2006-09-18 03:02:52 +04:00
|
|
|
#include "cache.h"
|
|
|
|
#include "grep.h"
|
2007-06-05 06:36:11 +04:00
|
|
|
#include "xdiff-interface.h"
|
2006-09-18 03:02:52 +04:00
|
|
|
|
log --author/--committer: really match only with name part
When we tried to find commits done by AUTHOR, the first implementation
tried to pattern match a line with "^author .*AUTHOR", which later was
enhanced to strip leading caret and look for "^author AUTHOR" when the
search pattern was anchored at the left end (i.e. --author="^AUTHOR").
This had a few problems:
* When looking for fixed strings (e.g. "git log -F --author=x --grep=y"),
the regexp internally used "^author .*x" would never match anything;
* To match at the end (e.g. "git log --author='google.com>$'"), the
generated regexp has to also match the trailing timestamp part the
commit header lines have. Also, in order to determine if the '$' at
the end means "match at the end of the line" or just a literal dollar
sign (probably backslash-quoted), we would need to parse the regexp
ourselves.
An earlier alternative tried to make sure that a line matches "^author "
(to limit by field name) and the user supplied pattern at the same time.
While it solved the -F problem by introducing a special override for
matching the "^author ", it did not solve the trailing timestamp nor tail
match problem. It also would have matched every commit if --author=author
was asked for, not because the author's email part had this string, but
because every commit header line that talks about the author begins with
that field name, regardleses of who wrote it.
Instead of piling more hacks on top of hacks, this rethinks the grep
machinery that is used to look for strings in the commit header, and makes
sure that (1) field name matches literally at the beginning of the line,
followed by a SP, and (2) the user supplied pattern is matched against the
remainder of the line, excluding the trailing timestamp data.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-09-05 09:15:02 +04:00
|
|
|
void append_header_grep_pattern(struct grep_opt *opt, enum grep_header_field field, const char *pat)
|
|
|
|
{
|
|
|
|
struct grep_pat *p = xcalloc(1, sizeof(*p));
|
|
|
|
p->pattern = pat;
|
|
|
|
p->origin = "header";
|
|
|
|
p->no = 0;
|
|
|
|
p->token = GREP_PATTERN_HEAD;
|
|
|
|
p->field = field;
|
|
|
|
*opt->pattern_tail = p;
|
|
|
|
opt->pattern_tail = &p->next;
|
|
|
|
p->next = NULL;
|
|
|
|
}
|
|
|
|
|
2006-09-18 03:02:52 +04:00
|
|
|
void append_grep_pattern(struct grep_opt *opt, const char *pat,
|
|
|
|
const char *origin, int no, enum grep_pat_token t)
|
|
|
|
{
|
|
|
|
struct grep_pat *p = xcalloc(1, sizeof(*p));
|
|
|
|
p->pattern = pat;
|
|
|
|
p->origin = origin;
|
|
|
|
p->no = no;
|
|
|
|
p->token = t;
|
|
|
|
*opt->pattern_tail = p;
|
|
|
|
opt->pattern_tail = &p->next;
|
|
|
|
p->next = NULL;
|
|
|
|
}
|
|
|
|
|
2009-01-10 02:18:34 +03:00
|
|
|
static int isregexspecial(int c)
|
|
|
|
{
|
|
|
|
return isspecial(c) || c == '$' || c == '(' || c == ')' || c == '+' ||
|
|
|
|
c == '.' || c == '^' || c == '{' || c == '|';
|
|
|
|
}
|
|
|
|
|
|
|
|
static int is_fixed(const char *s)
|
|
|
|
{
|
|
|
|
while (!isregexspecial(*s))
|
|
|
|
s++;
|
|
|
|
return !*s;
|
|
|
|
}
|
|
|
|
|
2006-09-18 03:02:52 +04:00
|
|
|
static void compile_regexp(struct grep_pat *p, struct grep_opt *opt)
|
|
|
|
{
|
2009-01-10 02:18:34 +03:00
|
|
|
int err;
|
|
|
|
|
|
|
|
if (opt->fixed || is_fixed(p->pattern))
|
|
|
|
p->fixed = 1;
|
|
|
|
if (opt->regflags & REG_ICASE)
|
|
|
|
p->fixed = 0;
|
|
|
|
if (p->fixed)
|
|
|
|
return;
|
|
|
|
|
|
|
|
err = regcomp(&p->regexp, p->pattern, opt->regflags);
