grep: refactor next_match() and match_one_pattern() for external use
These changes are made in preparation of, the colorization support for the "git log" subcommands that, rely on regex functionality (i.e. "--author", "--committer" and "--grep"). These changes are necessary primarily because match_one_pattern() expects header lines to be prefixed, however, in pretty, the prefixes are stripped from the lines because the name-email pairs need to go through additional parsing, before they can be printed and because next_match() doesn't handle the case of "ctx == GREP_CONTEXT_HEAD" at all. So, teach next_match() how to handle the new case and move match_one_pattern()'s core logic to headerless_match_one_pattern() while preserving match_one_pattern()'s uses that depend on the additional processing. Signed-off-by: Hamza Mahfooz <someguy@effective-light.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Hamza Mahfooz committed
Sep 29, 2021 at 07:57 UTC
3f566c4e695a6df8237c34b7c1f34f0832b7e575
2 files changed
+58
-30
grep.c
+49
-30
@@ -944,10 +944,10 @@ static struct {
944
{ "reflog ", 7 },
945
};
946
947
-static int match_one_pattern(struct grep_pat *p,
948
- const char *bol, const char *eol,
949
- enum grep_context ctx,
950
- regmatch_t *pmatch, int eflags)
947
+static int headerless_match_one_pattern(struct grep_pat *p,
948
+ const char *bol, const char *eol,
949
+ enum grep_context ctx,
950
+ regmatch_t *pmatch, int eflags)
951
{
952
int hit = 0;
953
const char *start = bol;
@@ -956,25 +956,6 @@ static int match_one_pattern(struct grep_pat *p,
956
((p->token == GREP_PATTERN_HEAD) != (ctx == GREP_CONTEXT_HEAD)))
957
return 0;
958
959
- if (p->token == GREP_PATTERN_HEAD) {
960
- const char *field;
961
- size_t len;
962
- assert(p->field < ARRAY_SIZE(header_field));
963
- field = header_field[p->field].field;
964
- len = header_field[p->field].len;
965
- if (strncmp(bol, field, len))
966
- return 0;
967
- bol += len;
968
- switch (p->field) {
969
- case GREP_HEADER_AUTHOR:
970
- case GREP_HEADER_COMMITTER:
971
- strip_timestamp(bol, &eol);
972
- break;
973
- default:
974
- break;
975
- }
976
- }
977
-
959
again:
960
hit = patmatch(p, bol, eol, pmatch, eflags);
961
@@ -1025,6 +1006,36 @@ static int match_one_pattern(struct grep_pat *p,
1006
return hit;
1007
}
1008
1009
+static int match_one_pattern(struct grep_pat *p,
1010
+ const char *bol, const char *eol,
1011
+ enum grep_context ctx, regmatch_t *pmatch,
1012
+ int eflags)
1013
+{
1014
+ const char *field;
1015
+ size_t len;
1016
+
1017
+ if (p->token == GREP_PATTERN_HEAD) {
1018
+ assert(p->field < ARRAY_SIZE(header_field));
1019
+ field = header_field[p->field].field;
1020
+ len = header_field[p->field].len;
1021
+ if (strncmp(bol, field, len))
1022
+ return 0;
1023
+ bol += len;
1024
+
1025
+ switch (p->field) {
1026
+ case GREP_HEADER_AUTHOR:
1027
+ case GREP_HEADER_COMMITTER:
1028
+ strip_timestamp(bol, &eol);
1029
+ break;
1030
+ default:
1031
+ break;
1032
+ }
1033
+ }
1034
+
1035
+ return headerless_match_one_pattern(p, bol, eol, ctx, pmatch, eflags);
1036
+}
1037
+
1038
+
1039
static int match_expr_eval(struct grep_opt *opt, struct grep_expr *x,
1040
const char *bol, const char *eol,
1041
enum grep_context ctx, ssize_t *col,
@@ -1143,7 +1154,7 @@ static int match_next_pattern(struct grep_pat *p,
1154
{
1155
regmatch_t match;
1156
1146
- if (!match_one_pattern(p, bol, eol, ctx, &match, eflags))
1157
+ if (!headerless_match_one_pattern(p, bol, eol, ctx, &match, eflags))
1158
return 0;
1159
if (match.rm_so < 0 || match.rm_eo < 0)
1160
return 0;
@@ -1158,19 +1169,26 @@ static int match_next_pattern(struct grep_pat *p,
1169
return 1;
1170
}
1171
1161
-static int next_match(struct grep_opt *opt,
1162
- const char *bol, const char *eol,
1163
- enum grep_context ctx, regmatch_t *pmatch, int eflags)
1172
+int grep_next_match(struct grep_opt *opt,
1173
+ const char *bol, const char *eol,
1174
+ enum grep_context ctx, regmatch_t *pmatch,
1175
+ enum grep_header_field field, int eflags)
1176
{
1177
struct grep_pat *p;
1178
int hit = 0;
1179
1180
pmatch->rm_so = pmatch->rm_eo = -1;
1181
if (bol < eol) {
1170
- for (p = opt->pattern_list; p; p = p->next) {
1182
+ for (p = ((ctx == GREP_CONTEXT_HEAD)
1183
+ ? opt->header_list : opt->pattern_list);
1184
+ p; p = p->next) {
1185
switch (p->token) {
1172
- case GREP_PATTERN: /* atom */
1186
case GREP_PATTERN_HEAD:
1187
+ if ((field != GREP_HEADER_FIELD_MAX) &&
1188
+ (p->field != field))
1189
+ continue;
1190
+ /* fall thru */
1191
+ case GREP_PATTERN: /* atom */
1192
case GREP_PATTERN_BODY:
1193
hit |= match_next_pattern(p, bol, eol, ctx,
1194
pmatch, eflags);
@@ -1261,7 +1279,8 @@ static void show_line(struct grep_opt *opt,
1279
else if (sign == '=')
1280
line_color = opt->colors[GREP_COLOR_FUNCTION];
1281
}
1264
- while (next_match(opt, bol, eol, ctx, &match, eflags)) {
1282
+ while (grep_next_match(opt, bol, eol, ctx, &match,
1283
+ GREP_HEADER_FIELD_MAX, eflags)) {
1284
if (match.rm_so == match.rm_eo)
1285
break;
1286
grep.h
+9
@@ -191,6 +191,15 @@ void compile_grep_patterns(struct grep_opt *opt);
191
void free_grep_patterns(struct grep_opt *opt);
192
int grep_buffer(struct grep_opt *opt, const char *buf, unsigned long size);
193
194
+/* The field parameter is only used to filter header patterns
195
+ * (where appropriate). If filtering isn't desirable
196
+ * GREP_HEADER_FIELD_MAX should be supplied.
197
+ */
198
+int grep_next_match(struct grep_opt *opt,
199
+ const char *bol, const char *eol,
200
+ enum grep_context ctx, regmatch_t *pmatch,
201
+ enum grep_header_field field, int eflags);
202
+
203
struct grep_source {
204
char *name;
205