HEAD ref")); diff --combined builtin/fast-export.c index ef4481615f,05d161f19f..92b4624a4b --- a/builtin/fast-export.c +++ b/builtin/fast-export.c @@@ -17,7 -17,6 +17,7 @@@ #include "utf8.h" #include "parse-options.h" #include "quote.h" +#include "remote.h" static const char *fast_export_usage[] = { N_("git fast-export [rev-list-opts]"), @@@ -32,8 -31,6 +32,8 @@@ static int use_done_feature static int no_data; static int full_tree; static struct string_list extra_refs = STRING_LIST_INIT_NODUP; +static struct refspec *refspecs; +static int refspecs_nr; static int parse_opt_signed_tag_mode(const struct option *opt, const char *arg, int unset) @@@ -282,6 -279,7 +282,7 @@@ static const char *find_encoding(const static void handle_commit(struct commit *commit, struct rev_info *rev) { int saved_output_format = rev->diffopt.output_format; + const char *commit_buffer; const char *author, *author_end, *committer, *committer_end; const char *encoding, *message; char *reencoded = NULL; @@@ -291,7 -289,8 +292,8 @@@ rev->diffopt.output_format = DIFF_FORMAT_CALLBACK; parse_commit_or_die(commit); - author = strstr(commit->buffer, "\nauthor "); + commit_buffer = get_commit_buffer(commit, NULL); + author = strstr(commit_buffer, "\nauthor "); if (!author) die ("Could not find author in commit %s", sha1_to_hex(commit->object.sha1)); @@@ -338,6 -337,7 +340,7 @@@ ? strlen(message) : 0), reencoded ? reencoded : message ? message : ""); free(reencoded); + unuse_commit_buffer(commit, commit_buffer); for (i = 0, p = commit->parents; p; p = p->next) { int mark = get_object_mark(&p->item->object); @@@ -528,15 -528,6 +531,15 @@@ static void get_tags_and_duplicates(str if (dwim_ref(e->name, strlen(e->name), sha1, &full_name) != 1) continue; + if (refspecs) { + char *private; + private = apply_refspecs(refspecs, refspecs_nr, full_name); + if (private) { + free(full_name); + full_name = private; + } + } + commit = get_commit(e, full_name); if (!commit) { warning("%s: Unexpected object of type %s, skipping.", @@@ -673,19 -664,6 +676,19 @@@ static void import_marks(char *input_fi fclose(f); } +static void handle_deletes(void) +{ + int i; + for (i = 0; i < refspecs_nr; i++) { + struct refspec *refspec = &refspecs[i]; + if (*refspec->src) + continue; + + printf("reset %s\nfrom %s\n\n", + refspec->dst, sha1_to_hex(null_sha1)); + } +} + int cmd_fast_export(int argc, const char **argv, const char *prefix) { struct rev_info revs; @@@ -693,7 -671,6 +696,7 @@@ struct commit *commit; char *export_filename = NULL, *import_filename = NULL; uint32_t lastimportid; + struct string_list refspecs_list = STRING_LIST_INIT_NODUP; struct option options[] = { OPT_INTEGER(0, "progress", &progress, N_("show progress after objects")), @@@ -714,8 -691,6 +717,8 @@@ OPT_BOOL(0, "use-done-feature", &use_done_feature, N_("Use the done feature to terminate the stream")), OPT_BOOL(0, "no-data", &no_data, N_("Skip output of blob data")), + OPT_STRING_LIST(0, "refspec", &refspecs_list, N_("refspec"), + N_("Apply refspec to exported refs")), OPT_END() }; @@@ -729,27 -704,11 +732,27 @@@ revs.topo_order = 1; revs.show_source = 1; revs.rewrite_parents = 1; + argc = parse_options(argc, argv, prefix, options, fast_export_usage, + PARSE_OPT_KEEP_ARGV0 | PARSE_OPT_KEEP_UNKNOWN); argc = setup_revisions(argc, argv, &revs, NULL); - argc = parse_options(argc, argv, prefix, options, fast_export_usage, 0); if (argc > 1) usage_with_options (fast_export_usage, options); + if (refspecs_list.nr) { + const char **refspecs_str; + int i; + + refspecs_str = xmalloc(sizeof(*refspecs_str) * refspecs_list.nr); + for (i = 0; i < refspecs_list.nr; i++) + refspecs_str[i] = refspecs_list.items[i].string; + + refspecs_nr = refspecs_list.nr; + refspecs = parse_fetch_refspec(refspecs_nr, refspecs_str); + + string_list_clear(&refspecs_list, 1); + free(refspecs_str); + } + if (use_done_feature) printf("feature done\n"); @@@ -777,7 -736,6 +780,7 @@@ } handle_tags_and_duplicates(); + handle_deletes(); if (export_filename && lastimportid != last_idnum) export_marks(export_filename); @@@ -785,7 -743,5 +788,7 @@@ if (use_done_feature) printf("done\n"); + free_refspec(refspecs_nr, refspecs); + return 0; } diff --combined builtin/index-pack.c index 18f57de58b,459b9f07bb..8b3bd29dbc --- a/builtin/index-pack.c +++ b/builtin/index-pack.c @@@ -40,13 -40,17 +40,13 @@@ struct base_data int ofs_first, ofs_last; }; -#if !defined(NO_PTHREADS) && defined(NO_THREAD_SAFE_PREAD) -/* pread() emulation is not thread-safe. Disable threading. */ -#define NO_PTHREADS -#endif - struct thread_local { #ifndef NO_PTHREADS pthread_t thread; #endif struct base_data *base_cache; size_t base_cache_used; + int pack_fd; }; /* @@@ -87,8 -91,7 +87,8 @@@ static off_t consumed_bytes static unsigned deepest_delta; static git_SHA_CTX input_ctx; static uint32_t input_crc32; -static int input_fd, output_fd, pack_fd; +static int input_fd, output_fd; +static const char *curr_pack; #ifndef NO_PTHREADS @@@ -131,7 -134,6 +131,7 @@@ static inline void unlock_mutex(pthread */ static void init_thread(void) { + int i; init_recursive_mutex(&read_mutex); pthread_mutex_init(&counter_mutex, NULL); pthread_mutex_init(&work_mutex, NULL); @@@ -139,18 -141,11 +139,18 @@@ pthread_mutex_init(&deepest_delta_mutex, NULL); pthread_key_create(&key, NULL); thread_data = xcalloc(nr_threads, sizeof(*thread_data)); + for (i = 0; i < nr_threads; i++) { + thread_data[i].pack_fd = open(curr_pack, O_RDONLY); + if (thread_data[i].pack_fd == -1) + die_errno(_("unable to open %s"), curr_pack); + } + threads_active = 1; } static void cleanup_thread(void) { + int i; if (!threads_active) return; threads_active = 0; @@@ -159,8 -154,6 +159,8 @@@ pthread_mutex_destroy(&work_mutex); if (show_stat) pthread_mutex_destroy(&deepest_delta_mutex); + for (i = 0; i < nr_threads; i++) + close(thread_data[i].pack_fd); pthread_key_delete(key); free(thread_data); } @@@ -207,13 -200,8 +207,13 @@@ static unsigned check_object(struct obj if (!(obj->flags & FLAG_CHECKED)) { unsigned long size; int type = sha1_object_info(obj->sha1, &size); - if (type != obj->type || type <= 0) - die(_("object of unexpected type")); + if (type <= 0) + die(_("did not receive expected object %s"), + sha1_to_hex(obj->sha1)); + if (type != obj->type) + die(_("object %s: expected type %s, found %s"), + sha1_to_hex(obj->sha1), + typename(obj->type), typename(type)); obj->flags |= FLAG_CHECKED; return 1; } @@@ -300,13 -288,13 +300,13 @@@ static const char *open_pack_file(cons output_fd = open(pack_name, O_CREAT|O_EXCL|O_RDWR, 0600); if (output_fd < 0) die_errno(_("unable to create '%s'"), pack_name); - pack_fd = output_fd; + nothread_data.pack_fd = output_fd; } else { input_fd = open(pack_name, O_RDONLY); if (input_fd < 0) die_errno(_("cannot open packfile '%s'"), pack_name); output_fd = -1; - pack_fd = input_fd; + nothread_data.pack_fd = input_fd; } git_SHA1_Init(&input_ctx); return pack_name; @@@ -554,7 -542,7 +554,7 @@@ static void *unpack_data(struct object_ do { ssize_t n = (len < 64*1024) ? len : 64*1024; - n = pread(pack_fd, inbuf, n, from); + n = xpread(get_thread_data()->pack_fd, inbuf, n, from); if (n < 0) die_errno(_("cannot pread pack file")); if (!n) @@@ -786,7 -774,8 +786,8 @@@ static void sha1_object(const void *dat } if (obj->type == OBJ_COMMIT) { struct commit *commit = (struct commit *) obj; - commit->buffer = NULL; + if (detach_commit_buffer(commit, NULL) != data) + die("BUG: parse_object_buffer transmogrified our buffer"); } obj->flags |= FLAG_CHECKED; } @@@ -1502,7 -1491,7 +1503,7 @@@ static void show_pack_info(int stat_onl int cmd_index_pack(int argc, const char **argv, const char *prefix) { int i, fix_thin_pack = 0, verify = 0, stat_only = 0; - const char *curr_pack, *curr_index; + const char *curr_index; const char *index_name = NULL, *pack_name = NULL; const char *keep_name = NULL, *keep_msg = NULL; char *index_name_buf = NULL, *keep_name_buf = NULL; diff --combined builtin/log.c index a7ba211731,c599eacf72..92063dfc48 --- a/builtin/log.c +++ b/builtin/log.c @@@ -63,8 -63,6 +63,8 @@@ static int parse_decoration_style(cons return DECORATE_FULL_REFS; else if (!strcmp(value, "short")) return DECORATE_SHORT_REFS; + else if (!strcmp(value, "auto")) + return (isatty(1) || pager_in_use()) ? DECORATE_SHORT_REFS : 0; return -1; } @@@ -160,9 -158,13 +160,9 @@@ static void cmd_log_init_finish(int arg if (rev->show_notes) init_display_notes(&rev->notes_opt); - if (rev->diffopt.pickaxe || rev->diffopt.filter) + if (rev->diffopt.pickaxe || rev->diffopt.filter || + DIFF_OPT_TST(&rev->diffopt, FOLLOW_RENAMES)) rev->always_show_header = 0; - if (DIFF_OPT_TST(&rev->diffopt, FOLLOW_RENAMES)) { - rev->always_show_header = 0; - if (rev->diffopt.pathspec.nr != 1) - usage("git logs can only follow renames on one pathname at a time"); - } if (source) rev->show_source = 1; @@@ -347,8 -349,7 +347,7 @@@ static int cmd_log_walk(struct rev_inf rev->max_count++; if (!rev->reflog_info) { /* we allow cycles in reflog ancestry */ - free(commit->buffer); - commit->buffer = NULL; + free_commit_buffer(commit); } free_commit_list(commit->parents); commit->parents = NULL; @@@ -671,7 -672,6 +670,7 @@@ static void add_header(const char *valu static int thread; static int do_signoff; static const char *signature = git_version_string; +static const char *signature_file; static int config_cover_letter; enum { @@@ -741,8 -741,6 +740,8 @@@ static int git_format_config(const cha } if (!strcmp(var, "format.signature")) return git_config_string(&signature, var, value); + if (!strcmp(var, "format.signaturefile")) + return git_config_pathname(&signature_file, var, value); if (!strcmp(var, "format.coverletter")) { if (value && !strcasecmp(value, "auto")) { config_cover_letter = COVER_AUTO; @@@ -845,13 -843,8 +844,13 @@@ static void gen_message_id(struct rev_i static void print_signature(void) { - if (signature && *signature) - printf("-- \n%s\n\n", signature); + if (!signature || !*signature) + return; + + printf("-- \n%s", signature); + if (signature[strlen(signature)-1] != '\n') + putchar('\n'); + putchar('\n'); } static void add_branch_description(struct strbuf *buf, const char *branch_name) @@@ -925,9 -918,12 +924,12 @@@ static void make_cover_letter(struct re log_write_email_headers(rev, head, &pp.subject, &pp.after_subject, &need_8bit_cte); - for (i = 0; !need_8bit_cte && i < nr; i++) - if (has_non_ascii(list[i]->buffer)) + for (i = 0; !need_8bit_cte && i < nr; i++) { + const char *buf = get_commit_buffer(list[i], NULL); + if (has_non_ascii(buf)) need_8bit_cte = 1; + unuse_commit_buffer(list[i], buf); + } if (!branch_name) branch_name = find_branch_name(rev); @@@ -1236,8 -1232,6 +1238,8 @@@ int cmd_format_patch(int argc, const ch PARSE_OPT_OPTARG, thread_callback }, OPT_STRING(0, "signature", &signature, N_("signature"), N_("add a signature")), + OPT_FILENAME(0, "signature-file", &signature_file, + N_("add a signature from a file")), OPT__QUIET(&quiet, N_("don't print the patch filenames")), OPT_END() }; @@@ -1455,18 -1449,6 +1457,18 @@@ cover_letter = (config_cover_letter == COVER_ON); } + if (!signature) { + ; /* --no-signature inhibits all signatures */ + } else if (signature && signature != git_version_string) { + ; /* non-default signature already set */ + } else if (signature_file) { + struct strbuf buf = STRBUF_INIT; + + if (strbuf_read_file(&buf, signature_file, 128) < 0) + die_errno(_("unable to read signature file '%s'"), signature_file); + signature = strbuf_detach(&buf, NULL); + } + if (in_reply_to || thread || cover_letter) rev.ref_message_ids = xcalloc(1, sizeof(struct string_list)); if (in_reply_to) { @@@ -1528,8 -1510,7 +1530,7 @@@ reopen_stdout(rev.numbered_files ? NULL : commit, NULL, &rev, quiet)) die(_("Failed to create output files")); shown = log_tree_commit(&rev, commit); - free(commit->buffer); - commit->buffer = NULL; + free_commit_buffer(commit); /* We put one extra blank line between formatted * patches and this flag is used by log-tree code diff --combined builtin/merge.c index 428ca247bd,8763b2efa2..b49c310866 --- a/builtin/merge.c +++ b/builtin/merge.c @@@ -63,7 -63,7 +63,7 @@@ static int verbosity static int allow_rerere_auto; static int abort_current_merge; static int show_progress = -1; -static int default_to_upstream; +static int default_to_upstream = 1; static const char *sign_commit; static struct strategy all_strategy[] = { @@@ -398,7 -398,7 +398,7 @@@ static void finish(struct commit *head_ const char *argv_gc_auto[] = { "gc", "--auto", NULL }; update_ref(reflog_message.buf, "HEAD", new_head, head, 0, - DIE_ON_ERR); + UPDATE_REFS_DIE_ON_ERR); /* * We ignore errors in 'gc --auto', since the * user should see them. @@@ -852,8 -852,8 +852,8 @@@ static int merge_trivial(struct commit parent->next->item = remoteheads->item; parent->next->next = NULL; prepare_to_commit(remoteheads); - if (commit_tree(&merge_msg, result_tree, parent, result_commit, NULL, - sign_commit)) + if (commit_tree(merge_msg.buf, merge_msg.len, result_tree, parent, + result_commit, NULL, sign_commit)) die(_("failed to write commit object")); finish(head, remoteheads, result_commit, "In-index merge"); drop_save(); @@@ -877,8 -877,8 +877,8 @@@ static int finish_automerge(struct comm commit_list_insert(head, &parents); strbuf_addch(&merge_msg, '\n'); prepare_to_commit(remoteheads); - if (commit_tree(&merge_msg, result_tree, parents, result_commit, - NULL, sign_commit)) + if (commit_tree(merge_msg.buf, merge_msg.len, result_tree, parents, + result_commit, NULL, sign_commit)) die(_("failed to