]>
Commit | Line | Data |
---|---|---|
1 | #include "cache.h" | |
2 | #include "tree.h" | |
3 | #include "tree-walk.h" | |
4 | #include "object-store.h" | |
5 | ||
6 | static int score_missing(unsigned mode) | |
7 | { | |
8 | int score; | |
9 | ||
10 | if (S_ISDIR(mode)) | |
11 | score = -1000; | |
12 | else if (S_ISLNK(mode)) | |
13 | score = -500; | |
14 | else | |
15 | score = -50; | |
16 | return score; | |
17 | } | |
18 | ||
19 | static int score_differs(unsigned mode1, unsigned mode2) | |
20 | { | |
21 | int score; | |
22 | ||
23 | if (S_ISDIR(mode1) != S_ISDIR(mode2)) | |
24 | score = -100; | |
25 | else if (S_ISLNK(mode1) != S_ISLNK(mode2)) | |
26 | score = -50; | |
27 | else | |
28 | score = -5; | |
29 | return score; | |
30 | } | |
31 | ||
32 | static int score_matches(unsigned mode1, unsigned mode2) | |
33 | { | |
34 | int score; | |
35 | ||
36 | /* Heh, we found SHA-1 collisions between different kind of objects */ | |
37 | if (S_ISDIR(mode1) != S_ISDIR(mode2)) | |
38 | score = -100; | |
39 | else if (S_ISLNK(mode1) != S_ISLNK(mode2)) | |
40 | score = -50; | |
41 | ||
42 | else if (S_ISDIR(mode1)) | |
43 | score = 1000; | |
44 | else if (S_ISLNK(mode1)) | |
45 | score = 500; | |
46 | else | |
47 | score = 250; | |
48 | return score; | |
49 | } | |
50 | ||
51 | static void *fill_tree_desc_strict(struct tree_desc *desc, | |
52 | const struct object_id *hash) | |
53 | { | |
54 | void *buffer; | |
55 | enum object_type type; | |
56 | unsigned long size; | |
57 | ||
58 | buffer = read_object_file(hash, &type, &size); | |
59 | if (!buffer) | |
60 | die("unable to read tree (%s)", oid_to_hex(hash)); | |
61 | if (type != OBJ_TREE) | |
62 | die("%s is not a tree", oid_to_hex(hash)); | |
63 | init_tree_desc(desc, buffer, size); | |
64 | return buffer; | |
65 | } | |
66 | ||
67 | static int base_name_entries_compare(const struct name_entry *a, | |
68 | const struct name_entry *b) | |
69 | { | |
70 | return base_name_compare(a->path, tree_entry_len(a), a->mode, | |
71 | b->path, tree_entry_len(b), b->mode); | |
72 | } | |
73 | ||
74 | /* | |
75 | * Inspect two trees, and give a score that tells how similar they are. | |
76 | */ | |
77 | static int score_trees(const struct object_id *hash1, const struct object_id *hash2) | |
78 | { | |
79 | struct tree_desc one; | |
80 | struct tree_desc two; | |
81 | void *one_buf = fill_tree_desc_strict(&one, hash1); | |
82 | void *two_buf = fill_tree_desc_strict(&two, hash2); | |
83 | int score = 0; | |
84 | ||
85 | for (;;) { | |
86 | int cmp; | |
87 | ||
88 | if (one.size && two.size) | |
89 | cmp = base_name_entries_compare(&one.entry, &two.entry); | |
90 | else if (one.size) | |
91 | /* two lacks this entry */ | |
92 | cmp = -1; | |
93 | else if (two.size) | |
94 | /* two has more entries */ | |
95 | cmp = 1; | |
96 | else | |
97 | break; | |
98 | ||
99 | if (cmp < 0) { | |
100 | /* path1 does not appear in two */ | |
101 | score += score_missing(one.entry.mode); | |
102 | update_tree_entry(&one); | |
103 | } else if (cmp > 0) { | |
104 | /* path2 does not appear in one */ | |
105 | score += score_missing(two.entry.mode); | |
106 | update_tree_entry(&two); | |
107 | } else { | |
108 | /* path appears in both */ | |
109 | if (!oideq(&one.entry.oid, &two.entry.oid)) { | |
110 | /* they are different */ | |
111 | score += score_differs(one.entry.mode, | |
112 | two.entry.mode); | |
113 | } else { | |
114 | /* same subtree or blob */ | |
115 | score += score_matches(one.entry.mode, | |
116 | two.entry.mode); | |
117 | } | |
118 | update_tree_entry(&one); | |
119 | update_tree_entry(&two); | |
120 | } | |
121 | } | |
122 | free(one_buf); | |
123 | free(two_buf); | |
124 | return score; | |
125 | } | |
126 | ||
127 | /* | |
128 | * Match one itself and its subtrees with two and pick the best match. | |
129 | */ | |
130 | static void match_trees(const struct object_id *hash1, | |
