Re: [PATCH] [SIGNED-OFF] remotes-hg: bugfix for fetching non local remotes

From: Junio C Hamano <gitster@pobox.com>
Date: Jul 24, 2013, 22:51 UTC
Message-ID: <7vhafj606p.fsf@alter.siamese.dyndns.org>
In-Reply-To: <665576E7-6083-4535-9CE0-773236A0E9ED@joernhees.de>

Jörn Hees <dev@joernhees.de> writes:

Show 17 quoted lines

> On 24 Jul 2013, at 17:20, Junio C Hamano <gitster@pobox.com> wrote:
>
>> 	Goes and looks...
>> 
>> OK, the tracking branches for these are created under refs/hg/*
>> using the same name.
>> ...
>> A refname shouldn't begin or end with a dot, because the range
>> ... ellided a correct description of the reason behind rule, which is
>> ... irrelevant to the topic
>
> I think there is a slight misunderstanding here:
>
> .git/hg/<remote_name> will be the actual directory for a hg::
> remote, which will then use mercurial internal magic to refer to
> the shared repo .git/hg/.shared in case the remote is not
> somewhere on the local filesystem, otherwise that path is used.

Yes, it is not a "slight" but a "huge" misunderstanding.

I saw the caller of get_repo(url, alias) doing a

    'refs/hg/%s' % alias

immediately after the call, and somehow assumed that the proposal would result in stuffing .shared to the "alias", which is not the case. The hg/$name directories may be used to store Hg repositories for 'alias' that is passed by the caller, but hg/.shared thing you guys were discussing is internal to get_repo() implementation and there won't be refs/hg/.shared created because of this change.