zwon has asked for the wisdom of the Perl Monks concerning the following question:

Hello everybody!

I'm looking for the tool to implement internationalisation support for a website and so far it seems like Locale::Maketext::Lexicon is the way to go except one problem -- it doesn't have support for plurals out of the box and I couldn't find any ready solution for this problem. So before I start working on my own solution I want to ask the monks if I missed the existing one. There's a quant function provided by Locale::Maketext, but it is not good enough even for English, and not suitable for Slavic languages at all. So what I missing is ngettext(3) functionality. Considering that module seems quite popular, I hope that somebody has already found workaround and is willing to share his wisdom. The examples of the messages I have problem with:

  • Comment on Locale::Maketext::Lexicon and plural forms

Replies are listed 'Best First'.
Re: Locale::Maketext::Lexicon and plural forms
by Anonymous Monk on Mar 23, 2011 at 19:10 UTC
[reply]

      These modules have their own problems unfortunately. They require to set LC_MESSAGES which mean that error messages in web-server error log also will be localised and I certainly don't want that. Also it looks like I can set LC_MESSAGES only once per process, and I need to be able to use different locales for every request. No, the last statement is actually wrong.

[reply]
        Are you sure that there won't be problems with using the process locale in a web server setting? For example, if users A and B are being served by the same Apache process, A is an English user and B is a Russian user, and B's request changes the process' locale before A's request finishes; isn't it possible that A would then get a page back in the wrong language? I know this is a problem in PHP ( http://php.net/manual/en/function.setlocale.php ), and I know that POSIX 2008 implemented a workaround via uselocale ( http://www.manpagez.com/man/3/uselocale/ ).
[reply]

Back to Seekers of Perl Wisdom