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

[PATCH 05/15] repack_without_ref(): reimplement using do_for_each_ref_in_array()

From
mhagger@alum.mit.edu <mhagger@alum.mit.edu>
Date
Apr 10, 2012, 05:30 UTC
Message-ID
<1334035827-20331-6-git-send-email-mhagger@alum.mit.edu>
In-Reply-To
<1334035827-20331-1-git-send-email-mhagger@alum.mit.edu>
From: Michael Haggerty <mhagger@alum.mit.edu>

It costs a bit of boilerplate, but it means that the function can be ignorant of how cached refs are stored.

Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
---
 refs.c |   51 ++++++++++++++++++++++++++++++---------------------
 1 file changed, 30 insertions(+), 21 deletions(-)
diff --git a/refs.c b/refs.c
index f9a1b57..78dbda6 100644
--- a/refs.c
+++ b/refs.c
@@ -1333,36 +1333,45 @@ struct ref_lock *lock_any_ref_for_update(const char *refname,
 	return lock_ref_sha1_basic(refname, old_sha1, flags, NULL);
 }
 
+struct repack_without_ref_sb {
+	const char *refname;
+	int fd;
+};
+
+static int repack_without_ref_fn(const char *refname, const unsigned char *sha1,
+				 int flags, void *cb_data)
+{
+	struct repack_without_ref_sb *data = cb_data;
+	char line[PATH_MAX + 100];
+	int len;
+
+	if (!strcmp(data->refname, refname))
+		return 0;
+	len = snprintf(line, sizeof(line), "%s %s\n",
+		       sha1_to_hex(sha1), refname);
+	/* this should not happen but just being defensive */
+	if (len > sizeof(line))
+		die("too long a refname '%s'", refname);
+	write_or_die(data->fd, line, len);
+	return 0;
+}
+
 static struct lock_file packlock;
 
 static int repack_without_ref(const char *refname)
 {
-	struct ref_array *packed;
-	int fd, i;
-
-	packed = get_packed_refs(get_ref_cache(NULL));
+	struct repack_without_ref_sb data;
+	struct ref_array *packed = get_packed_refs(get_ref_cache(NULL));
+	sort_ref_array(packed);
 	if (search_ref_array(packed, refname) == NULL)
 		return 0;
-	fd = hold_lock_file_for_update(&packlock, git_path("packed-refs"), 0);
-	if (fd < 0) {
+	data.refname = refname;
+	data.fd = hold_lock_file_for_update(&packlock, git_path("packed-refs"), 0);
+	if (data.fd < 0) {
 		unable_to_lock_error(git_path("packed-refs"), errno);
 		return error("cannot delete '%s' from packed refs", refname);
 	}
-
-	for (i = 0; i < packed->nr; i++) {
-		char line[PATH_MAX + 100];
-		int len;
-		struct ref_entry *ref = packed->refs[i];
-
-		if (!strcmp(refname, ref->name))
-			continue;
-		len = snprintf(line, sizeof(line), "%s %s\n",
-			       sha1_to_hex(ref->sha1), ref->name);
-		/* this should not happen but just being defensive */
-		if (len > sizeof(line))
-			die("too long a refname '%s'", ref->name);
-		write_or_die(fd, line, len);
-	}
+	do_for_each_ref_in_array(packed, 0, "", repack_without_ref_fn, 0, 0, &data);
 	return commit_lock_file(&packlock);
 }
 
-- 
1.7.10
Previous: mhagger@alum.mit.eduNext: mhagger@alum.mit.edu
Message 6 of 18 in “Hierarchical reference cache (once again)”
  1. 00/15 Hierarchical reference cache (once again)mhagger@alum.mit.edu, Apr 10, 2012
  2. 01/15 refs.c: reorder definitions more logicallymhagger@alum.mit.edu, Apr 10, 2012
  3. 02/15 refs: manage current_ref within do_one_ref()mhagger@alum.mit.edu, Apr 10, 2012
  4. 03/15 do_for_each_ref_in_array(): new functionmhagger@alum.mit.edu, Apr 10, 2012
  5. 04/15 do_for_each_ref_in_arrays(): new functionmhagger@alum.mit.edu, Apr 10, 2012
  6. 05/15 repack_without_ref(): reimplement using do_for_each_ref_in_array()mhagger@alum.mit.edu, Apr 10, 2012
  7. 06/15 names_conflict(): new function, extracted from is_refname_available()mhagger@alum.mit.edu, Apr 10, 2012
  8. 07/15 names_conflict(): simplify implementationmhagger@alum.mit.edu, Apr 10, 2012
  9. 08/15 is_refname_available(): reimplement using do_for_each_ref_in_array()mhagger@alum.mit.edu, Apr 10, 2012
  10. 09/15 free_ref_entry(): new functionmhagger@alum.mit.edu, Apr 10, 2012
  11. 10/15 check_refname_component(): return 0 for zero-length componentsmhagger@alum.mit.edu, Apr 10, 2012
  12. 11/15 struct ref_entry: nest the value part in a unionmhagger@alum.mit.edu, Apr 10, 2012
  13. 12/15 refs.c: rename ref_array -> ref_dirmhagger@alum.mit.edu, Apr 10, 2012
  14. 13/15 sort_ref_dir(): simplify logicmhagger@alum.mit.edu, Apr 10, 2012
  15. 14/15 refs: store references hierarchicallymhagger@alum.mit.edu, Apr 10, 2012
  16. 15/15 do_for_each_ref(): only iterate over the subtree that was requestedmhagger@alum.mit.edu, Apr 10, 2012
  17. Jeff KingApr 12, 2012
  18. Junio C HamanoApr 12, 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.