Merge branch 'sb/submodule-update-in-c' into pu

"git submodule update" is getting rewritten piece-by-piece into C.

It seems to pass its own self-tests standalone, but seems to break
horribly when merged to 'pu'.

* sb/submodule-update-in-c:
  submodule--helper: introduce new update-module-mode helper
  builtin/submodule--helper: factor out method to update a single submodule
  builtin/submodule--helper: store update_clone information in a struct
  builtin/submodule--helper: factor out submodule updating
  git-submodule.sh: rename unused variables
  git-submodule.sh: align error reporting for update mode to use path

Author: gitster

Diff for: builtin/submodule--helper.c

@@ -1446,6 +1446,72 @@ static int module_clone(int argc, const char **argv, const char *prefix)
 	return 0;
 }
 
+static void determine_submodule_update_strategy(struct repository *r,
+						int just_cloned,
+						const char *path,
+						const char *update,
+						struct submodule_update_strategy *out)
+{
+	const struct submodule *sub = submodule_from_path(r, &null_oid, path);
+	char *key;
+	const char *val;
+
+	key = xstrfmt("submodule.%s.update", sub->name);
+
+	if (update) {
+		trace_printf("parsing update");
+		if (parse_submodule_update_strategy(update, out) < 0)
+			die(_("Invalid update mode '%s' for submodule path '%s'"),
+				update, path);
+	} else if (!repo_config_get_string_const(r, key, &val)) {
+		if (parse_submodule_update_strategy(val, out) < 0)
+			die(_("Invalid update mode '%s' configured for submodule path '%s'"),
+				val, path);
+	} else if (sub->update_strategy.type != SM_UPDATE_UNSPECIFIED) {
+		trace_printf("loaded thing");
+		out->type = sub->update_strategy.type;
+		out->command = sub->update_strategy.command;
+	} else
+		out->type = SM_UPDATE_CHECKOUT;
+
+	if (just_cloned &&
+	    (out->type == SM_UPDATE_MERGE ||
+	     out->type == SM_UPDATE_REBASE ||
+	     out->type == SM_UPDATE_NONE))
+		out->type = SM_UPDATE_CHECKOUT;
+
+	free(key);
+}
+
+static int module_update_module_mode(int argc, const char **argv, const char *prefix)
+{
+	const char *path, *update = NULL;
+	int just_cloned;
+	struct submodule_update_strategy update_strategy = { .type = SM_UPDATE_CHECKOUT };
+
+	if (argc < 3 || argc > 4)
+		die("submodule--helper update-module-clone expects <just-cloned> <path> [<update>]");
+
+	just_cloned = git_config_int("just_cloned", argv[1]);
+	path = argv[2];
+
+	if (argc == 4)
+		update = argv[3];
+
+	determine_submodule_update_strategy(the_repository,
+					    just_cloned, path, update,
+					    &update_strategy);
+	fputs(submodule_strategy_to_string(&update_strategy), stdout);
+
+	return 0;
+}
+
+struct update_clone_data {
+	const struct submodule *sub;
+	struct object_id oid;
+	unsigned just_cloned;
+};
+
 struct submodule_update_clone {
 	/* index into 'list', the list of submodules to look into for cloning */
 	int current;
@@ -1465,20 +1531,23 @@ struct submodule_update_clone {
 	const char *recursive_prefix;
 	const char *prefix;
 
-	/* Machine-readable status lines to be consumed by git-submodule.sh */
-	struct string_list projectlines;
+	/* to be consumed by git-submodule.sh */
+	struct update_clone_data *update_clone;
+	int update_clone_nr; int update_clone_alloc;
 
 	/* If we want to stop as fast as possible and return an error */
 	unsigned quickstop : 1;
 
 	/* failed clones to be retried again */
 	const struct cache_entry **failed_clones;
 	int failed_clones_nr, failed_clones_alloc;
+
+	int max_jobs;
 };
 #define SUBMODULE_UPDATE_CLONE_INIT {0, MODULE_LIST_INIT, 0, \
 	SUBMODULE_UPDATE_STRATEGY_INIT, 0, 0, -1, STRING_LIST_INIT_DUP, 0, \
 	NULL, NULL, NULL, \
-	STRING_LIST_INIT_DUP, 0, NULL, 0, 0}
+	NULL, 0, 0, 0, NULL, 0, 0, 0}
 
 
 static void next_submodule_warn_missing(struct submodule_update_clone *suc,
@@ -1572,11 +1641,12 @@ static int prepare_to_clone_next_submodule(const struct cache_entry *ce,
 	strbuf_addf(&sb, "%s/.git", ce->name);
 	needs_cloning = !file_exists(sb.buf);
 
-	strbuf_reset(&sb);
-	strbuf_addf(&sb, "%06o %s %d %d\t%s\n", ce->ce_mode,
-			oid_to_hex(&ce->oid), ce_stage(ce),
-			needs_cloning, ce->name);
-	string_list_append(&suc->projectlines, sb.buf);
+	ALLOC_GROW(suc->update_clone, suc->update_clone_nr + 1,
+		   suc->update_clone_alloc);
+	oidcpy(&suc->update_clone[suc->update_clone_nr].oid, &ce->oid);
+	suc->update_clone[suc->update_clone_nr].just_cloned = needs_cloning;
+	suc->update_clone[suc->update_clone_nr].sub = sub;
+	suc->update_clone_nr++;
 
 	if (!needs_cloning)
 		goto cleanup;
@@ -1717,11 +1787,44 @@ static int git_update_clone_config(const char *var, const char *value,
 	return 0;
 }
 
