git log/diff: add -G<regexp> that greps in the patch text
Teach "-G<regexp>" that is similar to "-S<regexp> --pickaxe-regexp" to the "git diff" family of commands. This limits the diff queue to filepairs whose patch text actually has an added or a deleted line that matches the given regexp. Unlike "-S<regexp>", changing other parts of the line that has a substring that matches the given regexp IS counted as a change, as such a change would appear as one deletion followed by one addition in a patch text. Unlike -S (pickaxe) that is intended to be used to quickly detect a commit that changes the number of occurrences of hits between the preimage and the postimage to serve as a part of larger toolchain, this is meant to be used as the top-level Porcelain feature. The implementation unfortunately has to run "diff" twice if you are running "log" family of commands to produce patches in the final output (e.g. "git log -p" or "git format-patch"). I think we _could_ cache the result in-core if we wanted to, but that would require larger surgery to the diffcore machinery (i.e. adding an extra pointer in the filepair structure to keep a pointer to a strbuf around, stuff the textual diff to the strbuf inside diffgrep_consume(), and make use of it in later stages when it is available) and it may not be worth it. Signed-off-by: Junio C Hamano <gitster@pobox.com>
Junio C Hamano committed
Aug 23, 2010 at 10:17 UTC
f506b8e8b5fb231e63b69668b6b7516f12b32228
4 files changed
+164
-4
Documentation/diff-options.txt
+5
-1
@@ -284,8 +284,12 @@ ifndef::git-format-patch[]
284
appearing in diff output; see the 'pickaxe' entry in
285
linkgit:gitdiffcore[7] for more details.
286
287
+-G<regex>::
288
+ Look for differences whose added or removed line matches
289
+ the given <regex>.
290
+
291
--pickaxe-all::
288
- When `-S` finds a change, show all the changes in that
292
+ When `-S` or `-G` finds a change, show all the changes in that
293
changeset, not just the files that contain the change
294
in <string>.
295
diff.c
+7
-2
@@ -3268,12 +3268,17 @@ int diff_opt_parse(struct diff_options *options, const char **av, int ac)
3268
}
3269
else if ((argcount = short_opt('S', av, &optarg))) {
3270
options->pickaxe = optarg;
3271
+ options->pickaxe_opts |= DIFF_PICKAXE_KIND_S;
3272
+ return argcount;
3273
+ } else if ((argcount = short_opt('G', av, &optarg))) {
3274
+ options->pickaxe = optarg;
3275
+ options->pickaxe_opts |= DIFF_PICKAXE_KIND_G;
3276
return argcount;
3277
}
3278
else if (!strcmp(arg, "--pickaxe-all"))
3274
- options->pickaxe_opts = DIFF_PICKAXE_ALL;
3279
+ options->pickaxe_opts |= DIFF_PICKAXE_ALL;
3280
else if (!strcmp(arg, "--pickaxe-regex"))
3276
- options->pickaxe_opts = DIFF_PICKAXE_REGEX;
3281
+ options->pickaxe_opts |= DIFF_PICKAXE_REGEX;
3282
else if ((argcount = short_opt('O', av, &optarg))) {
3283
options->orderfile = optarg;
3284
return argcount;
diff.h
+3
@@ -238,6 +238,9 @@ extern int diff_setup_done(struct diff_options *);
238
#define DIFF_PICKAXE_ALL 1
239
#define DIFF_PICKAXE_REGEX 2
240
241
+#define DIFF_PICKAXE_KIND_S 4 /* traditional plumbing counter */
242
+#define DIFF_PICKAXE_KIND_G 8 /* grep in the patch */
243
+
244
extern void diffcore_std(struct diff_options *);
245
extern void diffcore_fix_diff_index(struct diff_options *);
246
diffcore-pickaxe.c
+149
-1
@@ -1,9 +1,148 @@
1
/*
2
* Copyright (C) 2005 Junio C Hamano
3
+ * Copyright (C) 2010 Google Inc.
4
*/
5
#include "cache.h"
6
#include "diff.h"
7
#include "diffcore.h"
8
+#include "xdiff-interface.h"
9
+
10
+struct diffgrep_cb {
11
+ regex_t *regexp;
12
+ int hit;
13
+};
14
+
15
+static void diffgrep_consume(void *priv, char *line, unsigned long len)
16
+{
17
+ struct diffgrep_cb *data = priv;
18
+ regmatch_t regmatch;
19
+ int hold;
20
+
21
+ if (line[0] != '+' && line[0] != '-')
22
+ return;
23
+ if (data->hit)
24
+ /*
25
+ * NEEDSWORK: we should have a way to terminate the
26
+ * caller early.
