word diff: handle zero length matches

If find_word_boundaries() encounters a zero length match (which can be
caused by matching a newline or using '*' instead of '+' in the regex)
we stop splitting the input into words which generates an inaccurate
diff. To fix this increment the start point when there is a zero
length match and try a new match. This is safe as posix regular
expressions always return the longest available match so a zero length
match means there are no longer matches available from the current
position.

Commit bf82940dbf (color-words: enable REG_NEWLINE to help user,
2009-01-17) prevented matching newlines in negated character classes
but it is still possible for the user to have an explicit newline
match in the regex which could cause a zero length match.

One could argue that having explicit newline matches or using '*'
rather than '+' are user errors but it seems to be better to work
round them than produce inaccurate diffs.

Signed-off-by: Phillip Wood <phillip.wood@dunelm.org.uk>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This commit is contained in:
Phillip Wood 2021-05-04 09:27:34 +00:00 committed by Junio C Hamano
parent 48bf2fa8ba
commit 0324e8fc6b
2 changed files with 12 additions and 3 deletions

10
diff.c
View File

@ -2053,7 +2053,7 @@ static void fn_out_diff_words_aux(void *priv,
static int find_word_boundaries(mmfile_t *buffer, regex_t *word_regex,
int *begin, int *end)
{
if (word_regex && *begin < buffer->size) {
while (word_regex && *begin < buffer->size) {
regmatch_t match[1];
if (!regexec_buf(word_regex, buffer->ptr + *begin,
buffer->size - *begin, 1, match, 0)) {
@ -2061,9 +2061,13 @@ static int find_word_boundaries(mmfile_t *buffer, regex_t *word_regex,
'\n', match[0].rm_eo - match[0].rm_so);
*end = p ? p - buffer->ptr : match[0].rm_eo + *begin;
*begin += match[0].rm_so;
return *begin >= *end;
if (*begin == *end)
(*begin)++;
else
return *begin > *end;
} else {
return -1;
}
return -1;
}
/* find the next word */

View File

@ -184,6 +184,11 @@ test_expect_success 'word diff with a regular expression' '
word_diff --color-words="[a-z]+"
'
test_expect_success 'word diff with zero length matches' '
cp expect.letter-runs-are-words expect &&
word_diff --color-words="[a-z${LF}]*"
'
test_expect_success 'set up a diff driver' '
git config diff.testdriver.wordRegex "[^[:space:]]" &&
cat <<-\EOF >.gitattributes