regex: use regexec_buf()
The new regexec_buf() function operates on buffers with an explicitly specified length, rather than NUL-terminated strings. We need to use this function whenever the buffer we want to pass to regexec(3) may have been mmap(2)ed (and is hence not NUL-terminated). Note: the original motivation for this patch was to fix a bug where `git diff -G <regex>` would crash. This patch converts more callers, though, some of which allocated to construct NUL-terminated strings, or worse, modified buffers to temporarily insert NULs while calling regexec(3). By converting them to use regexec_buf(), the code has become much cleaner. Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Johannes Schindelin committed
Sep 21, 2016 at 20:24 UTC
b7d36ffca02c23f545d6e098d78180e6e72dfd8d
5 files changed
+17
-33
diff.c
+2
-1
@@ -941,7 +941,8 @@ static int find_word_boundaries(mmfile_t *buffer, regex_t *word_regex,
941
{
942
if (word_regex && *begin < buffer->size) {
943
regmatch_t match[1];
944
- if (!regexec(word_regex, buffer->ptr + *begin, 1, match, 0)) {
944
+ if (!regexec_buf(word_regex, buffer->ptr + *begin,
945
+ buffer->size - *begin, 1, match, 0)) {
946
char *p = memchr(buffer->ptr + *begin + match[0].rm_so,
947
'\n', match[0].rm_eo - match[0].rm_so);
948
*end = p ? p - buffer->ptr : match[0].rm_eo + *begin;
diffcore-pickaxe.c
+8
-10
@@ -21,7 +21,6 @@ static void diffgrep_consume(void *priv, char *line, unsigned long len)
21
{
22
struct diffgrep_cb *data = priv;
23
regmatch_t regmatch;
24
- int hold;
24
25
if (line[0] != '+' && line[0] != '-')
26
return;
@@ -31,11 +30,8 @@ static void diffgrep_consume(void *priv, char *line, unsigned long len)
30
* caller early.
31
*/
32
return;
34
- /* Yuck -- line ought to be "const char *"! */
35
- hold = line[len];
36
- line[len] = '\0';
37
- data->hit = !regexec(data->regexp, line + 1, 1, ®match, 0);
38
- line[len] = hold;
33
+ data->hit = !regexec_buf(data->regexp, line + 1, len - 1, 1,
34
+ ®match, 0);
35
}
36
37
static int diff_grep(mmfile_t *one, mmfile_t *two,
@@ -48,9 +44,11 @@ static int diff_grep(mmfile_t *one, mmfile_t *two,
44
xdemitconf_t xecfg;
45
46
if (!one)
51
- return !regexec(regexp, two->ptr, 1, ®match, 0);
47
+ return !regexec_buf(regexp, two->ptr, two->size,
48
+ 1, ®match, 0);
49
if (!two)
53
- return !regexec(regexp, one->ptr, 1, ®match, 0);
50
+ return !regexec_buf(regexp, one->ptr, one->size,
51
+ 1, ®match, 0);
52
53
/*
54
* We have both sides; need to run textual diff and see if
@@ -81,8 +79,8 @@ static unsigned int contains(mmfile_t *mf, regex_t *regexp, kwset_t kws)
79
regmatch_t regmatch;
80
int flags = 0;
81
84
- assert(data[sz] == '\0');
85
- while (*data && !regexec(regexp, data, 1, ®match, flags)) {
82
+ while (*data &&
83
+ !regexec_buf(regexp, data, sz, 1, ®match, flags)) {
84
flags |= REG_NOTBOL;
85
data += regmatch.rm_eo;
86
if (*data && regmatch.rm_so == regmatch.rm_eo)
grep.c
+2
-12
@@ -848,17 +848,6 @@ static int fixmatch(struct grep_pat *p, char *line, char *eol,
848
}
849
}
850
851
-static int regmatch(const regex_t *preg, char *line, char *eol,
852
- regmatch_t *match, int eflags)
853
-{
854
-#ifdef REG_STARTEND
855
- match->rm_so = 0;
856
- match->rm_eo = eol - line;
857
- eflags |= REG_STARTEND;
858
-#endif
859
- return regexec(preg, line, 1, match, eflags);
860
-}
861
-
851
static int patmatch(struct grep_pat *p, char *line, char *eol,
852
regmatch_t *match, int eflags)
853
{
@@ -869,7 +858,8 @@ static int patmatch(struct grep_pat *p, char *line, char *eol,
858
else if (p->pcre_regexp)
859
hit = !pcrematch(p, line, eol, match, eflags);
860
else
872
- hit = !regmatch(&p->regexp, line, eol, match, eflags);
861
+ hit = !regexec_buf(&p->regexp, line, eol - line, 1, match,
862
+ eflags);
863
864
return hit;
865
}
t/t4062-diff-pickaxe.sh
+1
-1
@@ -14,7 +14,7 @@ test_expect_success setup '
14
test_tick &&
15
git commit -m "A 4k file"
16
'
17
-test_expect_failure '-G matches' '
17
+test_expect_success '-G matches' '
18
git diff --name-only -G "^0{4096}$" HEAD^ >out &&
19
test 4096-zeroes.txt = "$(cat out)"
20
'
xdiff-interface.c
+4
-9
@@ -216,11 +216,10 @@ struct ff_regs {
216
static long ff_regexp(const char *line, long len,
217
char *buffer, long buffer_size, void *priv)
218
{
219
- char *line_buffer;
219
struct ff_regs *regs = priv;
220
regmatch_t pmatch[2];
221
int i;
223
- int result = -1;
222
+ int result;
223
224
/* Exclude terminating newline (and cr) from matching */
225
if (len > 0 && line[len-1] == '\n') {
@@ -230,18 +229,16 @@ static long ff_regexp(const char *line, long len,
229
len--;
230
}
231
233
- line_buffer = xstrndup(line, len); /* make NUL terminated */
234
-
232
for (i = 0; i < regs->nr; i++) {
233
struct ff_reg *reg = regs->array + i;
237
- if (!regexec(®->re, line_buffer, 2, pmatch, 0)) {
234
+ if (!regexec_buf(®->re, line, len, 2, pmatch, 0)) {
235
if (reg->negate)
239
- goto fail;
236
+ return -1;
237
break;
238
}
239
}
240
if (regs->nr <= i)
244
- goto fail;
241
+ return -1;
242
i = pmatch[1].rm_so >= 0 ? 1 : 0;
243
line += pmatch[i].rm_so;
244
result = pmatch[i].rm_eo - pmatch[i].rm_so;
@@ -250,8 +247,6 @@ static long ff_regexp(const char *line, long len,
247
while (result > 0 && (isspace(line[result - 1])))
248
result--;
249
memcpy(buffer, line, result);
253
- fail:
254
- free(line_buffer);
250
return result;
251
}
252