[PATCH v3 0/6] Add PCRE support to git-grep

@@ -925,9 +925,11 @@ int cmd_grep(int argc, const char **argv, const char *prefix)
 
 	if (!opt.pattern_list)
 		die(_("no pattern given."));
+	if (opt.regflags != REG_NEWLINE && opt.pcre)
+		die(_("cannot mix --extended-regexp and --perl-regexp"));
 	if (!opt.fixed && opt.ignore_case)
 		opt.regflags |= REG_ICASE;
-	if ((opt.regflags != REG_NEWLINE) && opt.fixed)
+	if ((opt.regflags != REG_NEWLINE || opt.pcre) && opt.fixed)
 		die(_("cannot mix --fixed-strings and regexp"));
 
 #ifndef NO_PTHREADS

@@ -637,4 +637,20 @@ test_expect_success LIBPCRE 'grep -P -w pattern' '
 	test_cmp expected actual
 '
 
+test_expect_success LIBPCRE 'grep -P -F returns error' '
+	test_expect_code 128 git grep -P -F main
+'
+
+test_expect_success LIBPCRE 'grep -P -E returns error' '
+	test_expect_code 128 git grep -P -E main
+'
+
+test_expect_failure LIBPCRE 'grep -P -G returns error' '
+	test_expect_code 128 git grep -P -G main
+'
+
+test_expect_failure LIBPCRE 'grep -P -E -G returns error' '
+	test_expect_code 128 git grep -P -E -G main
+'
+
 test_done

@@ -61,6 +61,7 @@ NO_INET_PTON=@NO_INET_PTON@
 NO_ICONV=@NO_ICONV@
 OLD_ICONV=@OLD_ICONV@
 NO_REGEX=@NO_REGEX@
+USE_LIBPCRE=@USE_LIBPCRE@
 NO_DEFLATE_BOUND=@NO_DEFLATE_BOUND@
 INLINE=@INLINE@
 SOCKLEN_T=@SOCKLEN_T@

@@ -220,6 +220,27 @@ AS_HELP_STRING([--with-openssl],[use OpenSSL library (default is YES)])
 AS_HELP_STRING([],              [ARG can be prefix for openssl library and headers]),\
 GIT_PARSE_WITH(openssl))
 #
+# Define USE_LIBPCRE if you have and want to use libpcre. git-grep will be
+# able to use Perl-compatible regular expressions.
+#
+# Define LIBPCREDIR=/foo/bar if your libpcre header and library files are in
+# /foo/bar/include and /foo/bar/lib directories.
+#
+AC_ARG_WITH(libpcre,
+AS_HELP_STRING([--with-libpcre],[support Perl-compatible regexes (default is NO)])
+AS_HELP_STRING([],           [ARG can be also prefix for libpcre library and headers]),
+if test "$withval" = "no"; then \
+	USE_LIBPCRE=; \
+elif test "$withval" = "yes"; then \
+	USE_LIBPCRE=YesPlease; \
+else
+	USE_LIBPCRE=YesPlease; \
+	LIBPCREDIR=$withval; \
+	AC_MSG_NOTICE([Setting LIBPCREDIR to $withval]); \
+	GIT_CONF_APPEND_LINE(LIBPCREDIR=$withval); \
+fi \
+)
+#
 # Define NO_CURL if you do not have curl installed.  git-http-pull and
 # git-http-push are not built, and you cannot use http:// and https://
 # transports.

@@ -435,6 +456,25 @@ AC_SUBST(NEEDS_SSL_WITH_CRYPTO)
 AC_SUBST(NO_OPENSSL)
 
 #
+# Define USE_LIBPCRE if you have and want to use libpcre. git-grep will be
+# able to use Perl-compatible regular expressions.
+#
+
+if test -n "$USE_LIBPCRE"; then
+
+GIT_STASH_FLAGS($LIBPCREDIR)
+
+AC_CHECK_LIB([pcre], [pcre_version],
+[USE_LIBPCRE=YesPlease],
+[USE_LIBPCRE=])
+
+GIT_UNSTASH_FLAGS($LIBPCREDIR)
+
+AC_SUBST(USE_LIBPCRE)
+
+fi
+
+#
 # Define NO_CURL if you do not have libcurl installed.  git-http-pull and
 # git-http-push are not built, and you cannot use http:// and https://
 # transports.

