Re: (was Re: [ Pozilla #7 ] gtranslator R0.36)

on 3/12/01 11:25 AM, Christian Rose at wrote:

> Fatih Demir wrote:
>>> You should probably change the name of
>>> "data/mime/" to "data/mime/",
>>> and also change the reference in po/ accordingly.
>>> That will make that file be processed correctly by xml-i18n-tools and
>>> the strings in that file turn up in the pot (which I think should be the
>>> case).
>> This is not possible as is the base for
>> and created by configure.
> Hmm, why should this file be created by configure? Isn't
> what's used by configure to create gtranslator.keys
> ?

I see now. The file is created by configure with the
appropriate full paths compiled in. It's correct to do it that way, but
xml-i18n-tools is not really set up well for that sort of thing.

We've done things to make it so you can use xml-i18n-tools on a tree from
cvs without doing steps like "autogen" or "configure", but this is one case
we don't handle at the moment.

>> And then xml-i18n-tools operate on the file -- it has got to be like it's
>> now.
> As it is now, xml-i18n-update won't recognize the
> file, and those messages won't, as a result of that, end up in the pot
> and therefore not in the translations. The only way I could make them
> end up in the pot was by renaming the mime keys file to
> in my copy of the source.
> I'm cc:ing this to a xml-i18n-tools developer who probably knows the
> answer to this.

To handle a case like this, we'd have to make a minor enhancement to the
xml-i18n-tools script. As things stand, the tools won't work for translators
who are trying to work on gtranslator without first configuring it (which is
normal, I think).

(I added a cc to the other xml-i18n-tools maintainers.)

    -- Darin

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]