diffcore: add a pickaxe option to find a specific blob
Sometimes users are given a hash of an object and they want to identify it further (ex.: Use verify-pack to find the largest blobs, but what are these? or [1]) One might be tempted to extend git-describe to also work with blobs, such that `git describe <blob-id>` gives a description as '<commit-ish>:<path>'. This was implemented at [2]; as seen by the sheer number of responses (>110), it turns out this is tricky to get right. The hard part to get right is picking the correct 'commit-ish' as that could be the commit that (re-)introduced the blob or the blob that removed the blob; the blob could exist in different branches. Junio hinted at a different approach of solving this problem, which this patch implements. Teach the diff machinery another flag for restricting the information to what is shown. For example: $ ./git log --oneline --find-object=v2.0.0:Makefile b2feb64309 Revert the whole "ask curl-config" topic for now 47fbfded53 i18n: only extract comments marked with "TRANSLATORS:" we observe that the Makefile as shipped with 2.0 was appeared in v1.9.2-471-g47fbfded53 and in v2.0.0-rc1-5-gb2feb6430b. The reason why these commits both occur prior to v2.0.0 are evil merges that are not found using this new mechanism. [1] https://stackoverflow.com/questions/223678/which-commit-has-this-blob [2] https://public-inbox.org/git/20171028004419.10139-1-sbeller@google.com/ Signed-off-by: Stefan Beller <sbeller@google.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Stefan Beller committed
Jan 4, 2018 at 14:50 UTC
15af58c1adba431c216e2a45fa0d22944560ba02
6 files changed
+135
-20
Documentation/diff-options.txt
+10
@@ -492,6 +492,15 @@ occurrences of that string did not change).
492
See the 'pickaxe' entry in linkgit:gitdiffcore[7] for more
493
information.
494
495
+--find-object=<object-id>::
496
+ Look for differences that change the number of occurrences of
497
+ the specified object. Similar to `-S`, just the argument is different
498
+ in that it doesn't search for a specific string but for a specific
499
+ object id.
500
++
501
+The object can be a blob or a submodule commit. It implies the `-t` option in
502
+`git-log` to also find trees.
503
+
504
--pickaxe-all::
505
When `-S` or `-G` finds a change, show all the changes in that
506
changeset, not just the files that contain the change
@@ -500,6 +509,7 @@ information.
509
--pickaxe-regex::
510
Treat the <string> given to `-S` as an extended POSIX regular
511
expression to match.
512
+
513
endif::git-format-patch[]
514
515
-O<orderfile>::
diff.c
+20
-1
@@ -4082,6 +4082,7 @@ void diff_setup(struct diff_options *options)
4082
options->interhunkcontext = diff_interhunk_context_default;
4083
options->ws_error_highlight = ws_error_highlight_default;
4084
options->flags.rename_empty = 1;
4085
+ options->objfind = NULL;
4086
4087
/* pathchange left =NULL by default */
4088
options->change = diff_change;
@@ -4487,6 +4488,23 @@ static int parse_ws_error_highlight_opt(struct diff_options *opt, const char *ar
4488
return 1;
4489
}
4490
4491
