merge-ort, diffcore-rename: tweak dirs_removed and relevant_source type

A future change will want to take advantage of being able to store a bit
more information about removed directories and relevant sources.  Since
strset does not allow storing any data, and since we only need to store
an enum value for each, switch these over to using strintmap.  No
functional changes yet.

Signed-off-by: Elijah Newren <[email protected]>

Author: newren

diffcore-rename.c

@@ -371,7 +371,7 @@ struct dir_rename_info {
 	struct strintmap idx_map;
 	struct strmap dir_rename_guess;
 	struct strmap *dir_rename_count;
-	struct strset *relevant_source_dirs;
+	struct strintmap *relevant_source_dirs;
 	unsigned setup;
 };
 
@@ -429,7 +429,7 @@ static void increment_count(struct dir_rename_info *info,
 }
 
 static void update_dir_rename_counts(struct dir_rename_info *info,
-				     struct strset *dirs_removed,
+				     struct strintmap *dirs_removed,
 				     const char *oldname,
 				     const char *newname)
 {
@@ -458,7 +458,7 @@ static void update_dir_rename_counts(struct dir_rename_info *info,
 		/* Get old_dir, skip if its directory isn't relevant. */
 		dirname_munge(old_dir);
 		if (info->relevant_source_dirs &&
-		    !strset_contains(info->relevant_source_dirs, old_dir))
+		    !strintmap_contains(info->relevant_source_dirs, old_dir))
 			break;
 
 		/* Get new_dir */
@@ -503,7 +503,7 @@ static void update_dir_rename_counts(struct dir_rename_info *info,
 			}
 		}
 
-		if (strset_contains(dirs_removed, old_dir))
+		if (strintmap_contains(dirs_removed, old_dir))
 			increment_count(info, old_dir, new_dir);
 		else
 			break;
@@ -521,8 +521,8 @@ static void update_dir_rename_counts(struct dir_rename_info *info,
 }
 
 static void initialize_dir_rename_info(struct dir_rename_info *info,
-				       struct strset *relevant_sources,
-				       struct strset *dirs_removed,
+				       struct strintmap *relevant_sources,
+				       struct strintmap *dirs_removed,
 				       struct strmap *dir_rename_count)
 {
 	struct hashmap_iter iter;
@@ -548,12 +548,13 @@ static void initialize_dir_rename_info(struct dir_rename_info *info,
 		info->relevant_source_dirs = dirs_removed; /* might be NULL */
 	} else {
 		info->relevant_source_dirs = xmalloc(sizeof(struct strintmap));
-		strset_init(info->relevant_source_dirs);
-		strset_for_each_entry(relevant_sources, &iter, entry) {
+		strintmap_init(info->relevant_source_dirs, 0 /* unused */);
+		strintmap_for_each_entry(relevant_sources, &iter, entry) {
 			char *dirname = get_dirname(entry->key);
 			if (!dirs_removed ||
-			    strset_contains(dirs_removed, dirname))
-				strset_add(info->relevant_source_dirs, dirname);
+			    strintmap_contains(dirs_removed, dirname))
+				strintmap_set(info->relevant_source_dirs,
+					      dirname, 0 /* value irrelevant */);
 			free(dirname);
 		}
 	}
@@ -617,7 +618,7 @@ void partial_clear_dir_rename_count(struct strmap *dir_rename_count)
 }
 
 static void cleanup_dir_rename_info(struct dir_rename_info *info,
-				    struct strset *dirs_removed,
+				    struct strintmap *dirs_removed,
 				    int keep_dir_rename_count)
 {
 	struct hashmap_iter iter;
@@ -635,7 +636,7 @@ static void cleanup_dir_rename_info(struct dir_rename_info *info,
 	/* relevant_source_dirs */
 	if (info->relevant_source_dirs &&
 	    info->relevant_source_dirs != dirs_removed) {
-		strset_clear(info->relevant_source_dirs);
+		strintmap_clear(info->relevant_source_dirs);
 		FREE_AND_NULL(info->relevant_source_dirs);
 	}
 
