treewide: rename tree to maybe_tree

derrickstolee · gitster · commit 891435d55da8 · 2018-04-11T10:47:16.000+09:00
Using the commit-graph file to walk commit history removes the large
cost of parsing commits during the walk. This exposes a performance
issue: lookup_tree() takes a large portion of the computation time,
even when Git never uses those trees.

In anticipation of lazy-loading these trees, rename the 'tree' member
of struct commit to 'maybe_tree'. This serves two purposes: it hints
at the future role of possibly being NULL even if the commit has a
valid tree, and it allows for unambiguous transformation from simple
member access (i.e. commit-&gt;maybe_tree) to method access.

Signed-off-by: Derrick Stolee &lt;dstolee@microsoft.com&gt;
Signed-off-by: Junio C Hamano &lt;gitster@pobox.com&gt;
diff --git a/blame.c b/blame.c
@@ -553,10 +553,10 @@ static struct blame_origin *find_origin(struct commit *parent,
 	diff_setup_done(&diff_opts);
 
 	if (is_null_oid(&origin->commit->object.oid))
-		do_diff_cache(&parent->tree->object.oid, &diff_opts);
+		do_diff_cache(&parent->maybe_tree->object.oid, &diff_opts);
 	else
-		diff_tree_oid(&parent->tree->object.oid,
-			      &origin->commit->tree->object.oid,
+		diff_tree_oid(&parent->maybe_tree->object.oid,
+			      &origin->commit->maybe_tree->object.oid,
 			      "", &diff_opts);
 	diffcore_std(&diff_opts);
 
@@ -622,10 +622,10 @@ static struct blame_origin *find_rename(struct commit *parent,
 	diff_setup_done(&diff_opts);
 
 	if (is_null_oid(&origin->commit->object.oid))
-		do_diff_cache(&parent->tree->object.oid, &diff_opts);
+		do_diff_cache(&parent->maybe_tree->object.oid, &diff_opts);
 	else
-		diff_tree_oid(&parent->tree->object.oid,
-			      &origin->commit->tree->object.oid,
+		diff_tree_oid(&parent->maybe_tree->object.oid,
+			      &origin->commit->maybe_tree->object.oid,
 			      "", &diff_opts);
 	diffcore_std(&diff_opts);
 
@@ -1257,10 +1257,10 @@ static void find_copy_in_parent(struct blame_scoreboard *sb,
 		diff_opts.flags.find_copies_harder = 1;
 
 	if (is_null_oid(&target->commit->object.oid))
-		do_diff_cache(&parent->tree->object.oid, &diff_opts);
+		do_diff_cache(&parent->maybe_tree->object.oid, &diff_opts);
 	else
-		diff_tree_oid(&parent->tree->object.oid,
-			      &target->commit->tree->object.oid,
+		diff_tree_oid(&parent->maybe_tree->object.oid,
+			      &target->commit->maybe_tree->object.oid,
 			      "", &diff_opts);
 
 	if (!diff_opts.flags.find_copies_harder)
diff --git a/builtin/checkout.c b/builtin/checkout.c
@@ -484,7 +484,7 @@ static int merge_working_tree(const struct checkout_opts *opts,
 
 	resolve_undo_clear();
 	if (opts->force) {
-		ret = reset_tree(new_branch_info->commit->tree, opts, 1, writeout_error);
+		ret = reset_tree(new_branch_info->commit->maybe_tree, opts, 1, writeout_error);
 		if (ret)
 			return ret;
 	} else {
@@ -570,18 +570,18 @@ static int merge_working_tree(const struct checkout_opts *opts,
 			o.verbosity = 0;
 			work = write_tree_from_memory(&o);
 
