Re: po compilation breakage

["Ben A. Hetland" <Ben A Hetland chem sintef no>]

Hubert Figuiere wrote:
> 
> I can't finish building Dia because of these errors in a .po file:
> 
> file=./echo zh_TW.Big5 | sed 's,.*/,,'.gmo \
>   && rm -f $file && PATH=../src:$PATH /usr/bin/msgfmt -o $file zh_TW.Big5.po
> zh_TW.Big5.po:161: illegal control sequence
> zh_TW.Big5.po:382: illegal control sequence
> zh_TW.Big5.po:1592: illegal control sequence
> zh_TW.Big5.po:1907: illegal control sequence
> zh_TW.Big5.po:1911: illegal control sequence
> zh_TW.Big5.po:1915: illegal control sequence
> zh_TW.Big5.po:2050: illegal control sequence
> zh_TW.Big5.po:2808: illegal control sequence
> found 8 fatal errors
> 
> It looks like msgfmt is being confused by the Big5 chars used... because on
> several occasion there are '\' that are just part of the multibyte charset.
> What is the solution ?

One solution is to replace the single '\' with an escaped '\\', since
the \ is the standard escape character in C strings.

At least this will get rid of the errors, but I don't know what the
actual effect on the output of the localized language is -- if any.

Maybe the translator [software] just doesn't account for C escape
sequences?


-+-Ben-+-