summary | shortlog | log | commit | commitdiff | tree
raw | patch | inline | side by side (parent: 2b6a541)
raw | patch | inline | side by side (parent: 2b6a541)
author | Thomas Rast <trast@student.ethz.ch> | |
Sat, 17 Jan 2009 16:29:46 +0000 (17:29 +0100) | ||
committer | Junio C Hamano <gitster@pobox.com> | |
Sat, 17 Jan 2009 18:43:24 +0000 (10:43 -0800) |
We silently truncate a match at the newline, which may lead to
unexpected behaviour, e.g., when matching "<[^>]*>" against
<foo
bar>
since then "<foo" becomes a word (and "bar>" doesn't!) even though the
regex said only angle-bracket-delimited things can be words.
To alleviate the problem slightly, use REG_NEWLINE so that negated
classes can't match a newline. Of course newlines can still be
matched explicitly.
Signed-off-by: Thomas Rast <trast@student.ethz.ch>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
unexpected behaviour, e.g., when matching "<[^>]*>" against
<foo
bar>
since then "<foo" becomes a word (and "bar>" doesn't!) even though the
regex said only angle-bracket-delimited things can be words.
To alleviate the problem slightly, use REG_NEWLINE so that negated
classes can't match a newline. Of course newlines can still be
matched explicitly.
Signed-off-by: Thomas Rast <trast@student.ethz.ch>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff.c | patch | blob | history |
index 9fb3d0df31895e81a02d6bb67b9c4e0c13640155..00c661f82e744b3114db6cbd4a94d9c44d10a564 100644 (file)
--- a/diff.c
+++ b/diff.c
ecbdata.diff_words->word_regex = (regex_t *)
xmalloc(sizeof(regex_t));
if (regcomp(ecbdata.diff_words->word_regex,
- o->word_regex, REG_EXTENDED))
+ o->word_regex,
+ REG_EXTENDED | REG_NEWLINE))
die ("Invalid regular expression: %s",
o->word_regex);
}