git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox.com>
To: Miklos Vajna <vmiklos@frugalware.org>
Cc: Jeff King <peff@peff.net>, Brandon Casey <casey@nrlssc.navy.mil>,
	git@vger.kernel.org
Subject: Re: [PATCH 1/2] Fix git branch -m for symrefs.
Date: Sat, 25 Oct 2008 11:31:01 -0700	[thread overview]
Message-ID: <7v8wsca5ne.fsf@gitster.siamese.dyndns.org> (raw)
In-Reply-To: <a96243124c555cbc4081f733b348252ac200bd53.1224939436.git.vmiklos@frugalware.org> (Miklos Vajna's message of "Sat, 25 Oct 2008 14:58:41 +0200")

Miklos Vajna <vmiklos@frugalware.org> writes:

> -int delete_ref(const char *refname, const unsigned char *sha1)
> +int delete_ref(const char *refname, const unsigned char *sha1, int flags)
>  {
>  	struct ref_lock *lock;
> -	int err, i, ret = 0, flag = 0;
> +	int err, i = 0, ret = 0, flag = 0;
> +	char *path;
>  
>  	lock = lock_ref_sha1_basic(refname, sha1, 0, &flag);
>  	if (!lock)
>  		return 1;
>  	if (!(flag & REF_ISPACKED)) {

Two variables flag vs flags is a bit confusing, isn't it?  How about
naming the new one "delopt" or something?

The new variable "char *path" at the toplevel can be confined in the scope
of this if () {} block and probably can become "const char *", right?

> @@ -964,10 +971,15 @@ int rename_ref(const char *oldref, const char *newref, const char *logmsg)
>  	struct ref_lock *lock;
>  	struct stat loginfo;
>  	int log = !lstat(git_path("logs/%s", oldref), &loginfo);
> +	const char *symref = NULL;
> +	int is_symref = 0;
>  
>  	if (S_ISLNK(loginfo.st_mode))
>  		return error("reflog for %s is a symlink", oldref);

Possible bug in the context.  When there is no reflog for the ref being
renamed, lstat would fail; it doesn't feel right to have this S_ISLNK()
before checking the result of the lstat which is in "log".

> +	symref = resolve_ref(oldref, orig_sha1, 0, &flag);
> +	if (flag & REF_ISSYMREF)
> +		is_symref = 1;
>  	if (!resolve_ref(oldref, orig_sha1, 1, &flag))
>  		return error("refname %s not found", oldref);

Do we really need two calls to resolve_ref()?  Your new call calls it
without must-exist bit --- why?  Immediately after that, the existing call
will barf if it does not exist anyway.

I agree it is good to have symref aware delete_ref(), but I am not sure
supporting symref in rename_ref() is either needed or necessarily a good
idea.  You also need to worry about a symref pointing at a branch yet to
be born.

In the meantime, I think we should just check (flag & REF_ISSYMREF) after
the existing resolve_ref() we can see in the context above, and error out
saying you cannot rename a symref, and do nothing else.

  parent reply	other threads:[~2008-10-25 18:32 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-10-22  0:23 [PATCH] Implement git remote mv Miklos Vajna
2008-10-22 16:52 ` Brandon Casey
2008-10-23  1:18   ` Miklos Vajna
2008-10-23  3:52     ` Jeff King
2008-10-23 12:56       ` [PATCH] Implement git remote rename Miklos Vajna
2008-10-24 23:33         ` Junio C Hamano
2008-10-25 12:58           ` [PATCH 0/2] Fixes for git branch -m / update-ref --no-deref -d Miklos Vajna
2008-10-25 12:58             ` [PATCH 1/2] Fix git branch -m for symrefs Miklos Vajna
2008-10-25 12:58               ` [PATCH 2/2] Fix git update-ref --no-deref -d Miklos Vajna
2008-10-25 18:31               ` Junio C Hamano [this message]
2008-10-26  2:33                 ` [PATCH 0/3] symref rename/delete fixes Miklos Vajna
2008-10-26  2:33                   ` [PATCH 1/3] Fix git branch -m for symrefs Miklos Vajna
2008-10-26  2:33                     ` [PATCH 2/3] rename_ref(): handle the case when the reflog of a ref does not exist Miklos Vajna
2008-10-26  2:33                       ` [PATCH 3/3] Fix git update-ref --no-deref -d Miklos Vajna
2008-10-27  5:31                   ` [PATCH 0/3] symref rename/delete fixes Junio C Hamano
2008-10-27  8:50                     ` Miklos Vajna
2008-10-27 19:50                       ` Miklos Vajna
2008-10-27 19:50                         ` [PATCH 1/3] Disallow git branch -m for symrefs Miklos Vajna
2008-10-27 19:50                           ` [PATCH 2/3] rename_ref(): handle the case when the reflog of a ref does not exist Miklos Vajna
2008-10-27 19:50                             ` [PATCH 3/3] Fix git update-ref --no-deref -d Miklos Vajna
2008-10-28 23:45                         ` [PATCH 0/3] symref rename/delete fixes Miklos Vajna
2008-10-29  0:05                           ` [PATCH] git branch -m: forbid renaming of a symref Miklos Vajna
2008-11-03 18:26           ` [PATCH] Implement git remote rename Miklos Vajna
2008-11-10 20:42           ` Miklos Vajna
2008-11-10 20:43             ` [PATCH 1/4] remote: add a new 'origin' variable to the struct Miklos Vajna
2008-11-10 20:43               ` [PATCH 2/4] git-remote rename: support remotes->config migration Miklos Vajna
2008-11-10 20:43                 ` [PATCH 3/4] git-remote rename: support branches->config migration Miklos Vajna
2008-11-10 20:43                   ` [PATCH 4/4] git-remote: document the migration feature of the rename subcommand Miklos Vajna
2008-11-12  0:49                   ` [PATCH 3/4] git-remote rename: support branches->config migration Junio C Hamano
2008-11-12  2:01                     ` Miklos Vajna
2008-11-12  4:22                       ` Junio C Hamano
2008-11-12 17:11                         ` [PATCH v2 0/4] " Miklos Vajna
2008-11-12 17:11                           ` [PATCH 1/4] remote: add a new 'origin' variable to the struct Miklos Vajna
2008-11-12 17:11                             ` [PATCH 2/4] git-remote rename: support remotes->config migration Miklos Vajna
2008-11-12 17:11                               ` [PATCH 3/4] git-remote rename: support branches->config migration Miklos Vajna
2008-11-12 17:11                                 ` [PATCH 4/4] git-remote: document the migration feature of the rename subcommand Miklos Vajna

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=7v8wsca5ne.fsf@gitster.siamese.dyndns.org \
    --to=gitster@pobox.com \
    --cc=casey@nrlssc.navy.mil \
    --cc=git@vger.kernel.org \
    --cc=peff@peff.net \
    --cc=vmiklos@frugalware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).