Merge branch 'jc/fetch-raw-sha1'

Allows requests to fetch objects at any tip of refs (including
hidden ones).  It seems that there may be use cases even outside
Gerrit (e.g. $gmane/215701).

* jc/fetch-raw-sha1:
  fetch: fetch objects by their exact SHA-1 object names
  upload-pack: optionally allow fetching from the tips of hidden refs
  fetch: use struct ref to represent refs to be fetched
  parse_fetch_refspec(): clarify the codeflow a bit

10 files changed, 194 insertions, 79 deletions

diff --git a/Documentation/config.txt b/Documentation/config.txt
index 975d3d18d5..c1f435f6df 100644
--- a/Documentation/config.txt
+++ b/Documentation/config.txt
@@ -2123,7 +2123,13 @@ uploadpack.hiderefs::
 	are under the hierarchies listed on the value of this
 	variable is excluded, and is hidden from `git ls-remote`,
 	`git fetch`, etc.  An attempt to fetch a hidden ref by `git
-	fetch` will fail.
+	fetch` will fail.  See also `uploadpack.allowtipsha1inwant`.
+
+uploadpack.allowtipsha1inwant::
+	When `uploadpack.hiderefs` is in effect, allow `upload-pack`
+	to accept a fetch request that asks for an object at the tip
+	of a hidden ref (by default, such a request is rejected).
+	see also `uploadpack.hiderefs`.
 
 url.<base>.insteadOf::
 	Any URL that starts with this value will be rewritten to
diff --git a/builtin/fetch-pack.c b/builtin/fetch-pack.c
index 940ae35dc2..670e81fd99 100644
--- a/builtin/fetch-pack.c
+++ b/builtin/fetch-pack.c
@@ -7,12 +7,31 @@ static const char fetch_pack_usage[] =
 "[--include-tag] [--upload-pack=<git-upload-pack>] [--depth=<n>] "
 "[--no-progress] [-v] [<host>:]<directory> [<refs>...]";
 
+static void add_sought_entry_mem(struct ref ***sought, int *nr, int *alloc,
+				 const char *name, int namelen)
+{
+	struct ref *ref = xcalloc(1, sizeof(*ref) + namelen + 1);
+
+	memcpy(ref->name, name, namelen);
+	ref->name[namelen] = '\0';
+	(*nr)++;
+	ALLOC_GROW(*sought, *nr, *alloc);
+	(*sought)[*nr - 1] = ref;
+}
+
+static void add_sought_entry(struct ref ***sought, int *nr, int *alloc,
+			     const char *string)
+{
+	add_sought_entry_mem(sought, nr, alloc, string, strlen(string));
+}
+
 int cmd_fetch_pack(int argc, const char **argv, const char *prefix)
 {
 	int i, ret;
 	struct ref *ref = NULL;
 	const char *dest = NULL;
-	struct string_list sought = STRING_LIST_INIT_DUP;
+	struct ref **sought = NULL;
+	int nr_sought = 0, alloc_sought = 0;
 	int fd[2];
 	char *pack_lockfile = NULL;
 	char **pack_lockfile_ptr = NULL;
@@ -94,7 +113,7 @@ int cmd_fetch_pack(int argc, const char **argv, const char *prefix)
 	 * refs from the standard input:
 	 */
 	for (; i < argc; i++)
-		string_list_append(&sought, xstrdup(argv[i]));
+		add_sought_entry(&sought, &nr_sought, &alloc_sought, argv[i]);
 	if (args.stdin_refs) {
 		if (args.stateless_rpc) {
 			/* in stateless RPC mode we use pkt-line to read
@@ -107,14 +126,14 @@ int cmd_fetch_pack(int argc, const char **argv, const char *prefix)
 					break;
 				if (line[n-1] == '\n')
 					n--;
-				string_list_append(&sought, xmemdupz(line, n));
+				add_sought_entry_mem(&sought, &nr_sought,  &alloc_sought, line, n);
 			}
 		}
 		else {
 			/* read from stdin one ref per line, until EOF */
 			struct strbuf line = STRBUF_INIT;
 			while (strbuf_getline(&line, stdin, '\n') != EOF)
-				string_list_append(&sought, strbuf_detach(&line, NULL));
+				add_sought_entry(&sought, &nr_sought, &alloc_sought, line.buf);
 			strbuf_release(&line);
 		}
 	}
@@ -131,7 +150,7 @@ int cmd_fetch_pack(int argc, const char **argv, const char *prefix)
 	get_remote_heads(fd[0], &ref, 0, NULL);
 