+static void update_submodule(struct update_clone_data *ucd)
+{
+	fprintf(stdout, "dummy %s %d\t%s\n",
+		oid_to_hex(&ucd->oid),
+		ucd->just_cloned,
+		ucd->sub->path);
+}
+
+static int update_submodules(struct submodule_update_clone *suc)
+{
+	int i;
+
+	run_processes_parallel(suc->max_jobs,
+			       update_clone_get_next_task,
+			       update_clone_start_failure,
+			       update_clone_task_finished,
+			       suc);
+
+	/*
+	 * We saved the output and put it out all at once now.
+	 * That means:
+	 * - the listener does not have to interleave their (checkout)
+	 *   work with our fetching.  The writes involved in a
+	 *   checkout involve more straightforward sequential I/O.
+	 * - the listener can avoid doing any work if fetching failed.
+	 */
+	if (suc->quickstop)
+		return 1;
+
+	for (i = 0; i < suc->update_clone_nr; i++)
+		update_submodule(&suc->update_clone[i]);
+
+	return 0;
+}
+
 static int update_clone(int argc, const char **argv, const char *prefix)
 {
 	const char *update = NULL;
-	int max_jobs = 1;
-	struct string_list_item *item;
 	struct pathspec pathspec;
 	struct submodule_update_clone suc = SUBMODULE_UPDATE_CLONE_INIT;
 
@@ -1743,7 +1846,7 @@ static int update_clone(int argc, const char **argv, const char *prefix)
 		OPT_STRING(0, "depth", &suc.depth, "<depth>",
 			   N_("Create a shallow clone truncated to the "
 			      "specified number of revisions")),
-		OPT_INTEGER('j', "jobs", &max_jobs,
+		OPT_INTEGER('j', "jobs", &suc.max_jobs,
 			    N_("parallel jobs")),
 		OPT_BOOL(0, "recommend-shallow", &suc.recommend_shallow,
 			    N_("whether the initial clone should follow the shallow recommendation")),
@@ -1759,8 +1862,8 @@ static int update_clone(int argc, const char **argv, const char *prefix)
 	};
 	suc.prefix = prefix;
 
-	update_clone_config_from_gitmodules(&max_jobs);
-	git_config(git_update_clone_config, &max_jobs);
+	update_clone_config_from_gitmodules(&suc.max_jobs);
+	git_config(git_update_clone_config, &suc.max_jobs);
 
 	argc = parse_options(argc, argv, prefix, module_update_clone_options,
 			     git_submodule_helper_usage, 0);
@@ -1775,27 +1878,7 @@ static int update_clone(int argc, const char **argv, const char *prefix)
 	if (pathspec.nr)
 		suc.warn_if_uninitialized = 1;
 
-	run_processes_parallel(max_jobs,
-			       update_clone_get_next_task,
-			       update_clone_start_failure,
-			       update_clone_task_finished,
-			       &suc);
-
-	/*
-	 * We saved the output and put it out all at once now.
-	 * That means:
-	 * - the listener does not have to interleave their (checkout)
-	 *   work with our fetching.  The writes involved in a
-	 *   checkout involve more straightforward sequential I/O.
-	 * - the listener can avoid doing any work if fetching failed.
-	 */
-	if (suc.quickstop)
-		return 1;
-
-	for_each_string_list_item(item, &suc.projectlines)
-		fprintf(stdout, "%s", item->string);
-
-	return 0;
+	return update_submodules(&suc);
 }
 
 static int resolve_relative_path(int argc, const char **argv, const char *prefix)
@@ -2041,6 +2124,7 @@ static struct cmd_struct commands[] = {
 	{"list", module_list, 0},
 	{"name", module_name, 0},
 	{"clone", module_clone, 0},
+	{"update-module-mode", module_update_module_mode, 0},
 	{"update-clone", update_clone, 0},
 	{"connect-gitdir-workingtree", connect_gitdir_workingtree, 0},
 	{"relative-path", resolve_relative_path, 0},

Diff for: git-submodule.sh

@@ -531,31 +531,17 @@ cmd_update()
 		"$@" || echo "#unmatched" $?
 	} | {
 	err=
-	while read -r mode sha1 stage just_cloned sm_path
+	while read -r quickabort sha1 just_cloned sm_path
 	do
-		die_if_unmatched "$mode" "$sha1"
+		die_if_unmatched "$quickabort" "$sha1"
 
-		name=$(git submodule--helper name "$sm_path") || exit
-		if ! test -z "$update"
-		then
-			update_module=$update
-		else
-			update_module=$(git config submodule."$name".update)
-			if test -z "$update_module"
-			then
-				update_module="checkout"
-			fi
-		fi
+		update_module=$(git submodule--helper update-module-mode $just_cloned "$sm_path" $update)
 
 		displaypath=$(git submodule--helper relative-path "$prefix$sm_path" "$wt_prefix")
 
 		if test $just_cloned -eq 1
 		then
 			subsha1=
-			case "$update_module" in
-			merge | rebase | none)
-				update_module=checkout ;;
-			esac
 		else
 			subsha1=$(sanitize_submodule_env; cd "$sm_path" &&
 				git rev-parse --verify HEAD) ||
@@ -632,7 +618,7 @@ cmd_update()
 				must_die_on_failure=yes
 				;;
 			*)
-				die "$(eval_gettext "Invalid update mode '$update_module' for submodule '$name'")"
+				die "$(eval_gettext "Invalid update mode '$update_module' for submodule path '$path'")"
 			esac
 
 			if (sanitize_submodule_env; cd "$sm_path" && $command "$sha1")