|
2006-09-18 03:02:52 +04:00
|
|
|
if (err) {
|
|
|
|
char errbuf[1024];
|
|
|
|
char where[1024];
|
|
|
|
if (p->no)
|
|
|
|
sprintf(where, "In '%s' at %d, ",
|
|
|
|
p->origin, p->no);
|
|
|
|
else if (p->origin)
|
|
|
|
sprintf(where, "%s, ", p->origin);
|
|
|
|
else
|
|
|
|
where[0] = 0;
|
|
|
|
regerror(err, &p->regexp, errbuf, 1024);
|
|
|
|
regfree(&p->regexp);
|
|
|
|
die("%s'%s': %s", where, p->pattern, errbuf);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2006-09-28 04:50:52 +04:00
|
|
|
static struct grep_expr *compile_pattern_or(struct grep_pat **);
|
2006-09-18 03:02:52 +04:00
|
|
|
static struct grep_expr *compile_pattern_atom(struct grep_pat **list)
|
|
|
|
{
|
|
|
|
struct grep_pat *p;
|
|
|
|
struct grep_expr *x;
|
|
|
|
|
|
|
|
p = *list;
|
|
|
|
switch (p->token) {
|
|
|
|
case GREP_PATTERN: /* atom */
|
2006-09-20 23:39:46 +04:00
|
|
|
case GREP_PATTERN_HEAD:
|
|
|
|
case GREP_PATTERN_BODY:
|
2006-09-18 03:02:52 +04:00
|
|
|
x = xcalloc(1, sizeof (struct grep_expr));
|
|
|
|
x->node = GREP_NODE_ATOM;
|
|
|
|
x->u.atom = p;
|
|
|
|
*list = p->next;
|
|
|
|
return x;
|
|
|
|
case GREP_OPEN_PAREN:
|
|
|
|
*list = p->next;
|
2006-09-28 04:50:52 +04:00
|
|
|
x = compile_pattern_or(list);
|
2006-09-18 03:02:52 +04:00
|
|
|
if (!x)
|
|
|
|
return NULL;
|
|
|
|
if (!*list || (*list)->token != GREP_CLOSE_PAREN)
|
|
|
|
die("unmatched parenthesis");
|
|
|
|
*list = (*list)->next;
|
|
|
|
return x;
|
|
|
|
default:
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct grep_expr *compile_pattern_not(struct grep_pat **list)
|
|
|
|
{
|
|
|
|
struct grep_pat *p;
|
|
|
|
struct grep_expr *x;
|
|
|
|
|
|
|
|
p = *list;
|
|
|
|
switch (p->token) {
|
|
|
|
case GREP_NOT:
|
|
|
|
if (!p->next)
|
|
|
|
die("--not not followed by pattern expression");
|
|
|
|
*list = p->next;
|
|
|
|
x = xcalloc(1, sizeof (struct grep_expr));
|
|
|
|
x->node = GREP_NODE_NOT;
|
|
|
|
x->u.unary = compile_pattern_not(list);
|
|
|
|
if (!x->u.unary)
|
|
|
|
die("--not followed by non pattern expression");
|
|
|
|
return x;
|
|
|
|
default:
|
|
|
|
return compile_pattern_atom(list);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct grep_expr *compile_pattern_and(struct grep_pat **list)
|
|
|
|
{
|
|
|
|
struct grep_pat *p;
|
|
|
|
struct grep_expr *x, *y, *z;
|
|
|
|
|
|
|
|
x = compile_pattern_not(list);
|
|
|
|
p = *list;
|
|
|
|
if (p && p->token == GREP_AND) {
|
|
|
|
if (!p->next)
|
|
|
|
die("--and not followed by pattern expression");
|
|
|
|
*list = p->next;
|
|
|
|
y = compile_pattern_and(list);
|
|
|
|
if (!y)
|
|
|
|
die("--and not followed by pattern expression");
|
|
|
|
z = xcalloc(1, sizeof (struct grep_expr));
|
|
|
|
z->node = GREP_NODE_AND;
|
|
|
|
z->u.binary.left = x;
|
|
|
|
z->u.binary.right = y;
|
|
|
|
return z;
|
|
|
|
}
|
|
|
|
return x;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct grep_expr *compile_pattern_or(struct grep_pat **list)
|
|
|
|
{
|
|
|
|
struct grep_pat *p;
|
|
|
|
struct grep_expr *x, *y, *z;
|
|
|
|
|
|
|
|
x = compile_pattern_and(list);
|
|
|
|
p = *list;
|
|
|
|
if (x && p && p->token != GREP_CLOSE_PAREN) {
|
|
|
|
y = compile_pattern_or(list);
|
|
|
|
if (!y)
|
|
|
|
die("not a pattern expression %s", p->pattern);
|
|
|
|
z = xcalloc(1, sizeof (struct grep_expr));
|
|
|
|
z->node = GREP_NODE_OR;
|
|
|
|
z->u.binary.left = x;
|
|
|
|
z->u.binary.right = y;
|
|
|
|
return z;
|
|
|
|
}
|
|
|
|
return x;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct grep_expr *compile_pattern_expr(struct grep_pat **list)
|
|
|
|
{
|
|
|
|
return compile_pattern_or(list);
|
|
|
|
}
|
|
|
|
|
|
|
|
void compile_grep_patterns(struct grep_opt *opt)
