Re: [bug-gettext] xgettext --from-code does not work

[jaroslav . fojtik]

Dear Daiki Ueno,

> > the message "please specify the source encoding through --from-code"
> > is totally
> > frustrating, because the valid codepage is specified and it does not
> > work.
> >
> > ================================================================
> > xgettext -k_ "--from-code=WINDOWS-1250" -o ref.po ../sources.cc/*.cc
> > ./dump.c
> > ../sources.cc/igettext.cc:155: warning: Empty msgid.  It is reserved
> > by GNU gettext:
> >                                         gettext("") returns the header
> > entry with
> >                                         meta information, not the
> > empty string.
> > xgettext: Non-ASCII string at ../sources.cc/wp2latex.cc:438.
> >           Please specify the source encoding through --from-code.
> > make: *** [ref.po] Error 1
> 
> Briefly looking at the source, maybe --from-code=CP1250 will work.
> 
> There could be couple of improvements here: (1) in the error message,
> explicitly state that the specified encoding is not valid, and (2) use
> the same encoding scheme as iconv, and/or add an option to list
> available encodings.

Yes, you are right. When I explicitly specify encoding, the error message is 
totall mess.

I have not found any list of valid encodings supported by xgettext.
There should be some way, how to get them listed.

It is a good idea to be compatible with iconv names.

regards
   J.Fojtik




> 
> Regards,
> --
> Daiki Ueno


------------------------------------------------------------------------------------------
evolving systems consulting s.r.o.
Ús. armády 14
160 00 Praha 6
Czech Republic
www.evolvsys.cz 
ESABD code: 58020