Merge branch 'mh/notes-string-list'

Improve the asymptotic performance of the cat_sort_uniq notes merge
strategy.

* mh/notes-string-list:
  string_list_add_refs_from_colon_sep(): use string_list_split()
  notes: fix handling of colon-separated values
  combine_notes_cat_sort_uniq(): sort and dedup lines all at once
  Initialize sort_uniq_list using named constant
  string_list: add a function string_list_remove_empty_items()

Documentation/technical/api-string-list.txt

@@ -38,7 +38,8 @@ member (you need this if you add things later) and you should set the
   `unsorted_string_list_delete_item`.
 
 . Can remove items not matching a criterion from a sorted or unsorted
-  list using `filter_string_list`.
+  list using `filter_string_list`, or remove empty strings using
+  `string_list_remove_empty_items`.
 
 . Finally it should free the list using `string_list_clear`.
 
@@ -75,6 +76,12 @@ Functions
 	to be deleted.  Preserve the order of the items that are
 	retained.
 
+`string_list_remove_empty_items`::
+
+	Remove any empty strings from the list.  If free_util is true,
+	call free() on the util members of any items that have to be
+	deleted.  Preserve the order of the items that are retained.
+
 `string_list_longest_prefix`::
 
 	Return the longest string within a string_list that is a

notes.c

@@ -848,15 +848,16 @@ int combine_notes_ignore(unsigned char *cur_sha1,
 	return 0;
 }
 
-static int string_list_add_note_lines(struct string_list *sort_uniq_list,
+/*
+ * Add the lines from the named object to list, with trailing
+ * newlines removed.
+ */
+static int string_list_add_note_lines(struct string_list *list,
 				      const unsigned char *sha1)
 {
 	char *data;
 	unsigned long len;
 	enum object_type t;
-	struct strbuf buf = STRBUF_INIT;
-	struct strbuf **lines = NULL;
-	int i, list_index;
 
 	if (is_null_sha1(sha1))
 		return 0;
@@ -868,24 +869,14 @@ static int string_list_add_note_lines(struct string_list *sort_uniq_list,
 		return t != OBJ_BLOB || !data;
 	}
 
-	strbuf_attach(&buf, data, len, len + 1);
+	/*
-	lines = strbuf_split(&buf, '\n');
+	 * If the last line of the file is EOL-terminated, this will
-
+	 * add an empty string to the list.  But it will be removed
-	for (i = 0; lines[i]; i++) {
+	 * later, along with any empty strings that came from empty
-		if (lines[i]->buf[lines[i]->len - 1] == '\n')
+	 * lines within the file.
-			strbuf_setlen(lines[i], lines[i]->len - 1);
+	 */
-		if (!lines[i]->len)
+	string_list_split(list, data, '\n', -1);
-			continue; /* skip empty lines */
+	free(data);
-		list_index = string_list_find_insert_index(sort_uniq_list,
-							   lines[i]->buf, 0);
-		if (list_index < 0)
-			continue; /* skip duplicate lines */
-		string_list_insert_at_index(sort_uniq_list, list_index,
-					    lines[i]->buf);
-	}
-
-	strbuf_list_free(lines);
-	strbuf_release(&buf);
 	return 0;
 }
 
@@ -901,7 +892,7 @@ static int string_list_join_lines_helper(struct string_list_item *item,
 int combine_notes_cat_sort_uniq(unsigned char *cur_sha1,
 		const unsigned char *new_sha1)
 {
-	struct string_list sort_uniq_list = { NULL, 0, 0, 1 };
+	struct string_list sort_uniq_list = STRING_LIST_INIT_DUP;
 	struct strbuf buf = STRBUF_INIT;
 	int ret = 1;
 
@@ -910,6 +901,9 @@ int combine_notes_cat_sort_uniq(unsigned char *cur_sha1,
 		goto out;
 	if (string_list_add_note_lines(&sort_uniq_list, new_sha1))
 		goto out;
+	string_list_remove_empty_items(&sort_uniq_list, 0);
+	sort_string_list(&sort_uniq_list);
+	string_list_remove_duplicates(&sort_uniq_list, 0);
 
 	/* create a new blob object from sort_uniq_list */
 	if (for_each_string_list(&sort_uniq_list,
@@ -949,23 +943,18 @@ void string_list_add_refs_by_glob(struct string_list *list, const char *glob)
 void string_list_add_refs_from_colon_sep(struct string_list *list,
 					 const char *globs)
 {
-	struct strbuf globbuf = STRBUF_INIT;
+	struct string_list split = STRING_LIST_INIT_NODUP;
-	struct strbuf **split;
+	char *globs_copy = xstrdup(globs);
 	int i;
 
-	strbuf_addstr(&globbuf, globs);
+	string_list_split_in_place(&split, globs_copy, ':', -1);
-	split = strbuf_split(&globbuf, ':');
+	string_list_remove_empty_items(&split, 0);
 
-	for (i = 0; split[i]; i++) {
+	for (i = 0; i < split.nr; i++)
-		if (!split[i]->len)
+		string_list_add_refs_by_glob(list, split.items[i].string);
-			continue;
-		if (split[i]->buf[split[i]->len-1] == ':')
-			strbuf_setlen(split[i], split[i]->len-1);
-		string_list_add_refs_by_glob(list, split[i]->buf);
-	}
 
-	strbuf_list_free(split);
+	string_list_clear(&split, 0);
-	strbuf_release(&globbuf);
+	free(globs_copy);
 }
 
 static int notes_display_config(const char *k, const char *v, void *cb)

string-list.c

@@ -136,6 +136,15 @@ void filter_string_list(struct string_list *list, int free_util,
 	list->nr = dst;
 }
 
+static int item_is_not_empty(struct string_list_item *item, void *unused)
+{
+	return *item->string != '\0';
+}
+
+void string_list_remove_empty_items(struct string_list *list, int free_util) {
+	filter_string_list(list, free_util, item_is_not_empty, NULL);
+}
+
 char *string_list_longest_prefix(const struct string_list *prefixes,
 				 const char *string)
 {

string-list.h

@@ -38,6 +38,13 @@ int for_each_string_list(struct string_list *list,
 void filter_string_list(struct string_list *list, int free_util,
 			string_list_each_func_t want, void *cb_data);
 
+/*
+ * Remove any empty strings from the list.  If free_util is true, call
+ * free() on the util members of any items that have to be deleted.
+ * Preserve the order of the items that are retained.
+ */
+void string_list_remove_empty_items(struct string_list *list, int free_util);
+
 /*
  * Return the longest string in prefixes that is a prefix (in the
  * sense of prefixcmp()) of string, or NULL if no such prefix exists.