131 | const struct object_id *hash2, | |
132 | int *best_score, | |
133 | char **best_match, | |
134 | const char *base, | |
135 | int recurse_limit) | |
136 | { | |
137 | struct tree_desc one; | |
138 | void *one_buf = fill_tree_desc_strict(&one, hash1); | |
139 | ||
140 | while (one.size) { | |
141 | const char *path; | |
142 | const struct object_id *elem; | |
143 | unsigned short mode; | |
144 | int score; | |
145 | ||
146 | elem = tree_entry_extract(&one, &path, &mode); | |
147 | if (!S_ISDIR(mode)) | |
148 | goto next; | |
149 | score = score_trees(elem, hash2); | |
150 | if (*best_score < score) { | |
151 | free(*best_match); | |
152 | *best_match = xstrfmt("%s%s", base, path); | |
153 | *best_score = score; | |
154 | } | |
155 | if (recurse_limit) { | |
156 | char *newbase = xstrfmt("%s%s/", base, path); | |
157 | match_trees(elem, hash2, best_score, best_match, | |
158 | newbase, recurse_limit - 1); | |
159 | free(newbase); | |
160 | } | |
161 | ||
162 | next: | |
163 | update_tree_entry(&one); | |
164 | } | |
165 | free(one_buf); | |
166 | } | |
167 | ||
168 | /* | |
169 | * A tree "oid1" has a subdirectory at "prefix". Come up with a tree object by | |
170 | * replacing it with another tree "oid2". | |
171 | */ | |
172 | static int splice_tree(const struct object_id *oid1, const char *prefix, | |
173 | const struct object_id *oid2, struct object_id *result) | |
174 | { | |
175 | char *subpath; | |
176 | int toplen; | |
177 | char *buf; | |
178 | unsigned long sz; | |
179 | struct tree_desc desc; | |
180 | unsigned char *rewrite_here; | |
181 | const struct object_id *rewrite_with; | |
182 | struct object_id subtree; | |
183 | enum object_type type; | |
184 | int status; | |
185 | ||
186 | subpath = strchrnul(prefix, '/'); | |
187 | toplen = subpath - prefix; | |
188 | if (*subpath) | |
189 | subpath++; | |
190 | ||
191 | buf = read_object_file(oid1, &type, &sz); | |
192 | if (!buf) | |
193 | die("cannot read tree %s", oid_to_hex(oid1)); | |
194 | init_tree_desc(&desc, buf, sz); | |
195 | ||
196 | rewrite_here = NULL; | |
197 | while (desc.size) { | |
198 | const char *name; | |
199 | unsigned short mode; | |
200 | ||
201 | tree_entry_extract(&desc, &name, &mode); | |
202 | if (strlen(name) == toplen && | |
203 | !memcmp(name, prefix, toplen)) { | |
204 | if (!S_ISDIR(mode)) | |
205 | die("entry %s in tree %s is not a tree", name, | |
206 | oid_to_hex(oid1)); | |
207 | ||
208 | /* | |
209 | * We cast here for two reasons: | |
210 | * | |
211 | * - to flip the "char *" (for the path) to "unsigned | |
212 | * char *" (for the hash stored after it) | |
213 | * | |
214 | * - to discard the "const"; this is OK because we | |
215 | * know it points into our non-const "buf" | |
216 | */ | |
217 | rewrite_here = (unsigned char *)(desc.entry.path + | |
218 | strlen(desc.entry.path) + | |
219 | 1); | |
220 | break; | |
221 | } | |
222 | update_tree_entry(&desc); | |
223 | } | |
224 | if (!rewrite_here) | |
225 | die("entry %.*s not found in tree %s", toplen, prefix, | |
226 | oid_to_hex(oid1)); | |
227 | if (*subpath) { | |
228 | struct object_id tree_oid; | |
229 | hashcpy(tree_oid.hash, rewrite_here); | |
230 | status = splice_tree(&tree_oid, subpath, oid2, &subtree); | |
231 | if (status) | |
232 | return status; | |
233 | rewrite_with = &subtree; | |
234 | } else { | |
235 | rewrite_with = oid2; | |
236 | } | |
237 | hashcpy(rewrite_here, rewrite_with->hash); | |
238 | status = write_object_file(buf, sz, tree_type, result); | |
239 | free(buf); | |
240 | return status; | |
241 | } | |
242 | ||
243 | /* | |
244 | * We are trying to come up with a merge between one and two that | |
245 | * results in a tree shape similar to one. The tree two might | |
246 | * correspond to a subtree of one, in which case it needs to be | |
247 | * shifted down by prefixing otherwise empty directories. On the | |
