diff options
Diffstat (limited to 'match-trees.c')
-rw-r--r-- | match-trees.c | 229 |
1 files changed, 135 insertions, 94 deletions
diff --git a/match-trees.c b/match-trees.c index 0fd6df7d6e..396b7338df 100644 --- a/match-trees.c +++ b/match-trees.c @@ -47,78 +47,69 @@ static int score_matches(unsigned mode1, unsigned mode2, const char *path) return score; } +static void *fill_tree_desc_strict(struct tree_desc *desc, + const struct object_id *hash) +{ + void *buffer; + enum object_type type; + unsigned long size; + + buffer = read_sha1_file(hash->hash, &type, &size); + if (!buffer) + die("unable to read tree (%s)", oid_to_hex(hash)); + if (type != OBJ_TREE) + die("%s is not a tree", oid_to_hex(hash)); + init_tree_desc(desc, buffer, size); + return buffer; +} + +static int base_name_entries_compare(const struct name_entry *a, + const struct name_entry *b) +{ + return base_name_compare(a->path, tree_entry_len(a), a->mode, + b->path, tree_entry_len(b), b->mode); +} + /* * Inspect two trees, and give a score that tells how similar they are. */ -static int score_trees(const unsigned char *hash1, const unsigned char *hash2) +static int score_trees(const struct object_id *hash1, const struct object_id *hash2) { struct tree_desc one; struct tree_desc two; - void *one_buf, *two_buf; + void *one_buf = fill_tree_desc_strict(&one, hash1); + void *two_buf = fill_tree_desc_strict(&two, hash2); int score = 0; - enum object_type type; - unsigned long size; - one_buf = read_sha1_file(hash1, &type, &size); - if (!one_buf) - die("unable to read tree (%s)", sha1_to_hex(hash1)); - if (type != OBJ_TREE) - die("%s is not a tree", sha1_to_hex(hash1)); - init_tree_desc(&one, one_buf, size); - two_buf = read_sha1_file(hash2, &type, &size); - if (!two_buf) - die("unable to read tree (%s)", sha1_to_hex(hash2)); - if (type != OBJ_TREE) - die("%s is not a tree", sha1_to_hex(hash2)); - init_tree_desc(&two, two_buf, size); - while (one.size | two.size) { - const unsigned char *elem1 = elem1; - const unsigned char *elem2 = elem2; - const char *path1 = path1; - const char *path2 = path2; - unsigned mode1 = mode1; - unsigned mode2 = mode2; + for (;;) { + struct name_entry e1, e2; + int got_entry_from_one = tree_entry(&one, &e1); + int got_entry_from_two = tree_entry(&two, &e2); int cmp; - if (one.size) - elem1 = tree_entry_extract(&one, &path1, &mode1); - if (two.size) - elem2 = tree_entry_extract(&two, &path2, &mode2); - - if (!one.size) { - /* two has more entries */ - score += score_missing(mode2, path2); - update_tree_entry(&two); - continue; - } - if (!two.size) { + if (got_entry_from_one && got_entry_from_two) + cmp = base_name_entries_compare(&e1, &e2); + else if (got_entry_from_one) /* two lacks this entry */ - score += score_missing(mode1, path1); - update_tree_entry(&one); - continue; - } - cmp = base_name_compare(path1, strlen(path1), mode1, - path2, strlen(path2), mode2); - if (cmp < 0) { + cmp = -1; + else if (got_entry_from_two) + /* two has more entries */ + cmp = 1; + else + break; + + if (cmp < 0) /* path1 does not appear in two */ - score += score_missing(mode1, path1); - update_tree_entry(&one); - continue; - } - else if (cmp > 0) { + score += score_missing(e1.mode, e1.path); + else if (cmp > 0) /* path2 does not appear in one */ - score += score_missing(mode2, path2); - update_tree_entry(&two); - continue; - } - else if (hashcmp(elem1, elem2)) + score += score_missing(e2.mode, e2.path); + else if (oidcmp(e1.oid, e2.oid)) /* they are different */ - score += score_differs(mode1, mode2, path1); + score += score_differs(e1.mode, e2.mode, e1.path); else /* same subtree or blob */ - score += score_matches(mode1, mode2, path1); - update_tree_entry(&one); - update_tree_entry(&two); + score += score_matches(e1.mode, e2.mode, e1.path); } free(one_buf); free(two_buf); @@ -128,28 +119,19 @@ static int score_trees(const unsigned char *hash1, const unsigned char *hash2) /* * Match one itself and its subtrees with two and pick the best match. */ -static void match_trees(const unsigned char *hash1, - const unsigned char *hash2, +static void match_trees(const struct object_id *hash1, + const struct object_id *hash2, int *best_score, char **best_match, const char *base, int recurse_limit) { struct tree_desc one; - void *one_buf; - enum object_type type; - unsigned long size; - - one_buf = read_sha1_file(hash1, &type, &size); - if (!one_buf) - die("unable to read tree (%s)", sha1_to_hex(hash1)); - if (type != OBJ_TREE) - die("%s is not a tree", sha1_to_hex(hash1)); - init_tree_desc(&one, one_buf, size); + void *one_buf = fill_tree_desc_strict(&one, hash1); while (one.size) { const char *path; - const unsigned char *elem; + const struct object_id *elem; unsigned mode; int score; @@ -158,17 +140,12 @@ static void match_trees(const unsigned char *hash1, goto next; score = score_trees(elem, hash2); if (*best_score < score) { - char *newpath; - newpath = xmalloc(strlen(base) + strlen(path) + 