Bug #98495 for Expect: Failed test 'the number' -- random test results

Sun Aug 31 22:18:53 2014 ANDK [...] cpan.org - Ticket created

Subject:

Failed test 'the number' -- random test results

Thu Sep 04 21:24:24 2014 ANDK [...] cpan.org - Broken in 1.31 added

Thu Sep 04 21:26:51 2014 ANDK [...] cpan.org - Correspondence added

The failing tests are also in 1.31. One Sample report: http://www.cpantesters.org/cpan/report/45504987

Sun Sep 14 13:39:01 2014 SREZIC [...] cpan.org - Correspondence added

On 2014-09-04 21:26:51, ANDK wrote: Show quoted text

> The failing tests are also in 1.31. One Sample report: > > http://www.cpantesters.org/cpan/report/45504987

It looks it depends whether "bc" is installed. With the linux bc it fails, without the failing tests are skipped. BTW, FreeBSD's bc does not seem to be problematic.

Sun Sep 14 13:39:01 2014 The RT System itself - Status changed from 'new' to 'open'

Sun Sep 14 13:48:47 2014 SZABGAB [...] cpan.org - Correspondence added

Subject:	Re: [rt.cpan.org #98495] Failed test 'the number' -- random test results
Date:	Sun, 14 Sep 2014 20:48:39 +0300
To:	bug-Expect [...] rt.cpan.org
From:	Gabor Szabo <SZABGAB [...] cpan.org>

As I understand the test is skipped if bc is not installed, but on systems there it is installed there are differences. In some systems the expect catches the echo in other system it catches the result. I'll further refine the test.

Sun Sep 14 23:51:13 2014 andreas.koenig.7os6VVqR [...] franz.ak.mind.de - Correspondence added

CC:	ANDK [...] cpan.org
Subject:	Re: [rt.cpan.org #98495] Failed test 'the number' -- random test results
Date:	Mon, 15 Sep 2014 05:50:55 +0200
To:	bug-Expect [...] rt.cpan.org
From:	Andreas Koenig <andreas.koenig.7os6VVqR [...] franz.ak.mind.de>

Show quoted text

>>>>> On Sun, 14 Sep 2014 13:39:01 -0400, "Slaven_Rezic via RT" <bug-Expect@rt.cpan.org> said:

Show quoted text

> It looks it depends whether "bc" is installed. With the linux bc it > fails, without the failing tests are skipped.

No, there are fails and passes with same perl, same smoker, same bc. http://www.cpantesters.org/cpan/report/45519498 http://www.cpantesters.org/cpan/report/45500963 -- andreas

Mon Sep 15 00:26:23 2014 SZABGAB [...] cpan.org - Correspondence added

Subject:	Re: [rt.cpan.org #98495] Failed test 'the number' -- random test results
Date:	Mon, 15 Sep 2014 07:26:15 +0300
To:	bug-Expect [...] rt.cpan.org
From:	Gabor Szabo <SZABGAB [...] cpan.org>

On Mon, Sep 15, 2014 at 6:51 AM, (Andreas J. Koenig) via RT < bug-Expect@rt.cpan.org> wrote: Show quoted text

> Queue: Expect > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=98495 > >

> >>>>> On Sun, 14 Sep 2014 13:39:01 -0400, "Slaven_Rezic via RT" <

> bug-Expect@rt.cpan.org> said: >

> > It looks it depends whether "bc" is installed. With the linux bc it > > fails, without the failing tests are skipped.

> > No, there are fails and passes with same perl, same smoker, same bc. > > http://www.cpantesters.org/cpan/report/45519498 > http://www.cpantesters.org/cpan/report/45500963 > >

Now that's a good catch. So what is the difference between those two? Gabor

Mon Sep 15 00:33:50 2014 SZABGAB [...] cpan.org - Correspondence added

Subject:	Re: [rt.cpan.org #98495] Failed test 'the number' -- random test results
Date:	Mon, 15 Sep 2014 07:33:41 +0300
To:	bug-Expect [...] rt.cpan.org
From:	Gabor Szabo <SZABGAB [...] cpan.org>

On Mon, Sep 15, 2014 at 7:26 AM, Gabor Szabo <SZABGAB@cpan.org> wrote: Show quoted text

> > > > On Mon, Sep 15, 2014 at 6:51 AM, (Andreas J. Koenig) via RT <

bug-Expect@rt.cpan.org> wrote: Show quoted text

>> >> Queue: Expect >> Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=98495 > >>

>> >>>>> On Sun, 14 Sep 2014 13:39:01 -0400, "Slaven_Rezic via RT" <

bug-Expect@rt.cpan.org> said: Show quoted text

>>

>> > It looks it depends whether "bc" is installed. With the linux bc it >> > fails, without the failing tests are skipped.

>> >> No, there are fails and passes with same perl, same smoker, same bc. >> >> http://www.cpantesters.org/cpan/report/45519498 >> http://www.cpantesters.org/cpan/report/45500963 >>

> > Now that's a good catch. So what is the difference between those two? > > Gabor > >

So far I noticed that the failing had ExtUtils::MakeMaker 6.99_09 while the passing had ExtUtils::MakeMaker 6.98 but I doubt that will have an impact on the tests. PERL5_CPANPLUS_IS_RUNNING and PERL5_CPAN_IS_RUNNING had different numbers, but if I am not mistaken these are process IDs. Andreas, have these ran on the same machine? As I can understand it is either ExtUtils::MakeMaker or more likely some timing issue. Gabor

Sun Oct 26 01:39:22 2014 ANDK [...] cpan.org - Correspondence added

It would be very helpful if this test would be disabled. It may have to do with Expect but as long as the test is giving random results it is counterproductive to have it. Please make a stable release without it. Next development steps should be done in dev releases. Thanks,

Sun Oct 26 02:05:33 2014 SZABGAB [...] cpan.org - Correspondence added

Subject:	Re: [rt.cpan.org #98495] Failed test 'the number' -- random test results
Date:	Sun, 26 Oct 2014 08:05:25 +0200
To:	bug-Expect [...] rt.cpan.org
From:	Gabor Szabo <SZABGAB [...] cpan.org>

1.32 was uploaded to CPAN which is skipping the whole t/02-bc.t

Sun Oct 26 03:16:38 2014 ANDK [...] cpan.org - Correspondence added

On 20141026T060533Z, SZABGAB wrote: Show quoted text

> 1.32 was uploaded to CPAN which is skipping the whole t/02-bc.t

Wow, that was fast, THANKS!