Comment 1 juha 2008-12-10 14:12:14 CET

Is it the same if you set 
export LANG=cs_CZ.UTF-8
also?
Why do you have that differently than the other settings?

What do you get if you do:
date +%x

Can you compile and run if I create you a small c program to test the 
language settings?

I've tried this:

hrubi@janhruban ~ $ export LC_ALL=cs_CZ.UTF-8 LANG=cs_CZ.UTF-8
hrubi@janhruban ~ $ locale
LANG=cs_CZ.UTF-8
LC_CTYPE="cs_CZ.UTF-8"
LC_NUMERIC="cs_CZ.UTF-8"
LC_TIME="cs_CZ.UTF-8"
LC_COLLATE="cs_CZ.UTF-8"
LC_MONETARY="cs_CZ.UTF-8"
LC_MESSAGES="cs_CZ.UTF-8"
LC_PAPER="cs_CZ.UTF-8"
LC_NAME="cs_CZ.UTF-8"
LC_ADDRESS="cs_CZ.UTF-8"
LC_TELEPHONE="cs_CZ.UTF-8"
LC_MEASUREMENT="cs_CZ.UTF-8"
LC_IDENTIFICATION="cs_CZ.UTF-8"
LC_ALL=cs_CZ.UTF-8
hrubi@janhruban ~ $ orage 

** (orage:17691): WARNING **: Failed to connect to the D-BUS session bus: Failed to connect to socket /tmp/dbus-LamykVLx6N: Spojení odmítnuto

** (orage:17691): WARNING **: Default timezone set to floating. Do not use timezones when setting appointments, it does not make sense without proper local timezone.

** (orage:17691): CRITICAL **: convert_to_timezone: builtin timezone Central Europe Standard Time not found, conversion failed.

** (orage:17691): CRITICAL **: convert_to_timezone: builtin timezone Central Europe Standard Time not found, conversion failed.
** Message: Orage **: Sestavit seznam výstrah: Přidáno 1 výstrah. Zpracováno 3 výstrah.
** Message: Orage **:   Bylo nalezeno 3 výstrah, z nichž 1 je aktivních. (Bylo prohledáno 104826 opakujících se výstrah.)

** ERROR **: Orage: orage_i18_date_to_tm_date wrong format (10.12.2008)
aborting...
Neúspěšně ukončen (SIGABRT)


So the result is the same. (If you wonder about custom locale settings, I just don't like czech speaking system, so LANG and LC_MESSAGES are set to en_US, but I want time, currentcy etc. to be in czech)



date +%x gives:
10.12.2008


I've tried custom program derived from orage piece of code:

#include <stdio.h>
#define __USE_XOPEN
#include <time.h>
#include <locale.h>

int main(void)
{
    //setlocale(LC_TIME, "sk_SK.UTF-8");
    setlocale(LC_TIME, "cs_CZ.UTF-8");

    struct tm tm_date = {0,0,0,0,0,0,0,0,0};
    const char *i18_date = "10.12.2008";

    if (strptime(i18_date, "%x", &tm_date) == NULL)
    {
        printf("error\n");
        return 1;
    }
    else
    {
        printf("success\n");
        return 0;
    }
}

It prints "error" cs_CZ.UTF-8 LC_CTIME. However if I comment the czech locale settings and supply it with slovak one (which date +%x is the same as czech) it prints out success.

So probably my thoughts about buggy glibc were right?


I'd be glad to further cooperate..

Comment 3 juha 2008-12-11 07:06:34 CET

Yes, this is clearly glibc problem.

I run your test program and it works in my environment with your locale.

I tested in Ubuntu 8.04, which has glibc 2.7.

Thanks for your effort, but I can not fix this, try to upgrade your system.

Comment 4 juha 2008-12-11 07:07:20 CET

glibc issue