forked from Softcatala/git-po
-
Notifications
You must be signed in to change notification settings - Fork 0
/
blame.c
1864 lines (1672 loc) · 49.6 KB
/
blame.c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
#include "cache.h"
#include "refs.h"
#include "cache-tree.h"
#include "mergesort.h"
#include "diff.h"
#include "diffcore.h"
#include "tag.h"
#include "blame.h"
void blame_origin_decref(struct blame_origin *o)
{
if (o && --o->refcnt <= 0) {
struct blame_origin *p, *l = NULL;
if (o->previous)
blame_origin_decref(o->previous);
free(o->file.ptr);
/* 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");
}
}
/*
* Given a commit and a path in it, create a new origin structure.
* The callers that add blame to the scoreboard should use
* get_origin() to obtain shared, refcounted copy instead of calling
* this function directly.
*/
static struct blame_origin *make_origin(struct commit *commit, const char *path)
{
struct blame_origin *o;
FLEX_ALLOC_STR(o, path, path);
o->commit = commit;
o->refcnt = 1;
o->next = commit->util;
commit->util = o;
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 blame_origin *get_origin(struct commit *commit, const char *path)
{
struct blame_origin *o, *l;
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 blame_origin_incref(o);
}
}
return make_origin(commit, path);
}
static void verify_working_tree_path(struct commit *work_tree, const char *path)
{
struct commit_list *parents;
int pos;
for (parents = work_tree->parents; parents; parents = parents->next) {
const struct object_id *commit_oid = &parents->item->object.oid;
struct object_id blob_oid;
unsigned mode;
if (!get_tree_entry(commit_oid->hash, path, blob_oid.hash, &mode) &&
sha1_object_info(blob_oid.hash, NULL) == OBJ_BLOB)
return;
}
pos = cache_name_pos(path, strlen(path));
if (pos >= 0)
; /* path is in the index */
else if (-1 - pos < active_nr &&
!strcmp(active_cache[-1 - pos]->name, path))
; /* path is in the index, unmerged */
else
die("no such path '%s' in HEAD", path);
}
static struct commit_list **append_parent(struct commit_list **tail, const struct object_id *oid)
{
struct commit *parent;
parent = lookup_commit_reference(oid);
if (!parent)
die("no such commit %s", oid_to_hex(oid));
return &commit_list_insert(parent, tail)->next;
}
static void append_merge_parents(struct commit_list **tail)
{
int merge_head;
struct strbuf line = STRBUF_INIT;
merge_head = open(git_path_merge_head(), O_RDONLY);
if (merge_head < 0) {
if (errno == ENOENT)
return;
die("cannot open '%s' for reading", git_path_merge_head());
}
while (!strbuf_getwholeline_fd(&line, merge_head, '\n')) {
struct object_id oid;
if (line.len < GIT_SHA1_HEXSZ || get_oid_hex(line.buf, &oid))
die("unknown line in '%s': %s", git_path_merge_head(), line.buf);
tail = append_parent(tail, &oid);
}
close(merge_head);
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.