+static int parse_objfind_opt(struct diff_options *opt, const char *arg)
4492
+{
4493
+ struct object_id oid;
4494
+
4495
+ if (get_oid(arg, &oid))
4496
+ return error("unable to resolve '%s'", arg);
4497
+
4498
+ if (!opt->objfind)
4499
+ opt->objfind = xcalloc(1, sizeof(*opt->objfind));
4500
+
4501
+ opt->pickaxe_opts |= DIFF_PICKAXE_KIND_OBJFIND;
4502
+ opt->flags.recursive = 1;
4503
+ opt->flags.tree_in_recursive = 1;
4504
+ oidset_insert(opt->objfind, &oid);
4505
+ return 1;
4506
+}
4507
+
4508
int diff_opt_parse(struct diff_options *options,
4509
const char **av, int ac, const char *prefix)
4510
{
@@ -4736,7 +4754,8 @@ int diff_opt_parse(struct diff_options *options,
4754
else if ((argcount = short_opt('O', av, &optarg))) {
4755
options->orderfile = prefix_filename(prefix, optarg);
4756
return argcount;
4739
- }
4757
+ } else if (skip_prefix(arg, "--find-object=", &arg))
4758
+ return parse_objfind_opt(options, arg);
4759
else if ((argcount = parse_long_opt("diff-filter", av, &optarg))) {
4760
int offending = parse_diff_filter_opt(optarg, options);
4761
if (offending)
diff.h
+7
-1
@@ -7,6 +7,7 @@
7
#include "tree-walk.h"
8
#include "pathspec.h"
9
#include "object.h"
10
+#include "oidset.h"
11
12
struct rev_info;
13
struct diff_options;
@@ -173,6 +174,8 @@ struct diff_options {
174
enum diff_words_type word_diff;
175
enum diff_submodule_format submodule_format;
176
177
+ struct oidset *objfind;
178
+
179
/* this is set by diffcore for DIFF_FORMAT_PATCH */
180
int found_changes;
181
@@ -325,8 +328,11 @@ extern void diff_setup_done(struct diff_options *);
328
329
#define DIFF_PICKAXE_KIND_S 4 /* traditional plumbing counter */
330
#define DIFF_PICKAXE_KIND_G 8 /* grep in the patch */
331
+#define DIFF_PICKAXE_KIND_OBJFIND 16 /* specific object IDs */
332
329
-#define DIFF_PICKAXE_KINDS_MASK (DIFF_PICKAXE_KIND_S | DIFF_PICKAXE_KIND_G)
333
+#define DIFF_PICKAXE_KINDS_MASK (DIFF_PICKAXE_KIND_S | \
334
+ DIFF_PICKAXE_KIND_G | \
335
+ DIFF_PICKAXE_KIND_OBJFIND)
336
337
#define DIFF_PICKAXE_IGNORE_CASE 32
338
diffcore-pickaxe.c
+27
-18
@@ -124,13 +124,20 @@ static int pickaxe_match(struct diff_filepair *p, struct diff_options *o,
124
mmfile_t mf1, mf2;
125
int ret;
126
127
- if (!o->pickaxe[0])
128
- return 0;
129
-
127
/* ignore unmerged */
128
if (!DIFF_FILE_VALID(p->one) && !DIFF_FILE_VALID(p->two))
129
return 0;
130
131
+ if (o->objfind) {
132
+ return (DIFF_FILE_VALID(p->one) &&
133
+ oidset_contains(o->objfind, &p->one->oid)) ||
134
+ (DIFF_FILE_VALID(p->two) &&
135
+ oidset_contains(o->objfind, &p->two->oid));
136
+ }
137
+
138
+ if (!o->pickaxe[0])
139
+ return 0;
140
+
141
if (o->flags.allow_textconv) {
142
textconv_one = get_textconv(p->one);
143
textconv_two = get_textconv(p->two);
@@ -226,20 +233,22 @@ void diffcore_pickaxe(struct diff_options *o)
233
cflags |= REG_ICASE;
234
regcomp_or_die(®ex, needle, cflags);
235
regexp = ®ex;
229
- } else if (o->pickaxe_opts & DIFF_PICKAXE_IGNORE_CASE &&
230
- has_non_ascii(needle)) {
231
- struct strbuf sb = STRBUF_INIT;
232
- int cflags = REG_NEWLINE | REG_ICASE;
233
-
234
- basic_regex_quote_buf(&sb, needle);
235
