Japanese Characters in .po files

[Dave Watson]

Hi ..

I'm a newbie to the translation world and I'm not sure where to address
this question/bug to but I thought I would start here, as I'm not sure
it's a bug, or just my lack of understanding of how gettext and/or GNU
Natural Language Support (NLS) is supposed to work .. 

I have just recently setup the NLS support for a web page in a number of
languages .. I got a translator to try and translate the same English
phrases used in a PHP Web Application. I got the translator to put the
translation string in UTF-8 charaters as that is the only character set
I have that will support Japaneses/Korean etc..

When I try to create the .mo files I get the following errors: 


msgfmt -o AdminBanner.mo AdminBanner.po
AdminBanner.po:24:13: invalid multibyte sequence
AdminBanner.po:24:15: invalid multibyte sequence
AdminBanner.po:24:17: invalid multibyte sequence
AdminBanner.po:24:19: invalid multibyte sequence
AdminBanner.po:24:20: invalid multibyte sequence
AdminBanner.po:24:21: invalid multibyte sequence
AdminBanner.po:24:22: invalid multibyte sequence
AdminBanner.po:24:23: invalid multibyte sequence
AdminBanner.po:24:25: invalid multibyte sequence
AdminBanner.po:24:27: invalid multibyte sequence
AdminBanner.po:24:28: invalid multibyte sequence
AdminBanner.po:24:29: invalid multibyte sequence
AdminBanner.po:24:31: invalid multibyte sequence
msgfmt: found 13 fatal errors

I have been trying to find out if UTF-8 multi-byte characters are
supported in the gettext system and how to put the multi byte sequences
into the .po files ..

I have not yet been able to find any donumentation that defines if/how
this would be supported

Any help that you could offer would be appreciated ..

Thanks in advance for your help 

Dave W 

-- 
Dave Watson
NorthSeas AMT
e-mail archiving without software

http://www.NorthSeasAMT.com
155 Terence Matthews Cres., Unit 1
Ottawa, ON K2M 2A8
Phone: 613-591-1045 x32
Toll Free: 1-888-591-1045