*/
static struct commit *fake_working_tree_commit(struct diff_options *opt,
const char *path,
const char *contents_from)
{
struct commit *commit;
struct blame_origin *origin;
struct commit_list **parent_tail, *parent;
struct object_id head_oid;
struct strbuf buf = STRBUF_INIT;
const char *ident;
time_t now;
int size, len;
struct cache_entry *ce;
unsigned mode;
struct strbuf msg = STRBUF_INIT;
read_cache();
time(&now);
commit = alloc_commit_node();
commit->object.parsed = 1;
commit->date = now;
parent_tail = &commit->parents;
if (!resolve_ref_unsafe("HEAD", RESOLVE_REF_READING, head_oid.hash, NULL))
die("no such ref: HEAD");
parent_tail = append_parent(parent_tail, &head_oid);
append_merge_parents(parent_tail);
verify_working_tree_path(commit, path);
origin = make_origin(commit, path);
ident = fmt_ident("Not Committed Yet", "not.committed.yet", NULL, 0);
strbuf_addstr(&msg, "tree 0000000000000000000000000000000000000000\n");
for (parent = commit->parents; parent; parent = parent->next)
strbuf_addf(&msg, "parent %s\n",
oid_to_hex(&parent->item->object.oid));
strbuf_addf(&msg,
"author %s\n"
"committer %s\n\n"
"Version of %s from %s\n",
ident, ident, path,
(!contents_from ? path :
(!strcmp(contents_from, "-") ? "standard input" : contents_from)));
set_commit_buffer_from_strbuf(commit, &msg);
if (!contents_from || strcmp("-", contents_from)) {
struct stat st;
const char *read_from;
char *buf_ptr;
unsigned long buf_len;
if (contents_from) {
if (stat(contents_from, &st) < 0)
die_errno("Cannot stat '%s'", contents_from);
read_from = contents_from;
}
else {
if (lstat(path, &st) < 0)
die_errno("Cannot lstat '%s'", path);
read_from = path;
}
mode = canon_mode(st.st_mode);
switch (st.st_mode & S_IFMT) {
case S_IFREG:
if (DIFF_OPT_TST(opt, ALLOW_TEXTCONV) &&
textconv_object(read_from, mode, &null_oid, 0, &buf_ptr, &buf_len))
strbuf_attach(&buf, buf_ptr, buf_len, buf_len + 1);
else if (strbuf_read_file(&buf, read_from, st.st_size) != st.st_size)
die_errno("cannot open or read '%s'", read_from);
break;
case S_IFLNK:
if (strbuf_readlink(&buf, read_from, st.st_size) < 0)
die_errno("cannot readlink '%s'", read_from);
break;
default:
die("unsupported file type %s", read_from);
}
}
else {
/* Reading from stdin */
mode = 0;
if (strbuf_read(&buf, 0, 0) < 0)
die_errno("failed to read from stdin");
}
convert_to_git(&the_index, 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_oid.hash);
/*
* Read the current index, replace the path entry with
* origin->blob_sha1 without mucking with its mode or type
* bits; we are not going to write this index out -- we just
* want to run "diff-index --cached".
*/
discard_cache();
read_cache();
len = strlen(path);
if (!mode) {
int pos = cache_name_pos(path, len);
if (0 <= pos)
mode = active_cache[pos]->ce_mode;
else
/* Let's not bother reading from HEAD tree */
mode = S_IFREG | 0644;
}
size = cache_entry_size(len);
ce = xcalloc(1, size);
oidcpy(&ce->oid, &origin->blob_oid);
memcpy(ce->name, path, len);
ce->ce_flags = create_ce_flags(0);
ce->ce_namelen = len;
ce->ce_mode = create_ce_mode(mode);
add_cache_entry(ce, ADD_CACHE_OK_TO_ADD|ADD_CACHE_OK_TO_REPLACE);
cache_tree_invalidate_path(&the_index, path);
return commit;
}
static int diff_hunks(mmfile_t *file_a, mmfile_t *file_b,
xdl_emit_hunk_consume_func_t hunk_func, void *cb_data, int xdl_opts)
{
xpparam_t xpp = {0};
xdemitconf_t xecfg = {0};
xdemitcb_t ecb = {NULL};
xpp.flags = xdl_opts;
xecfg.hunk_func = hunk_func;
ecb.priv = cb_data;
return xdi_diff(file_a, file_b, &xpp, &xecfg, &ecb);
}
/*
* Given an origin, prepare mmfile_t structure to be used by the
* diff machinery
*/
static void fill_origin_blob(struct diff_options *opt,
struct blame_origin *o, mmfile_t *file, int *num_read_blob)
{
if (!o->file.ptr) {
enum object_type type;
unsigned long file_size;
(*num_read_blob)++;
if (DIFF_OPT_TST(opt, ALLOW_TEXTCONV) &&
textconv_object(o->path, o->mode, &o->blob_oid, 1, &file->ptr, &file_size))
;
else
file->ptr = read_sha1_file(o->blob_oid.hash, &type,
&file_size);
file->size = file_size;
if (!file->ptr)
die("Cannot read blob %s for path %s",
oid_to_hex(&o->blob_oid),
o->path);
o->file = *file;
}
else
*file = o->file;
}
static void drop_origin_blob(struct blame_origin *o)
{
if (o->file.ptr) {
FREE_AND_NULL(o->file.ptr);
}
}
/*
* 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;
}
void blame_sort_final(struct blame_scoreboard *sb)
{
sb->ent = llist_mergesort(sb->ent, get_next_blame, set_next_blame,
compare_blame_final);
}
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);
}
/*
* For debugging -- origin is refcounted, and this asserts that
* we do not underflow.