 	ref = fetch_pack(&args, fd, conn, ref, dest,
-			 &sought, pack_lockfile_ptr);
+			 sought, nr_sought, pack_lockfile_ptr);
 	if (pack_lockfile) {
 		printf("lock %s\n", pack_lockfile);
 		fflush(stdout);
@@ -141,7 +160,7 @@ int cmd_fetch_pack(int argc, const char **argv, const char *prefix)
 	if (finish_connect(conn))
 		return 1;
 
-	ret = !ref || sought.nr;
+	ret = !ref;
 
 	/*
 	 * If the heads to pull were given, we should have consumed
@@ -149,8 +168,13 @@ int cmd_fetch_pack(int argc, const char **argv, const char *prefix)
 	 * remote no-such-ref' would silently succeed without issuing
 	 * an error.
 	 */
-	for (i = 0; i < sought.nr; i++)
-		error("no such remote ref %s", sought.items[i].string);
+	for (i = 0; i < nr_sought; i++) {
+		if (!sought[i] || sought[i]->matched)
+			continue;
+		error("no such remote ref %s", sought[i]->name);
+		ret = 1;
+	}
+
 	while (ref) {
 		printf("%s %s\n",
 		       sha1_to_hex(ref->old_sha1), ref->name);
diff --git a/cache.h b/cache.h
index e493563f4c..6818d87fa0 100644
--- a/cache.h
+++ b/cache.h
@@ -1017,7 +1017,8 @@ struct ref {
 		force:1,
 		forced_update:1,
 		merge:1,
-		deletion:1;
+		deletion:1,
+		matched:1;
 	enum {
 		REF_STATUS_NONE = 0,
 		REF_STATUS_OK,
diff --git a/fetch-pack.c b/fetch-pack.c
index 6d8926a550..cef8fde61b 100644
--- a/fetch-pack.c
+++ b/fetch-pack.c
@@ -36,7 +36,7 @@ static int marked;
 #define MAX_IN_VAIN 256
 
 static struct commit_list *rev_list;
-static int non_common_revs, multi_ack, use_sideband;
+static int non_common_revs, multi_ack, use_sideband, allow_tip_sha1_in_want;
 
 static void rev_list_push(struct commit *commit, int mark)
 {
@@ -520,47 +520,37 @@ static void mark_recent_complete_commits(struct fetch_pack_args *args,
 	}
 }
 
-static int non_matching_ref(struct string_list_item *item, void *unused)
-{
-	if (item->util) {
-		item->util = NULL;
-		return 0;
-	}
-	else
-		return 1;
-}
-
 static void filter_refs(struct fetch_pack_args *args,
-			struct ref **refs, struct string_list *sought)
+			struct ref **refs,
+			struct ref **sought, int nr_sought)
 {
 	struct ref *newlist = NULL;
 	struct ref **newtail = &newlist;
 	struct ref *ref, *next;
-	int sought_pos;
+	int i;
 
-	sought_pos = 0;
+	i = 0;
 	for (ref = *refs; ref; ref = next) {
 		int keep = 0;
 		next = ref->next;
+
 		if (!memcmp(ref->name, "refs/", 5) &&
 		    check_refname_format(ref->name + 5, 0))
 			; /* trash */
 		else {
-			while (sought_pos < sought->nr) {
-				int cmp = strcmp(ref->name, sought->items[sought_pos].string);
+			while (i < nr_sought) {
+				int cmp = strcmp(ref->name, sought[i]->name);
 				if (cmp < 0)
 					break; /* definitely do not have it */
 				else if (cmp == 0) {
 					keep = 1; /* definitely have it */
-					sought->items[sought_pos++].util = "matched";
-					break;
+					sought[i]->matched = 1;
 				}
-				else
-					sought_pos++; /* might have it; keep looking */
+				i++;
 			}
 		}
 
-		if (! keep && args->fetch_all &&
+		if (!keep && args->fetch_all &&
 		    (!args->depth || prefixcmp(ref->name, "refs/tags/")))
 			keep = 1;
 
