[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#29638: Same five tests fail with 1.15 on RHEL 7.4
From: |
Mathieu Lirzin |
Subject: |
bug#29638: Same five tests fail with 1.15 on RHEL 7.4 |
Date: |
Fri, 05 Jan 2018 04:08:15 +0100 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/25.3 (gnu/linux) |
Eric Blake <address@hidden> writes:
> On 01/04/2018 07:49 PM, Mathieu Lirzin wrote:
>
>> for example from Automake 1.15.1 build directory the following command
>> is supposed to work:
>>
>> --8<---------------cut here---------------start------------->8---
>> $ t/wrap/automake-1.15 --vers
>> automake (GNU automake) 1.15.1
>> Copyright (C) 2017 Free Software Foundation, Inc.
>> License GPLv2+: GNU GPL version 2 or later
>> <http://gnu.org/licenses/gpl-2.0.html>
>> This is free software: you are free to change and redistribute it.
>> There is NO WARRANTY, to the extent permitted by law.
>>
>> Written by Tom Tromey <address@hidden>
>> and Alexandre Duret-Lutz <address@hidden>.
>> --8<---------------cut here---------------end--------------->8---
>>
>> According to your logs this doesn't work on your system. My impression
>> is that those failing tests are checking the edge cases of Getopt::Long
>> which is system dependent and not the functional behavior of Automake.
>> As a consequence it seems reasonable to narrow the tests to more
>> conservative Getopt::Long behaviors.
>>
>> WDYT?
>
> If I understand GNU Coding Standards, we really do want to make sure
> unambiguous abbreviations of long options work.
I am unaware of such GCS recommandation. Do you have a pointer to the
part of the standards suggesting that?
> I'd argue that if not all versions of perl Getopt::Long are working
> the way the testsuite currently expects, that we should instead keep
> the test unchanged and find ways to work around the broken perl module
> versions (perhaps by manually specifying all abbreviations as explicit
> options ourselves, rather than relying on Getopt::Long to do it for
> us).
This could indeed be done, however I am not convinced by the usefulness
of such workaround.
> At the same time, once we do ascertain which version of
> Getopt::Long you are using, it may be worth reporting the flaw in that
> version to your distro vendor, as Automake is not the only software
> that would have to work around that particular weakness.
Agreed.
--
Mathieu Lirzin
GPG: F2A3 8D7E EB2B 6640 5761 070D 0ADE E100 9460 4D37