*/
static void sanity_check_refcnt(struct blame_scoreboard *sb)
{
int baa = 0;
struct blame_entry *ent;
for (ent = sb->ent; ent; ent = ent->next) {
/* Nobody should have zero or negative refcnt */
if (ent->suspect->refcnt <= 0) {
fprintf(stderr, "%s in %s has negative refcnt %d\n",
ent->suspect->path,
oid_to_hex(&ent->suspect->commit->object.oid),
ent->suspect->refcnt);
baa = 1;
}
}
if (baa)
sb->on_sanity_fail(sb, baa);
}
/*
* If two blame entries that are next to each other came from
* contiguous lines in the same origin (i.e. <commit, path> pair),
* merge them together.
*/
void blame_coalesce(struct blame_scoreboard *sb)
{
struct blame_entry *ent, *next;
for (ent = sb->ent; ent && (next = ent->next); ent = next) {
if (ent->suspect == next->suspect &&
ent->s_lno + ent->num_lines == next->s_lno) {
ent->num_lines += next->num_lines;
ent->next = next->next;
blame_origin_decref(next->suspect);
free(next);
ent->score = 0;
next = ent; /* again */
}
}
if (sb->debug) /* sanity */
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 blame_scoreboard *sb, struct blame_origin *porigin,
struct blame_entry *sorted)
{
if (porigin->suspects)
porigin->suspects = blame_merge(porigin->suspects, sorted);
else {
struct blame_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);
}
}
/*
* Fill the blob_sha1 field of an origin if it hasn't, so that later
* call to fill_origin_blob() can use it to locate the data. blob_sha1
* for an origin is also used to pass the blame for the entire file to
* the parent to detect the case where a child's blob is identical to
* that of its parent's.
*
* This also fills origin->mode for corresponding tree path.
*/
static int fill_blob_sha1_and_mode(struct blame_origin *origin)
{
if (!is_null_oid(&origin->blob_oid))
return 0;
if (get_tree_entry(origin->commit->object.oid.hash,
origin->path,
origin->blob_oid.hash, &origin->mode))
goto error_out;
if (sha1_object_info(origin->blob_oid.hash, NULL) != OBJ_BLOB)
goto error_out;
return 0;
error_out:
oidclr(&origin->blob_oid);
origin->mode = S_IFINVALID;
return -1;
}
/*
* We have an origin -- check if the same path exists in the
* parent and return an origin structure to represent it.
*/
static struct blame_origin *find_origin(struct commit *parent,
struct blame_origin *origin)
{
struct blame_origin *porigin;
struct diff_options diff_opts;
const char *paths[2];
/* 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.
*/
return blame_origin_incref (porigin);
}
/* See if the origin->path is different between parent
* and origin first. Most of the time they are the
* same and diff-tree is fairly efficient about this.
*/
diff_setup(&diff_opts);
DIFF_OPT_SET(&diff_opts, RECURSIVE);
diff_opts.detect_rename = 0;
diff_opts.output_format = DIFF_FORMAT_NO_OUTPUT;
paths[0] = origin->path;
paths[1] = NULL;
parse_pathspec(&diff_opts.pathspec,
PATHSPEC_ALL_MAGIC & ~PATHSPEC_LITERAL,
PATHSPEC_LITERAL_PATH, "", paths);
diff_setup_done(&diff_opts);
if (is_null_oid(&origin->commit->object.oid))
do_diff_cache(&parent->tree->object.oid, &diff_opts);
else
diff_tree_oid(&parent->tree->object.oid,
&origin->commit->tree->object.oid,
"", &diff_opts);
diffcore_std(&diff_opts);
if (!diff_queued_diff.nr) {
/* The path is the same as parent */
porigin = get_origin(parent, origin->path);
oidcpy(&porigin->blob_oid, &origin->blob_oid);
porigin->mode = origin->mode;
} else {
/*
* Since origin->path is a pathspec, if the parent
* commit had it as a directory, we will see a whole
* bunch of deletion of files in the directory that we
* do not care about.
