MIME-Version: 1.0
In-Reply-To: <87txxvuyn4.fsf@servo.finestructure.net>
References: <1340656899-5644-1-git-send-email-ethan@betacantrips.com>
	<87txxvuyn4.fsf@servo.finestructure.net>
Date: Thu, 28 Jun 2012 14:39:55 -0400
Message-ID:
 <CAOJ+Ob0=N=AgZ+HWu_2U3yP1ro+Cfv5JmoJzb+2BxzndV-9mwQ@mail.gmail.com>
Subject: Re: [RFC PATCH 00/14] modular mail stores based on URIs
From: Ethan <ethan.glasser.camp@gmail.com>
To: Jameson Graef Rollins <jrollins@finestructure.net>
Content-Type: multipart/alternative; boundary=20cf3071ce1472ccaa04c38ca8f2
Cc: notmuch@notmuchmail.org
Precedence: list

--20cf3071ce1472ccaa04c38ca8f2
Content-Type: text/plain; charset=ISO-8859-1

It is pretty big and there are a couple places where the series could be
simplified, the first patch in particular. I will break it out and resubmit
piecewise but I'd like to know how to address these particular issues:

1. Are URIs the way to specify individual messages, despite bremner's
concerns about too much of the API being strings? Is adding another library
is the easiest way to parse URIs?

2. Is it OK to break maildir relocatability, or is it worth it to pass more
config to the library (perhaps by adding it to the notmuch_database_t
object)?

3. Is a global variable in the library acceptable? (I don't see any
others.) If not, how to store mailstore state?

The patch series is really more like a very rough draft to try to give
these concerns a context (specifically, mbox support).

Ethan

--20cf3071ce1472ccaa04c38ca8f2
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

It is pretty big and there are a couple places where the series could be si=
mplified, the first patch in particular. I will break it out and resubmit p=
iecewise but I&#39;d like to know how to address these particular issues:<b=
r>

<br>1. Are URIs  the way to specify individual messages, despite bremner=
9;s concerns about too much of the API being strings? Is adding another lib=
rary is the easiest way to parse URIs?<br><br>2. Is it OK to break maildir =
relocatability, or  is it worth it to pass more config to the library (perh=
aps by adding it to the notmuch_database_t object)?<br>

<br>3. Is a global variable in the library  acceptable? (I don&#39;t see an=
y others.) If not, how to store mailstore state?<br><br>The patch series is=
 really more like a very rough draft to try to give these concerns a contex=
t (specifically, mbox support).<br>
<br>Ethan<br><br>

--20cf3071ce1472ccaa04c38ca8f2--