@@ -573,7 +563,21 @@ static void filter_refs(struct fetch_pack_args *args,
 		}
 	}
 
-	filter_string_list(sought, 0, non_matching_ref, NULL);
+	/* Append unmatched requests to the list */
+	if (allow_tip_sha1_in_want) {
+		for (i = 0; i < nr_sought; i++) {
+			ref = sought[i];
+			if (ref->matched)
+				continue;
+			if (get_sha1_hex(ref->name, ref->old_sha1))
+				continue;
+
+			ref->matched = 1;
+			*newtail = ref;
+			ref->next = NULL;
+			newtail = &ref->next;
+		}
+	}
 	*refs = newlist;
 }
 
@@ -583,7 +587,8 @@ static void mark_alternate_complete(const struct ref *ref, void *unused)
 }
 
 static int everything_local(struct fetch_pack_args *args,
-			    struct ref **refs, struct string_list *sought)
+			    struct ref **refs,
+			    struct ref **sought, int nr_sought)
 {
 	struct ref *ref;
 	int retval;
@@ -637,7 +642,7 @@ static int everything_local(struct fetch_pack_args *args,
 		}
 	}
 
-	filter_refs(args, refs, sought);
+	filter_refs(args, refs, sought, nr_sought);
 
 	for (retval = 1, ref = *refs; ref ; ref = ref->next) {
 		const unsigned char *remote = ref->old_sha1;
@@ -767,10 +772,17 @@ static int get_pack(struct fetch_pack_args *args,
 	return 0;
 }
 
+static int cmp_ref_by_name(const void *a_, const void *b_)
+{
+	const struct ref *a = *((const struct ref **)a_);
+	const struct ref *b = *((const struct ref **)b_);
+	return strcmp(a->name, b->name);
+}
+
 static struct ref *do_fetch_pack(struct fetch_pack_args *args,
 				 int fd[2],
 				 const struct ref *orig_ref,
-				 struct string_list *sought,
+				 struct ref **sought, int nr_sought,
 				 char **pack_lockfile)
 {
 	struct ref *ref = copy_ref_list(orig_ref);
@@ -779,6 +791,7 @@ static struct ref *do_fetch_pack(struct fetch_pack_args *args,
 	int agent_len;
 
 	sort_ref_list(&ref, ref_compare_name);
+	qsort(sought, nr_sought, sizeof(*sought), cmp_ref_by_name);
 
 	if (is_repository_shallow() && !server_supports("shallow"))
 		die("Server does not support shallow clients");
@@ -808,6 +821,11 @@ static struct ref *do_fetch_pack(struct fetch_pack_args *args,
 			fprintf(stderr, "Server supports side-band\n");
 		use_sideband = 1;
 	}
+	if (server_supports("allow-tip-sha1-in-want")) {
+		if (args->verbose)
+			fprintf(stderr, "Server supports allow-tip-sha1-in-want\n");
+		allow_tip_sha1_in_want = 1;
+	}
 	if (!server_supports("thin-pack"))
 		args->use_thin_pack = 0;
 	if (!server_supports("no-progress"))
@@ -827,7 +845,7 @@ static struct ref *do_fetch_pack(struct fetch_pack_args *args,
 				agent_len, agent_feature);
 	}
 
-	if (everything_local(args, &ref, sought)) {
+	if (everything_local(args, &ref, sought, nr_sought)) {
 		packet_flush(fd[1]);
 		goto all_done;
 	}
@@ -890,11 +908,32 @@ static void fetch_pack_setup(void)
 	did_setup = 1;
 }
 
+static int remove_duplicates_in_refs(struct ref **ref, int nr)
+{
+	struct string_list names = STRING_LIST_INIT_NODUP;
+	int src, dst;
+
+	for (src = dst = 0; src < nr; src++) {
+		struct string_list_item *item;
+		item = string_list_insert(&names, ref[src]->name);
+		if (item->util)
+			continue; /* already have it */
+		item->util = ref[src];
+		if (src != dst)
+			ref[dst] = ref[src];
+		dst++;
+	}
+	for (src = dst; src < nr; src++)
+		ref[src] = NULL;
+	string_list_clear(&names, 0);
+	return dst;
+}
+
 struct ref *fetch_pack(struct fetch_pack_args *args,
 		       int fd[], struct child_process *conn,
 		       const struct ref *ref,
 		       const char *dest,
-		       struct string_list *sought,
+		       struct ref **sought, int nr_sought,
 		       char **pack_lockfile)
 {
 	struct stat st;
@@ -906,16 +945,14 @@ struct ref *fetch_pack(struct fetch_pack_args *args,
 			st.st_mtime = 0;
 	}
 
