remote: allow specifying refs to prefetch

When using 'git fetch --prefetch', all fetchable refs are prefetched
by default. In large repositories with many refs, this can lead to
unnecessary network traffic and increased disk space use.

Introduce a new configuration option 'remote.<name>.prefetchref' that
allows users to specify specific ref patterns to be prefetched during
a 'git fetch --prefetch' operation.

The 'prefetchref' option accepts a space-separated list of ref
patterns (e.g., 'refs/heads/main !refs/heads/feature/*'). When the
'--prefetch' option is used with 'git fetch', only the refs matching
these patterns will be prefetched.

Signed-off-by: Shubham Kanodia <shubham.kanodia10@gmail.com>

Documentation/config/remote.txt

@@ -33,6 +33,13 @@ remote.<name>.fetch::
 	The default set of "refspec" for linkgit:git-fetch[1]. See
 	linkgit:git-fetch[1].
 
+remote.<name>.prefetchref::
+	Specify the refs to be prefetched when fetching from this
+	remote. The value is a space-separated list of ref patterns
+	(e.g., "refs/heads/main !refs/heads/develop*"). This can be
+	used to optimize fetch operations by specifying exactly which
+	refs should be prefetched.
+
 remote.<name>.push::
 	The default set of "refspec" for linkgit:git-push[1]. See
 	linkgit:git-push[1].

builtin/fetch.c

@@ -38,6 +38,7 @@
 #include "trace.h"
 #include "trace2.h"
 #include "bundle-uri.h"
+#include "wildmatch.h"
 
 #define FORCED_UPDATES_DELAY_WARNING_IN_MS (10 * 1000)
 
@@ -485,6 +486,32 @@ static void filter_prefetch_refspec(struct refspec *rs)
 	}
 }
 
+static int pattern_matches_ref(const char *pattern, const char *refname)
+{
+	if (strchr(pattern, '*'))
+		return match_refspec_name_with_pattern(pattern, refname, NULL, NULL) != 0;
+	return strcmp(pattern, refname) == 0;
+}
+
+static int matches_prefetch_refs(const char *refname, const struct string_list *prefetch_refs)
+{
+	int has_positive = 0, matched_positive = 0, matched_negative = 0;
+
+	for (int i = 0; i < prefetch_refs->nr; i++) {
+		const char *pattern = prefetch_refs->items[i].string;
+		int is_negative = (*pattern == '!');
+		if (is_negative) pattern++;
+		else has_positive = 1;
+
+		if (pattern_matches_ref(pattern, refname)) {
+			if (is_negative) matched_negative = 1;
+			else matched_positive = 1;
+		}
+	}
+
+	return has_positive ? (matched_positive && !matched_negative) : !matched_negative;
+}
+
 static struct ref *get_ref_map(struct remote *remote,
 			       const struct ref *remote_refs,
 			       struct refspec *rs,
@@ -502,6 +529,7 @@ static struct ref *get_ref_map(struct remote *remote,
 	int existing_refs_populated = 0;
 
 	filter_prefetch_refspec(rs);
+
 	if (remote)
 		filter_prefetch_refspec(&remote->fetch);
 
@@ -610,6 +638,27 @@ static struct ref *get_ref_map(struct remote *remote,
 	else
 		ref_map = apply_negative_refspecs(ref_map, &remote->fetch);
 
+	/**
+	 * Filter out advertised refs that we don't want to fetch during
+	 * prefetch if a prefetchref config is set
+	 */
+
+	struct ref *prefetch_filtered_ref_map = NULL, **ref_map_tail = &prefetch_filtered_ref_map;
+	struct ref *ref, *next;
+
+	for (ref = ref_map; ref; ref = next) {
+		next = ref->next;
+		ref->next = NULL;
+
+		if (matches_prefetch_refs(ref->name, &remote->prefetch_refs)) {
+			*ref_map_tail = ref;
+			ref_map_tail = &ref->next;
+		} else {
+			free_one_ref(ref);
+		}
+	}
+	ref_map = prefetch_filtered_ref_map;
+
 	ref_map = ref_remove_duplicates(ref_map);
 