|
|
|
|
{
|
|
|
|
struct grep_pat *p;
|
|
|
|
|
2006-09-28 04:50:52 +04:00
|
|
|
if (opt->all_match)
|
|
|
|
opt->extended = 1;
|
|
|
|
|
2006-09-18 03:02:52 +04:00
|
|
|
for (p = opt->pattern_list; p; p = p->next) {
|
2006-09-20 23:39:46 +04:00
|
|
|
switch (p->token) {
|
|
|
|
case GREP_PATTERN: /* atom */
|
|
|
|
case GREP_PATTERN_HEAD:
|
|
|
|
case GREP_PATTERN_BODY:
|
2009-01-10 02:18:34 +03:00
|
|
|
compile_regexp(p, opt);
|
2006-09-20 23:39:46 +04:00
|
|
|
break;
|
|
|
|
default:
|
2006-09-18 03:02:52 +04:00
|
|
|
opt->extended = 1;
|
2006-09-20 23:39:46 +04:00
|
|
|
break;
|
|
|
|
}
|
2006-09-18 03:02:52 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
if (!opt->extended)
|
|
|
|
return;
|
|
|
|
|
|
|
|
/* Then bundle them up in an expression.
|
|
|
|
* A classic recursive descent parser would do.
|
|
|
|
*/
|
|
|
|
p = opt->pattern_list;
|
|
|
|
opt->pattern_expression = compile_pattern_expr(&p);
|
|
|
|
if (p)
|
|
|
|
die("incomplete pattern expression: %s", p->pattern);
|
|
|
|
}
|
|
|
|
|
2006-09-28 03:27:10 +04:00
|
|
|
static void free_pattern_expr(struct grep_expr *x)
|
|
|
|
{
|
|
|
|
switch (x->node) {
|
|
|
|
case GREP_NODE_ATOM:
|
|
|
|
break;
|
|
|
|
case GREP_NODE_NOT:
|
|
|
|
free_pattern_expr(x->u.unary);
|
|
|
|
break;
|
|
|
|
case GREP_NODE_AND:
|
|
|
|
case GREP_NODE_OR:
|
|
|
|
free_pattern_expr(x->u.binary.left);
|
|
|
|
free_pattern_expr(x->u.binary.right);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
free(x);
|
|
|
|
}
|
|
|
|
|
|
|
|
void free_grep_patterns(struct grep_opt *opt)
|
|
|
|
{
|
|
|
|
struct grep_pat *p, *n;
|
|
|
|
|
|
|
|
for (p = opt->pattern_list; p; p = n) {
|
|
|
|
n = p->next;
|
|
|
|
switch (p->token) {
|
|
|
|
case GREP_PATTERN: /* atom */
|
|
|
|
case GREP_PATTERN_HEAD:
|
|
|
|
case GREP_PATTERN_BODY:
|
|
|
|
regfree(&p->regexp);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
free(p);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!opt->extended)
|
|
|
|
return;
|
|
|
|
free_pattern_expr(opt->pattern_expression);
|
|
|
|
}
|
|
|
|
|
2006-09-18 03:02:52 +04:00
|
|
|
static char *end_of_line(char *cp, unsigned long *left)
|
|
|
|
{
|
|
|
|
unsigned long l = *left;
|
|
|
|
while (l && *cp != '\n') {
|
|
|
|
l--;
|
|
|
|
cp++;
|
|
|
|
}
|
|
|
|
*left = l;
|
|
|
|
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)
|
|
|
|
{
|
2008-10-01 20:11:15 +04:00
|
|
|
if (opt->null_following_name)
|
|
|
|
sign = '\0';
|
2006-09-18 03:02:52 +04:00
|
|
|
if (opt->pathname)
|
|
|
|
printf("%s%c", name, sign);
|
|
|
|
if (opt->linenum)
|
|
|
|
printf("%d%c", lno, sign);
|
|
|
|
printf("%.*s\n", (int)(eol-bol), bol);
|
|
|
|
}
|
|
|
|
|
2008-10-01 20:11:15 +04:00
|
|
|
static void show_name(struct grep_opt *opt, const char *name)
|
|
|
|
{
|
|
|
|
printf("%s%c", name, opt->null_following_name ? '\0' : '\n');
|
|
|
|
}
|
|
|
|
|
2006-09-18 03:02:52 +04:00
|
|
|
static int fixmatch(const char *pattern, char *line, regmatch_t *match)
|
|
|
|
{
|
|
|
|
char *hit = strstr(line, pattern);
|
|
|
|
if (!hit) {
|
|
|
|
match->rm_so = match->rm_eo = -1;
|
|
|
|
return REG_NOMATCH;
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
match->rm_so = hit - line;
|
|
|
|
match->rm_eo = match->rm_so + strlen(pattern);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
log --author/--committer: really match only with name part
When we tried to find commits done by AUTHOR, the first implementation
tried to pattern match a line with "^author .*AUTHOR", which later was
enhanced to strip leading caret and look for "^author AUTHOR" when the
search pattern was anchored at the left end (i.e. --author="^AUTHOR").