-	if (sought->nr) {
-		sort_string_list(sought);
-		string_list_remove_duplicates(sought, 0);
-	}
+	if (nr_sought)
+		nr_sought = remove_duplicates_in_refs(sought, nr_sought);
 
 	if (!ref) {
 		packet_flush(fd[1]);
 		die("no matching remote head");
 	}
-	ref_cpy = do_fetch_pack(args, fd, ref, sought, pack_lockfile);
+	ref_cpy = do_fetch_pack(args, fd, ref, sought, nr_sought, pack_lockfile);
 
 	if (args->depth > 0) {
 		static struct lock_file lock;
diff --git a/fetch-pack.h b/fetch-pack.h
index cb148719bf..dc5266c970 100644
--- a/fetch-pack.h
+++ b/fetch-pack.h
@@ -20,17 +20,16 @@ struct fetch_pack_args {
 };
 
 /*
- * sought contains the full names of remote references that should be
- * updated from.  On return, the names that were found on the remote
- * will have been removed from the list.  The util members of the
- * string_list_items are used internally; they must be NULL on entry
- * (and will be NULL on exit).
+ * sought represents remote references that should be updated from.
+ * On return, the names that were found on the remote will have been
+ * marked as such.
  */
 struct ref *fetch_pack(struct fetch_pack_args *args,
 		       int fd[], struct child_process *conn,
 		       const struct ref *ref,
 		       const char *dest,
-		       struct string_list *sought,
+		       struct ref **sought,
+		       int nr_sought,
 		       char **pack_lockfile);
 
 #endif
diff --git a/remote.c b/remote.c
index e53a6eb776..174e48eb74 100644
--- a/remote.c
+++ b/remote.c
@@ -15,6 +15,7 @@ static struct refspec s_tag_refspec = {
 	0,
 	1,
 	0,
+	0,
 	"refs/tags/*",
 	"refs/tags/*"
 };
@@ -538,7 +539,7 @@ static struct refspec *parse_refspec_internal(int nr_refspec, const char **refsp
 
 		/*
 		 * Before going on, special case ":" (or "+:") as a refspec
-		 * for matching refs.
+		 * for pushing matching refs.
 		 */
 		if (!fetch && rhs == lhs && rhs[1] == '\0') {
 			rs[i].matching = 1;
@@ -565,26 +566,25 @@ static struct refspec *parse_refspec_internal(int nr_refspec, const char **refsp
 		flags = REFNAME_ALLOW_ONELEVEL | (is_glob ? REFNAME_REFSPEC_PATTERN : 0);
 
 		if (fetch) {
-			/*
-			 * LHS
-			 * - empty is allowed; it means HEAD.
-			 * - otherwise it must be a valid looking ref.
-			 */
+			unsigned char unused[40];
+
+			/* LHS */
 			if (!*rs[i].src)
-				; /* empty is ok */
-			else if (check_refname_format(rs[i].src, flags))
+				; /* empty is ok; it means "HEAD" */
+			else if (llen == 40 && !get_sha1_hex(rs[i].src, unused))
+				rs[i].exact_sha1 = 1; /* ok */
+			else if (!check_refname_format(rs[i].src, flags))
+				; /* valid looking ref is ok */
+			else
 				goto invalid;
-			/*
-			 * RHS
-			 * - missing is ok, and is same as empty.
-			 * - empty is ok; it means not to store.
-			 * - otherwise it must be a valid looking ref.
-			 */
+			/* RHS */
 			if (!rs[i].dst)
-				; /* ok */
+				; /* missing is ok; it is the same as empty */
 			else if (!*rs[i].dst)
-				; /* ok */
-			else if (check_refname_format(rs[i].dst, flags))
+				; /* empty is ok; it means "do not store" */
+			else if (!check_refname_format(rs[i].dst, flags))
+				; /* valid looking ref is ok */
+			else
 				goto invalid;
 		} else {
 			/*
@@ -1466,7 +1466,12 @@ int get_fetch_map(const struct ref *remote_refs,
 	} else {
 		const char *name = refspec->src[0] ? refspec->src : "HEAD";
 