 	for (rm = ref_map; rm; rm = rm->next) {

refspec.c

@@ -281,3 +281,31 @@ void refspec_ref_prefixes(const struct refspec *rs,
 		}
 	}
 }
+
+int match_refspec_name_with_pattern(const char *key, const char *name,
+				   const char *value, char **result)
+{
+	const char *kstar = strchr(key, '*');
+	size_t klen;
+	size_t ksuffixlen;
+	size_t namelen;
+	int ret;
+	if (!kstar)
+		die(_("key '%s' of pattern had no '*'"), key);
+	klen = kstar - key;
+	ksuffixlen = strlen(kstar + 1);
+	namelen = strlen(name);
+	ret = !strncmp(name, key, klen) && namelen >= klen + ksuffixlen &&
+		!memcmp(name + namelen - ksuffixlen, kstar + 1, ksuffixlen);
+	if (ret && value) {
+		struct strbuf sb = STRBUF_INIT;
+		const char *vstar = strchr(value, '*');
+		if (!vstar)
+			die(_("value '%s' of pattern has no '*'"), value);
+		strbuf_add(&sb, value, vstar - value);
+		strbuf_add(&sb, name + klen, namelen - klen - ksuffixlen);
+		strbuf_addstr(&sb, vstar + 1);
+		*result = strbuf_detach(&sb, NULL);
+	}
+	return ret;
+}

refspec.h

@@ -73,4 +73,6 @@ struct strvec;
 void refspec_ref_prefixes(const struct refspec *rs,
 			  struct strvec *ref_prefixes);
 
+int match_refspec_name_with_pattern(const char *key, const char *name,
+                            const char *value, char **result);
 #endif /* REFSPEC_H */

remote.c

@@ -141,6 +141,7 @@ static struct remote *make_remote(struct remote_state *remote_state,
 	ret->prune = -1;  /* unspecified */
 	ret->prune_tags = -1;  /* unspecified */
 	ret->name = xstrndup(name, len);
+	string_list_init_dup(&ret->prefetch_refs);
 	refspec_init(&ret->push, REFSPEC_PUSH);
 	refspec_init(&ret->fetch, REFSPEC_FETCH);
 
@@ -166,6 +167,7 @@ static void remote_clear(struct remote *remote)
 	free((char *)remote->uploadpack);
 	FREE_AND_NULL(remote->http_proxy);
 	FREE_AND_NULL(remote->http_proxy_authmethod);
+	string_list_clear(&remote->prefetch_refs, 0);
 }
 
 static void add_merge(struct branch *branch, const char *name)
@@ -456,6 +458,12 @@ static int handle_config(const char *key, const char *value,
 		remote->prune = git_config_bool(key, value);
 	else if (!strcmp(subkey, "prunetags"))
 		remote->prune_tags = git_config_bool(key, value);
+	else if (!strcmp(subkey, "prefetchref")) {
+		if (!value)
+			return config_error_nonbool(key);
+		string_list_split(&remote->prefetch_refs, value, ' ', -1);
+		return 0;
+	}
 	else if (!strcmp(subkey, "url")) {
 		if (!value)
 			return config_error_nonbool(key);
@@ -868,39 +876,11 @@ struct strvec *push_url_of_remote(struct remote *remote)
 	return remote->pushurl.nr ? &remote->pushurl : &remote->url;
 }
 
-static int match_name_with_pattern(const char *key, const char *name,
-				   const char *value, char **result)
-{
-	const char *kstar = strchr(key, '*');
-	size_t klen;
-	size_t ksuffixlen;
-	size_t namelen;
-	int ret;
-	if (!kstar)
-		die(_("key '%s' of pattern had no '*'"), key);
-	klen = kstar - key;
-	ksuffixlen = strlen(kstar + 1);
-	namelen = strlen(name);
-	ret = !strncmp(name, key, klen) && namelen >= klen + ksuffixlen &&
-		!memcmp(name + namelen - ksuffixlen, kstar + 1, ksuffixlen);
-	if (ret && value) {
-		struct strbuf sb = STRBUF_INIT;
-		const char *vstar = strchr(value, '*');
-		if (!vstar)
-			die(_("value '%s' of pattern has no '*'"), value);
-		strbuf_add(&sb, value, vstar - value);
-		strbuf_add(&sb, name + klen, namelen - klen - ksuffixlen);
-		strbuf_addstr(&sb, vstar + 1);
-		*result = strbuf_detach(&sb, NULL);
-	}
-	return ret;
-}
-
 static int refspec_match(const struct refspec_item *refspec,
 			 const char *name)
 {
 	if (refspec->pattern)
-		return match_name_with_pattern(refspec->src, name, NULL, NULL);
+		return match_refspec_name_with_pattern(refspec->src, name, NULL, NULL);
 