write commit object")); strbuf_addf(&buf, "Merge made by the '%s' strategy.", wt_strategy); finish(head, remoteheads, result_commit, buf.buf); @@@ -1222,7 -1222,7 +1222,7 @@@ int cmd_merge(int argc, const char **ar die(_("%s - not something we can merge"), argv[0]); read_empty(remote_head->object.sha1, 0); update_ref("initial pull", "HEAD", remote_head->object.sha1, - NULL, 0, DIE_ON_ERR); + NULL, 0, UPDATE_REFS_DIE_ON_ERR); goto done; } else { struct strbuf merge_names = STRBUF_INIT; @@@ -1339,7 -1339,7 +1339,7 @@@ } update_ref("updating ORIG_HEAD", "ORIG_HEAD", head_commit->object.sha1, - NULL, 0, DIE_ON_ERR); + NULL, 0, UPDATE_REFS_DIE_ON_ERR); if (remoteheads && !common) ; /* No common ancestors found. We need a real merge. */ diff --combined builtin/reset.c index f368266762,6bd6245821..850d53229a --- a/builtin/reset.c +++ b/builtin/reset.c @@@ -93,7 -93,7 +93,7 @@@ static int reset_index(const unsigned c static void print_new_head_line(struct commit *commit) { const char *hex, *body; - char *msg; + const char *msg; hex = find_unique_abbrev(commit->object.sha1, DEFAULT_ABBREV); printf(_("HEAD is now at %s"), hex); @@@ -109,7 -109,7 +109,7 @@@ } else printf("\n"); - logmsg_free(msg, commit); + unuse_commit_buffer(commit, msg); } static void update_index_from_diff(struct diff_queue_struct *q, @@@ -252,13 -252,11 +252,13 @@@ static int reset_refs(const char *rev, if (!get_sha1("HEAD", sha1_orig)) { orig = sha1_orig; set_reflog_message(&msg, "updating ORIG_HEAD", NULL); - update_ref(msg.buf, "ORIG_HEAD", orig, old_orig, 0, MSG_ON_ERR); + update_ref(msg.buf, "ORIG_HEAD", orig, old_orig, 0, + UPDATE_REFS_MSG_ON_ERR); } else if (old_orig) delete_ref("ORIG_HEAD", old_orig, 0); set_reflog_message(&msg, "updating HEAD", rev); - update_ref_status = update_ref(msg.buf, "HEAD", sha1, orig, 0, MSG_ON_ERR); + update_ref_status = update_ref(msg.buf, "HEAD", sha1, orig, 0, + UPDATE_REFS_MSG_ON_ERR); strbuf_release(&msg); return update_ref_status; } diff --combined commit.c index 881be3baa3,ebd7ad8465..4ff8077dbf --- a/commit.c +++ b/commit.c @@@ -17,7 -17,6 +17,6 @@@ static struct commit_extra_header *read int save_commit_buffer = 1; const char *commit_type = "commit"; - static int commit_count; static struct commit *check_commit(struct object *obj, const unsigned char *sha1, @@@ -64,7 -63,6 +63,6 @@@ struct commit *lookup_commit(const unsi struct object *obj = lookup_object(sha1); if (!obj) { struct commit *c = alloc_commit_node(); - c->index = commit_count++; return create_object(sha1, OBJ_COMMIT, c); } if (!obj->type) @@@ -247,6 -245,76 +245,76 @@@ int unregister_shallow(const unsigned c return 0; } + struct commit_buffer { + void *buffer; + unsigned long size; + }; + define_commit_slab(buffer_slab, struct commit_buffer); + static struct buffer_slab buffer_slab = COMMIT_SLAB_INIT(1, buffer_slab); + + void set_commit_buffer(struct commit *commit, void *buffer, unsigned long size) + { + struct commit_buffer *v = buffer_slab_at(&buffer_slab, commit); + v->buffer = buffer; + v->size = size; + } + + const void *get_cached_commit_buffer(const struct commit *commit, unsigned long *sizep) + { + struct commit_buffer *v = buffer_slab_at(&buffer_slab, commit); + if (sizep) + *sizep = v->size; + return v->buffer; + } + + const void *get_commit_buffer(const struct commit *commit, unsigned long *sizep) + { + const void *ret = get_cached_commit_buffer(commit, sizep); + if (!ret) { + enum object_type type; + unsigned long size; + ret = read_sha1_file(commit->object.sha1, &type, &size); + if (!ret) + die("cannot read commit object %s", + sha1_to_hex(commit->object.sha1)); + if (type != OBJ_COMMIT) + die("expected commit for %s, got %s", + sha1_to_hex(commit->object.sha1), typename(type)); + if (sizep) + *sizep = size; + } + return ret; + } + + void unuse_commit_buffer(const struct commit *commit, const void *buffer) + { + struct commit_buffer *v = buffer_slab_at(&buffer_slab, commit); + if (v->buffer != buffer) + free((void *)buffer); + } + + void free_commit_buffer(struct commit *commit) + { + struct commit_buffer *v = buffer_slab_at(&buffer_slab, commit); + free(v->buffer); + v->buffer = NULL; + v->size = 0; + } + + const void *detach_commit_buffer(struct commit *commit, unsigned long *sizep) + { + struct commit_buffer *v = buffer_slab_at(&buffer_slab, commit); + void *ret; + + ret = v->buffer; + if (sizep) + *sizep = v->size; + + v->buffer = NULL; + v->size = 0; + return ret; + } + int parse_commit_buffer(struct commit *item, const void *buffer, unsigned long size) { const char *tail = buffer; @@@ -324,7 -392,7 +392,7 @@@ int parse_commit(struct commit *item } ret = parse_commit_buffer(item, buffer, size); if (save_commit_buffer && !ret) { - item->buffer = buffer; + set_commit_buffer(item, buffer, size); return 0; } free(buffer); @@@ -539,22 -607,12 +607,12 @@@ static void record_author_date(struct a struct commit *commit) { const char *buf, *line_end, *ident_line; - char *buffer = NULL; + const char *buffer = get_commit_buffer(commit, NULL); struct ident_split ident; char *date_end; unsigned long date; - if (!commit->buffer) { - unsigned long size; - enum object_type type; - buffer = read_sha1_file(commit->object.sha1, &type, &size); - if (!buffer) - return; - } - - for (buf = commit->buffer ? commit->buffer : buffer; - buf; - buf = line_end + 1) { + for (buf = buffer; buf; buf = line_end + 1) { line_end = strchrnul(buf, '\n'); ident_line = skip_prefix(buf, "author "); if (!ident_line) { @@@ -575,7 -633,7 +633,7 @@@ *(author_date_slab_at(author_date, commit)) = date; fail_exit: - free(buffer); + unuse_commit_buffer(commit, buffer); } static int compare_commits_by_author_date(const void *a_, const void *b_, @@@ -1031,7 -1089,7 +1089,7 @@@ struct commit_list *reduce_heads(struc p->item->object.flags |= STALE; num_head++; } - array = xcalloc(sizeof(*array), num_head); + array = xcalloc(num_head, sizeof(*array)); for (p = heads, i = 0; p; p = p->next) { if (p->item->object.flags & STALE) { array[i++] = p->item; @@@ -1080,17 -1138,14 +1138,14 @@@ static int do_sign_commit(struct strbu return 0; } - int parse_signed_commit(const unsigned char *sha1, + int parse_signed_commit(const struct commit *commit, struct strbuf *payload, struct strbuf *signature) { + unsigned long size; - enum object_type type; - char *buffer = read_sha1_file(sha1, &type, &size); + const char *buffer = get_commit_buffer(commit, &size); int in_signature, saw_signature = -1; - char *line, *tail; - - if (!buffer || type != OBJ_COMMIT) - goto cleanup; + const char *line, *tail; line = buffer; tail = buffer + size; @@@ -1098,7 -1153,7 +1153,7 @@@ saw_signature = 0; while (line < tail) { const char *sig = NULL; - char *next = memchr(line, '\n', tail - line); + const char *next = memchr(line, '\n', tail - line); next = next ? next + 1 : tail; if (in_signature && line[0] == ' ') @@@ -1119,8 -1174,7 +1174,7 @@@ } line = next; } - cleanup: - free(buffer); + unuse_commit_buffer(commit, buffer); return saw_signature; } @@@ -1211,8 -1265,7 +1265,7 @@@ void check_commit_signature(const struc sigc->result = 'N'; - if (parse_signed_commit(commit->object.sha1, - &payload, &signature) <= 0) + if (parse_signed_commit(commit, &payload, &signature) <= 0) goto out; status = verify_signed_buffer(payload.buf, payload.len, signature.buf, signature.len, @@@ -1257,11 -1310,9 +1310,9 @@@ struct commit_extra_header *read_commit { struct commit_extra_header *extra = NULL; unsigned long size; - enum object_type type; - char *buffer = read_sha1_file(commit->object.sha1, &type, &size); - if (buffer && type == OBJ_COMMIT) - extra = read_commit_extra_header_lines(buffer, size, exclude); - free(buffer); + const char *buffer = get_commit_buffer(commit, &size); + extra = read_commit_extra_header_lines(buffer, size, exclude); + unuse_commit_buffer(commit, buffer); return extra; } @@@ -1344,7 -1395,8 +1395,8 @@@ void free_commit_extra_headers(struct c } } - int commit_tree(const struct strbuf *msg, const unsigned char *tree, + int commit_tree(const char *msg, size_t msg_len, + const unsigned char *tree, struct commit_list *parents, unsigned char *ret, const char *author, const char *sign_commit) { @@@ -1352,7 -1404,7 +1404,7 @@@ int result; append_merge_tag_headers(parents, &tail); - result = commit_tree_extended(msg, tree, parents, ret, + result = commit_tree_extended(msg, msg_len, tree, parents, ret, author, sign_commit, extra); free_commit_extra_headers(extra); return result; @@@ -1473,7 -1525,8 +1525,8 @@@ static const char commit_utf8_warn[] "You may want to amend it after fixing the message, or set the config\n" "variable i18n.commitencoding to the encoding your project uses.