In fact, for PO, "automatic" (I suppose that was your setup) means the file is considered to use the system encoding.
The charset is not used at all (e.g., "Content-Type: text/plain; charset=UTF-8\n"), which is I think the real bug.
Another issue is that, while the BOM is respected, it does not trigger UTF-8 recognition.
Often, when translating from English, we don't realise the charset was wrong already on reading.
As I suppose most .po files would be in UTF-8 (rather than ISO-8859-1), a workaround for you would be to set .po files as UTF-8 (in Filter parameters).
Didier
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm quite new to OmegaT, and struggled over this minor lack of functionality. The docs state that normally, file content is displayed as per encoding spec when present in the source file (e.g. html). But with .po-files, one must set options to the .po filter, rather then setting any global option (as I expected).
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
As a workaround I have put in a default encoding for .po files as UTF-8. This seems to be the current default encoding for .po files anyway. See the attached patch.
Logged In: NO
Sorry, just checked, the source file does have a byte order mark.
Logged In: YES
user_id=1343245
Originator: NO
In fact, for PO, "automatic" (I suppose that was your setup) means the file is considered to use the system encoding.
The charset is not used at all (e.g., "Content-Type: text/plain; charset=UTF-8\n"), which is I think the real bug.
Another issue is that, while the BOM is respected, it does not trigger UTF-8 recognition.
Often, when translating from English, we don't realise the charset was wrong already on reading.
As I suppose most .po files would be in UTF-8 (rather than ISO-8859-1), a workaround for you would be to set .po files as UTF-8 (in Filter parameters).
Didier
I'm quite new to OmegaT, and struggled over this minor lack of functionality. The docs state that normally, file content is displayed as per encoding spec when present in the source file (e.g. html). But with .po-files, one must set options to the .po filter, rather then setting any global option (as I expected).
I ran into this bug today.
As a workaround I have put in a default encoding for .po files as UTF-8. This seems to be the current default encoding for .po files anyway. See the attached patch.
Committed to SVN (/trunk).
Didier
Fix had been published in v3.1.0 update 1
The change is 6890e533