-			ret = reset_tree(new_branch_info->commit->tree, opts, 1,
+			ret = reset_tree(new_branch_info->commit->maybe_tree, opts, 1,
 					 writeout_error);
 			if (ret)
 				return ret;
 			o.ancestor = old_branch_info->name;
 			o.branch1 = new_branch_info->name;
 			o.branch2 = "local";
-			ret = merge_trees(&o, new_branch_info->commit->tree, work,
-				old_branch_info->commit->tree, &result);
+			ret = merge_trees(&o, new_branch_info->commit->maybe_tree, work,
+				old_branch_info->commit->maybe_tree, &result);
 			if (ret < 0)
 				exit(128);
-			ret = reset_tree(new_branch_info->commit->tree, opts, 0,
+			ret = reset_tree(new_branch_info->commit->maybe_tree, opts, 0,
 					 writeout_error);
 			strbuf_release(&o.obuf);
 			if (ret)
@@ -1002,7 +1002,7 @@ static int parse_branchname_arg(int argc, const char **argv,
 		*source_tree = parse_tree_indirect(rev);
 	} else {
 		parse_commit_or_die(new_branch_info->commit);
-		*source_tree = new_branch_info->commit->tree;
+		*source_tree = new_branch_info->commit->maybe_tree;
 	}
 
 	if (!*source_tree)                   /* case (1): want a tree */
diff --git a/builtin/diff.c b/builtin/diff.c
@@ -398,7 +398,7 @@ int cmd_diff(int argc, const char **argv, const char *prefix)
 		if (!obj)
 			die(_("invalid object '%s' given."), name);
 		if (obj->type == OBJ_COMMIT)
-			obj = &((struct commit *)obj)->tree->object;
+			obj = &((struct commit *)obj)->maybe_tree->object;
 
 		if (obj->type == OBJ_TREE) {
 			obj->flags |= flags;
diff --git a/builtin/fast-export.c b/builtin/fast-export.c
@@ -578,11 +578,11 @@ static void handle_commit(struct commit *commit, struct rev_info *rev,
 	    get_object_mark(&commit->parents->item->object) != 0 &&
 	    !full_tree) {
 		parse_commit_or_die(commit->parents->item);
-		diff_tree_oid(&commit->parents->item->tree->object.oid,
-			      &commit->tree->object.oid, "", &rev->diffopt);
+		diff_tree_oid(&commit->parents->item->maybe_tree->object.oid,
+			      &commit->maybe_tree->object.oid, "", &rev->diffopt);
 	}
 	else
-		diff_root_tree_oid(&commit->tree->object.oid,
+		diff_root_tree_oid(&commit->maybe_tree->object.oid,
 				   "", &rev->diffopt);
 
 	/* Export the referenced blobs, and remember the marks. */
diff --git a/builtin/log.c b/builtin/log.c
@@ -1064,8 +1064,8 @@ static void make_cover_letter(struct rev_info *rev, int use_stdout,
 
 	diff_setup_done(&opts);
 
-	diff_tree_oid(&origin->tree->object.oid,
-		      &head->tree->object.oid,
+	diff_tree_oid(&origin->maybe_tree->object.oid,
+		      &head->maybe_tree->object.oid,
 		      "", &opts);
 	diffcore_std(&opts);
 	diff_flush(&opts);
diff --git a/builtin/reflog.c b/builtin/reflog.c
@@ -153,7 +153,7 @@ static int commit_is_complete(struct commit *commit)
 		for (i = 0; i < found.nr; i++) {
 			struct commit *c =
 				(struct commit *)found.objects[i].item;
-			if (!tree_is_complete(&c->tree->object.oid)) {
+			if (!tree_is_complete(&c->maybe_tree->object.oid)) {
 				is_incomplete = 1;
 				c->object.flags |= INCOMPLETE;
 			}
diff --git a/commit-graph.c b/commit-graph.c
@@ -258,7 +258,7 @@ static int fill_commit_in_graph(struct commit *item, struct commit_graph *g, uin
 	item->graph_pos = pos;
 
 	hashcpy(oid.hash, commit_data);
-	item->tree = lookup_tree(&oid);
+	item->maybe_tree = lookup_tree(&oid);
 
 	date_high = get_be32(commit_data + g->hash_len + 8) & 0x3;
 	date_low = get_be32(commit_data + g->hash_len + 12);
@@ -369,7 +369,7 @@ static void write_graph_chunk_data(struct hashfile *f, int hash_len,
 		uint32_t packedDate[2];
 
 		parse_commit(*list);
-		hashwrite(f, (*list)->tree->object.oid.hash, hash_len);
+		hashwrite(f, (*list)->maybe_tree->object.oid.hash, hash_len);
 
 		parent = (*list)->parents;
 
