From: Michal Sojka <sojkam1@fel.cvut.cz>
To: Serge Z <triumhiz@yandex.ru>, notmuch@notmuchmail.org
Subject: Re: Searching through different charsets
In-Reply-To: <20120222171041.11455.92079@localhost>
References: <20120222171041.11455.92079@localhost>
User-Agent: Notmuch/0.11.1+210~g5c2fc0a (http://notmuchmail.org) Emacs/23.3.1
	(x86_64-pc-linux-gnu)
Date: Fri, 24 Feb 2012 01:31:35 +0100
Message-ID: <877gzd5axk.fsf@steelpick.2x.cz>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Precedence: list

On Wed, 22 Feb 2012, Serge Z wrote:
> 
> Hello!
> 
> I've got the following problem: fetched emails can be in different encodings.
> And searching a term typed in one encoding (system default) does not match the
> same term in another encoding.
> 
> The solution, as I see, can be in preprocessing each incoming email to
> "normalize" it and its encoding so that indexer will handle emails in system
> encoding only. Could you please suggest something?

I can confirm this issue and sending a patch with test case (marked as
broken) for this. I expect the fix to be quite simple because all
encoding/docoding stuff is already implemented in gmime which is used by
notmuch when indexing.

> 
> Another issue (not so much wanted but wanted too) is searching through html
> messages without matching html tags.

I don't know whether somebody works on this or nor.

> This problem looks to be solvable by properly configured run-mailcap. Is there
> such solution anywhere?

I don't think that run-mailcap has anything to do with notmuch.

-Michal