grep: support newline separated pattern list

Currently, patterns that contain newline characters don't match anything when given to git grep. Regular grep(1) interprets patterns as lists of newline separated search strings instead. Implement this functionality by creating and inserting extra grep_pat structures for patterns consisting of multiple lines when appending to the pattern lists. For simplicity, all pattern strings are duplicated. The original pattern is truncated in place to make it contain only the first line. Requested-by: Torne (Richard Coles) <torne@google.com> Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>

René Scharfe committed May 20, 2012 at 16:33 UTC 526a858a99ace6698823740374edc3e35b87901a
4 files changed +41 -3
Documentation/git-grep.txt
+3 -1
@@ -29,7 +29,9 @@ SYNOPSIS
29 DESCRIPTION
30 -----------
31 Look for specified patterns in the tracked files in the work tree, blobs
32 -registered in the index file, or blobs in given tree objects.
32 +registered in the index file, or blobs in given tree objects. Patterns
33 +are lists of one or more search expressions separated by newline
34 +characters. An empty string as search expression matches all lines.
35
36
37 CONFIGURATION
grep.c
+32 -1
@@ -9,7 +9,7 @@ static struct grep_pat *create_grep_pat(const char *pat, size_t patlen,
9 enum grep_header_field field)
10 {
11 struct grep_pat *p = xcalloc(1, sizeof(*p));
12 - p->pattern = pat;
12 + p->pattern = xmemdupz(pat, patlen);
13 p->patternlen = patlen;
14 p->origin = origin;
15 p->no = no;
@@ -23,6 +23,36 @@ static void do_append_grep_pat(struct grep_pat ***tail, struct grep_pat *p)
23 **tail = p;
24 *tail = &p->next;
25 p->next = NULL;
26 +
27 + switch (p->token) {
28 + case GREP_PATTERN: /* atom */
29 + case GREP_PATTERN_HEAD:
30 + case GREP_PATTERN_BODY:
31 + for (;;) {
32 + struct grep_pat *new_pat;
33 + size_t len = 0;
34 + char *cp = p->pattern + p->patternlen, *nl = NULL;
35 + while (++len <= p->patternlen) {
36 + if (*(--cp) == '\n') {
37 + nl = cp;
38 + break;
39 + }
40 + }
41 + if (!nl)
42 + break;
43 + new_pat = create_grep_pat(nl + 1, len - 1, p->origin,
44 + p->no, p->token, p->field);
45 + new_pat->next = p->next;
46 + if (!p->next)
47 + *tail = &new_pat->next;
48 + p->next = new_pat;
49 + *nl = '\0';
50 + p->patternlen -= len;
51 + }
52 + break;
53 + default:
54 + break;
55 + }
56 }
57
58 void append_header_grep_pattern(struct grep_opt *opt,
@@ -408,6 +438,7 @@ void free_grep_patterns(struct grep_opt *opt)
438 free_pcre_regexp(p);
439 else
440 regfree(&p->regexp);
441 + free(p->pattern);
442 break;
443 default:
444 break;
grep.h
+1 -1
@@ -35,7 +35,7 @@ struct grep_pat {
35 const char *origin;
36 int no;
37 enum grep_pat_token token;
38 - const char *pattern;
38 + char *pattern;
39 size_t patternlen;
40 enum grep_header_field field;
41 regex_t regexp;
t/t7810-grep.sh
+5
@@ -322,6 +322,11 @@ test_expect_success 'grep -f, multiple patterns' '
322 test_cmp expected actual
323 '
324
325 +test_expect_success 'grep, multiple patterns' '
326 + git grep "$(cat patterns)" >actual &&
327 + test_cmp expected actual
328 +'
329 +
330 cat >expected <<EOF
331 file:foo mmap bar
332 file:foo_mmap bar