X-Authentication-Warning: delorie.com: mail set sender to djgpp-workers-bounces using -f
Date: Thu, 30 Dec 2004 12:26:58 -0700
From: Brian Inglis <Brian DOT Inglis AT SystematicSw DOT ab DOT ca>
Subject: Re: strftime patch
In-reply-to: <200412301834.iBUIYugk012214@speedy.ludd.ltu.se>
To: djgpp-workers AT delorie DOT com
Message-id: <tel8t0dj2c3spfaag35281rejl7t3oqlqb@4ax.com>
Organization: Systematic Software
MIME-version: 1.0
X-Mailer: Forte Agent 1.93/32.576 English (American)
Content-type: text/plain; charset=us-ascii
References: <8ot7t0pskgfe03p1ii3l2gfvr0guvc0plf AT 4ax DOT com>
 <200412301834 DOT iBUIYugk012214 AT speedy DOT ludd DOT ltu DOT se>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by delorie.com id iBUJR1Li006924
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

On Thu, 30 Dec 2004 19:34:55 +0100 (CET), ams AT ludd DOT ltu DOT se wrote:

>According to Brian Inglis:
>> On Thu, 30 Dec 2004 13:03:13 +0100 (CET), ams AT ludd DOT ltu DOT se wrote:
>> 
>> >I'm getting failures from the strftim* test programs. OS WINDOWS 98
>> >English, Swedish "Regional settings". djtz*.zip NOT installed. No
>> >(what I think is) relevant environment variables set (LANG, LC_ALL,
>> >TZ).
>> >
>> >My (lack of) setup might very well why it fails, but if something
>> >are necessary for correct functioning the program should verify that.
>> >
>> >strftime:

>> Should work correctly as putenv(unconst("TZ=GMT0",char*) used! 
>> Myabe should add putenv()/getenv() test comparison or switch to using
>> setenv() as below? 

I was getting a bit paranoid here -- but maybe will add check. 

>If it's supposed to do the right thing, it's probably me that needs to
>debug it here where I have the problem.

It's possibly just default behaviour when no tz files present that
shows up the problem. 
I will look at the tz setting code, which should be hooked in via
mktime. 

>> >strftimt:

>> These tests currently assume tz files installed, could change to using
>> TZ variable as in strftime test program and strftimt test 31. 
>
>No, because this tests with tz files. It's not bad to have such a test
>too. Perhaps we should have two? One like today and one using TZ?

Good idea -- maybe I should change strftimt to use TZ variable and add
strftimz using tz files? 

>> Could try patching strftime test program to use setenv() instead of
>> putenv() and specify normal TZ env vars in strftimt. 
>
>Possible. Or I can unzip djtz*.zip.

Keep yourself clean of those: variety of test environments is useful. 

>If someone wants/needs my recompiled libc or the current source I
>have (remember, I had to manually apply some patches), I could zip it
>up and put it up for download. Specify what exactly you want in that
>case.