From: Elijah Newren <newren@gmail.com>
Date: Fri, 26 Jun 2009 04:45:13 GMT
Subject: Re: [PATCH v2] fast-export: Add a --tag-of-filtered-object option for newly dangling tags
Message-ID: <51419b2c0906252145pe174d1diddee9ed81413af74@mail.gmail.com>
In-Reply-To: <7vd48wazhu.fsf@alter.siamese.dyndns.org>

Hi,

Thanks for the detailed feedback and suggestions.  I'll send a new
patch series shortly, which I believe addresses all your suggestions
and requests (and opens up a new issue), but there is one item in
particular I wanted to respond to...

On Mon, Jun 22, 2009 at 12:34 PM, Junio C Hamano<gitster@pobox.com> wrote:
> Also, are you absolutely sure that your revs is always limited at this
> point?  Otherwise, the parents of this commit are queued in rev->list,
> expecting somebody else to later pick them up and further process, but
> there is nobody who does that in your codepath as far as I can see.  What
> will happen to these parent commits?

revs was not limited, but should have been.  Inside setup_revisions,
revs->limited is set if revs->topo_order is, but unfortunately we
weren't setting revs->topo_order until after the call to
setup_revisions.  So, as you suggest, my code caused the parents of
the rewritten commit to needlessly get queued up and left around, with
no one to process them.  Oops.


Elijah