Merge branch 'rs/grep-lookahead'

Fix 'git grep' regression on macOS by disabling lookahead when
encountering invalid UTF-8 byte sequences.

* rs/grep-lookahead:
  grep: disable lookahead on error

Author: ttaylorr

grep.c

@@ -906,15 +906,17 @@ static int patmatch(struct grep_pat *p,
 		    const char *line, const char *eol,
 		    regmatch_t *match, int eflags)
 {
-	int hit;
-
 	if (p->pcre2_pattern)
-		hit = !pcre2match(p, line, eol, match, eflags);
-	else
-		hit = !regexec_buf(&p->regexp, line, eol - line, 1, match,
-				   eflags);
+		return !pcre2match(p, line, eol, match, eflags);
 
-	return hit;
+	switch (regexec_buf(&p->regexp, line, eol - line, 1, match, eflags)) {
+	case 0:
+		return 1;
+	case REG_NOMATCH:
+		return 0;
+	default:
+		return -1;
+	}
 }
 
 static void strip_timestamp(const char *bol, const char **eol_p)
@@ -952,6 +954,8 @@ static int headerless_match_one_pattern(struct grep_pat *p,
 
  again:
 	hit = patmatch(p, bol, eol, pmatch, eflags);
+	if (hit < 0)
+		hit = 0;
 
 	if (hit && p->word_regexp) {
 		if ((pmatch[0].rm_so < 0) ||
@@ -1461,6 +1465,8 @@ static int look_ahead(struct grep_opt *opt,
 		regmatch_t m;
 
 		hit = patmatch(p, bol, bol + *left_p, &m, 0);
+		if (hit < 0)
+			return -1;
 		if (!hit || m.rm_so < 0 || m.rm_eo < 0)
 			continue;
 		if (earliest < 0 || m.rm_so < earliest)
@@ -1655,9 +1661,13 @@ static int grep_source_1(struct grep_opt *opt, struct grep_source *gs, int colle
 		if (try_lookahead
 		    && !(last_hit
 			 && (show_function ||
-			     lno <= last_hit + opt->post_context))
-		    && look_ahead(opt, &left, &lno, &bol))
-			break;
+			     lno <= last_hit + opt->post_context))) {
+			hit = look_ahead(opt, &left, &lno, &bol);
+			if (hit < 0)
+				try_lookahead = 0;
+			else if (hit)
+				break;
+		}
 		eol = end_of_line(bol, &left);
 
 		if ((ctx == GREP_CONTEXT_HEAD) && (eol == bol))

t/t7810-grep.sh

@@ -87,6 +87,7 @@ test_expect_success setup '
 	# Still a no-op.
 	function dummy() {}
 	EOF
+	printf "\200\nASCII\n" >invalid-utf8 &&
 	if test_have_prereq FUNNYNAMES
 	then
 		echo unusual >"\"unusual\" pathname" &&
@@ -534,6 +535,14 @@ do
 		test_cmp expected actual
 	'
 
+	test_expect_success "grep $L searches past invalid lines on UTF-8 locale" '
+		LC_ALL=en_US.UTF-8 git grep A. invalid-utf8 >actual &&
+		cat >expected <<-EOF &&
+		invalid-utf8:ASCII
+		EOF
+		test_cmp expected actual
+	'
+
 	test_expect_success FUNNYNAMES "grep $L should quote unusual pathnames" '
 		cat >expected <<-EOF &&
 		${HC}"\"unusual\" pathname":unusual