This had a few problems:
* When looking for fixed strings (e.g. "git log -F --author=x --grep=y"),
the regexp internally used "^author .*x" would never match anything;
* To match at the end (e.g. "git log --author='google.com>$'"), the
generated regexp has to also match the trailing timestamp part the
commit header lines have. Also, in order to determine if the '$' at
the end means "match at the end of the line" or just a literal dollar
sign (probably backslash-quoted), we would need to parse the regexp
ourselves.
An earlier alternative tried to make sure that a line matches "^author "
(to limit by field name) and the user supplied pattern at the same time.
While it solved the -F problem by introducing a special override for
matching the "^author ", it did not solve the trailing timestamp nor tail
match problem. It also would have matched every commit if --author=author
was asked for, not because the author's email part had this string, but
because every commit header line that talks about the author begins with
that field name, regardleses of who wrote it.
Instead of piling more hacks on top of hacks, this rethinks the grep
machinery that is used to look for strings in the commit header, and makes
sure that (1) field name matches literally at the beginning of the line,
followed by a SP, and (2) the user supplied pattern is matched against the
remainder of the line, excluding the trailing timestamp data.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-09-05 09:15:02 +04:00
|
|
|
static int strip_timestamp(char *bol, char **eol_p)
|
|
|
|
{
|
|
|
|
char *eol = *eol_p;
|
|
|
|
int ch;
|
|
|
|
|
|
|
|
while (bol < --eol) {
|
|
|
|
if (*eol != '>')
|
|
|
|
continue;
|
|
|
|
*eol_p = ++eol;
|
|
|
|
ch = *eol;
|
|
|
|
*eol = '\0';
|
|
|
|
return ch;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct {
|
|
|
|
const char *field;
|
|
|
|
size_t len;
|
|
|
|
} header_field[] = {
|
|
|
|
{ "author ", 7 },
|
|
|
|
{ "committer ", 10 },
|
|
|
|
};
|
|
|
|
|
2006-09-20 23:39:46 +04:00
|
|
|
static int match_one_pattern(struct grep_opt *opt, struct grep_pat *p, char *bol, char *eol, enum grep_context ctx)
|
2006-09-18 03:02:52 +04:00
|
|
|
{
|
|
|
|
int hit = 0;
|
log --author/--committer: really match only with name part
When we tried to find commits done by AUTHOR, the first implementation
tried to pattern match a line with "^author .*AUTHOR", which later was
enhanced to strip leading caret and look for "^author AUTHOR" when the
search pattern was anchored at the left end (i.e. --author="^AUTHOR").
This had a few problems:
* When looking for fixed strings (e.g. "git log -F --author=x --grep=y"),
the regexp internally used "^author .*x" would never match anything;
* To match at the end (e.g. "git log --author='google.com>$'"), the
generated regexp has to also match the trailing timestamp part the
commit header lines have. Also, in order to determine if the '$' at
the end means "match at the end of the line" or just a literal dollar
sign (probably backslash-quoted), we would need to parse the regexp
ourselves.
An earlier alternative tried to make sure that a line matches "^author "
(to limit by field name) and the user supplied pattern at the same time.
While it solved the -F problem by introducing a special override for
matching the "^author ", it did not solve the trailing timestamp nor tail
match problem. It also would have matched every commit if --author=author
was asked for, not because the author's email part had this string, but
because every commit header line that talks about the author begins with
that field name, regardleses of who wrote it.
Instead of piling more hacks on top of hacks, this rethinks the grep
machinery that is used to look for strings in the commit header, and makes
sure that (1) field name matches literally at the beginning of the line,
followed by a SP, and (2) the user supplied pattern is matched against the
remainder of the line, excluding the trailing timestamp data.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-09-05 09:15:02 +04:00
|
|
|
int saved_ch = 0;
|
2006-09-18 03:02:52 +04:00
|
|
|
regmatch_t pmatch[10];
|
|
|
|
|
2006-09-20 23:39:46 +04:00
|
|
|
if ((p->token != GREP_PATTERN) &&
|
|
|
|
((p->token == GREP_PATTERN_HEAD) != (ctx == GREP_CONTEXT_HEAD)))
|
|
|
|
return 0;
|
|
|
|
|
log --author/--committer: really match only with name part
When we tried to find commits done by AUTHOR, the first implementation
tried to pattern match a line with "^author .*AUTHOR", which later was
enhanced to strip leading caret and look for "^author AUTHOR" when the
search pattern was anchored at the left end (i.e. --author="^AUTHOR").
