Hi Anton,
this one looks to me that the encoding have been damaged. Can you first ensure that your translated sdf file is UTF8 encoded ( open in mozilla, view->character Encoding-> UTF8 )? After merging that sdf file into the source code look in some *.src files and check whether your strings are in language depended encoding.
Cheers, Ivo
Anton I. Danilov wrote:
Hello all,
I am building a Russian localized build under Linux (RedHat 9).
Everything goes OK all of the UI seems all right, but somewhere in the depths I find strings where *some* of the national characters are replaced with the pairs of question marks (like "??мя"), that is, two question marks per one national character.
If I try to extract localization string from the source code with localize -e, I see that these question marks are there, right at the places where I find them in the compiled binaries.
Only the part of the UI is malformed, but there is a huge amount of "infected" strings.
Russian binary installation kid, downloadable from http://ru.openoffice.ru/ is ok, no question marks at the same places.
Is there any clue to why can such thing happen? How to fix it?
Maybe, russian build is built with some other (patched?) version of the source code, not the one I downloaded from http://download.openoffice.org/1.1.4/source.html ?
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