-		ref_map = get_remote_ref(remote_refs, name);
+		if (refspec->exact_sha1) {
+			ref_map = alloc_ref(name);
+			get_sha1_hex(name, ref_map->old_sha1);
+		} else {
+			ref_map = get_remote_ref(remote_refs, name);
+		}
 		if (!missing_ok && !ref_map)
 			die("Couldn't find remote ref %s", name);
 		if (ref_map) {
diff --git a/remote.h b/remote.h
index 251d8fd965..f7b08f15de 100644
--- a/remote.h
+++ b/remote.h
@@ -62,6 +62,7 @@ struct refspec {
 	unsigned force : 1;
 	unsigned pattern : 1;
 	unsigned matching : 1;
+	unsigned exact_sha1 : 1;
 
 	char *src;
 	char *dst;
diff --git a/t/t5516-fetch-push.sh b/t/t5516-fetch-push.sh
index c31e5c1c52..6fd125aecf 100755
--- a/t/t5516-fetch-push.sh
+++ b/t/t5516-fetch-push.sh
@@ -1043,4 +1043,38 @@ do
 	'
 done
 
+test_expect_success 'fetch exact SHA1' '
+	mk_test heads/master hidden/one &&
+	git push testrepo master:refs/hidden/one &&
+	(
+		cd testrepo &&
+		git config transfer.hiderefs refs/hidden
+	) &&
+	check_push_result $the_commit hidden/one &&
+
+	mk_child child &&
+	(
+		cd child &&
+
+		# make sure $the_commit does not exist here
+		git repack -a -d &&
+		git prune &&
+		test_must_fail git cat-file -t $the_commit &&
+
+		# fetching the hidden object should fail by default
+		test_must_fail git fetch -v ../testrepo $the_commit:refs/heads/copy &&
+		test_must_fail git rev-parse --verify refs/heads/copy &&
+
+		# the server side can allow it to succeed
+		(
+			cd ../testrepo &&
+			git config uploadpack.allowtipsha1inwant true
+		) &&
+
+		git fetch -v ../testrepo $the_commit:refs/heads/copy &&
+		result=$(git rev-parse --verify refs/heads/copy) &&
+		test "$the_commit" = "$result"
+	)
+'
+
 test_done
diff --git a/transport.c b/transport.c
index 886ffd8b1e..bd290fa7d3 100644
--- a/transport.c
+++ b/transport.c
@@ -518,11 +518,9 @@ static int fetch_refs_via_pack(struct transport *transport,
 			       int nr_heads, struct ref **to_fetch)
 {
 	struct git_transport_data *data = transport->data;
-	struct string_list sought = STRING_LIST_INIT_DUP;
 	const struct ref *refs;
 	char *dest = xstrdup(transport->url);
 	struct fetch_pack_args args;
-	int i;
 	struct ref *refs_tmp = NULL;
 
 	memset(&args, 0, sizeof(args));
@@ -536,9 +534,6 @@ static int fetch_refs_via_pack(struct transport *transport,
 	args.no_progress = !transport->progress;
 	args.depth = data->options.depth;
 
-	for (i = 0; i < nr_heads; i++)
-		string_list_append(&sought, to_fetch[i]->name);
-
 	if (!data->got_remote_heads) {
 		connect_setup(transport, 0, 0);
 		get_remote_heads(data->fd[0], &refs_tmp, 0, NULL);
@@ -547,7 +542,8 @@ static int fetch_refs_via_pack(struct transport *transport,
 
 	refs = fetch_pack(&args, data->fd, data->conn,
 			  refs_tmp ? refs_tmp : transport->remote_refs,
-			  dest, &sought, &transport->pack_lockfile);
+			  dest, to_fetch, nr_heads,
+			  &transport->pack_lockfile);
 	close(data->fd[0]);
 	close(data->fd[1]);
 	if (finish_connect(data->conn))
@@ -557,7 +553,6 @@ static int fetch_refs_via_pack(struct transport *transport,
 
 	free_refs(refs_tmp);
 
