Date: Sat, 12 Nov 2011 11:35:02 -0500
From: Austin Clements <amdragon@MIT.EDU>
To: David Bremner <david@tethera.net>
Subject: Re: [PATCH 6/6] emacs: make `notmuch-search-operate-all' operate on
	threads, not messages
Message-ID: <20111112163502.GE2658@mit.edu>
References: <1310313335-4159-1-git-send-email-pieter@praet.org>
	<1310313335-4159-7-git-send-email-pieter@praet.org>
	<87fwht2u9k.fsf@rocinante.cs.unb.ca>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <87fwht2u9k.fsf@rocinante.cs.unb.ca>
User-Agent: Mutt/1.5.21 (2010-09-15)
Cc: Notmuch Mail <notmuch@notmuchmail.org>
Precedence: list

Quoth David Bremner on Nov 12 at 11:13 am:
> On Sun, 10 Jul 2011 17:55:35 +0200, Pieter Praet <pieter@praet.org> wrote:
> > In order to be consistent with `notmuch-search-{add,remove}-tag' ("+"/"-"),
> > `notmuch-search-operate-all' ("*") should operate on matching threads
> > instead of matching messages.
> > 
> 
> As far as I can tell, the follow-up series for the race condition kindof
> got stalled.  Am I right in thinking this bug fix should still be
> applied? I didn't see any review/feedback on the list.

We concluded that fixing the tagging race correctly was actually a lot
of work, which should be done but hasn't yet.  We have to add message
IDs or docids to the search results, which is difficult to do with the
current text format, so rather than further entrenching ourselves, we
should first we should migrate Emacs to using the JSON-based search
output.

However, this series doesn't actually have much to do with the race.
I think the question here is whether notmuch-search-operate-all should
affect only matched messages or entire threads.  It seems to me it
should affect all threads, since that's what you're seeing visually,
but other people may disagree.

The test patches seem reasonable, though they could use a little
review before being pushed.