bug-gnu-utils
[Top][All Lists]
Advanced

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

Re: Only ASCII char in msgid


From: Bruno Haible
Subject: Re: Only ASCII char in msgid
Date: Tue, 29 Oct 2002 13:58:15 +0100 (CET)

Michal Cernoevic writes:
> Hello,

Hello, please send mails to mailing lists in *plain text* please.
Wordprocessor generated HTML is unwelcome in these areas of the
internet.

> I've got a problem with the new version of gettext. I am a czech
> programmer and of course when I am programming then in Czech and all
> texts are Czech texts. Then, when I try to use gettext and make a
> *.po file, the new gettext is protesting the msgid must be in ASCII
> code, which is impossible, because English is for me the foreign
> language to which I need translation. I don't want to distribute my
> programs over the whole world, but we use them for example here:

Really, English is recommended nowadays for the msgids, because it is
much easier to find a translator from English to, say, Spanish than
from Czech to Spanish. Long ago, I did the same as you do, namely put
all messages in German into my programs. It was a mistake. Then I
changed to English, and now people have translated these to Spanish
and Dutch. Never would I have found a translator from German to
Spanish.

The requirement for ASCII comes from
  1. from the requirement that the translator need to view the
     original and the translation in the same text editor.
     If you are writing your messages in ISO-8859-2, how could a
     Russian translator add his translations in the same text file?
     If however, you are already using UTF-8, then tell me and I can
     change xgettext accordingly.
  2. English doesn't need many non-ASCII characters.

> There is another problem with programming languages. Preps it should
> be made in the same way as is in ctags. I use xgettext for example
> for *.4GL programs too, and xgettext sends warnings, that such
> extensions are unknown, it will take C. If there would be option
> like ---langmap=c++:.c.ec.EC.h then it would be better.

xgettext already has a --language option. If you have multiple
languages in a project, you can use 'msgcat' to combine the results of
multiple xgettext runs. Therefore there's no need for a --langmap
option. But thanks for the suggestion anyway.

Bruno




reply via email to

[Prev in Thread] Current Thread [Next in Thread]