bug-gawk
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [bug-gawk] gawk-4.0.0 sortu regression test failure


From: Aharon Robbins
Subject: Re: [bug-gawk] gawk-4.0.0 sortu regression test failure
Date: Thu, 01 Mar 2012 23:36:15 +0200
User-agent: Heirloom mailx 12.4 7/29/08

Hi. Thanks for the note.

This sounds like either a locale problem or something weird in your
system's qsort() library routine.  For the former, try doing

        export LC_ALL=C

before running "make check".  If that doesn't fix it, try finding a
different version of qsort() to add to the code and see if that helps.

In the worst case, we may need to correspond directly, offline, to work
on it more.

You may also wish to install git and pull down the current code
in the gawk-4.0-stable branch and see if that is any better.

Thanks!

Arnold

> From: "Fowler,Jerry" <address@hidden>
> To: "address@hidden" <address@hidden>
> Date: Tue, 28 Feb 2012 15:49:30 -0600
> Subject: [bug-gawk] gawk-4.0.0 sortu regression test failure
>
> I have built the gawk-4.0.0 source code on an AIX 6.1 machine with v10.1
> of the XL C/C++ compiler successfully.  When I run the supplied regression
> test, all test are successful except the sortu test.  An excerpt of the
> log of the regression test is as follows:
>
> shadow
> sortfor
> sortu
> ./sortu.ok _sortu differ: char 340, line 16
> gmake[1]: [sortu] Error 1 (ignored)
> splitarg4
> This test could fail on slow machines or on a minute boundary,
> so if it does, double check the actual results:
> strftime
> strtonum
> switch2
>
> When I diff the two files I get the following output:
>
> d_srdmak:address@hidden:/cerner/w_standard/surround/code/ots/gawk-4.0.0/test> 
> diff sortu.ok _sortu
> 16c16
> < [1]       :barz      barz
> ---
> > [1]       :barz      Zebra
> 18c18
> < [3]       :Zebra     Zebra
> ---
> > [3]       :Zebra     barz
> 34c34
> < [1]       :barz      barz
> ---
> > [1]       :barz      Zebra
> 36c36
> < [3]       :Zebra     Zebra
> ---
> > [3]       :Zebra     barz
>
> How do I troubleshoot this problem and determine why this test is failing?
>
> Jerry Fowler
> Senior Technology Architect, CernerWorks Technologies
> address@hidden<mailto:address@hidden> | 816-201-1476 direct
> www.cerner.com<http://www.cerner.com/>



reply via email to

[Prev in Thread] Current Thread [Next in Thread]