Re: Question about 'branch -d' safety

From: Jonathan Nieder <jrnieder@gmail.com>
Date: Jul 10, 2010, 22:17 UTC
Message-ID: <20100710221724.GA3891@burratino>
In-Reply-To: <m3lj9jknlr.fsf@localhost.localdomain>

Jakub Narebski wrote:

> Jonathan Nieder <jrnieder@gmail.com> writes:

Show 14 quoted lines

>> +++ b/refs.c
>> @@ -1084,7 +1084,6 @@ int delete_ref(const char *refname, const unsigned char *sha1, int delopt)
>>  	 */
>>  	ret |= repack_without_ref(refname);
>>  
>> -	unlink_or_warn(git_path("logs/%s", lock->ref_name));
>>  	invalidate_cached_refs();
>>  	unlock_ref(lock);
>>  	return ret;
>
> It's unfortunately not so easy.  The problem you have to solve is D/F
> conflict: if you have 'foo/bar' branch, you can't create 'foo' branch,
> but after deleting 'foo/bar' you want to be able to create 'foo'
> branch and reflog for 'foo' branch.

Thanks for the pointer.  Here are some more, for the interested.

 - http://thread.gmane.org/gmane.comp.version-control.git/144250/focus=145366
   the dead_refs namespace

 - http://thread.gmane.org/gmane.comp.version-control.git/73838/focus=73877 
   why to make and why not to make the change