*/
int i;
struct diff_filepair *p = NULL;
for (i = 0; i < diff_queued_diff.nr; i++) {
const char *name;
p = diff_queued_diff.queue[i];
name = p->one->path ? p->one->path : p->two->path;
if (!strcmp(name, origin->path))
break;
}
if (!p)
die("internal error in blame::find_origin");
switch (p->status) {
default:
die("internal error in blame::find_origin (%c)",
p->status);
case 'M':
porigin = get_origin(parent, origin->path);
oidcpy(&porigin->blob_oid, &p->one->oid);
porigin->mode = p->one->mode;
break;
case 'A':
case 'T':
/* Did not exist in parent, or type changed */
break;
}
}
diff_flush(&diff_opts);
clear_pathspec(&diff_opts.pathspec);
return porigin;
}
/*
* We have an origin -- find the path that corresponds to it in its
* parent and return an origin structure to represent it.
*/
static struct blame_origin *find_rename(struct commit *parent,
struct blame_origin *origin)
{
struct blame_origin *porigin = NULL;
struct diff_options diff_opts;
int i;
diff_setup(&diff_opts);
DIFF_OPT_SET(&diff_opts, RECURSIVE);
diff_opts.detect_rename = DIFF_DETECT_RENAME;
diff_opts.output_format = DIFF_FORMAT_NO_OUTPUT;
diff_opts.single_follow = origin->path;
diff_setup_done(&diff_opts);
if (is_null_oid(&origin->commit->object.oid))
do_diff_cache(&parent->tree->object.oid, &diff_opts);
else
diff_tree_oid(&parent->tree->object.oid,
&origin->commit->tree->object.oid,
"", &diff_opts);
diffcore_std(&diff_opts);
for (i = 0; i < diff_queued_diff.nr; i++) {
struct diff_filepair *p = diff_queued_diff.queue[i];
if ((p->status == 'R' || p->status == 'C') &&
!strcmp(p->two->path, origin->path)) {
porigin = get_origin(parent, p->one->path);
oidcpy(&porigin->blob_oid, &p->one->oid);
porigin->mode = p->one->mode;
break;
}
}
diff_flush(&diff_opts);
clear_pathspec(&diff_opts.pathspec);
return porigin;
}
/*
* Append a new blame entry to a given output queue.
*/
static void add_blame_entry(struct blame_entry ***queue,
const struct blame_entry *src)
{
struct blame_entry *e = xmalloc(sizeof(*e));
memcpy(e, src, sizeof(*e));
blame_origin_incref(e->suspect);
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 gets added to the given queue. The
* origin of dst loses a refcnt.
*/
static void dup_entry(struct blame_entry ***queue,
struct blame_entry *dst, struct blame_entry *src)
{
blame_origin_incref(src->suspect);
blame_origin_decref(dst->suspect);
memcpy(dst, src, sizeof(*src));
dst->next = **queue;
**queue = dst;
*queue = &dst->next;
}
const char *blame_nth_line(struct blame_scoreboard *sb, long lno)
{
return sb->final_buf + sb->lineno[lno];
}
/*
* It is known that lines between tlno to same came from parent, and e
* has an overlap with that range. it also is known that parent's
* line plno corresponds to e's line tlno.
*
* <---- e ----->
* <------>
* <------------>
* <------------>
* <------------------>
*
* Split e into potentially three parts; before this chunk, the chunk
* to be blamed for the parent, and after that portion.
*/
static void split_overlap(struct blame_entry *split,
struct blame_entry *e,
int tlno, int plno, int same,
struct blame_origin *parent)
{
int chunk_end_lno;
memset(split, 0, sizeof(struct blame_entry [3]));
if (e->s_lno < tlno) {
/* there is a pre-chunk part not blamed on parent */
split[0].suspect = blame_origin_incref(e->suspect);
split[0].lno = e->lno;
split[0].s_lno = e->s_lno;
split[0].num_lines = tlno - e->s_lno;
split[1].lno = e->lno + tlno - e->s_lno;
split[1].s_lno = plno;
}
else {
split[1].lno = e->lno;
split[1].s_lno = plno + (e->s_lno - tlno);
}
if (same < e->s_lno + e->num_lines) {
/* there is a post-chunk part not blamed on parent */
split[2].suspect = blame_origin_incref(e->suspect);
split[2].lno = e->lno + (same - e->s_lno);
split[2].s_lno = e->s_lno + (same - e->s_lno);
split[2].num_lines = e->s_lno + e->num_lines - same;
chunk_end_lno = split[2].lno;
}
else
chunk_end_lno = e->lno + e->num_lines;
split[1].num_lines = chunk_end_lno - split[1].lno;
/*
* if it turns out there is nothing to blame the parent for,
* forget about the splitting. !split[1].suspect signals this.
