From: Mark Walters <markwalters1009@gmail.com>
To: Jani Nikula <jani@nikula.org>, Carl Worth <cworth@cworth.org>,
	Austin Clements <aclements@csail.mit.edu>
Subject: Re: [PATCH 0/5] lib: make folder: prefix literal
In-Reply-To: <87a9eeou4v.fsf@nikula.org>
References: <cover.1389304779.git.jani@nikula.org>
	<87y525m649.fsf@awakening.csail.mit.edu>
	<87r47wfltb.fsf@nikula.org> <87iot8f4vg.fsf@nikula.org>
	<874n4rvcvo.fsf@yoom.home.cworth.org> <874n4mfw1x.fsf@nikula.org>
	<87k3dir3ci.fsf@yoom.home.cworth.org>
	<20140129204608.GE4375@mit.edu>
	<87bnyuqw60.fsf@yoom.home.cworth.org> <87a9eeou4v.fsf@nikula.org>
User-Agent: Notmuch/0.15.2+484~gfb59956 (http://notmuchmail.org) Emacs/23.4.1
	(x86_64-pc-linux-gnu)
Date: Thu, 30 Jan 2014 21:15:40 +0000
Message-ID: <8761p19nnn.fsf@qmul.ac.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: notmuch@notmuchmail.org
Precedence: list


Dear Jani

> That is the unicorn... many of the query improvements I have in mind
> depend on a custom query parser. So I'd like to have that. And a
> pony. But in the mean time, I'm left wondering whether I should pursue
> folder: as a boolean prefix, or try to figure out if there are
> improvements to be made as a probabilistic prefix, or just put this work
> on hold. With the db upgrade and upgrade tests, it's not exactly a
> trivial amount of work.

I think a boolean prefix sounds good. As you say there are lots of
things we could do with a custom parser but I don't think its worth
delaying things like this for it.

I also think it's not worth aiming for something perfect, just "good
enough". In a bike-shedding spirit I like the suggestion above (and on
irc) of a path: prefix which is basically literal. Either ^ or a
starting / to root the folder and a terminal $ or / to fix the end.

(I don't have a preference between these:  / avoids a clash with folder
name with a $ or ^ in them but $ ^ are more familiar.)

Best wishes

Mark