 	return !strcmp(refspec->src, name);
 }
@@ -969,7 +949,7 @@ static int query_matches_negative_refspec(struct refspec *rs, struct refspec_ite
 			const char *key = refspec->dst ? refspec->dst : refspec->src;
 			const char *value = refspec->src;
 
-			if (match_name_with_pattern(key, needle, value, &expn_name))
+			if (match_refspec_name_with_pattern(key, needle, value, &expn_name))
 				string_list_append_nodup(&reversed, expn_name);
 		} else if (refspec->matching) {
 			/* For the special matching refspec, any query should match */
@@ -1014,7 +994,7 @@ static void query_refspecs_multiple(struct refspec *rs,
 		if (!refspec->dst || refspec->negative)
 			continue;
 		if (refspec->pattern) {
-			if (match_name_with_pattern(key, needle, value, result))
+			if (match_refspec_name_with_pattern(key, needle, value, result))
 				string_list_append_nodup(results, *result);
 		} else if (!strcmp(needle, key)) {
 			string_list_append(results, value);
@@ -1043,7 +1023,7 @@ int query_refspecs(struct refspec *rs, struct refspec_item *query)
 		if (!refspec->dst || refspec->negative)
 			continue;
 		if (refspec->pattern) {
-			if (match_name_with_pattern(key, needle, value, result)) {
+			if (match_refspec_name_with_pattern(key, needle, value, result)) {
 				query->force = refspec->force;
 				return 0;
 			}
@@ -1456,9 +1436,9 @@ static char *get_ref_match(const struct refspec *rs, const struct ref *ref,
 			const char *dst_side = item->dst ? item->dst : item->src;
 			int match;
 			if (direction == FROM_SRC)
-				match = match_name_with_pattern(item->src, ref->name, dst_side, &name);
+				match = match_refspec_name_with_pattern(item->src, ref->name, dst_side, &name);
 			else
-				match = match_name_with_pattern(dst_side, ref->name, item->src, &name);
+				match = match_refspec_name_with_pattern(dst_side, ref->name, item->src, &name);
 			if (match) {
 				matching_refs = i;
 				break;
@@ -2076,7 +2056,7 @@ static struct ref *get_expanded_map(const struct ref *remote_refs,
 
 		if (strchr(ref->name, '^'))
 			continue; /* a dereference item */
-		if (match_name_with_pattern(refspec->src, ref->name,
+		if (match_refspec_name_with_pattern(refspec->src, ref->name,
 					    refspec->dst, &expn_name) &&
 		    !ignore_symref_update(expn_name, &scratch)) {
 			struct ref *cpy = copy_ref(ref);

remote.h

@@ -5,6 +5,7 @@
 #include "hashmap.h"
 #include "refspec.h"
 #include "strvec.h"
+#include "string-list.h"
 
 struct option;
 struct transport_ls_refs_options;
@@ -77,6 +78,8 @@ struct remote {
 
 	struct refspec fetch;
 
+	struct string_list prefetch_refs;
+
 	/*
 	 * The setting for whether to fetch tags (as a separate rule from the
 	 * configured refspecs);

t/t7900-maintenance.sh

@@ -245,6 +245,91 @@ test_expect_success 'prefetch multiple remotes' '
 	test_subcommand git fetch remote2 $fetchargs <skip-remote1.txt
 '
 
