![[tokkee]](http://tokkee.org/images/avatar.png){kind=avatar}
diff --git a/builtin-grep.c b/builtin-grep.c
index eb821b41b95ad128d12e2362f9048ff810fa225e..c89ee33a1ede4e032cae850315ee2b2185a64457 100644 (file)
--- a/builtin-grep.c
+++ b/builtin-grep.c
for (i = 0; paths[i]; i++) {
const char *match = paths[i];
int matchlen = strlen(match);
- const char *slash, *cp;
+ const char *cp, *meta;
if ((matchlen <= namelen) &&
!strncmp(name, match, matchlen) &&
if (name[namelen-1] != '/')
continue;
- /* We are being asked if the name directory is worth
+ /* We are being asked if the directory ("name") is worth
* descending into.
*
* Find the longest leading directory name that does
* not have metacharacter in the pathspec; the name
* we are looking at must overlap with that directory.
*/
- for (cp = match, slash = NULL; cp - match < matchlen; cp++) {
+ for (cp = match, meta = NULL; cp - match < matchlen; cp++) {
char ch = *cp;
- if (ch == '/')
- slash = cp;
- if (ch == '*' || ch == '[')
+ if (ch == '*' || ch == '[' || ch == '?') {
+ meta = cp;
break;
+ }
}
- if (!slash)
- slash = match; /* toplevel */
- else
- slash++;
- if (namelen <= slash - match) {
+ if (!meta)
+ meta = cp; /* fully literal */
+
+ if (namelen <= meta - match) {
/* Looking at "Documentation/" and
* the pattern says "Documentation/howto/", or
- * "Documentation/diff*.txt".
+ * "Documentation/diff*.txt". The name we
+ * have should match prefix.
*/
if (!memcmp(match, name, namelen))
return 1;
+ continue;
}
- else {
+
+ if (meta - match < namelen) {
/* Looking at "Documentation/howto/" and
- * the pattern says "Documentation/h*".
+ * the pattern says "Documentation/h*";
+ * match up to "Do.../h"; this avoids descending
+ * into "Documentation/technical/".
*/
- if (!memcmp(match, name, slash - match))
+ if (!memcmp(match, name, meta - match))
return 1;
+ continue;
}
}
return 0;
}
-struct grep_opt {
+struct grep_pat {
+ struct grep_pat *next;
const char *pattern;
regex_t regexp;
+};
+
+struct grep_opt {
+ struct grep_pat *pattern_list;
+ struct grep_pat **pattern_tail;
+ regex_t regexp;
unsigned linenum:1;
unsigned invert:1;
unsigned name_only:1;
+ unsigned unmatch_name_only:1;
+ unsigned count:1;
+ unsigned word_regexp:1;
+#define GREP_BINARY_DEFAULT 0
+#define GREP_BINARY_NOMATCH 1
+#define GREP_BINARY_TEXT 2
+ unsigned binary:2;
int regflags;
unsigned pre_context;
unsigned post_context;
};
+static void add_pattern(struct grep_opt *opt, const char *pat)
+{
+ struct grep_pat *p = xcalloc(1, sizeof(*p));
+ p->pattern = pat;
+ *opt->pattern_tail = p;
+ opt->pattern_tail = &p->next;
+ p->next = NULL;
+}
+
+static void compile_patterns(struct grep_opt *opt)
+{
+ struct grep_pat *p;
+ for (p = opt->pattern_list; p; p = p->next) {
+ int err = regcomp(&p->regexp, p->pattern, opt->regflags);
+ if (err) {
+ char errbuf[1024];
+ regerror(err, &p->regexp, errbuf, 1024);
+ regfree(&p->regexp);
+ die("'%s': %s", p->pattern, errbuf);
+ }
+ }
+}
+
static char *end_of_line(char *cp, unsigned long *left)
{
unsigned long l = *left;
return cp;
}
+static int word_char(char ch)
+{
+ return isalnum(ch) || ch == '_';
+}
+
static void show_line(struct grep_opt *opt, const char *bol, const char *eol,