@@ -657,7 +658,7 @@ static void cleanup_dir_rename_info(struct dir_rename_info *info,
 			const char *source_dir = entry->key;
 			struct strintmap *counts = entry->value;
 
-			if (!strset_contains(dirs_removed, source_dir)) {
+			if (!strintmap_contains(dirs_removed, source_dir)) {
 				string_list_append(&to_remove, source_dir);
 				strintmap_clear(counts);
 				continue;
@@ -747,8 +748,8 @@ static int find_basename_matches(struct diff_options *options,
 				 int minimum_score,
 				 int num_src,
 				 struct dir_rename_info *info,
-				 struct strset *relevant_sources,
-				 struct strset *dirs_removed)
+				 struct strintmap *relevant_sources,
+				 struct strintmap *dirs_removed)
 {
 	/*
 	 * When I checked in early 2020, over 76% of file renames in linux
@@ -837,7 +838,7 @@ static int find_basename_matches(struct diff_options *options,
 
 		/* Skip irrelevant sources */
 		if (relevant_sources &&
-		    !strset_contains(relevant_sources, filename))
+		    !strintmap_contains(relevant_sources, filename))
 			continue;
 
 		/* Find out if this basename is unique among sources */
@@ -963,7 +964,7 @@ static int find_renames(struct diff_score *mx,
 			int minimum_score,
 			int copies,
 			struct dir_rename_info *info,
-			struct strset *dirs_removed)
+			struct strintmap *dirs_removed)
 {
 	int count = 0, i;
 
@@ -988,7 +989,7 @@ static int find_renames(struct diff_score *mx,
 }
 
 static void remove_unneeded_paths_from_src(int detecting_copies,
-					   struct strset *interesting)
+					   struct strintmap *interesting)
 {
 	int i, new_num_src;
 
@@ -1030,7 +1031,7 @@ static void remove_unneeded_paths_from_src(int detecting_copies,
 			continue;
 
 		/* If we don't care about the source path, skip it */
-		if (interesting && !strset_contains(interesting, one->path))
+		if (interesting && !strintmap_contains(interesting, one->path))
 			continue;
 
 		if (new_num_src < i)
@@ -1043,8 +1044,8 @@ static void remove_unneeded_paths_from_src(int detecting_copies,
 }
 
 void diffcore_rename_extended(struct diff_options *options,
-			      struct strset *relevant_sources,
-			      struct strset *dirs_removed,
+			      struct strintmap *relevant_sources,
+			      struct strintmap *dirs_removed,
 			      struct strmap *dir_rename_count)
 {
 	int detect_rename = options->detect_rename;

diffcore.h

@@ -8,8 +8,8 @@
 
 struct diff_options;
 struct repository;
+struct strintmap;
 struct strmap;
-struct strset;
 struct userdiff_driver;
 
 /* This header file is internal between diff.c and its diff transformers
@@ -166,8 +166,8 @@ void partial_clear_dir_rename_count(struct strmap *dir_rename_count);
 void diffcore_break(struct repository *, int);
 void diffcore_rename(struct diff_options *);
 void diffcore_rename_extended(struct diff_options *options,
-			      struct strset *relevant_sources,
-			      struct strset *dirs_removed,
+			      struct strintmap *relevant_sources,
+			      struct strintmap *dirs_removed,
 			      struct strmap *dir_rename_count);
 void diffcore_merge_broken(void);
 void diffcore_pickaxe(struct diff_options *);

merge-ort.c

@@ -74,7 +74,7 @@ struct rename_info {
 	/*
 	 * dirs_removed: directories removed on a given side of history.
 	 */
-	struct strset dirs_removed[3];
+	struct strintmap dirs_removed[3];
 