This had a few problems:
* When looking for fixed strings (e.g. "git log -F --author=x --grep=y"),
the regexp internally used "^author .*x" would never match anything;
* To match at the end (e.g. "git log --author='google.com>$'"), the
generated regexp has to also match the trailing timestamp part the
commit header lines have. Also, in order to determine if the '$' at
the end means "match at the end of the line" or just a literal dollar
sign (probably backslash-quoted), we would need to parse the regexp
ourselves.
An earlier alternative tried to make sure that a line matches "^author "
(to limit by field name) and the user supplied pattern at the same time.
While it solved the -F problem by introducing a special override for
matching the "^author ", it did not solve the trailing timestamp nor tail
match problem. It also would have matched every commit if --author=author
was asked for, not because the author's email part had this string, but
because every commit header line that talks about the author begins with
that field name, regardleses of who wrote it.
Instead of piling more hacks on top of hacks, this rethinks the grep
machinery that is used to look for strings in the commit header, and makes
sure that (1) field name matches literally at the beginning of the line,
followed by a SP, and (2) the user supplied pattern is matched against the
remainder of the line, excluding the trailing timestamp data.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-09-05 09:15:02 +04:00
|
|
|
if (p->token == GREP_PATTERN_HEAD) {
|
|
|
|
const char *field;
|
|
|
|
size_t len;
|
|
|
|
assert(p->field < ARRAY_SIZE(header_field));
|
|
|
|
field = header_field[p->field].field;
|
|
|
|
len = header_field[p->field].len;
|
|
|
|
if (strncmp(bol, field, len))
|
|
|
|
return 0;
|
|
|
|
bol += len;
|
|
|
|
saved_ch = strip_timestamp(bol, &eol);
|
|
|
|
}
|
|
|
|
|
2006-09-18 03:02:52 +04:00
|
|
|
again:
|
2009-01-10 02:18:34 +03:00
|
|
|
if (!p->fixed) {
|
2006-09-18 03:02:52 +04:00
|
|
|
regex_t *exp = &p->regexp;
|
|
|
|
hit = !regexec(exp, bol, ARRAY_SIZE(pmatch),
|
|
|
|
pmatch, 0);
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
hit = !fixmatch(p->pattern, bol, pmatch);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (hit && opt->word_regexp) {
|
|
|
|
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");
|
|
|
|
|
|
|
|
/* Match beginning must be either beginning of the
|
|
|
|
* line, or at word boundary (i.e. the last char must
|
|
|
|
* not be a word char). Similarly, match end must be
|
|
|
|
* either end of the line, or at word boundary
|
|
|
|
* (i.e. the next char must not be a word char).
|
|
|
|
*/
|
2009-01-10 02:08:40 +03:00
|
|
|
if ( ((pmatch[0].rm_so == 0) ||
|
2006-09-18 03:02:52 +04:00
|
|
|
!word_char(bol[pmatch[0].rm_so-1])) &&
|
|
|
|
((pmatch[0].rm_eo == (eol-bol)) ||
|
|
|
|
!word_char(bol[pmatch[0].rm_eo])) )
|
|
|
|
;
|
|
|
|
else
|
|
|
|
hit = 0;
|
|
|
|
|
|
|
|
if (!hit && pmatch[0].rm_so + bol + 1 < eol) {
|
|
|
|
/* There could be more than one match on the
|
|
|
|
* line, and the first match might not be
|
|
|
|
* strict word match. But later ones could be!
|
2009-01-10 02:08:40 +03:00
|
|
|
* Forward to the next possible start, i.e. the
|
|
|
|
* next position following a non-word char.
|
2006-09-18 03:02:52 +04:00
|
|
|
*/
|
|
|
|
bol = pmatch[0].rm_so + bol + 1;
|
2009-01-10 02:08:40 +03:00
|
|
|
while (word_char(bol[-1]) && bol < eol)
|
|
|
|
bol++;
|
|
|
|
if (bol < eol)
|
|
|
|
goto again;
|
2006-09-18 03:02:52 +04:00
|
|
|
}
|
|
|
|
}
|
log --author/--committer: really match only with name part
When we tried to find commits done by AUTHOR, the first implementation
tried to pattern match a line with "^author .*AUTHOR", which later was
enhanced to strip leading caret and look for "^author AUTHOR" when the
search pattern was anchored at the left end (i.e. --author="^AUTHOR").