const char *name, unsigned lno, char sign)
{
printf("%.*s\n", (int)(eol-bol), bol);
}
+/*
+ * NEEDSWORK: share code with diff.c
+ */
+#define FIRST_FEW_BYTES 8000
+static int buffer_is_binary(const char *ptr, unsigned long size)
+{
+ if (FIRST_FEW_BYTES < size)
+ size = FIRST_FEW_BYTES;
+ if (memchr(ptr, 0, size))
+ return 1;
+ return 0;
+}
+
static int grep_buffer(struct grep_opt *opt, const char *name,
char *buf, unsigned long size)
{
} *prev = NULL, *pcl;
unsigned last_hit = 0;
unsigned last_shown = 0;
+ int binary_match_only = 0;
const char *hunk_mark = "";
+ unsigned count = 0;
+
+ if (buffer_is_binary(buf, size)) {
+ switch (opt->binary) {
+ case GREP_BINARY_DEFAULT:
+ binary_match_only = 1;
+ break;
+ case GREP_BINARY_NOMATCH:
+ return 0; /* Assume unmatch */
+ break;
+ default:
+ break;
+ }
+ }
if (opt->pre_context)
prev = xcalloc(opt->pre_context, sizeof(*prev));
while (left) {
regmatch_t pmatch[10];
char *eol, ch;
- int hit;
+ int hit = 0;
+ struct grep_pat *p;
eol = end_of_line(bol, &left);
ch = *eol;
*eol = 0;
- hit = !regexec(&opt->regexp, bol, ARRAY_SIZE(pmatch),
- pmatch, 0);
+ for (p = opt->pattern_list; p; p = p->next) {
+ regex_t *exp = &p->regexp;
+ hit = !regexec(exp, bol, ARRAY_SIZE(pmatch),
+ pmatch, 0);
+
+ if (hit && opt->word_regexp) {
+ /* Match beginning must be either
+ * beginning of the line, or at word
+ * boundary (i.e. the last char must
+ * not be alnum or underscore).
+ */
+ if ((pmatch[0].rm_so < 0) ||
+ (eol - bol) <= pmatch[0].rm_so ||
+ (pmatch[0].rm_eo < 0) ||
+ (eol - bol) < pmatch[0].rm_eo)
+ die("regexp returned nonsense");
+ if (pmatch[0].rm_so != 0 &&
+ word_char(bol[pmatch[0].rm_so-1]))
+ continue; /* not a word boundary */
+ if ((eol-bol) < pmatch[0].rm_eo &&
+ word_char(bol[pmatch[0].rm_eo]))
+ continue; /* not a word boundary */
+ }
+ if (hit)
+ break;
+ }
+ /* "grep -v -e foo -e bla" should list lines
+ * that do not have either, so inversion should
+ * be done outside.
+ */
if (opt->invert)
hit = !hit;
+ if (opt->unmatch_name_only) {
+ if (hit)
+ return 0;
+ goto next_line;
+ }
if (hit) {
+ count++;
+ if (binary_match_only) {
+ printf("Binary file %s matches\n", name);
+ return 1;
+ }
if (opt->name_only) {
printf("%s\n", name);
return 1;
}
/* Hit at this line. If we haven't shown the
* pre-context lines, we would need to show them.
+ * When asked to do "count", this still show
+ * the context which is nonsense, but the user
+ * deserves to get that ;-).
*/
if (opt->pre_context) {
unsigned from;
}
if (last_shown && lno != last_shown + 1)
printf(hunk_mark);
- show_line(opt, bol, eol, name, lno, ':');
+ if (!opt->count)
+ show_line(opt, bol, eol, name, lno, ':');
last_shown = last_hit = lno;
}
else if (last_hit &&
prev->bol = bol;
prev->eol = eol;
}
+
+ next_line:
*eol = ch;
bol = eol + 1;
+ if (!left)
+ break;
left--;
lno++;
}
+
+ if (opt->unmatch_name_only) {
+ /* We did not see any hit, so we want to show this */
+ printf("%s\n", name);
+ return 1;
+ }
+
+ /* NEEDSWORK:
+ * The real "grep -c foo *.c" gives many "bar.c:0" lines,
+ * which feels mostly useless but sometimes useful. Maybe
+ * make it another option? For now suppress them.