@@ -59,6 +59,21 @@ struct grep_opt *grep_opt_dup(const struct grep_opt *opt)
 	return ret;
 }
 
+static NORETURN void compile_regexp_failed(const struct grep_pat *p,
+		const char *error)
+{
+	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;
+
+	die("%s'%s': %s", where, p->pattern, error);
+}
+
 static void compile_regexp(struct grep_pat *p, struct grep_opt *opt)
 {
 	int err;

@@ -73,17 +88,9 @@ static void compile_regexp(struct grep_pat *p, struct grep_opt *opt)
 	err = regcomp(&p->regexp, p->pattern, opt->regflags);
 	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);
+		compile_regexp_failed(p, errbuf);
 	}
 }
 

@@ -587,6 +587,11 @@ use these, and "test_set_prereq" for how to define your own.
    Test is not run by root user, and an attempt to write to an
    unwritable file is expected to fail correctly.
 
+ - LIBPCRE
+
+   Git was compiled with USE_LIBPCRE=YesPlease. Wrap any tests
+   that use git-grep --perl-regexp or git-grep -P in these.
+
 Tips for Writing Tests
 ----------------------
 

@@ -26,6 +26,12 @@ test_expect_success setup '
 		echo foo mmap bar_mmap
 		echo foo_mmap bar mmap baz
 	} >file &&
+	{
+		echo Hello world
+		echo HeLLo world
+		echo Hello_world
+		echo HeLLo_world
+	} >hello_world &&
 	echo vvv >v &&
 	echo ww w >w &&
 	echo x x xx x >x &&

@@ -599,4 +605,36 @@ test_expect_success 'grep -e -- -- path' '
 	test_cmp expected actual
 '
 
+cat >expected <<EOF
+hello.c:int main(int argc, const char **argv)
+hello.c:	printf("Hello world.\n");
+EOF
+
+test_expect_success LIBPCRE 'grep --perl-regexp pattern' '
+	git grep --perl-regexp "\p{Ps}.*?\p{Pe}" hello.c >actual &&
+	test_cmp expected actual
+'
+
+test_expect_success LIBPCRE 'grep -P pattern' '
+	git grep -P "\p{Ps}.*?\p{Pe}" hello.c >actual &&
+	test_cmp expected actual
+'
+
+test_expect_success LIBPCRE 'grep -P -i pattern' '
+	{
+		echo "hello.c:	printf(\"Hello world.\n\");"
+	} >expected &&
+	git grep -P -i "PRINTF\([^\d]+\)" hello.c >actual &&
+	test_cmp expected actual
+'
+
+test_expect_success LIBPCRE 'grep -P -w pattern' '
+	{
+		echo "hello_world:Hello world"
+		echo "hello_world:HeLLo world"
+	} >expected &&
+	git grep -P -w "He((?i)ll)o" hello_world >actual &&
+	test_cmp expected actual
+'
+
 test_done

@@ -1069,6 +1069,7 @@ esac
 
 test -z "$NO_PERL" && test_set_prereq PERL
 test -z "$NO_PYTHON" && test_set_prereq PYTHON
+test -n "$USE_LIBPCRE" && test_set_prereq LIBPCRE
 
 # Can we rely on git's output in the C locale?
 if test -n "$GETTEXT_POISON"

@@ -12,6 +12,7 @@ SYNOPSIS
 'git grep' [-a | --text] [-I] [-i | --ignore-case] [-w | --word-regexp]
 	   [-v | --invert-match] [-h|-H] [--full-name]
 	   [-E | --extended-regexp] [-G | --basic-regexp]
+	   [-P | --perl-regexp]
 	   [-F | --fixed-strings] [-n | --line-number]
 	   [-l | --files-with-matches] [-L | --files-without-match]
 	   [(-O | --open-files-in-pager) [<pager>]]

@@ -97,6 +98,11 @@ OPTIONS
 	Use POSIX extended/basic regexp for patterns.  Default
 	is to use basic regexp.
 
