[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: gettext-0.10.35, couple of issues
From: |
Gisli Ottarsson |
Subject: |
Re: gettext-0.10.35, couple of issues |
Date: |
20 Apr 2001 21:45:15 -0400 |
User-agent: |
Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (Cuyahoga Valley) |
>>>>> On Fri, 20 Apr 2001 20:24:13 +0200 (CEST), Bruno Haible said:
BH> Gisli Ottarsson wrote on 2001-01-23:
>> #: t1.c:21
>> msgid "Integer %d before the float %e\n"
>> msgstr "Yo buss dis. Float %1$e aftuh de integer %2$d. Sheeit!\n"
>>
>> Here, the translator intended to first cite the float, which is
>> the second argument in the argument list and should have written
>> %2$e. msgfmt does not complain about this detectable error.
BH> msgfmt would have complained if this message were marked with
BH> #, c-format
BH> Why is this tag missing?
I am not sure, although I shall try to investigate. Perhaps you can
tell me now xgettext detects a c-format. If it is the occurrence of
the printf family of functions as a keyword, then this would explain
it. In my application I register cAsMsg as a keyword for xgettext and
don't use printf directly. cAsMsg is a message constructor which uses
printf style arguments.
BH> This is fixed in gettext-0.10.37. Thanks for the backtrace.
No, thank you!
Gisli
PS: Can someone in this forum cite a reference for a good way to use
gettext or a gettext equivalent in C++? The lack of a debate of the
poor state of I18N with iostream makes me wonder if I am missing
something.