248 | * other hand, it could cover tree one and we might need to pick a | |
249 | * subtree of it. | |
250 | */ | |
251 | void shift_tree(struct repository *r, | |
252 | const struct object_id *hash1, | |
253 | const struct object_id *hash2, | |
254 | struct object_id *shifted, | |
255 | int depth_limit) | |
256 | { | |
257 | char *add_prefix; | |
258 | char *del_prefix; | |
259 | int add_score, del_score; | |
260 | ||
261 | /* | |
262 | * NEEDSWORK: this limits the recursion depth to hardcoded | |
263 | * value '2' to avoid excessive overhead. | |
264 | */ | |
265 | if (!depth_limit) | |
266 | depth_limit = 2; | |
267 | ||
268 | add_score = del_score = score_trees(hash1, hash2); | |
269 | add_prefix = xcalloc(1, 1); | |
270 | del_prefix = xcalloc(1, 1); | |
271 | ||
272 | /* | |
273 | * See if one's subtree resembles two; if so we need to prefix | |
274 | * two with a few fake trees to match the prefix. | |
275 | */ | |
276 | match_trees(hash1, hash2, &add_score, &add_prefix, "", depth_limit); | |
277 | ||
278 | /* | |
279 | * See if two's subtree resembles one; if so we need to | |
280 | * pick only subtree of two. | |
281 | */ | |
282 | match_trees(hash2, hash1, &del_score, &del_prefix, "", depth_limit); | |
283 | ||
284 | /* Assume we do not have to do any shifting */ | |
285 | oidcpy(shifted, hash2); | |
286 | ||
287 | if (add_score < del_score) { | |
288 | /* We need to pick a subtree of two */ | |
289 | unsigned short mode; | |
290 | ||
291 | if (!*del_prefix) | |
292 | return; | |
293 | ||
294 | if (get_tree_entry(r, hash2, del_prefix, shifted, &mode)) | |
295 | die("cannot find path %s in tree %s", | |
296 | del_prefix, oid_to_hex(hash2)); | |
297 | return; | |
298 | } | |
299 | ||
300 | if (!*add_prefix) | |
301 | return; | |
302 | ||
303 | splice_tree(hash1, add_prefix, hash2, shifted); | |
304 | } | |
305 | ||
306 | /* | |
307 | * The user says the trees will be shifted by this much. | |
308 | * Unfortunately we cannot fundamentally tell which one to | |
309 | * be prefixed, as recursive merge can work in either direction. | |
310 | */ | |
311 | void shift_tree_by(struct repository *r, | |
312 | const struct object_id *hash1, | |
313 | const struct object_id *hash2, | |
314 | struct object_id *shifted, | |
315 | const char *shift_prefix) | |
316 | { | |
317 | struct object_id sub1, sub2; | |
318 | unsigned short mode1, mode2; | |
319 | unsigned candidate = 0; | |
320 | ||
321 | /* Can hash2 be a tree at shift_prefix in tree hash1? */ | |
322 | if (!get_tree_entry(r, hash1, shift_prefix, &sub1, &mode1) && | |
323 | S_ISDIR(mode1)) | |
324 | candidate |= 1; | |
325 | ||
326 | /* Can hash1 be a tree at shift_prefix in tree hash2? */ | |
327 | if (!get_tree_entry(r, hash2, shift_prefix, &sub2, &mode2) && | |
328 | S_ISDIR(mode2)) | |
329 | candidate |= 2; | |
330 | ||
331 | if (candidate == 3) { | |
332 | /* Both are plausible -- we need to evaluate the score */ | |
333 | int best_score = score_trees(hash1, hash2); | |
334 | int score; | |
335 | ||
336 | candidate = 0; | |
337 | score = score_trees(&sub1, hash2); | |
338 | if (score > best_score) { | |
339 | candidate = 1; | |
340 | best_score = score; | |
341 | } | |
342 | score = score_trees(&sub2, hash1); | |
343 | if (score > best_score) | |
344 | candidate = 2; | |
345 | } | |
346 | ||
347 | if (!candidate) { | |
348 | /* Neither is plausible -- do not shift */ | |
349 | oidcpy(shifted, hash2); | |
350 | return; | |
351 | } | |
352 | ||
353 | if (candidate == 1) | |
354 | /* | |
355 | * shift tree2 down by adding shift_prefix above it | |
356 | * to match tree1. | |
357 | */ | |
358 | splice_tree(hash1, shift_prefix, hash2, shifted); | |
359 | else | |
360 | /* | |
361 | * shift tree2 up by removing shift_prefix from it | |
362 | * to match tree1. | |
363 | */ | |
364 | oidcpy(shifted, &sub2); | |
365 | } |