+-P::
+--perl-regexp::
+	Use Perl-compatible regexp for patterns. Requires libpcre to be
+	compiled in.
+
 -F::
 --fixed-strings::
 	Use fixed strings for patterns (don't interpret pattern

@@ -24,6 +24,12 @@ all::
 # Define NO_OPENSSL environment variable if you do not have OpenSSL.
 # This also implies BLK_SHA1.
 #
+# Define USE_LIBPCRE if you have and want to use libpcre. git-grep will be
+# able to use Perl-compatible regular expressions.
+#
+# Define LIBPCREDIR=/foo/bar if your libpcre header and library files are in
+# /foo/bar/include and /foo/bar/lib directories.
+#
 # Define NO_CURL if you do not have libcurl installed.  git-http-pull and
 # git-http-push are not built, and you cannot use http:// and https://
 # transports.

@@ -1251,6 +1257,15 @@ ifdef NO_LIBGEN_H
 	COMPAT_OBJS += compat/basename.o
 endif
 
+ifdef USE_LIBPCRE
+	BASIC_CFLAGS += -DUSE_LIBPCRE
+	ifdef LIBPCREDIR
+		BASIC_CFLAGS += -I$(LIBPCREDIR)/include
+		EXTLIBS += -L$(LIBPCREDIR)/$(lib) $(CC_LD_DYNPATH)$(LIBPCREDIR)/$(lib)
+	endif
+	EXTLIBS += -lpcre
+endif
+
 ifdef NO_CURL
 	BASIC_CFLAGS += -DNO_CURL
 	REMOTE_CURL_PRIMARY =

@@ -781,6 +781,8 @@ int cmd_grep(int argc, const char **argv, const char *prefix)
 			REG_EXTENDED),
 		OPT_BOOLEAN('F', "fixed-strings", &opt.fixed,
 			"interpret patterns as fixed strings"),
+		OPT_BOOLEAN('P', "perl-regexp", &opt.pcre,
+				"use Perl-compatible regular expressions"),
 		OPT_GROUP(""),
 		OPT_BOOLEAN('n', "line-number", &opt.linenum, "show line numbers"),
 		OPT_NEGBIT('h', NULL, &opt.pathname, "don't show filenames", 1),

@@ -1487,6 +1487,7 @@ _git_grep ()
 			--text --ignore-case --word-regexp --invert-match
 			--full-name --line-number
 			--extended-regexp --basic-regexp --fixed-strings
+			--perl-regexp
 			--files-with-matches --name-only
 			--files-without-match
 			--max-depth

@@ -74,6 +74,69 @@ static NORETURN void compile_regexp_failed(const struct grep_pat *p,
 	die("%s'%s': %s", where, p->pattern, error);
 }
 
