Re: [RFC/PATCH] git-remote-mediawiki: reset private ref after non-dumb push

From: Felipe Contreras <felipe.contreras@gmail.com>
Date: Aug 23, 2013, 19:52 UTC
Message-ID: <CAMP44s281qkdMXUM-2P6T+emFajASX6Jyj4nk8_cU5xYJYvdLA@mail.gmail.com>
In-Reply-To: <vpqbo4o3jba.fsf@anie.imag.fr>

On Fri, Aug 23, 2013 at 3:25 AM, Matthieu Moy <Matthieu.Moy@grenoble-inp.fr> wrote:

Show 7 quoted lines

> This is assuming you follow the scheme
>
>   git -> local repo for other vcs -> remote repo for other vcs
>
> which itself more or less assumes that the other VCS is a decentralized
> VCS. I understand this is a good idea for hg or bzr remote helpers, but
> not applicable for git-remote-mediawiki.

A centralized repository is a subset of decentralized workflows.

Show 8 quoted lines

> Back to my original problem: the point of dumb push is to make sure the
> next import will re-import the pushed revisions. if I use something
> other than the private namespace to keep track of the last imported,
> then I'll need to do a non-fast forward update to the private ref. I
> could do that by sending "feature force\n" it the beginning of the
> fast-import stream when importing, but that loses one safety feature
> (e.g. I detected the breakage thanks to the non-fast forward error
> message, while the tests incorrectly pass if I enable "force").

I don't know if a dumb push is the right thing to do here, but supposing it is, you can still report non-fast-forward errors:

https://github.com/felipec/git/commit/0f7f0a223d710d29a4f1a03fc27348a8690d8a19 https://github.com/felipec/git/commit/b67a8914bc1bb3ad23591875611165b84135aaf9

If it's too much hassle to find non-fast-forward situations by yourself, then perhaps it would make sense to update the remote namespace only when a certain feature has been flagged.

-- 
Felipe Contreras