-	string_list_clear(&sought, 0);
 	free(dest);
 	return (refs ? 0 : -1);
 }
diff --git a/upload-pack.c b/upload-pack.c
index 30146a04f7..35605310d2 100644
--- a/upload-pack.c
+++ b/upload-pack.c
@@ -26,6 +26,7 @@ static const char upload_pack_usage[] = "git upload-pack [--strict] [--timeout=<
 #define SHALLOW		(1u << 16)
 #define NOT_SHALLOW	(1u << 17)
 #define CLIENT_SHALLOW	(1u << 18)
+#define HIDDEN_REF	(1u << 19)
 
 static unsigned long oldest_have;
 
@@ -33,6 +34,7 @@ static int multi_ack;
 static int no_done;
 static int use_thin_pack, use_ofs_delta, use_include_tag;
 static int no_progress, daemon_mode;
+static int allow_tip_sha1_in_want;
 static int shallow_nr;
 static struct object_array have_obj;
 static struct object_array want_obj;
@@ -487,6 +489,12 @@ static int get_common_commits(void)
 	}
 }
 
+static int is_our_ref(struct object *o)
+{
+	return o->flags &
+		((allow_tip_sha1_in_want ? HIDDEN_REF : 0) | OUR_REF);
+}
+
 static void check_non_tip(void)
 {
 	static const char *argv[] = {
@@ -523,7 +531,7 @@ static void check_non_tip(void)
 		o = get_indexed_object(--i);
 		if (!o)
 			continue;
-		if (!(o->flags & OUR_REF))
+		if (!is_our_ref(o))
 			continue;
 		memcpy(namebuf + 1, sha1_to_hex(o->sha1), 40);
 		if (write_in_full(cmd.in, namebuf, 42) < 0)
@@ -532,7 +540,7 @@ static void check_non_tip(void)
 	namebuf[40] = '\n';
 	for (i = 0; i < want_obj.nr; i++) {
 		o = want_obj.objects[i].item;
-		if (o->flags & OUR_REF)
+		if (is_our_ref(o))
 			continue;
 		memcpy(namebuf, sha1_to_hex(o->sha1), 40);
 		if (write_in_full(cmd.in, namebuf, 41) < 0)
@@ -566,7 +574,7 @@ error:
 	/* Pick one of them (we know there at least is one) */
 	for (i = 0; i < want_obj.nr; i++) {
 		o = want_obj.objects[i].item;
-		if (!(o->flags & OUR_REF))
+		if (!is_our_ref(o))
 			die("git upload-pack: not our ref %s",
 			    sha1_to_hex(o->sha1));
 	}
@@ -646,7 +654,7 @@ static void receive_needs(void)
 			    sha1_to_hex(sha1_buf));
 		if (!(o->flags & WANTED)) {
 			o->flags |= WANTED;
-			if (!(o->flags & OUR_REF))
+			if (!is_our_ref(o))
 				has_non_tip = 1;
 			add_object_array(o, NULL, &want_obj);
 		}
@@ -732,8 +740,10 @@ static int mark_our_ref(const char *refname, const unsigned char *sha1, int flag
 {
 	struct object *o = lookup_unknown_object(sha1);
 
-	if (ref_is_hidden(refname))
+	if (ref_is_hidden(refname)) {
+		o->flags |= HIDDEN_REF;
 		return 1;
+	}
 	if (!o)
 		die("git upload-pack: cannot find object %s:", sha1_to_hex(sha1));
 	o->flags |= OUR_REF;
@@ -752,9 +762,10 @@ static int send_ref(const char *refname, const unsigned char *sha1, int flag, vo
 		return 0;
 
 	if (capabilities)
-		packet_write(1, "%s %s%c%s%s agent=%s\n",
+		packet_write(1, "%s %s%c%s%s%s agent=%s\n",
 			     sha1_to_hex(sha1), refname_nons,
 			     0, capabilities,
+			     allow_tip_sha1_in_want ? " allow-tip-sha1-in-want" : "",
 			     stateless_rpc ? " no-done" : "",
 			     git_user_agent_sanitized());
 	else
@@ -788,6 +799,8 @@ static void upload_pack(void)
 
 static int upload_pack_config(const char *var, const char *value, void *unused)
 {
+	if (!strcmp("uploadpack.allowtipsha1inwant", var))
+		allow_tip_sha1_in_want = git_config_bool(var, value);
 	return parse_hide_refs_config(var, value, "uploadpack");
 }