Date: Tue, 21 Feb 2012 10:53:12 -0500
From: Austin Clements <amdragon@MIT.EDU>
To: Justus Winter <4winter@informatik.uni-hamburg.de>
Subject: Re: notmuch as a shared object aka library knigge
Message-ID: <20120221155312.GB30513@mit.edu>
References: <20120221002921.8534.57091@thinkbox.jade-hamburg.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20120221002921.8534.57091@thinkbox.jade-hamburg.de>
User-Agent: Mutt/1.5.21 (2010-09-15)
Cc: notmuch mailing list <notmuch@notmuchmail.org>
Precedence: list

Quoth Justus Winter on Feb 21 at  1:29 am:
> Hi fellow notmuchrs,
> 
> while going through the python bindings I recently came across the
> following note in the documentation for the Database.get_directory
> function [0]:
> 
> ~~~ snip ~~~
> Warning
> 
> This call needs a writable database in Database.MODE.READ_WRITE
> mode. The underlying library will exit the program if this method is
> used on a read-only database!
> ~~~ snap ~~~

This is a bug and should be thought of as such.  INTERNAL_ERROR should
only be used for internal library inconsistencies (e.g., things that
should never ever happen) and the fact that it's leaking out here (and
easy to trigger) is simply a mistake.

This hasn't been fixed because it derives from an interface flaw.
What should notmuch_database_get_directory do on a read-only database?
It's specified to *create* the directory document if it doesn't exist,
which is the problem.  We could of course bandage this up and make it
return an error if you request a non-existent directory on a read-only
database, but that's an inconsistent interface.  I think we were
hoping to tweak the interface so you can tell it whether or not you
want to create the directory, independent of the database being
read/write or read-only, but no one has gotten around to that.  As
always, patches welcome!