First login?
Lost password?
 
 
 
mailing-list for TeXmacs Users

Text archives Help


Re: [TeXmacs] spell check


Chronological Thread 
  • From: Kemal Ilgar Eroglu <address@hidden>
  • To: "address@hidden" <address@hidden>
  • Subject: Re: [TeXmacs] spell check
  • Date: Fri, 6 Apr 2012 14:42:16 -0700 (PDT)
  • Authentication-results: spamcheck-ng1.citic74.net; dkim=pass (1024-bit key) header.i=@yahoo.com; dkim-asp=none
  • Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:In-Reply-To:MIME-Version:Content-Type; b=yUDcXpoNW5+YYXTgo4h6iYajeEfNzf8dvBVab2Xm3Y3qnIdxOUQqOMKPQqBLuZSj50+Y70vjoEA7SYw43BLZtemXeHYxaUybFFQmHA1LLAr4XwTG0xjupeTelGjuWpLL+lzX9U7bHevR9ulJ2gFWG6ZgMsS5denxA7ggUJ/QShU=;
Hi,

I made a little test with Turkish. It doesn't readily exist in the menus so I modified some files to add it (and I hope I did it correctly). Spell check generally seems to work, but I noticed the parser seems to get stuck at the dotless i letter. For example, the word "burası" is recognized by aspell when run from the command line, but when entered into TeXmacs the spell checker only takes the "buras" part, not including the dotless i at the end. This is the only problem I've observed so far.

Of course, since it is an agglutinative language, building a word-list for Turkish is a hopeless effort and aspell is bound to do a poor job. I have a little question: Not that I have the ability to write the code, but perhaps I can push some people I know to do it: Would TeXmacs be able to use a Java-based spell checker? Because the best open-source one for Turkish (also used by Open Office) is called "zemberek" and is Java-based.

Thanks for all the improvements by the way,
Ilgar


From: François Poulain <address@hidden>
To: address@hidden
Sent: Tuesday, April 3, 2012 11:29 AM
Subject: Re: [TeXmacs] spell check

Dear all,

Le Mon, 2 Apr 2012 15:01:35 +0200,
François Poulain <address@hidden> a écrit :

> So, I propose to migrate the code to aspell, with a better language
> management, with an optional dictionary managment and to use unicode
> for communication. This will greatly simplify the plugin.

I pushed on svn the patch, with some improvement (thanks to Jan
Lellmann).

Could some people test it in their native languages ? Especially for
non latin language (asian and cyrillic), because it really not easy for
me to test it (whereas it's rather easy to test italian spell checking
even if I do not know any word from italian ... ).


Thanks.
François

--
François Poulain <address@hidden>



Archive powered by MHonArc 2.6.19.

Top of page