This had a few problems:
* When looking for fixed strings (e.g. "git log -F --author=x --grep=y"),
the regexp internally used "^author .*x" would never match anything;
* To match at the end (e.g. "git log --author='google.com>$'"), the
generated regexp has to also match the trailing timestamp part the
commit header lines have. Also, in order to determine if the '$' at
the end means "match at the end of the line" or just a literal dollar
sign (probably backslash-quoted), we would need to parse the regexp
ourselves.
An earlier alternative tried to make sure that a line matches "^author "
(to limit by field name) and the user supplied pattern at the same time.
While it solved the -F problem by introducing a special override for
matching the "^author ", it did not solve the trailing timestamp nor tail
match problem. It also would have matched every commit if --author=author
was asked for, not because the author's email part had this string, but
because every commit header line that talks about the author begins with
that field name, regardleses of who wrote it.
Instead of piling more hacks on top of hacks, this rethinks the grep
machinery that is used to look for strings in the commit header, and makes
sure that (1) field name matches literally at the beginning of the line,
followed by a SP, and (2) the user supplied pattern is matched against the
remainder of the line, excluding the trailing timestamp data.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2008-09-05 09:15:02 +04:00
|
|
|
if (p->token == GREP_PATTERN_HEAD && saved_ch)
|
|
|
|
*eol = saved_ch;
|
2006-09-18 03:02:52 +04:00
|
|
|
return hit;
|
|
|
|
}
|
|
|
|
|
2006-09-28 04:50:52 +04:00
|
|
|
static int match_expr_eval(struct grep_opt *o,
|
2006-09-18 03:02:52 +04:00
|
|
|
struct grep_expr *x,
|
2006-09-20 23:39:46 +04:00
|
|
|
char *bol, char *eol,
|
2006-09-28 04:50:52 +04:00
|
|
|
enum grep_context ctx,
|
|
|
|
int collect_hits)
|
2006-09-18 03:02:52 +04:00
|
|
|
{
|
2006-09-28 04:50:52 +04:00
|
|
|
int h = 0;
|
|
|
|
|
2006-09-18 03:02:52 +04:00
|
|
|
switch (x->node) {
|
|
|
|
case GREP_NODE_ATOM:
|
2006-09-28 04:50:52 +04:00
|
|
|
h = match_one_pattern(o, x->u.atom, bol, eol, ctx);
|
2006-09-18 03:02:52 +04:00
|
|
|
break;
|
|
|
|
case GREP_NODE_NOT:
|
2006-09-28 04:50:52 +04:00
|
|
|
h = !match_expr_eval(o, x->u.unary, bol, eol, ctx, 0);
|
|
|
|
break;
|
2006-09-18 03:02:52 +04:00
|
|
|
case GREP_NODE_AND:
|
2006-09-28 04:50:52 +04:00
|
|
|
if (!collect_hits)
|
|
|
|
return (match_expr_eval(o, x->u.binary.left,
|
|
|
|
bol, eol, ctx, 0) &&
|
|
|
|
match_expr_eval(o, x->u.binary.right,
|
|
|
|
bol, eol, ctx, 0));
|
|
|
|
h = match_expr_eval(o, x->u.binary.left, bol, eol, ctx, 0);
|
|
|
|
h &= match_expr_eval(o, x->u.binary.right, bol, eol, ctx, 0);
|
|
|
|
break;
|
2006-09-18 03:02:52 +04:00
|
|
|
case GREP_NODE_OR:
|
2006-09-28 04:50:52 +04:00
|
|
|
if (!collect_hits)
|
|
|
|
return (match_expr_eval(o, x->u.binary.left,
|
|
|
|
bol, eol, ctx, 0) ||
|
|
|
|
match_expr_eval(o, x->u.binary.right,
|
|
|
|
bol, eol, ctx, 0));
|
|
|
|
h = match_expr_eval(o, x->u.binary.left, bol, eol, ctx, 0);
|
|
|
|
x->u.binary.left->hit |= h;
|
|
|
|
h |= match_expr_eval(o, x->u.binary.right, bol, eol, ctx, 1);
|
|
|
|
break;
|
|
|
|
default:
|
2009-01-04 21:38:41 +03:00
|
|
|
die("Unexpected node type (internal error) %d", x->node);
|
2006-09-18 03:02:52 +04:00
|
|
|
}
|
2006-09-28 04:50:52 +04:00
|
|
|
if (collect_hits)
|
|
|
|
x->hit |= h;
|
|
|
|
return h;
|
2006-09-18 03:02:52 +04:00
|
|
|
}
|
|
|
|
|
2006-09-20 23:39:46 +04:00
|
|
|
static int match_expr(struct grep_opt *opt, char *bol, char *eol,
|
2006-09-28 04:50:52 +04:00
|
|
|