diff --git a/commit.c b/commit.c
@@ -335,7 +335,7 @@ int parse_commit_buffer(struct commit *item, const void *buffer, unsigned long s
 	if (get_sha1_hex(bufptr + 5, parent.hash) < 0)
 		return error("bad tree pointer in commit %s",
 			     oid_to_hex(&item->object.oid));
-	item->tree = lookup_tree(&parent);
+	item->maybe_tree = lookup_tree(&parent);
 	bufptr += tree_entry_len + 1; /* "tree " + "hex sha1" + "\n" */
 	pptr = &item->parents;
 
diff --git a/commit.h b/commit.h
@@ -22,7 +22,7 @@ struct commit {
 	unsigned int index;
 	timestamp_t date;
 	struct commit_list *parents;
-	struct tree *tree;
+	struct tree *maybe_tree;
 	uint32_t graph_pos;
 };
 
diff --git a/fsck.c b/fsck.c
@@ -396,9 +396,9 @@ static int fsck_walk_commit(struct commit *commit, void *data, struct fsck_optio
 
 	name = get_object_name(options, &commit->object);
 	if (name)
-		put_object_name(options, &commit->tree->object, "%s:", name);
+		put_object_name(options, &commit->maybe_tree->object, "%s:", name);
 
-	result = options->walk((struct object *)commit->tree, OBJ_TREE, data, options);
+	result = options->walk((struct object *)commit->maybe_tree, OBJ_TREE, data, options);
 	if (result < 0)
 		return result;
 	res = result;
@@ -772,7 +772,7 @@ static int fsck_commit_buffer(struct commit *commit, const char *buffer,
 	err = fsck_ident(&buffer, &commit->object, options);
 	if (err)
 		return err;
-	if (!commit->tree) {
+	if (!commit->maybe_tree) {
 		err = report(options, &commit->object, FSCK_MSG_BAD_TREE, "could not load commit's tree %s", sha1_to_hex(tree_sha1));
 		if (err)
 			return err;
diff --git a/http-push.c b/http-push.c
@@ -1330,7 +1330,7 @@ static int get_delta(struct rev_info *revs, struct remote_lock *lock)
 	int count = 0;
 
 	while ((commit = get_revision(revs)) != NULL) {
-		p = process_tree(commit->tree, p);
+		p = process_tree(commit->maybe_tree, p);
 		commit->object.flags |= LOCAL;
 		if (!(commit->object.flags & UNINTERESTING))
 			count += add_send_request(&commit->object, lock);
diff --git a/line-log.c b/line-log.c
@@ -817,8 +817,8 @@ static void queue_diffs(struct line_log_data *range,
 	assert(commit);
 
 	DIFF_QUEUE_CLEAR(&diff_queued_diff);
-	diff_tree_oid(parent ? &parent->tree->object.oid : NULL,
-		      &commit->tree->object.oid, "", opt);
+	diff_tree_oid(parent ? &parent->maybe_tree->object.oid : NULL,
+		      &commit->maybe_tree->object.oid, "", opt);
 	if (opt->detect_rename) {
 		filter_diffs_for_paths(range, 1);
 		if (diff_might_be_rename())
diff --git a/list-objects.c b/list-objects.c
@@ -195,7 +195,7 @@ static void mark_edge_parents_uninteresting(struct commit *commit,
 		struct commit *parent = parents->item;
 		if (!(parent->object.flags & UNINTERESTING))
 			continue;
-		mark_tree_uninteresting(parent->tree);
+		mark_tree_uninteresting(parent->maybe_tree);
 		if (revs->edge_hint && !(parent->object.flags & SHOWN)) {
 			parent->object.flags |= SHOWN;
 			show_edge(parent);
@@ -212,7 +212,7 @@ void mark_edges_uninteresting(struct rev_info *revs, show_edge_fn show_edge)
 		struct commit *commit = list->item;
 
