[debian-freedict] About language codes and wrongly named dictionaries

Jimmy O'Regan joregan at gmail.com
Thu May 9 20:52:41 UTC 2013

On 9 May 2013 21:31, Sebastian Humenda <shumenda at gmx.de> wrote:
> Hi,
> with the database of all dictionaries from
>   http://www.freedict.org/freedict-database.xml
> and using the table of ISO language codes which Jimmy posted, I'm now nearly
> there to translate e. g. lat-deu into „Latin-German”.
> Two problems arose:
> 1. Welsh seems to have two codes: cym and wel. I have a solution which
>   recognizes both, but out of curiosity, why are there two?

That's ISO-639-2: the 'B' code is derived from the English name, the
'T' code usually from the native name ('Cymraeg'). ISO-639-3 usually
follows the 'T' code. There's a little more on Wikipedia:

> 2. There's a dictionary  with 'scr' in it in the XML file which is not in the
>   SVN. Has a move taken place and the XML file was not updated? @michael,
>   I guess you have that in a script, could you update the file?

I think 'scr' is the deprecated code for 'Serbo-Croatian', which was
replaced by 'hbs' ('hrvatski/bosanski/srpski').

<Sefam> Are any of the mentors around?
<jimregan> yes, they're the ones trolling you

More information about the Pkg-FreeDict-discuss mailing list