enum grep_context ctx, int collect_hits)
|
2006-09-18 03:02:52 +04:00
|
|
|
{
|
|
|
|
struct grep_expr *x = opt->pattern_expression;
|
2006-09-28 04:50:52 +04:00
|
|
|
return match_expr_eval(opt, x, bol, eol, ctx, collect_hits);
|
2006-09-18 03:02:52 +04:00
|
|
|
}
|
|
|
|
|
2006-09-20 23:39:46 +04:00
|
|
|
static int match_line(struct grep_opt *opt, char *bol, char *eol,
|
2006-09-28 04:50:52 +04:00
|
|
|
enum grep_context ctx, int collect_hits)
|
2006-09-18 03:02:52 +04:00
|
|
|
{
|
|
|
|
struct grep_pat *p;
|
|
|
|
if (opt->extended)
|
2006-09-28 04:50:52 +04:00
|
|
|
return match_expr(opt, bol, eol, ctx, collect_hits);
|
|
|
|
|
|
|
|
/* we do not call with collect_hits without being extended */
|
2006-09-18 03:02:52 +04:00
|
|
|
for (p = opt->pattern_list; p; p = p->next) {
|
2006-09-20 23:39:46 +04:00
|
|
|
if (match_one_pattern(opt, p, bol, eol, ctx))
|
2006-09-18 03:02:52 +04:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2006-09-28 04:50:52 +04:00
|
|
|
static int grep_buffer_1(struct grep_opt *opt, const char *name,
|
|
|
|
char *buf, unsigned long size, int collect_hits)
|
2006-09-18 03:02:52 +04:00
|
|
|
{
|
|
|
|
char *bol = buf;
|
|
|
|
unsigned long left = size;
|
|
|
|
unsigned lno = 1;
|
|
|
|
struct pre_context_line {
|
|
|
|
char *bol;
|
|
|
|
char *eol;
|
|
|
|
} *prev = NULL, *pcl;
|
|
|
|
unsigned last_hit = 0;
|
|
|
|
unsigned last_shown = 0;
|
|
|
|
int binary_match_only = 0;
|
|
|
|
const char *hunk_mark = "";
|
|
|
|
unsigned count = 0;
|
2006-09-20 23:39:46 +04:00
|
|
|
enum grep_context ctx = GREP_CONTEXT_HEAD;
|
2006-09-18 03:02:52 +04:00
|
|
|
|
|
|
|
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));
|
|
|
|
if (opt->pre_context || opt->post_context)
|
|
|
|
hunk_mark = "--\n";
|
|
|
|
|
|
|
|
while (left) {
|
|
|
|
char *eol, ch;
|
2006-09-28 04:50:52 +04:00
|
|
|
int hit;
|
2006-09-18 03:02:52 +04:00
|
|
|
|
|
|
|
eol = end_of_line(bol, &left);
|
|
|
|
ch = *eol;
|
|
|
|
*eol = 0;
|
|
|
|
|
2006-09-20 23:39:46 +04:00
|
|
|
if ((ctx == GREP_CONTEXT_HEAD) && (eol == bol))
|
|
|
|
ctx = GREP_CONTEXT_BODY;
|
|
|
|
|
2006-09-28 04:50:52 +04:00
|
|
|
hit = match_line(opt, bol, eol, ctx, collect_hits);
|
2006-09-18 03:02:52 +04:00
|
|
|
*eol = ch;
|
|
|
|
|
2006-09-28 04:50:52 +04:00
|
|
|
if (collect_hits)
|
|
|
|
goto next_line;
|
|
|
|
|
2006-09-18 03:02:52 +04:00
|
|
|
/* "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 (opt->status_only)
|
|
|
|
return 1;
|
|
|
|
if (binary_match_only) {
|
|
|
|
printf("Binary file %s matches\n", name);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
if (opt->name_only) {
|
2008-10-01 20:11:15 +04:00
|
|
|
show_name(opt, name);
|
2006-09-18 03:02:52 +04:00
|
|
|
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 (opt->pre_context < lno)
|
|
|
|
from = lno - opt->pre_context;
|
|
|
|
else
|
|
|
|
from = 1;
|
|
|
|
if (from <= last_shown)
|
|
|
|
from = last_shown + 1;
|
|
|
|
if (last_shown && from != last_shown + 1)
|
2008-11-11 00:07:52 +03:00
|
|
|
fputs(hunk_mark, stdout);
|
2006-09-18 03:02:52 +04:00
|
|
|
while (from < lno) {
|
|
|
|
pcl = &prev[lno-from-1];
|
|
|
|
show_line(opt, pcl->bol, pcl->eol,
|
|
|
|
name, from, '-');
|
|
|
|
from++;
|
|
|
|
}
|
|
|
|
last_shown = lno-1;
|
|
|
|
}
|
|
|
|
if (last_shown && lno != last_shown + 1)
|
2008-11-11 00:07:52 +03:00
|
|
|
fputs(hunk_mark, stdout);
|
2006-09-18 03:02:52 +04:00
|
|
|
if (!opt->count)
|
|
|
|
show_line(opt, bol, eol, name, lno, ':');
|
|
|
|
last_shown = last_hit = lno;
|
|
|
|
}
|
|
|
|
else if (last_hit &&
|
|
|
|
lno <= last_hit + opt->post_context) {
|
|
|
|
/* If the last hit is within the post context,
|
|
|
|
* we need to show this line.