- regcomp_or_die(®ex, sb.buf, cflags);
236
- strbuf_release(&sb);
237
- regexp = ®ex;
238
- } else {
239
- kws = kwsalloc(o->pickaxe_opts & DIFF_PICKAXE_IGNORE_CASE
240
- ? tolower_trans_tbl : NULL);
241
- kwsincr(kws, needle, strlen(needle));
242
- kwsprep(kws);
236
+ } else if (opts & DIFF_PICKAXE_KIND_S) {
237
+ if (o->pickaxe_opts & DIFF_PICKAXE_IGNORE_CASE &&
238
+ has_non_ascii(needle)) {
239
+ struct strbuf sb = STRBUF_INIT;
240
+ int cflags = REG_NEWLINE | REG_ICASE;
241
+
242
+ basic_regex_quote_buf(&sb, needle);
243
+ regcomp_or_die(®ex, sb.buf, cflags);
244
+ strbuf_release(&sb);
245
+ regexp = ®ex;
246
+ } else {
247
+ kws = kwsalloc(o->pickaxe_opts & DIFF_PICKAXE_IGNORE_CASE
248
+ ? tolower_trans_tbl : NULL);
249
+ kwsincr(kws, needle, strlen(needle));
250
+ kwsprep(kws);
251
+ }
252
}
253
254
/* Might want to warn when both S and G are on; I don't care... */
@@ -248,7 +257,7 @@ void diffcore_pickaxe(struct diff_options *o)
257
258
if (regexp)
259
regfree(regexp);
251
- else
260
+ if (kws)
261
kwsfree(kws);
262
return;
263
}
revision.c
+3
@@ -2412,6 +2412,9 @@ int setup_revisions(int argc, const char **argv, struct rev_info *revs, struct s
2412
revs->diffopt.flags.follow_renames)
2413
revs->diff = 1;
2414
2415
+ if (revs->diffopt.objfind)
2416
+ revs->simplify_history = 0;
2417
+
2418
if (revs->topo_order)
2419
revs->limited = 1;
2420
t/t4064-diff-oidfind.sh
new
+68
@@ -0,0 +1,68 @@
1
+#!/bin/sh
2
+
3
+test_description='test finding specific blobs in the revision walking'
4
+. ./test-lib.sh
5
+
6
+test_expect_success 'setup ' '
7
+ git commit --allow-empty -m "empty initial commit" &&
8
+
9
+ echo "Hello, world!" >greeting &&
10
+ git add greeting &&
11
+ git commit -m "add the greeting blob" && # borrowed from Git from the Bottom Up
12
+ git tag -m "the blob" greeting $(git rev-parse HEAD:greeting) &&
13
+
14
+ echo asdf >unrelated &&
15
+ git add unrelated &&
16
+ git commit -m "unrelated history" &&
17
+
18
+ git revert HEAD^ &&
19
+
20
+ git commit --allow-empty -m "another unrelated commit"
21
+'
22
+
23
+test_expect_success 'find the greeting blob' '
24
+ cat >expect <<-EOF &&
25
+ Revert "add the greeting blob"
26
+ add the greeting blob
27
+ EOF
28
+
29
+ git log --format=%s --find-object=greeting^{blob} >actual &&
30
+
31
+ test_cmp expect actual
32
+'
33
+
34
+test_expect_success 'setup a tree' '
35
+ mkdir a &&
36
+ echo asdf >a/file &&
37
+ git add a/file &&
38
+ git commit -m "add a file in a subdirectory"
39
+'
40
+
41
+test_expect_success 'find a tree' '
42
+ cat >expect <<-EOF &&
43
+ add a file in a subdirectory
44
+ EOF
45
+
46
+ git log --format=%s -t --find-object=HEAD:a >actual &&
47
+
48
+ test_cmp expect actual
49
+'
50
+
51
+test_expect_success 'setup a submodule' '
52
+ test_create_repo sub &&
53
+ test_commit -C sub sub &&
54
+ git submodule add ./sub sub &&
55
+ git commit -a -m "add sub"
56
+'
57
+
58
+test_expect_success 'find a submodule' '
59
+ cat >expect <<-EOF &&
60
+ add sub
61
+ EOF
62
+
63
+ git log --format=%s --find-object=HEAD:sub >actual &&
64
+
65
+ test_cmp expect actual
66
+'
67
+
68
+test_done