27
+ */
28
+ return;
29
+ /* Yuck -- line ought to be "const char *"! */
30
+ hold = line[len];
31
+ line[len] = '\0';
32
+ data->hit = !regexec(data->regexp, line + 1, 1, ®match, 0);
33
+ line[len] = hold;
34
+}
35
+
36
+static void fill_one(struct diff_filespec *one,
37
+ mmfile_t *mf, struct userdiff_driver **textconv)
38
+{
39
+ if (DIFF_FILE_VALID(one)) {
40
+ *textconv = get_textconv(one);
41
+ mf->size = fill_textconv(*textconv, one, &mf->ptr);
42
+ } else {
43
+ memset(mf, 0, sizeof(*mf));
44
+ }
45
+}
46
+
47
+static int diff_grep(struct diff_filepair *p, regex_t *regexp, struct diff_options *o)
48
+{
49
+ regmatch_t regmatch;
50
+ struct userdiff_driver *textconv_one = NULL;
51
+ struct userdiff_driver *textconv_two = NULL;
52
+ mmfile_t mf1, mf2;
53
+ int hit;
54
+
55
+ if (diff_unmodified_pair(p))
56
+ return 0;
57
+
58
+ fill_one(p->one, &mf1, &textconv_one);
59
+ fill_one(p->two, &mf2, &textconv_two);
60
+
61
+ if (!mf1.ptr) {
62
+ if (!mf2.ptr)
63
+ return 0; /* ignore unmerged */
64
+ /* created "two" -- does it have what we are looking for? */
65
+ hit = !regexec(regexp, p->two->data, 1, ®match, 0);
66
+ } else if (!mf2.ptr) {
67
+ /* removed "one" -- did it have what we are looking for? */
68
+ hit = !regexec(regexp, p->one->data, 1, ®match, 0);
69
+ } else {
70
+ /*
71
+ * We have both sides; need to run textual diff and see if
72
+ * the pattern appears on added/deleted lines.
73
+ */
74
+ struct diffgrep_cb ecbdata;
75
+ xpparam_t xpp;
76
+ xdemitconf_t xecfg;
77
+
78
+ memset(&xpp, 0, sizeof(xpp));
79
+ memset(&xecfg, 0, sizeof(xecfg));
80
+ ecbdata.regexp = regexp;
81
+ ecbdata.hit = 0;
82
+ xecfg.ctxlen = o->context;
83
+ xecfg.interhunkctxlen = o->interhunkcontext;
84
+ xdi_diff_outf(&mf1, &mf2, diffgrep_consume, &ecbdata,
85
+ &xpp, &xecfg);
86
+ hit = ecbdata.hit;
87
+ }
88
+ if (textconv_one)
89
+ free(mf1.ptr);
90
+ if (textconv_two)
91
+ free(mf2.ptr);
92
+ return hit;
93
+}
94
+
95
+static void diffcore_pickaxe_grep(struct diff_options *o)
96
+{
97
+ struct diff_queue_struct *q = &diff_queued_diff;
98
+ int i, has_changes, err;
99
+ regex_t regex;
100
+ struct diff_queue_struct outq;
101
+ outq.queue = NULL;
102
+ outq.nr = outq.alloc = 0;
103
+
104
+ err = regcomp(®ex, o->pickaxe, REG_EXTENDED | REG_NEWLINE);
105
+ if (err) {
106
+ char errbuf[1024];
107
+ regerror(err, ®ex, errbuf, 1024);
108
+ regfree(®ex);
109
+ die("invalid log-grep regex: %s", errbuf);
110
+ }
111
+
112
+ if (o->pickaxe_opts & DIFF_PICKAXE_ALL) {
113
+ /* Showing the whole changeset if needle exists */
114
+ for (i = has_changes = 0; !has_changes && i < q->nr; i++) {
115
+ struct diff_filepair *p = q->queue[i];
116
+ if (diff_grep(p, ®ex, o))
117
+ has_changes++;
118
+ }
119
+ if (has_changes)
120
+ return; /* do not munge the queue */
121
+
122
+ /*
123
+ * Otherwise we will clear the whole queue by copying
124
+ * the empty outq at the end of this function, but
125
+ * first clear the current entries in the queue.
126
+ */
127
+ for (i = 0; i < q->nr; i++)
128
+ diff_free_filepair(q->queue[i]);
129
+ } else {
130
+ /* Showing only the filepairs that has the needle */
131
+ for (i = 0; i < q->nr; i++) {
132
+ struct diff_filepair *p = q->queue[i];
133
+ if (diff_grep(p, ®ex, o))
134
+ diff_q(&outq, p);
135
+ else
136
+ diff_free_filepair(p);
137
+ }
138
+ }
139
+
140
+ regfree(®ex);
141
+
142
+ free(q->queue);
143
+ *q = outq;
144
+ return;
145
+}
146
147
static unsigned int contains(struct diff_filespec *one,
148
const char *needle, unsigned long len,
@@ -48,7 +187,7 @@ static unsigned int contains(struct diff_filespec *one,
187
return cnt;
188
}
189
51
-void diffcore_pickaxe(struct diff_options *o)
190
+static void diffcore_pickaxe_count(struct diff_options *o)
191
{
192
const char *needle = o->pickaxe;
193
int opts = o->pickaxe_opts;
@@ -138,3 +277,12 @@ void diffcore_pickaxe(struct diff_options *o)
277
*q = outq;
278
return;
279
}
280
+
281
+void diffcore_pickaxe(struct diff_options *o)
282
+{
283
+ /* Might want to warn when both S and G are on; I don't care... */
284
+ if (o->pickaxe_opts & DIFF_PICKAXE_KIND_G)
285
+ return diffcore_pickaxe_grep(o);
286
+ else
287
+ return diffcore_pickaxe_count(o);
288
+}