I'm trying to load .po files with poedit, but I keep getting the following error:
Error converting the output to UTF-8
I'm running poedit on a ubuntu distribution if that makes any difference.
Anyone knows if there is a workarround?
I use poedit for software translations for Opera, and I never had such problems. PO files are text files, with some twists. They will handle multi-line translations, check for flags such as fuzzy. I must say that for now, I use POedit under Windows, as the SVN software I use (Tortoise SVN)only runs under Windows, and I'm too lazy to find the equivalent under Linux.
You should be able to load them in any UTF8-compliant editor. I do not know all the editors under Linux, although I would very much like to know more about this. One of the characteristics of UTF-8 files is that all accented letters are represented in two bytes, instead of one, with a special one as the fist byte.
At this point, your best bet would be to search for UTF-8 editors under linux. I'm sure OpenOffice should be able to do the trick. Try to load your file in this editor, and save it back, perhaps under a different name. Also, there are several lines at the beginning that tell about the translated language, author, revision number, etc. I kinda remember having seen warnings about such missing fields.
Are you sure UTF-8 is actually the problem?
I've always found the error messages to be a bit dodgy in poEdit - on OSX at least I regularly get 'import failed' messages when actually its succeeded. The whole package is a bit shoddy, I've been looking for a replacement for a while.
Stack Overflow has a similar question that lists a bunch of alternatives to poEdit.
Doesn't directly answer your question, but it might help you get round the problem!