Bug #58130 for App-cpanminus:

Fri Jun 04 10:36:32 2010 ribasushi [...] leporine.io - Ticket created

cpanm sets local $ENV{AUTOMATED_TESTING} = 1; during the test() phase. This is an environment variable reserved for smokers only, consider removing it. <ribasushi> avar: wrt RT#56698 - I am pretty sure AUTOMATED_TESTING is something that smokers set, not cpan clients <avar> I just used cpanm <ribasushi> bottom line is, if you did a cpan DBIx::Class or similar, this test should *not* have executed in the 1st place <LeoNerd> Out of interests, what sort of tests might those be? Anything I should consider doing? <ribasushi> LeoNerd: it depends, for example SQL::Abstract has a test that is not crucial for actual users, and can take up to a minute on an old-ish cpu <ribasushi> so I made it smoker-only, as I still want some feedback, but don't want to tax installers <ribasushi> LeoNerd: and on the contrary - DBIC has a test that tells you "this redhat's perl sucks cock, performance will be abysmall" <ribasushi> however I don't run it on smokers because - well, what do I care <Alias_> Some things add extra dependencies on AUTOMATED_TESTING <avar> ribasushi: file a cpanm bug :)

Fri Jun 04 15:38:55 2010 miyagawa [...] gmail.com - Correspondence added

Subject:	Re: [rt.cpan.org #58130]
Date:	Fri, 4 Jun 2010 12:38:46 -0700
To:	bug-App-cpanminus [...] rt.cpan.org
From:	Tatsuhiko Miyagawa <miyagawa [...] gmail.com>

At one point it stopped setting AUTOMATED_TESTING but that caused lots of test failures on distributions expecting STDIN to be open - which is not the case with cpanm. I filed a bunch of reports against those distributions but they never got fixed. The problem is there's no consensus in CPAN clients to say "Nobody is watching this test, so don't expect human to interfere with your test" and the closest thing we get is AUTOMATED_TESTING, unfortunately. I discussed this on #toolchain. http://github.com/miyagawa/cpanminus/commit/20a27551343775730ecf0b1e205ef38c849650ee On Fri, Jun 4, 2010 at 7:36 AM, Peter Rabbitson via RT <bug-App-cpanminus@rt.cpan.org> wrote: Show quoted text

> Fri Jun 04 10:36:32 2010: Request 58130 was acted upon. > Transaction: Ticket created by RIBASUSHI > Queue: App-cpanminus > Subject: (No subject given) > Broken in: (no value) > Severity: (no value) > Owner: Nobody > Requestors: ribasushi@cpan.org > Status: new > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=58130 > > > > cpanm sets local $ENV{AUTOMATED_TESTING} = 1; during the test() phase. > This is an environment variable reserved for smokers only, consider > removing it. > > <ribasushi> avar: wrt RT#56698 - I am pretty sure AUTOMATED_TESTING is > something that smokers set, not cpan clients > <avar> I just used cpanm > <ribasushi> bottom line is, if you did a cpan DBIx::Class or similar, > this test should *not* have executed in the 1st place > <LeoNerd> Out of interests, what sort of tests might those be? Anything > I should consider doing? > <ribasushi> LeoNerd: it depends, for example SQL::Abstract has a test > that is not crucial for actual users, and can take up to a minute on an > old-ish cpu > <ribasushi> so I made it smoker-only, as I still want some feedback, but > don't want to tax installers > <ribasushi> LeoNerd: and on the contrary - DBIC has a test that tells > you "this redhat's perl sucks cock, performance will be abysmall" > <ribasushi> however I don't run it on smokers because - well, what do I care > <Alias_> Some things add extra dependencies on AUTOMATED_TESTING > <avar> ribasushi: file a cpanm bug :) > >

-- Tatsuhiko Miyagawa

Fri Jun 04 15:38:57 2010 The RT System itself - Status changed from 'new' to 'open'

Sat Oct 30 15:20:23 2010 MIYAGAWA [...] cpan.org - Status changed from 'open' to 'resolved'

Bug #58130 for App-cpanminus:

Preferred bug tracker