1); - sprintf(newpath, "%s%s", base, path); free(*best_match); - *best_match = newpath; + *best_match = xstrfmt("%s%s", base, path); *best_score = score; } if (recurse_limit) { - char *newbase; - newbase = xmalloc(strlen(base) + strlen(path) + 2); - sprintf(newbase, "%s%s/", base, path); + char *newbase = xstrfmt("%s%s/", base, path); match_trees(elem, hash2, best_score, best_match, newbase, recurse_limit - 1); free(newbase); @@ -185,7 +162,7 @@ static void match_trees(const unsigned char *hash1, * tree object by replacing it with another tree "hash2". */ static int splice_tree(const unsigned char *hash1, - char *prefix, + const char *prefix, const unsigned char *hash2, unsigned char *result) { @@ -200,13 +177,10 @@ static int splice_tree(const unsigned char *hash1, enum object_type type; int status; - subpath = strchr(prefix, '/'); - if (!subpath) - toplen = strlen(prefix); - else { - toplen = subpath - prefix; + subpath = strchrnul(prefix, '/'); + toplen = subpath - prefix; + if (*subpath) subpath++; - } buf = read_sha1_file(hash1, &type, &sz); if (!buf) @@ -217,15 +191,15 @@ static int splice_tree(const unsigned char *hash1, while (desc.size) { const char *name; unsigned mode; - const unsigned char *sha1; + const struct object_id *oid; - sha1 = tree_entry_extract(&desc, &name, &mode); + oid = tree_entry_extract(&desc, &name, &mode); if (strlen(name) == toplen && !memcmp(name, prefix, toplen)) { if (!S_ISDIR(mode)) die("entry %s in tree %s is not a tree", name, sha1_to_hex(hash1)); - rewrite_here = (unsigned char *) sha1; + rewrite_here = (unsigned char *) oid->hash; break; } update_tree_entry(&desc); @@ -233,7 +207,7 @@ static int splice_tree(const unsigned char *hash1, if (!rewrite_here) die("entry %.*s not found in tree %s", toplen, prefix, sha1_to_hex(hash1)); - if (subpath) { + if (*subpath) { status = splice_tree(rewrite_here, subpath, hash2, subtree); if (status) return status; @@ -255,15 +229,22 @@ static int splice_tree(const unsigned char *hash1, * other hand, it could cover tree one and we might need to pick a * subtree of it. */ -void shift_tree(const unsigned char *hash1, - const unsigned char *hash2, - unsigned char *shifted, +void shift_tree(const struct object_id *hash1, + const struct object_id *hash2, + struct object_id *shifted, int depth_limit) { char *add_prefix; char *del_prefix; int add_score, del_score; + /* + * NEEDSWORK: this limits the recursion depth to hardcoded + * value '2' to avoid excessive overhead. + */ + if (!depth_limit) + depth_limit = 2; + add_score = del_score = score_trees(hash1, hash2); add_prefix = xcalloc(1, 1); del_prefix = xcalloc(1, 1); @@ -281,7 +262,7 @@ void shift_tree(const unsigned char *hash1, match_trees(hash2, hash1, &del_score, &del_prefix, "", depth_limit); /* Assume we do not have to do any shifting */ - hashcpy(shifted, hash2); + oidcpy(shifted, hash2); if (add_score < del_score) { /* We need to pick a subtree of two */ @@ -290,14 +271,74 @@ void shift_tree(const unsigned char *hash1, if (!*del_prefix) return; - if (get_tree_entry(hash2, del_prefix, shifted, &mode)) + if (get_tree_entry(hash2->hash, del_prefix, shifted->hash, &mode)) die("cannot find path %s in tree %s", - del_prefix, sha1_to_hex(hash2)); + del_prefix, oid_to_hex(hash2)); return; } if (!*add_prefix) return; - splice_tree(hash1, add_prefix, hash2, shifted); + splice_tree(hash1->hash, add_prefix, hash2->hash, shifted->hash); +} + +/* + * The user says the trees will be shifted by this much. + * Unfortunately we cannot fundamentally tell which one to + * be prefixed, as recursive merge can work in either direction. + */ +void shift_tree_by(const struct object_id *hash1, + const struct object_id *hash2, + struct object_id *shifted, + const char *shift_prefix) +{ + struct object_id sub1, sub2; + unsigned mode1, mode2; + unsigned candidate = 0; + + /* Can hash2 be a tree at shift_prefix in tree hash1? */ + if (!get_tree_entry(hash1->hash, shift_prefix, sub1.hash, &mode1) && + S_ISDIR(mode1)) + candidate |= 1; + + /* Can hash1 be a tree at shift_prefix in tree hash2? */ + if (!get_tree_entry(hash2->hash, shift_prefix, sub2.hash, &mode2) && + S_ISDIR(mode2)) + candidate |= 2; + + if (candidate == 3) { + /* Both are plausible -- we need to evaluate the score */ + int best_score = score_trees(hash1, hash2); + int score; + + candidate = 0; + score = score_trees(&sub1, hash2); + if (score > best_score) { + candidate = 1; + best_score = score; + } + score = score_trees(&sub2, hash1); + if (score > best_score) + candidate = 2; + } + + if (!candidate) { + /* Neither is plausible -- do not shift */ + oidcpy(shifted, hash2); + return; + } + + if (candidate == 1) + /* + * shift tree2 down by adding shift_prefix above it + * to match tree1. + */ + splice_tree(hash1->hash, shift_prefix, hash2->hash, shifted->hash); + else + /* + * shift tree2 up by removing shift_prefix from it + * to match tree1. + */ + oidcpy(shifted, &sub2); } |