@matheusbsbs / matheus-bagliano / commits / 04deccda11

log: re-encode commit messages before grepping

If you run "git log --grep=foo", we will run your regex on the literal bytes of the commit message. This can provide confusing results if the commit message is not in the same encoding as your grep expression (or worse, you have commits in multiple encodings, in which case your regex would need to be written to match either encoding). On top of this, we might also be grepping in the commit's notes, which are already re-encoded, potentially leading to grepping in a buffer with mixed encodings concatenated. This is insanity, but most people never noticed, because their terminal and their commit encodings all match. Instead, let's massage the to-be-grepped commit into a standardized encoding. There is not much point in adding a flag for "this is the encoding I expect my grep pattern to match"; the only sane choice is for it to use the log output encoding. That is presumably what the user's terminal is using, and it means that the patterns found by the grep will match the output produced by git. As a bonus, this fixes a potential segfault in commit_match when commit->buffer is NULL, as we now build on logmsg_reencode, which handles reading the commit buffer from disk if necessary. The segfault can be triggered with: git commit -m 'text1' --allow-empty git commit -m 'text2' --allow-empty git log --graph --no-walk --grep 'text2' which arguably does not make any sense (--graph inherently wants a connected history, and by --no-walk the command line is telling us to show discrete points in history without connectivity), and we probably should forbid the combination, but that is a separate issue. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Jeff King committed Feb 11, 2013 at 15:59 UTC 04deccda112881f87bb514445aa2e539f77f6095

2 files changed +78 -7

revision.c

+20 -7

@@ -2268,7 +2268,10 @@ static int commit_rewrite_person(struct strbuf *buf, const char *what, struct st

2268 static int commit_match(struct commit *commit, struct rev_info *opt)

2269 {

2270 int retval;

2271 + const char *encoding;

2272 + char *message;

2273 struct strbuf buf = STRBUF_INIT;

2274 +

2275 if (!opt->grep_filter.pattern_list && !opt->grep_filter.header_list)

2276 return 1;

2277

@@ -2279,13 +2282,23 @@ static int commit_match(struct commit *commit, struct rev_info *opt)

2282 strbuf_addch(&buf, '\n');

2283 }

2284

2285 + /*

2286 + * We grep in the user's output encoding, under the assumption that it

2287 + * is the encoding they are most likely to write their grep pattern

2288 + * for. In addition, it means we will match the "notes" encoding below,

2289 + * so we will not end up with a buffer that has two different encodings

2290 + * in it.

2291 + */

2292 + encoding = get_log_output_encoding();

2293 + message = logmsg_reencode(commit, encoding);

2294 +

2295 /* Copy the commit to temporary if we are using "fake" headers */

2296 if (buf.len)

2284 - strbuf_addstr(&buf, commit->buffer);

2297 + strbuf_addstr(&buf, message);

2298

2299 if (opt->grep_filter.header_list && opt->mailmap) {

2300 if (!buf.len)

2288 - strbuf_addstr(&buf, commit->buffer);

2301 + strbuf_addstr(&buf, message);

2302

2303 commit_rewrite_person(&buf, "\nauthor ", opt->mailmap);

2304 commit_rewrite_person(&buf, "\ncommitter ", opt->mailmap);

@@ -2294,18 +2307,18 @@ static int commit_match(struct commit *commit, struct rev_info *opt)

2307 /* Append "fake" message parts as needed */

2308 if (opt->show_notes) {

2309 if (!buf.len)

2297 - strbuf_addstr(&buf, commit->buffer);

2298 - format_display_notes(commit->object.sha1, &buf,

2299 - get_log_output_encoding(), 1);

2310 + strbuf_addstr(&buf, message);

2311 + format_display_notes(commit->object.sha1, &buf, encoding, 1);

2312 }

2313

2302 - /* Find either in the commit object, or in the temporary */

2314 + /* Find either in the original commit message, or in the temporary */

2315 if (buf.len)

2316 retval = grep_buffer(&opt->grep_filter, buf.buf, buf.len);

2317 else

2318 retval = grep_buffer(&opt->grep_filter,

2307 - commit->buffer, strlen(commit->buffer));

2319 + message, strlen(message));

2320 strbuf_release(&buf);

2321 + logmsg_free(message, commit);

2322 return retval;

2323 }

2324

t/t4210-log-i18n.sh new

+58

@@ -0,0 +1,58 @@

1 +#!/bin/sh

2 +

3 +test_description='test log with i18n features'

4 +. ./test-lib.sh

5 +

6 +# two forms of é

7 +utf8_e=$(printf '\303\251')

8 +latin1_e=$(printf '\351')

9 +

10 +test_expect_success 'create commits in different encodings' '

11 + test_tick &&

12 + cat >msg <<-EOF &&

13 + utf8

14 +

15 + t${utf8_e}st

16 + EOF

17 + git add msg &&

18 + git -c i18n.commitencoding=utf8 commit -F msg &&

19 + cat >msg <<-EOF &&

20 + latin1

21 +

22 + t${latin1_e}st

23 + EOF

24 + git add msg &&

25 + git -c i18n.commitencoding=ISO-8859-1 commit -F msg

26 +'

27 +

28 +test_expect_success 'log --grep searches in log output encoding (utf8)' '

29 + cat >expect <<-\EOF &&

30 + latin1

31 + utf8

32 + EOF

33 + git log --encoding=utf8 --format=%s --grep=$utf8_e >actual &&

34 + test_cmp expect actual

35 +'

36 +

37 +test_expect_success 'log --grep searches in log output encoding (latin1)' '

38 + cat >expect <<-\EOF &&

39 + latin1

40 + utf8

41 + EOF

42 + git log --encoding=ISO-8859-1 --format=%s --grep=$latin1_e >actual &&

43 + test_cmp expect actual

44 +'

45 +

46 +test_expect_success 'log --grep does not find non-reencoded values (utf8)' '

47 + >expect &&

48 + git log --encoding=utf8 --format=%s --grep=$latin1_e >actual &&

49 + test_cmp expect actual

50 +'

51 +

52 +test_expect_success 'log --grep does not find non-reencoded values (latin1)' '

53 + >expect &&

54 + git log --encoding=ISO-8859-1 --format=%s --grep=$utf8_e >actual &&

55 + test_cmp expect actual

56 +'

57 +

58 +test_done