|
|
|
|
*/
|
|
|
|
if (last_shown && lno != last_shown + 1)
|
2008-11-11 00:07:52 +03:00
|
|
|
fputs(hunk_mark, stdout);
|
2006-09-18 03:02:52 +04:00
|
|
|
show_line(opt, bol, eol, name, lno, '-');
|
|
|
|
last_shown = lno;
|
|
|
|
}
|
|
|
|
if (opt->pre_context) {
|
|
|
|
memmove(prev+1, prev,
|
|
|
|
(opt->pre_context-1) * sizeof(*prev));
|
|
|
|
prev->bol = bol;
|
|
|
|
prev->eol = eol;
|
|
|
|
}
|
|
|
|
|
|
|
|
next_line:
|
|
|
|
bol = eol + 1;
|
|
|
|
if (!left)
|
|
|
|
break;
|
|
|
|
left--;
|
|
|
|
lno++;
|
|
|
|
}
|
|
|
|
|
2006-09-28 03:27:10 +04:00
|
|
|
free(prev);
|
2006-09-28 04:50:52 +04:00
|
|
|
if (collect_hits)
|
|
|
|
return 0;
|
2006-09-28 03:27:10 +04:00
|
|
|
|
2006-09-18 03:02:52 +04:00
|
|
|
if (opt->status_only)
|
|
|
|
return 0;
|
|
|
|
if (opt->unmatch_name_only) {
|
|
|
|
/* We did not see any hit, so we want to show this */
|
2008-10-01 20:11:15 +04:00
|
|
|
show_name(opt, name);
|
2006-09-18 03:02:52 +04:00
|
|
|
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)
|
2008-10-01 20:11:15 +04:00
|
|
|
printf("%s%c%u\n", name,
|
|
|
|
opt->null_following_name ? '\0' : ':', count);
|
2006-09-18 03:02:52 +04:00
|
|
|
return !!last_hit;
|
|
|
|
}
|
|
|
|
|
2006-09-28 04:50:52 +04:00
|
|
|
static void clr_hit_marker(struct grep_expr *x)
|
|
|
|
{
|
|
|
|
/* All-hit markers are meaningful only at the very top level
|
|
|
|
* OR node.
|
|
|
|
*/
|
|
|
|
while (1) {
|
|
|
|
x->hit = 0;
|
|
|
|
if (x->node != GREP_NODE_OR)
|
|
|
|
return;
|
|
|
|
x->u.binary.left->hit = 0;
|
|
|
|
x = x->u.binary.right;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static int chk_hit_marker(struct grep_expr *x)
|
|
|
|
{
|
|
|
|
/* Top level nodes have hit markers. See if they all are hits */
|
|
|
|
while (1) {
|
|
|
|
if (x->node != GREP_NODE_OR)
|
|
|
|
return x->hit;
|
|
|
|
if (!x->u.binary.left->hit)
|
|
|
|
return 0;
|
|
|
|
x = x->u.binary.right;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
int grep_buffer(struct grep_opt *opt, const char *name, char *buf, unsigned long size)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* we do not have to do the two-pass grep when we do not check
|
|
|
|
* buffer-wide "all-match".
|
|
|
|
*/
|
|
|
|
if (!opt->all_match)
|
|
|
|
return grep_buffer_1(opt, name, buf, size, 0);
|
|
|
|
|
|
|
|
/* Otherwise the toplevel "or" terms hit a bit differently.
|
|
|
|
* We first clear hit markers from them.
|
|
|
|
*/
|
|
|
|
clr_hit_marker(opt->pattern_expression);
|
|
|
|
grep_buffer_1(opt, name, buf, size, 1);
|
|
|
|
|
|
|
|
if (!chk_hit_marker(opt->pattern_expression))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
return grep_buffer_1(opt, name, buf, size, 0);
|
|
|
|
}
|