+ */
+ if (opt->count && count)
+ printf("%s:%u\n", name, count);
return !!last_hit;
}
int cmd_grep(int argc, const char **argv, char **envp)
{
- int err;
int hit = 0;
int no_more_flags = 0;
int seen_noncommit = 0;
const char **paths = NULL;
memset(&opt, 0, sizeof(opt));
+ opt.pattern_tail = &opt.pattern_list;
opt.regflags = REG_NEWLINE;
/*
cached = 1;
continue;
}
+ if (!strcmp("-a", arg) ||
+ !strcmp("--text", arg)) {
+ opt.binary = GREP_BINARY_TEXT;
+ continue;
+ }
if (!strcmp("-i", arg) ||
!strcmp("--ignore-case", arg)) {
opt.regflags |= REG_ICASE;
continue;
}
+ if (!strcmp("-I", arg)) {
+ opt.binary = GREP_BINARY_NOMATCH;
+ continue;
+ }
if (!strcmp("-v", arg) ||
!strcmp("--invert-match", arg)) {
opt.invert = 1;
opt.name_only = 1;
continue;
}
- if (!strcmp("-A", arg) ||
- !strcmp("-B", arg) ||
- !strcmp("-C", arg)) {
+ if (!strcmp("-L", arg) ||
+ !strcmp("--files-without-match", arg)) {
+ opt.unmatch_name_only = 1;
+ continue;
+ }
+ if (!strcmp("-c", arg) ||
+ !strcmp("--count", arg)) {
+ opt.count = 1;
+ continue;
+ }
+ if (!strcmp("-w", arg) ||
+ !strcmp("--word-regexp", arg)) {
+ opt.word_regexp = 1;
+ continue;
+ }
+ if (!strncmp("-A", arg, 2) ||
+ !strncmp("-B", arg, 2) ||
+ !strncmp("-C", arg, 2) ||
+ (arg[0] == '-' && '1' <= arg[1] && arg[1] <= '9')) {
unsigned num;
- if (argc <= 1 ||
- sscanf(*++argv, "%u", &num) != 1)
+ const char *scan;
+ switch (arg[1]) {
+ case 'A': case 'B': case 'C':
+ if (!arg[2]) {
+ if (argc <= 1)
+ usage(builtin_grep_usage);
+ scan = *++argv;
+ argc--;
+ }
+ else
+ scan = arg + 2;
+ break;
+ default:
+ scan = arg + 1;
+ break;
+ }
+ if (sscanf(scan, "%u", &num) != 1)
usage(builtin_grep_usage);
- argc--;
switch (arg[1]) {
case 'A':
opt.post_context = num;
break;
+ default:
case 'C':
opt.post_context = num;
case 'B':
}
if (!strcmp("-e", arg)) {
if (1 < argc) {
- /* We probably would want to do
- * -e pat1 -e pat2 as well later...
- */
- if (opt.pattern)
- die("more than one pattern?");
- opt.pattern = *++argv;
+ add_pattern(&opt, argv[1]);
+ argv++;
argc--;
continue;
}
/* Either unrecognized option or a single pattern */
if (!no_more_flags && *arg == '-')
usage(builtin_grep_usage);
- if (!opt.pattern) {
- opt.pattern = arg;
+ if (!opt.pattern_list) {
+ add_pattern(&opt, arg);
break;
}
else {
break;
}
}
- if (!opt.pattern)
+ if (!opt.pattern_list)
die("no pattern given.");
- err = regcomp(&opt.regexp, opt.pattern, opt.regflags);
- if (err) {
- char errbuf[1024];
- regerror(err, &opt.regexp, errbuf, 1024);
- regfree(&opt.regexp);
- die("'%s': %s", opt.pattern, errbuf);
- }
+ compile_patterns(&opt);
tail = &object_list;
while (1 < argc) {
struct object *object;