+test_expect_success 'prefetch with positive prefetch ref patterns' '
+	test_create_repo filter-prefetch-positive &&
+	(
+		cd filter-prefetch-positive &&
+		test_commit initial &&
+		git clone . clone2 &&
+		git remote add remote2 "file://$(pwd)/clone2" &&
+
+		cd clone2 &&
+		git checkout -b feature && test_commit feature-commit-2 &&
+		git checkout -b wip/test && test_commit wip-test-commit-2 &&
+		git checkout -b topic/x && test_commit topic-x-commit-2 &&
+		git push -f origin feature wip/test topic/x &&
+		cd .. &&
+
+		git config remote.remote2.prefetchref "refs/heads/feature" &&
+		fetchargs="--prefetch --prune --no-tags --no-write-fetch-head \
+			--recurse-submodules=no --quiet" &&
+		GIT_TRACE2_EVENT="$(pwd)/prefetch-positive.txt" \
+			git maintenance run --task=prefetch 2>/dev/null &&
+		test_subcommand git fetch remote2 $fetchargs <prefetch-positive.txt &&
+
+		git rev-parse refs/prefetch/remotes/remote2/feature &&
+		test_must_fail git rev-parse refs/prefetch/remotes/remote2/wip/test &&
+		test_must_fail git rev-parse refs/prefetch/remotes/remote2/topic/x
+	)
+'
+
+test_expect_success 'prefetch with negative prefetch ref patterns' '
+	test_create_repo filter-prefetch-negative &&
+	(
+		cd filter-prefetch-negative &&
+		test_commit initial &&
+		git clone . clone3 &&
+		git remote add remote3 "file://$(pwd)/clone3" &&
+		cat .git/config &&
+
+		cd clone3 &&
+		git checkout -b feature && test_commit feature-commit-3 &&
+		git checkout -b wip/test && test_commit wip-test-commit-3 &&
+		git checkout -b topic/x && test_commit topic-x-commit-3 &&
+		git push -f origin feature wip/test topic/x &&
+		cd .. &&
+
+		git config remote.remote3.prefetchref "!refs/heads/wip/*" &&
+		fetchargs="--prefetch --prune --no-tags --no-write-fetch-head \
+			--recurse-submodules=no --quiet" &&
+		GIT_TRACE2_EVENT="$(pwd)/prefetch-negative.txt" \
+			git maintenance run --task=prefetch 2>/dev/null &&
+		test_subcommand git fetch remote3 $fetchargs <prefetch-negative.txt &&
+		git rev-parse refs/prefetch/remotes/remote3/feature &&
+		git rev-parse refs/prefetch/remotes/remote3/topic/x &&
+		test_must_fail git rev-parse refs/prefetch/remotes/remote3/wip/test
+	)
+'
+
+test_expect_success 'prefetch with positive & negative prefetch ref patterns' '
+	test_create_repo filter-prefetch-mixed &&
+	(
+		cd filter-prefetch-mixed &&
+		test_commit initial &&
+		git clone . clone4 &&
+		git remote add remote4 "file://$(pwd)/clone4" &&
+
+		cd clone4 &&
+		git checkout -b feature && test_commit feature-commit-4 &&
+		git checkout -b topic/x && test_commit topic-x-commit-4 &&
+		git checkout -b topic/y && test_commit topic-y-commit-4 &&
+		git push -f origin feature topic/x topic/y &&
+		cd .. &&
+
+		git config remote.remote4.prefetchref \
+			"refs/heads/topic/* !refs/heads/topic/y" &&
+		fetchargs="--prefetch --prune --no-tags --no-write-fetch-head \
+			--recurse-submodules=no --quiet" &&
+		GIT_TRACE2_EVENT="$(pwd)/prefetch-mixed.txt" \
+			git maintenance run --task=prefetch 2>/dev/null &&
+		test_subcommand git fetch remote4 $fetchargs <prefetch-mixed.txt &&
+
+		test_must_fail git rev-parse refs/prefetch/remotes/remote4/feature &&
+		test_must_fail git rev-parse refs/prefetch/remotes/remote4/topic/y &&
+		git rev-parse refs/prefetch/remotes/remote4/topic/x
+	)
+'
+
 test_expect_success 'loose-objects task' '
 	# Repack everything so we know the state of the object dir
 	git repack -adk &&