git-fetch: rewrite expand_ref_wildcard in C

This does not seem to make measurable improvement when dealing
with 1000 unpacked refs, but we would need something like it
if we were to do a full rewrite in C somedaoy.

Signed-off-by: Junio C Hamano <junkio@cox.net>

builtin-fetch--tool.c

@@ -323,6 +323,91 @@ static int parse_reflist(const char *reflist)
 	return 0;
 }
 
+static int expand_refs_wildcard(const char *ls_remote_result, int numrefs,
+				const char **refs)
+{
+	int i, matchlen, replacelen;
+	int found_one = 0;
+	const char *remote = *refs++;
+	numrefs--;
+
+	if (numrefs == 0) {
+		fprintf(stderr, "Nothing specified for fetching with remote.%s.fetch\n",
+			remote);
+		printf("empty\n");
+	}
+
+	for (i = 0; i < numrefs; i++) {
+		const char *ref = refs[i];
+		const char *lref = ref;
+		const char *colon;
+		const char *tail;
+		const char *ls;
+		const char *next;
+
+		if (*lref == '+')
+			lref++;
+		colon = strchr(lref, ':');
+		tail = lref + strlen(lref);
+		if (!(colon &&
+		      2 < colon - lref &&
+		      colon[-1] == '*' &&
+		      colon[-2] == '/' &&
+		      2 < tail - (colon + 1) &&
+		      tail[-1] == '*' &&
+		      tail[-2] == '/')) {
+			/* not a glob */
+			if (!found_one++)
+				printf("explicit\n");
+			printf("%s\n", ref);
+			continue;
+		}
+
+		/* glob */
+		if (!found_one++)
+			printf("glob\n");
+
+		/* lref to colon-2 is remote hierarchy name;
+		 * colon+1 to tail-2 is local.
+		 */
+		matchlen = (colon-1) - lref;
+		replacelen = (tail-1) - (colon+1);
+		for (ls = ls_remote_result; ls; ls = next) {
+			const char *eol;
+			unsigned char sha1[20];
+			int namelen;
+
+			while (*ls && isspace(*ls))
+				ls++;
+			next = strchr(ls, '\n');
+			eol = !next ? (ls + strlen(ls)) : next;
+			if (!memcmp("^{}", eol-3, 3))
+				continue;
+			if (get_sha1_hex(ls, sha1))
+				continue;
+			ls += 40;
+			while (ls < eol && isspace(*ls))
+				ls++;
+			/* ls to next (or eol) is the name.
+			 * is it identical to lref to colon-2?
+			 */
+			if ((eol - ls) <= matchlen ||
+			    strncmp(ls, lref, matchlen))
+				continue;
+
+			/* Yes, it is a match */
+			namelen = eol - ls;
+			if (lref != ref)
+				putchar('+');
+			printf("%.*s:%.*s%.*s\n",
+			       namelen, ls,
+			       replacelen, colon + 1,
+			       namelen - matchlen, ls + matchlen);
+		}
+	}
+	return 0;
+}
+
 int cmd_fetch__tool(int argc, const char **argv, const char *prefix)
 {
 	int verbose = 0;
@@ -380,6 +465,11 @@ int cmd_fetch__tool(int argc, const char **argv, const char *prefix)
 			return error("parse-reflist takes 1 arg");
 		return parse_reflist(argv[2]);
 	}
+	if (!strcmp("expand-refs-wildcard", argv[1])) {
+		if (argc < 4)
+			return error("expand-refs-wildcard takes at least 2 args");
+		return expand_refs_wildcard(argv[2], argc - 3, argv + 3);
+	}
 
 	return error("Unknown subcommand: %s", argv[1]);
 }

git-parse-remote.sh

@@ -81,51 +81,7 @@ get_remote_default_refs_for_push () {
 # is to help prevent randomly "globbed" ref from being chosen as
 # a merge candidate
 expand_refs_wildcard () {
-	remote="$1"
-	shift
-	first_one=yes
-	if test "$#" = 0
-	then
-		echo empty
-		echo >&2 "Nothing specified for fetching with remote.$remote.fetch"
-	fi
-	for ref
-	do
-		lref=${ref#'+'}
-		# a non glob pattern is given back as-is.
-		expr "z$lref" : 'zrefs/.*/\*:refs/.*/\*$' >/dev/null || {
-			if test -n "$first_one"
-			then
-				echo "explicit"
-				first_one=
-			fi
-			echo "$ref"
-			continue
-		}
-
-		# glob
-		if test -n "$first_one"
-		then
-			echo "glob"
-			first_one=
-		fi
-		from=`expr "z$lref" : 'z\(refs/.*/\)\*:refs/.*/\*$'`
-		to=`expr "z$lref" : 'zrefs/.*/\*:\(refs/.*/\)\*$'`
-		local_force=
-		test "z$lref" = "z$ref" || local_force='+'
-		echo "$ls_remote_result" |
-		sed -e '/\^{}$/d' |
-		(
-			IFS='	'
-			while read sha1 name
-			do
-				# ignore the ones that do not start with $from
-				mapped=${name#"$from"}
-				test "z$name" = "z$mapped" && continue
-				echo "${local_force}${name}:${to}${mapped}"
-			done
-		)
-	done
+	git fetch--tool expand-refs-wildcard "$ls_remote_result" "$@"
 }
 
 # Subroutine to canonicalize remote:local notation.