\n"; - int commit_tree_extended(const struct strbuf *msg, const unsigned char *tree, + int commit_tree_extended(const char *msg, size_t msg_len, + const unsigned char *tree, struct commit_list *parents, unsigned char *ret, const char *author, const char *sign_commit, struct commit_extra_header *extra) @@@ -1484,7 -1537,7 +1537,7 @@@ assert_sha1_type(tree, OBJ_TREE); - if (memchr(msg->buf, '\0', msg->len)) + if (memchr(msg, '\0', msg_len)) return error("a NUL byte in commit log message not allowed."); /* Not having i18n.commitencoding is the same as having utf-8 */ @@@ -1523,7 -1576,7 +1576,7 @@@ strbuf_addch(&buffer, '\n'); /* And add the comment */ - strbuf_addbuf(&buffer, msg); + strbuf_add(&buffer, msg, msg_len); /* And check the encoding */ if (encoding_is_utf8 && !verify_utf8(&buffer)) diff --combined merge-recursive.c index cab16fafb5,a9ab328923..d38a3b2eb5 --- a/merge-recursive.c +++ b/merge-recursive.c @@@ -40,7 -40,7 +40,7 @@@ static struct tree *shift_tree_object(s static struct commit *make_virtual_commit(struct tree *tree, const char *comment) { - struct commit *commit = xcalloc(1, sizeof(struct commit)); + struct commit *commit = alloc_commit_node(); struct merge_remote_desc *desc = xmalloc(sizeof(*desc)); desc->name = comment; @@@ -190,9 -190,11 +190,11 @@@ static void output_commit_title(struct printf(_("(bad commit)\n")); else { const char *title; - int len = find_commit_subject(commit->buffer, &title); + const char *msg = get_commit_buffer(commit, NULL); + int len = find_commit_subject(msg, &title); if (len) printf("%.*s\n", len, title); + unuse_commit_buffer(commit, msg); } } } @@@ -589,12 -591,6 +591,12 @@@ static int remove_file(struct merge_opt return -1; } if (update_working_directory) { + if (ignore_case) { + struct cache_entry *ce; + ce = cache_file_exists(path, strlen(path), ignore_case); + if (ce && ce_stage(ce) == 0) + return 0; + } if (remove_path(path)) return -1; } diff --combined notes-cache.c index 97dfd63c9b,25b20aa21f..c4e9bb7f6c --- a/notes-cache.c +++ b/notes-cache.c @@@ -48,7 -48,6 +48,6 @@@ int notes_cache_write(struct notes_cach { unsigned char tree_sha1[20]; unsigned char commit_sha1[20]; - struct strbuf msg = STRBUF_INIT; if (!c || !c->tree.initialized || !c->tree.ref || !*c->tree.ref) return -1; @@@ -57,12 -56,11 +56,11 @@@ if (write_notes_tree(&c->tree, tree_sha1)) return -1; - strbuf_attach(&msg, c->validity, - strlen(c->validity), strlen(c->validity) + 1); - if (commit_tree(&msg, tree_sha1, NULL, commit_sha1, NULL, NULL) < 0) + if (commit_tree(c->validity, strlen(c->validity), tree_sha1, NULL, + commit_sha1, NULL, NULL) < 0) return -1; if (update_ref("update notes cache", c->tree.ref, commit_sha1, NULL, - 0, QUIET_ON_ERR) < 0) + 0, UPDATE_REFS_QUIET_ON_ERR) < 0) return -1; return 0; diff --combined notes-utils.c index a0b1d7be98,f6891f7255..b64dc1b021 --- a/notes-utils.c +++ b/notes-utils.c @@@ -4,7 -4,8 +4,8 @@@ #include "notes-utils.h" void create_notes_commit(struct notes_tree *t, struct commit_list *parents, - const struct strbuf *msg, unsigned char *result_sha1) + const char *msg, size_t msg_len, + unsigned char *result_sha1) { unsigned char tree_sha1[20]; @@@ -25,7 -26,7 +26,7 @@@ /* else: t->ref points to nothing, assume root/orphan commit */ } - if (commit_tree(msg, tree_sha1, parents, result_sha1, NULL, NULL)) + if (commit_tree(msg, msg_len, tree_sha1, parents, result_sha1, NULL, NULL)) die("Failed to commit notes tree to database"); } @@@ -46,10 -47,9 +47,10 @@@ void commit_notes(struct notes_tree *t if (buf.buf[buf.len - 1] != '\n') strbuf_addch(&buf, '\n'); /* Make sure msg ends with newline */ - create_notes_commit(t, NULL, &buf, commit_sha1); + create_notes_commit(t, NULL, buf.buf, buf.len, commit_sha1); strbuf_insert(&buf, 0, "notes: ", 7); /* commit message starts at index 7 */ - update_ref(buf.buf, t->ref, commit_sha1, NULL, 0, DIE_ON_ERR); + update_ref(buf.buf, t->ref, commit_sha1, NULL, 0, + UPDATE_REFS_DIE_ON_ERR); strbuf_release(&buf); } diff --combined pretty.c index 924bc61004,b9fceedbb9..c93c14b080 --- a/pretty.c +++ b/pretty.c @@@ -274,7 -274,7 +274,7 @@@ static void add_rfc822_quoted(struct st enum rfc2047_type { RFC2047_SUBJECT, - RFC2047_ADDRESS, + RFC2047_ADDRESS }; static int is_rfc2047_special(char ch, enum rfc2047_type type) @@@ -393,8 -393,8 +393,8 @@@ static void add_rfc2047(struct strbuf * strbuf_addstr(sb, "?