+#ifdef USE_LIBPCRE
+static void compile_pcre_regexp(struct grep_pat *p, const struct grep_opt *opt)
+{
+	const char *error;
+	int erroffset;
+	int options = 0;
+
+	if (opt->ignore_case)
+		options |= PCRE_CASELESS;
+
+	p->pcre_regexp = pcre_compile(p->pattern, options, &error, &erroffset,
+			NULL);
+	if (!p->pcre_regexp)
+		compile_regexp_failed(p, error);
+
+	p->pcre_extra_info = pcre_study(p->pcre_regexp, 0, &error);
+	if (!p->pcre_extra_info && error)
+		die("%s", error);
+}
+
+static int pcrematch(struct grep_pat *p, const char *line, const char *eol,
+		regmatch_t *match, int eflags)
+{
+	int ovector[30], ret, flags = 0;
+
+	if (eflags & REG_NOTBOL)
+		flags |= PCRE_NOTBOL;
+
+	ret = pcre_exec(p->pcre_regexp, p->pcre_extra_info, line, eol - line,
+			0, flags, ovector, ARRAY_SIZE(ovector));
+	if (ret < 0 && ret != PCRE_ERROR_NOMATCH)
+		die("pcre_exec failed with error code %d", ret);
+	if (ret > 0) {
+		ret = 0;
+		match->rm_so = ovector[0];
+		match->rm_eo = ovector[1];
+	}
+
+	return ret;
+}
+
+static void free_pcre_regexp(struct grep_pat *p)
+{
+	pcre_free(p->pcre_regexp);
+	pcre_free(p->pcre_extra_info);
+}
+#else /* !USE_LIBPCRE */
+static void compile_pcre_regexp(struct grep_pat *p, const struct grep_opt *opt)
+{
+	die("cannot use Perl-compatible regexes when not compiled with USE_LIBPCRE");
+}
+
+static int pcrematch(struct grep_pat *p, const char *line, const char *eol,
+		regmatch_t *match, int eflags)
+{
+	return 1;
+}
+
+static void free_pcre_regexp(struct grep_pat *p)
+{
+}
+#endif /* !USE_LIBPCRE */
+
 static void compile_regexp(struct grep_pat *p, struct grep_opt *opt)
 {
 	int err;

@@ -85,6 +148,11 @@ static void compile_regexp(struct grep_pat *p, struct grep_opt *opt)
 	if (p->fixed)
 		return;
 
+	if (opt->pcre) {
+		compile_pcre_regexp(p, opt);
+		return;
+	}
+
 	err = regcomp(&p->regexp, p->pattern, opt->regflags);
 	if (err) {
 		char errbuf[1024];

@@ -327,7 +395,10 @@ void free_grep_patterns(struct grep_opt *opt)
 		case GREP_PATTERN: /* atom */
 		case GREP_PATTERN_HEAD:
 		case GREP_PATTERN_BODY:
-			regfree(&p->regexp);
+			if (p->pcre_regexp)
+				free_pcre_regexp(p);
+			else
+				regfree(&p->regexp);
 			break;
 		default:
 			break;

@@ -426,6 +497,8 @@ static int patmatch(struct grep_pat *p, char *line, char *eol,
 
 	if (p->fixed)
 		hit = !fixmatch(p, line, eol, match);
+	else if (p->pcre_regexp)
+		hit = !pcrematch(p, line, eol, match, eflags);
 	else
 		hit = !regmatch(&p->regexp, line, eol, match, eflags);
 

@@ -1,6 +1,12 @@
 #ifndef GREP_H
 #define GREP_H
 #include "color.h"
+#ifdef USE_LIBPCRE
+#include <pcre.h>
+#else
+typedef int pcre;
+typedef int pcre_extra;
+#endif
 
 enum grep_pat_token {
 	GREP_PATTERN,

@@ -33,6 +39,8 @@ struct grep_pat {
 	size_t patternlen;
 	enum grep_header_field field;
 	regex_t regexp;
+	pcre *pcre_regexp;
+	pcre_extra *pcre_extra_info;
 	unsigned fixed:1;
 	unsigned ignore_case:1;
 	unsigned word_regexp:1;

@@ -83,6 +91,7 @@ struct grep_opt {
 #define GREP_BINARY_TEXT	2
 	int binary;
 	int extended;
+	int pcre;
 	int relative;
 	int pathname;
 	int null_following_name;

@@ -898,7 +898,7 @@ static int grep_buffer_1(struct grep_opt *opt, const char *name,
 		int hit;
 
 		/*
-		 * look_ahead() skips quicly to the line that possibly
+		 * look_ahead() skips quickly to the line that possibly
 		 * has the next hit; don't call it if we need to do
 		 * something more than just skipping the current line
 		 * in response to an unmatch for the current line.  E.g.