git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH 3/4] pathspec: apply "*.c" optimization from exclude

From
Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Date
Nov 18, 2012, 09:13 UTC
Message-ID
<1353229989-13075-4-git-send-email-pclouds@gmail.com>
In-Reply-To
<1353229989-13075-1-git-send-email-pclouds@gmail.com>
-O2 build on linux-2.6, without the patch:
$ time git rev-list --quiet HEAD -- '*.c'

real 0m40.770s user 0m40.290s sys 0m0.256s

With the patch
$ time ~/w/git/git rev-list --quiet HEAD -- '*.c'

real 0m34.288s user 0m33.997s sys 0m0.205s

The above command is not supposed to be widely popular. It's chosen because it exercises pathspec matching a lot. The point is it cuts down matching time for popular patterns like *.c, which could be used as pathspec in other places.

Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
---
 cache.h     |  3 +++
 dir.c       | 17 +++++++++++++++--
 dir.h       |  1 +
 tree-walk.c |  6 ++++--
 4 files changed, 23 insertions(+), 4 deletions(-)
diff --git a/cache.h b/cache.h
index bf031f1..d18f584 100644
--- a/cache.h
+++ b/cache.h
@@ -473,6 +473,8 @@ extern int index_name_is_other(const struct index_state *, const char *, int);
 extern int ie_match_stat(const struct index_state *, struct cache_entry *, struct stat *, unsigned int);
 extern int ie_modified(const struct index_state *, struct cache_entry *, struct stat *, unsigned int);
 
+#define PSF_ONESTAR 1
+
 struct pathspec {
 	const char **raw; /* get_pathspec() result, not freed by free_pathspec() */
 	int nr;
@@ -483,6 +485,7 @@ struct pathspec {
 		const char *match;
 		int len;
 		int nowildcard_len;
+		int flags;
 	} *items;
 };
 
diff --git a/dir.c b/dir.c
index e4e6ca1..d00f240 100644
--- a/dir.c
+++ b/dir.c
@@ -46,6 +46,12 @@ inline int git_fnmatch(const char *pattern, const char *string,
 		pattern += prefix;
 		string += prefix;
 	}
+	if (flags & GF_ONESTAR) {
+		int pattern_len = strlen(++pattern);
+		int string_len = strlen(string);
+		return strcmp(pattern,
+			      string + string_len - pattern_len);
+	}
 	return fnmatch(pattern, string, fnm_flags);
 }
 
@@ -246,7 +252,9 @@ static int match_pathspec_item(const struct pathspec_item *item, int prefix,
 	}
 
 	if (item->nowildcard_len < item->len &&
-	    !git_fnmatch(match, name, 0, item->nowildcard_len - prefix))
+	    !git_fnmatch(match, name,
+			 item->flags & PSF_ONESTAR ? GF_ONESTAR : 0,
+			 item->nowildcard_len - prefix))
 		return MATCHED_FNMATCH;
 
 	return 0;
@@ -1446,8 +1454,13 @@ int init_pathspec(struct pathspec *pathspec, const char **paths)
 		item->match = path;
 		item->len = strlen(path);
 		item->nowildcard_len = simple_length(path);
-		if (item->nowildcard_len < item->len)
+		item->flags = 0;
+		if (item->nowildcard_len < item->len) {
 			pathspec->has_wildcard = 1;
+			if (path[item->nowildcard_len] == '*' &&
+			    no_wildcard(path + item->nowildcard_len + 1))
+				item->flags |= PSF_ONESTAR;
+		}
 	}
 
 	qsort(pathspec->items, pathspec->nr,
diff --git a/dir.h b/dir.h
index 4cd5074..590532b 100644
--- a/dir.h
+++ b/dir.h
@@ -143,6 +143,7 @@ extern int fnmatch_icase(const char *pattern, const char *string, int flags);
  * The prefix part of pattern must not contains wildcards.
  */
 #define GF_PATHNAME 1
+#define GF_ONESTAR  2
 
 extern int git_fnmatch(const char *pattern, const char *string,
 		       int flags, int prefix);
diff --git a/tree-walk.c b/tree-walk.c
index 2fcf3c0..42fe610 100644
--- a/tree-walk.c
+++ b/tree-walk.c
@@ -628,7 +628,8 @@ enum interesting tree_entry_interesting(const struct name_entry *entry,
 
 			if (item->nowildcard_len < item->len) {
 				if (!git_fnmatch(match + baselen, entry->path,
-						 0, item->nowildcard_len - baselen))
+						 item->flags & PSF_ONESTAR ? GF_ONESTAR : 0,
+						 item->nowildcard_len - baselen))
 					return entry_interesting;
 
 				/*
@@ -654,7 +655,8 @@ match_wildcards:
 		strbuf_add(base, entry->path, pathlen);
 
 		if (!git_fnmatch(match, base->buf + base_offset,
-				 0, item->nowildcard_len)) {
+				 item->flags & PSF_ONESTAR ? GF_ONESTAR : 0,
+				 item->nowildcard_len)) {
 			strbuf_setlen(base, base_offset + baselen);
 			return entry_interesting;
 		}
-- 
1.8.0.rc2.23.g1fb49df
Previous: Nguyen Thai Ngoc DuyNext: Junio C Hamano
Message 7 of 11 in “Some pathspec wildcard optimization”
  1. 0/4 Some pathspec wildcard optimizationNguyễn Thái Ngọc Duy, Nov 18, 2012
  2. 1/4 pathspec: save the non-wildcard length partNguyễn Thái Ngọc Duy, Nov 18, 2012
  3. Junio C HamanoNov 19, 2012
  4. 2/4 pathspec: do exact comparison on the leading non-wildcard partNguyễn Thái Ngọc Duy, Nov 18, 2012
  5. Junio C HamanoNov 19, 2012
  6. Nguyen Thai Ngoc DuyNov 20, 2012
  7. 3/4 pathspec: apply "*.c" optimization from excludeNguyễn Thái Ngọc Duy, Nov 18, 2012
  8. Junio C HamanoNov 19, 2012
  9. Nguyen Thai Ngoc DuyNov 20, 2012
  10. Junio C HamanoNov 20, 2012
  11. 4/4 tree_entry_interesting: do basedir compare on wildcard patterns when possibleNguyễn Thái Ngọc Duy, Nov 18, 2012

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.