 	/*
 	 * dir_rename_count: tracking where parts of a directory were renamed to
@@ -106,7 +106,7 @@ struct rename_info {
 	 * If neither of those are true, we can skip rename detection for
 	 * that path.
 	 */
-	struct strset relevant_sources[3];
+	struct strintmap relevant_sources[3];
 
 	/*
 	 * dir_rename_mask:
@@ -362,8 +362,8 @@ static void clear_or_reinit_internal_opts(struct merge_options_internal *opti,
 	int i;
 	void (*strmap_func)(struct strmap *, int) =
 		reinitialize ? strmap_partial_clear : strmap_clear;
-	void (*strset_func)(struct strset *) =
-		reinitialize ? strset_partial_clear : strset_clear;
+	void (*strintmap_func)(struct strintmap *) =
+		reinitialize ? strintmap_partial_clear : strintmap_clear;
 
 	/*
 	 * We marked opti->paths with strdup_strings = 0, so that we
@@ -395,15 +395,15 @@ static void clear_or_reinit_internal_opts(struct merge_options_internal *opti,
 
 	/* Free memory used by various renames maps */
 	for (i = MERGE_SIDE1; i <= MERGE_SIDE2; ++i) {
-		strset_func(&renames->dirs_removed[i]);
+		strintmap_func(&renames->dirs_removed[i]);
 
 		partial_clear_dir_rename_count(&renames->dir_rename_count[i]);
 		if (!reinitialize)
 			strmap_clear(&renames->dir_rename_count[i], 1);
 
 		strmap_func(&renames->dir_renames[i], 0);
 
-		strset_func(&renames->relevant_sources[i]);
+		strintmap_func(&renames->relevant_sources[i]);
 	}
 
 	if (!reinitialize) {
@@ -675,7 +675,7 @@ static void add_pair(struct merge_options *opt,
 		unsigned location_relevant = (dir_rename_mask == 0x07);
 
 		if (content_relevant || location_relevant)
-			strset_add(&renames->relevant_sources[side], pathname);
+			strintmap_set(&renames->relevant_sources[side], pathname, 1);
 	}
 
 	one = alloc_filespec(pathname);
@@ -731,9 +731,9 @@ static void collect_rename_info(struct merge_options *opt,
 		/* absent_mask = 0x07 - dirmask; sides = absent_mask/2 */
 		unsigned sides = (0x07 - dirmask)/2;
 		if (sides & 1)
-			strset_add(&renames->dirs_removed[1], fullname);
+			strintmap_set(&renames->dirs_removed[1], fullname, 1);
 		if (sides & 2)
-			strset_add(&renames->dirs_removed[2], fullname);
+			strintmap_set(&renames->dirs_removed[2], fullname, 1);
 	}
 
 	if (filemask == 0 || filemask == 7)
@@ -2162,7 +2162,7 @@ static inline int possible_side_renames(struct rename_info *renames,
 					unsigned side_index)
 {
 	return renames->pairs[side_index].nr > 0 &&
-	       !strset_empty(&renames->relevant_sources[side_index]);
+	       !strintmap_empty(&renames->relevant_sources[side_index]);
 }
 
 static inline int possible_renames(struct rename_info *renames)
@@ -3447,14 +3447,14 @@ static void merge_start(struct merge_options *opt, struct merge_result *result)
 	/* Initialization of various renames fields */
 	renames = &opt->priv->renames;
 	for (i = MERGE_SIDE1; i <= MERGE_SIDE2; i++) {
-		strset_init_with_options(&renames->dirs_removed[i],
-					 NULL, 0);
+		strintmap_init_with_options(&renames->dirs_removed[i],
+					    0, NULL, 0);
 		strmap_init_with_options(&renames->dir_rename_count[i],
 					 NULL, 1);
 		strmap_init_with_options(&renames->dir_renames[i],
 					 NULL, 0);
-		strset_init_with_options(&renames->relevant_sources[i],
-					 NULL, 0);
+		strintmap_init_with_options(&renames->relevant_sources[i],
+					    0, NULL, 0);
 	}
 
 	/*