*/
if (split[1].num_lines < 1)
return;
split[1].suspect = blame_origin_incref(parent);
}
/*
* split_overlap() divided an existing blame e into up to three parts
* in split. Any assigned blame is moved to queue to
* reflect the split.
*/
static void split_blame(struct blame_entry ***blamed,
struct blame_entry ***unblamed,
struct blame_entry *split,
struct blame_entry *e)
{
if (split[0].suspect && split[2].suspect) {
/* The first part (reuse storage for the existing entry e) */
dup_entry(unblamed, e, &split[0]);
/* The last part -- me */
add_blame_entry(unblamed, &split[2]);
/* ... and the middle part -- parent */
add_blame_entry(blamed, &split[1]);
}
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(blamed, e, &split[1]);
else if (split[0].suspect) {
/* me and then parent */
dup_entry(unblamed, e, &split[0]);
add_blame_entry(blamed, &split[1]);
}
else {
/* parent and then me */
dup_entry(blamed, e, &split[1]);
add_blame_entry(unblamed, &split[2]);
}
}
/*
* After splitting the blame, the origins used by the
* on-stack blame_entry should lose one refcnt each.
*/
static void decref_split(struct blame_entry *split)
{
int i;
for (i = 0; i < 3; i++)
blame_origin_decref(split[i].suspect);
}
/*
* 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 struct blame_entry *reverse_blame(struct blame_entry *head,
struct blame_entry *tail)
{
while (head) {
struct blame_entry *next = head->next;
head->next = tail;
tail = head;
head = next;
}
return tail;
}
/*
* Process one hunk from the patch between the current suspect for
* 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 blame_entry ***dstq, struct blame_entry ***srcq,
int tlno, int offset, int same,
struct blame_origin *parent)
{
struct blame_entry *e = **srcq;
struct blame_entry *samep = NULL, *diffp = NULL;
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
blame_origin_decref(e->suspect);
/* Pass blame for everything before the differing
* chunk to the parent */
e->suspect = blame_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 = blame_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 blame_origin *parent;
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;
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;
}
/*
* We are looking at the origin 'target' and aiming to pass blame
* for the lines it is suspected to its parent. Run diff to find
* which lines came from parent and pass blame for them.
*/
static void pass_blame_to_parent(struct blame_scoreboard *sb,
struct blame_origin *target,
struct blame_origin *parent)
{
mmfile_t file_p, file_o;
struct blame_chunk_cb_data d;
struct blame_entry *newdest = NULL;
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, &sb->num_read_blob);
fill_origin_blob(&sb->revs->diffopt, target, &file_o, &sb->num_read_blob);
sb->num_get_patch++;
if (diff_hunks(&file_p, &file_o, blame_chunk_cb, &d, sb->xdl_opts))
die("unable to generate diff (%s -> %s)",
oid_to_hex(&parent->commit->object.oid),
oid_to_hex(&target->commit->object.oid));
/* 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;
}
/*
* The lines in blame_entry after splitting blames many times can become
* very small and trivial, and at some point it becomes pointless to
* blame the parents. E.g. "\t\t}\n\t}\n\n" appears everywhere in any
* ordinary C program, and it is not worth to say it was copied from
* totally unrelated file in the parent.
*
* Compute how trivial the lines in the blame_entry are.
*/
unsigned blame_entry_score(struct blame_scoreboard *sb, struct blame_entry *e)
{
unsigned score;
const char *cp, *ep;
if (e->score)
return e->score;
score = 1;
cp = blame_nth_line(sb, e->lno);
ep = blame_nth_line(sb, e->lno + e->num_lines);
while (cp < ep) {
unsigned ch = *((unsigned char *)cp);
if (isalnum(ch))
score++;
cp++;
}
e->score = score;
return score;
}
/*