="); } -static const char *show_ident_date(const struct ident_split *ident, - enum date_mode mode) +const char *show_ident_date(const struct ident_split *ident, + enum date_mode mode) { unsigned long date = 0; long tz = 0; @@@ -606,29 -606,16 +606,16 @@@ static char *replace_encoding_header(ch return strbuf_detach(&tmp, NULL); } - char *logmsg_reencode(const struct commit *commit, - char **commit_encoding, - const char *output_encoding) + const char *logmsg_reencode(const struct commit *commit, + char **commit_encoding, + const char *output_encoding) { static const char *utf8 = "UTF-8"; const char *use_encoding; char *encoding; - char *msg = commit->buffer; + const char *msg = get_commit_buffer(commit, NULL); char *out; - if (!msg) { - enum object_type type; - unsigned long size; - - msg = read_sha1_file(commit->object.sha1, &type, &size); - if (!msg) - die("Cannot read commit object %s", - sha1_to_hex(commit->object.sha1)); - if (type != OBJ_COMMIT) - die("Expected commit for '%s', got %s", - sha1_to_hex(commit->object.sha1), typename(type)); - } - if (!output_encoding || !*output_encoding) { if (commit_encoding) *commit_encoding = @@@ -652,12 -639,13 +639,13 @@@ * Otherwise, we still want to munge the encoding header in the * result, which will be done by modifying the buffer. If we * are using a fresh copy, we can reuse it. But if we are using - * the cached copy from commit->buffer, we need to duplicate it - * to avoid munging commit->buffer. + * the cached copy from get_commit_buffer, we need to duplicate it + * to avoid munging the cached copy. */ - out = msg; - if (out == commit->buffer) - out = xstrdup(out); + if (msg == get_cached_commit_buffer(commit, NULL)) + out = xstrdup(msg); + else + out = (char *)msg; } else { /* @@@ -667,8 -655,8 +655,8 @@@ * copy, we can free it. */ out = reencode_string(msg, output_encoding, use_encoding); - if (out && msg != commit->buffer) - free(msg); + if (out) + unuse_commit_buffer(commit, msg); } /* @@@ -687,12 -675,6 +675,6 @@@ return out ? out : msg; } - void logmsg_free(char *msg, const struct commit *commit) - { - if (msg != commit->buffer) - free(msg); - } - static int mailmap_name(const char **email, size_t *email_len, const char **name, size_t *name_len) { @@@ -796,7 -778,7 +778,7 @@@ struct format_commit_context struct signature_check signature_check; enum flush_type flush_type; enum trunc_type truncate; - char *message; + const char *message; char *commit_encoding; size_t width, indent1, indent2; int auto_color; @@@ -1506,18 -1488,13 +1488,18 @@@ void format_commit_message(const struc context.commit = commit; context.pretty_ctx = pretty_ctx; context.wrap_start = sb->len; + /* + * convert a commit message to UTF-8 first + * as far as 'format_commit_item' assumes it in UTF-8 + */ context.message = logmsg_reencode(commit, &context.commit_encoding, - output_enc); + utf8); strbuf_expand(sb, format, format_commit_item, &context); rewrap_message_tail(sb, &context, 0, 0, 0); + /* then convert a commit message to an actual output encoding */ if (output_enc) { if (same_encoding(utf8, output_enc)) output_enc = NULL; @@@ -1536,7 -1513,7 +1518,7 @@@ } free(context.commit_encoding); - logmsg_free(context.message, commit); + unuse_commit_buffer(commit, context.message); free(context.signature_check.gpg_output); free(context.signature_check.signer); } @@@ -1705,7 -1682,7 +1687,7 @@@ void pretty_print_commit(struct pretty_ unsigned long beginning_of_body; int indent = 4; const char *msg; - char *reencoded; + const char *reencoded; const char *encoding; int need_8bit_cte = pp->need_8bit_cte; @@@ -1772,7 -1749,7 +1754,7 @@@ if (pp->fmt == CMIT_FMT_EMAIL && sb->len <= beginning_of_body) strbuf_addch(sb, '\n'); - logmsg_free(reencoded, commit); + unuse_commit_buffer(commit, reencoded); } void pp_commit_easy(enum cmit_fmt fmt, const struct commit *commit, diff --combined revision.c index 8351e794df,1cc91e5911..2571ada6bf --- a/revision.c +++ b/revision.c @@@ -1633,7 -1633,6 +1633,7 @@@ static int handle_revision_opt(struct r !strcmp(arg, "--reflog") || !strcmp(arg, "--not") || !strcmp(arg, "--no-walk") || !strcmp(arg, "--do-walk") || !strcmp(arg, "--bisect") || starts_with(arg, "--glob=") || + starts_with(arg, "--exclude=") || starts_with(arg, "--branches=") || starts_with(arg, "--tags=") || starts_with(arg, "--remotes=") || starts_with(arg, "--no-walk=")) { @@@ -1649,10 -1648,8 +1649,10 @@@ revs->skip_count = atoi(optarg); return argcount; } else if ((*arg == '-') && isdigit(arg[1])) { - /* accept -, like traditional "head" */ - revs->max_count = atoi(arg + 1); + /* accept -, like traditional "head" */ + if (strtol_i(arg + 1, 10, &revs->max_count) < 0 || + revs->max_count < 0) + die("'%s': not a non-negative integer", arg + 1); revs->no_walk = 0; } else if (!strcmp(arg, "-n")) { if (argc <= 1) @@@ -2791,7 -2788,7 +2791,7 @@@ static int commit_match(struct commit * { int retval; const char *encoding; - char *message; + const char *message; struct strbuf buf = STRBUF_INIT; if (!opt->grep_filter.pattern_list && !opt->grep_filter.header_list) @@@ -2833,14 -2830,21 +2833,21 @@@ format_display_notes(commit->object.sha1, &buf, encoding, 1); } - /* Find either in the original commit message, or in the temporary */ + /* + * Find either in the original commit message, or in the temporary. + * Note that we cast away the constness of "message" here. It is + * const because it may come from the cached commit buffer. That's OK, + * because we know that it is modifiable heap memory, and that while + * grep_buffer may modify it for speed, it will restore any + * changes before returning. + */ if (buf.len) retval = grep_buffer(&opt->grep_filter, buf.buf, buf.len); else retval = grep_buffer(&opt->grep_filter, - message, strlen(message)); + (char *)message, strlen(message)); strbuf_release(&buf); - logmsg_free(message, commit); + unuse_commit_buffer(commit, message); return retval; } diff --combined sequencer.c index 923047459b,bbaddcb05a..c513d7eeb4 --- a/sequencer.c +++ b/sequencer.c @@@ -116,39 -116,23 +116,23 @@@ static const char *action_name(const st return opts->action == REPLAY_REVERT ? "revert" : "cherry-pick"; } - static char *get_encoding(const char *message); - struct commit_message { char *parent_label; const char *label; const char *subject; - char *reencoded_message; const char *message; }; static int get_message(struct commit *commit, struct commit_message *out) { - const char *encoding; const char *abbrev, *subject; int abbrev_len, subject_len; char *q; - if (!commit->buffer) - return -1; - encoding = get_encoding(commit->buffer); - if (!encoding) - encoding = "UTF-8"; if (!git_commit_encoding) git_commit_encoding = "UTF-8"; - out->reencoded_message = NULL; - out->message = commit->buffer; - if (same_encoding(encoding, git_commit_encoding)) - out->reencoded_message = reencode_string(commit->buffer, - git_commit_encoding, encoding); - if (out->reencoded_message) - out->message = out->reencoded_message; - + out->message = logmsg_reencode(commit, NULL, git_commit_encoding); abbrev = find_unique_abbrev(commit->object.sha1, DEFAULT_ABBREV); abbrev_len = strlen(abbrev); @@@ -167,29 -151,10 +151,10 @@@ return 0; } - static void free_message(struct commit_message *msg) + static void free_message(struct commit *commit, struct commit_message *msg) { free(msg->parent_label); - free(msg->reencoded_message); - } - - static char *get_encoding(const char *message) - { - const char *p = message, *eol; - - while (*p && *p != '\n') { - for (eol = p + 1; *eol && *eol != '\n'; eol++) - ; /* do nothing */ - if (starts_with(p, "encoding ")) { - char *result = xmalloc(eol - 8 - p); - strlcpy(result, p + 9, eol - 8 - p); - return result; - } - p = eol; - if (*p == '\n') - p++; - } - return NULL; + unuse_commit_buffer(commit, msg->message); } static void write_cherry_pick_head(struct commit *commit, const char *pseudoref) @@@ -278,15 -243,11 +243,15 @@@ static int fast_forward_to(const unsign read_cache(); if (checkout_fast_forward(from, to, 1)) - exit(1); /* the callee should have complained already */ + exit(128); /* the callee should have complained already */ ref_lock = lock_any_ref_for_update("HEAD", unborn ? null_sha1 : from, 0, NULL); + if (!ref_lock) + return error(_("Failed to lock HEAD during fast_forward_to")); + strbuf_addf(&sb, "%s: fast-forward", action_name(opts)); ret = write_ref_sha1(ref_lock, to, sb.buf); + strbuf_release(&sb); return ret; } @@@ -489,7 -450,7 +454,7 @@@ static int do_pick_commit(struct commi unsigned char head[20]; struct commit *base, *next, *parent; const char *base_label, *next_label; - struct commit_message msg = { NULL, NULL, NULL, NULL, NULL }; + struct commit_message msg = { NULL, NULL, NULL, NULL }; char *defmsg = NULL; struct strbuf msgbuf = STRBUF_INIT; int res, unborn = 0, allow; @@@ -654,7 -615,7 +619,7 @@@ res = run_git_commit(defmsg, opts, allow); leave: - free_message(&msg); + free_message(commit, &msg); free(defmsg); return res; @@@ -701,10 -662,12 +666,12 @@@ static int format_todo(struct strbuf *b int subject_len; for (cur = todo_list; cur; cur = cur->next) { + const char *commit_buffer = get_commit_buffer(cur->item, NULL); sha1_abbrev = find_unique_abbrev(cur->item->object.sha1, DEFAULT_ABBREV); - subject_len = find_commit_subject(cur->item->buffer, &subject); + subject_len = find_commit_subject(commit_buffer, &subject); strbuf_addf(buf, "%s %s %.*s\n", action_str, sha1_abbrev, subject_len, subject); + unuse_commit_buffer(cur->item, commit_buffer); } return 0; }