Re: [Bug-gnubg] updated po files

[Petr Kadlec]

Hi!

> i.e., line 5647 from gnubg.c. It's the header of .gnubgautorc. I think
> the problem is that you've duplicated the header, so it's written in
> both Czech and English. 

Yes, I did that, thinking that a script file should contain both language 
versions. But that was probably not a good idea. Anyway:
1. That dual translation has been in the file since the first version, I have 
not changed it in this version.
2. The formatting strings used are "%s" and "%1$s" (they should refer to the 
same argument). As I look to them now, I think that they really are not 
correct. I am not sure, but if the position specifier is used, it must be used 
with all arguments, maybe. So that instead of that "%s", "%1$s" should be used.

OK. I presume that the double-language idea was not really great, so that I 
will delete the czech translation (anyone who is editing his rc file by hand is 
probably english-capable anyway...), which should definitely fix the problem. 
The reason why this problem has not appeared until now is probably that you 
have upgraded your gettext, hasn't you? If that version improved its error 
detection capabilities, it could report the argument specifier error which my 
gettext gladly accepted.

> That could be. What's the MD5 checksum of your cs.po (the one you've
> committed)?

Well, my file has a different checksum, but if I convert the line ends using 
fromdos, I get exactly the same hash, which proves that CVS converts line ends 
during transfer (which is good and well).


I will not get to my development machine until evening today or maybe tommorow. 
If you need to use the file, please, feel free to remove the czech translation 
of the header (and even commit that version, if you want to). Otherwise, I'll 
do that asap.

Thanks,
Petr