bug-texinfo
[Top][All Lists]
Advanced

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

Need help for the japanese document errors


From: Patrice Dumas
Subject: Need help for the japanese document errors
Date: Tue, 17 Nov 2009 16:05:23 +0100
User-agent: Mutt/1.5.20 (2009-06-14)

Hello,

This is the remaining bug first reported by Reinhold a while ago I would
like to solve. The japanese .texi is in utf8, and the utf8 in the document 
is handled rightly, but not the translated strings. The document is part of 
the test suite (and more precisely, it is the only file whose result is 
not in the repository, since it is wrong), in 
test/encodings/japanese.texi

When the strings were in the translations.pl file, I couldn't manage
to fix the issue without breaking something else. Now, with the gettext
framework the strings are separate from the code, in the file:
po_document/ja.utf-8.po

So it should be possible to fix this issue now. I even thought that the 
issue would solve itself magically once that a gettext-like framework was 
used. Unfortunately this is not the case and the document output hasn't 
changed a bit after switching to the new translation framework.

Maybe this is a known issue with an easy fix and is covered by the 
libintl-perl FAQ:
http://search.cpan.org/dist/libintl-perl/lib/Locale/libintlFAQ.pod#Why_do_the_gettext_functions_always_unset_the_utf-8_flag_on_the_strings_it_returns?
http://search.cpan.org/dist/libintl-perl/lib/Locale/libintlFAQ.pod#Can_I_set_the_utf-8_flag_on_strings_returned_by_the_gettext_family_of_functions?

But I cannot make sense of this.

To test, the best is to do:
cd test/encodings/
../run_test.sh japanese

the result is in out/japanese/japanese.html.

Help please!





For the record here is the original mail from Reinhold:

Recently, a Japanese guy decided to start translating the LilyPond
documentation to Japanese. It basically works with texi2html (texi2pdf doesn't
even handle Japanese glyphs at all!), however, we are running into a problem
with the navigation bars:

The strings are translated, but unfortunately they are not written out in
UTF-8 to the html file, but in some other encoding (I couldn't find out which,
none of the three major Japanese non-utf-8 encodings seems to fit!). For
example, the [Top] link in the navbars is displayed as [å†’é ­] (that's an a
with a ring diacritic above, a dagger, an apostrophe and an e with an acute
accent) instead of the correct two-glyph [冒頭].

An example (.texi file to be run as "texi2html japanese.texi" and the
resulting .html file) is attached.

--
Pat




reply via email to

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