 		if (commit->object.flags & UNINTERESTING) {
-			mark_tree_uninteresting(commit->tree);
+			mark_tree_uninteresting(commit->maybe_tree);
 			if (revs->edge_hint_aggressive && !(commit->object.flags & SHOWN)) {
 				commit->object.flags |= SHOWN;
 				show_edge(commit);
@@ -227,7 +227,7 @@ void mark_edges_uninteresting(struct rev_info *revs, show_edge_fn show_edge)
 			struct commit *commit = (struct commit *)obj;
 			if (obj->type != OBJ_COMMIT || !(obj->flags & UNINTERESTING))
 				continue;
-			mark_tree_uninteresting(commit->tree);
+			mark_tree_uninteresting(commit->maybe_tree);
 			if (!(obj->flags & SHOWN)) {
 				obj->flags |= SHOWN;
 				show_edge(commit);
@@ -300,8 +300,8 @@ static void do_traverse(struct rev_info *revs,
 		 * an uninteresting boundary commit may not have its tree
 		 * parsed yet, but we are not going to show them anyway
 		 */
-		if (commit->tree)
-			add_pending_tree(revs, commit->tree);
+		if (commit->maybe_tree)
+			add_pending_tree(revs, commit->maybe_tree);
 		show_commit(commit, show_data);
 
 		if (revs->tree_blobs_in_commit_order)
diff --git a/log-tree.c b/log-tree.c
@@ -806,7 +806,7 @@ static int log_tree_diff(struct rev_info *opt, struct commit *commit, struct log
 		return 0;
 
 	parse_commit_or_die(commit);
-	oid = &commit->tree->object.oid;
+	oid = &commit->maybe_tree->object.oid;
 
 	/* Root commit? */
 	parents = get_saved_parents(opt, commit);
@@ -831,7 +831,7 @@ static int log_tree_diff(struct rev_info *opt, struct commit *commit, struct log
 			 * we merged _in_.
 			 */
 			parse_commit_or_die(parents->item);
-			diff_tree_oid(&parents->item->tree->object.oid,
+			diff_tree_oid(&parents->item->maybe_tree->object.oid,
 				      oid, "", &opt->diffopt);
 			log_tree_diff_flush(opt);
 			return !opt->loginfo;
@@ -846,7 +846,7 @@ static int log_tree_diff(struct rev_info *opt, struct commit *commit, struct log
 		struct commit *parent = parents->item;
 
 		parse_commit_or_die(parent);
-		diff_tree_oid(&parent->tree->object.oid,
+		diff_tree_oid(&parent->maybe_tree->object.oid,
 			      oid, "", &opt->diffopt);
 		log_tree_diff_flush(opt);
 
diff --git a/merge-recursive.c b/merge-recursive.c
@@ -101,7 +101,7 @@ static struct commit *make_virtual_commit(struct tree *tree, const char *comment
 	struct commit *commit = alloc_commit_node();
 
 	set_merge_remote_desc(commit, comment, (struct object *)commit);
-	commit->tree = tree;
+	commit->maybe_tree = tree;
 	commit->object.parsed = 1;
 	return commit;
 }
@@ -2154,7 +2154,8 @@ int merge_recursive(struct merge_options *o,
 		read_cache();
 
 	o->ancestor = "merged common ancestors";
-	clean = merge_trees(o, h1->tree, h2->tree, merged_common_ancestors->tree,
+	clean = merge_trees(o, h1->maybe_tree, h2->maybe_tree,
+			    merged_common_ancestors->maybe_tree,
 			    &mrtree);
 	if (clean < 0) {
 		flush_output(o);
diff --git a/notes-merge.c b/notes-merge.c
@@ -600,14 +600,14 @@ int notes_merge(struct notes_merge_options *o,
 			printf("No merge base found; doing history-less merge\n");
 	} else if (!bases->next) {
 		base_oid = &bases->item->object.oid;
-		base_tree_oid = &bases->item->tree->object.oid;
+		base_tree_oid = &bases->item->maybe_tree->object.oid;
 		if (o->verbosity >= 4)
 			printf("One merge base found (%.7s)\n",
 			       oid_to_hex(base_oid));
 	} else {
 		/* TODO: How to handle multiple merge-bases? */
 		base_oid = &bases->item->object.oid;
-		base_tree_oid = &bases->item->tree->object.oid;
+		base_tree_oid = &bases->item->maybe_tree->object.oid;
 		if (o->verbosity >= 3)
 			printf("Multiple merge bases found. Using the first "
 				"(%.7s)\n", oid_to_hex(base_oid));
@@ -634,8 +634,8 @@ int notes_merge(struct notes_merge_options *o,
 		goto found_result;
 	}
 
-	result = merge_from_diffs(o, base_tree_oid, &local->tree->object.oid,
-				  &remote->tree->object.oid, local_tree);
+	result = merge_from_diffs(o, base_tree_oid, &local->maybe_tree->object.oid,
+				  &remote->maybe_tree->object.oid, local_tree);
 
 	if (result != 0) { /* non-trivial merge (with or without conflicts) */
 		/* Commit (partial) result */
diff --git a/packfile.c b/packfile.c
@@ -1925,7 +1925,7 @@ static int add_promisor_object(const struct object_id *oid,
 		struct commit *commit = (struct commit *) obj;
 		struct commit_list *parents = commit->parents;
 
-		oidset_insert(set, &commit->tree->object.oid);
+		oidset_insert(set, &commit->maybe_tree->object.oid);
 		for (; parents; parents = parents->next)
 			oidset_insert(set, &parents->item->object.oid);
 	} else if (obj->type == OBJ_TAG) {
diff --git a/pretty.c b/pretty.c
@@ -1161,10 +1161,10 @@ static size_t format_commit_one(struct strbuf *sb, /* in UTF-8 */
 		strbuf_addstr(sb, diff_get_color(c->auto_color, DIFF_RESET));
 		return 1;
 	case 'T':		/* tree hash */
-		strbuf_addstr(sb, oid_to_hex(&commit->tree->object.oid));
+		strbuf_addstr(sb, oid_to_hex(&commit->maybe_tree->object.oid));
 		return 1;
 	case 't':		/* abbreviated tree hash */
-		strbuf_add_unique_abbrev(sb, commit->tree->object.oid.hash,
+		strbuf_add_unique_abbrev(sb, commit->maybe_tree->object.oid.hash,
 					 c->pretty_ctx->abbrev);
 		return 1;
 	case 'P':		/* parent hashes */
diff --git a/ref-filter.c b/ref-filter.c
@@ -815,7 +815,7 @@ static void grab_commit_values(struct atom_value *val, int deref, struct object
 		if (deref)
 			name++;
 		if (!strcmp(name, "tree")) {
-			v->s = xstrdup(oid_to_hex(&commit->tree->object.oid));
+			v->s = xstrdup(oid_to_hex(&commit->maybe_tree->object.oid));
 		}
 		else if (!strcmp(name, "numparent")) {
 			v->value = commit_list_count(commit->parents);
diff --git a/revision.c b/revision.c
@@ -439,8 +439,8 @@ static void file_change(struct diff_options *options,
 static int rev_compare_tree(struct rev_info *revs,
 			    struct commit *parent, struct commit *commit)
 {
-	struct tree *t1 = parent->tree;
-	struct tree *t2 = commit->tree;
+	struct tree *t1 = parent->maybe_tree;
+	struct tree *t2 = commit->maybe_tree;
 
 	if (!t1)
 		return REV_TREE_NEW;
@@ -476,7 +476,7 @@ static int rev_compare_tree(struct rev_info *revs,
 static int rev_same_tree_as_empty(struct rev_info *revs, struct commit *commit)
 {
 	int retval;
-	struct tree *t1 = commit->tree;
+	struct tree *t1 = commit->maybe_tree;
 
 	if (!t1)
 		return 0;
@@ -614,7 +614,7 @@ static void try_to_simplify_commit(struct rev_info *revs, struct commit *commit)
 	if (!revs->prune)
 		return;
 
-	if (!commit->tree)
+	if (!commit->maybe_tree)
 		return;
 
 	if (!commit->parents) {
diff --git a/sequencer.c b/sequencer.c
diff --git a/sha1_name.c b/sha1_name.c
diff --git a/tree.c b/tree.c
diff --git a/walker.c b/walker.c

Original file line number	Diff line number	Diff line change
`@@ -153,7 +153,7 @@ static int commit_is_complete(struct commit *commit)`
`153`	`153`	`for (i = 0; i < found.nr; i++) {`
`154`	`154`	`struct commit *c =`
`155`	`155`	`(struct commit *)found.objects[i].item;`
`156`		`- if (!tree_is_complete(&c->tree->object.oid)) {`
	`156`	`+ if (!tree_is_complete(&c->maybe_tree->object.oid)) {`
`157`	`157`	`is_incomplete = 1;`
`158`	`158`	`c->object.flags \|= INCOMPLETE;`
`159`	`159`	`}`