Skip Menu |

This queue is for tickets about the Pod-Wrap CPAN distribution.

Report information
The Basics
Id: 99433
Status: open
Priority: 0/
Queue: Pod-Wrap

People
Owner: ether [...] cpan.org
Requestors: NGLENN [...] cpan.org
Cc:
AdminCc:

Bug Information
Severity: (no value)
Broken in: (no value)
Fixed in: (no value)



Subject: t\deparse_cmp.t hangs on Windows
This leads to there being zero test reports for Windows machines, since none of them can finish the installation process and report pass or failure. I am available to test your code on Windows if you do not have a Windows box available.
On 2014-10-11 08:00:26, NGLENN wrote: Show quoted text
> This leads to there being zero test reports for Windows machines, > since none of them can finish the installation process and report pass > or failure. > I am available to test your code on Windows if you do not have a > Windows box available.
If you are able to identify where the hang is taking place, that would be enormously helpful!
Subject: Re: [rt.cpan.org #99433] t\deparse_cmp.t hangs on Windows
Date: Tue, 21 Jul 2015 00:55:11 +0100
To: bug-Pod-Wrap [...] rt.cpan.org
From: Yuval Kogman <nuffin [...] cpan.org>
Oh dear, this test should probably just be skipped on windows it relies on forking and pipes and isn't really worth porting On 20 July 2015 at 22:44, Karen Etheridge via RT <bug-Pod-Wrap@rt.cpan.org> wrote: Show quoted text
> Queue: Pod-Wrap > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=99433 > > > On 2014-10-11 08:00:26, NGLENN wrote:
> > This leads to there being zero test reports for Windows machines, > > since none of them can finish the installation process and report pass > > or failure. > > I am available to test your code on Windows if you do not have a > > Windows box available.
> > > If you are able to identify where the hang is taking place, that would be > enormously helpful! >
Show quoted text
> If you are able to identify where the hang is taking place, that would > be enormously helpful!
I don't have the time to find the exact cause right now, but here's the output: prove -lv t\deparse_cmp.t .. 1..42 ok 1 - Create wrapper obj ok 2 # skip Couldn't deparse (An operation was attempted on something that is no t a socket. at t\deparse_cmp.t line 154. # ) ok 3 # skip Couldn't deparse (An operation was attempted on something that is no t a socket. at t\deparse_cmp.t line 154. # ) ... ok 41 # skip Couldn't deparse (An operation was attempted on something that is not a socket. at t\deparse_cmp.t line 154. # ) ok 42 # skip Couldn't deparse (An operation was attempted on something that is not a socket. at t\deparse_cmp.t line 154. # )
On 2015-07-20 17:44:13, ETHER wrote: Show quoted text
> On 2014-10-11 08:00:26, NGLENN wrote:
> > This leads to there being zero test reports for Windows machines, > > since none of them can finish the installation process and report > > pass > > or failure. > > I am available to test your code on Windows if you do not have a > > Windows box available.
> > > If you are able to identify where the hang is taking place, that would > be enormously helpful!
I also see hangs on a freebsd 9.2 box with perl 5.10.1 (and only in this configuration). The hang happens while trying to deparse B.pm. A possible problem could be mixing system perl and the currently used perl for the build. In my case, the following patch fixed the hanging test: diff --git a/t/deparse_cmp.t b/t/deparse_cmp.t index bb2164d..d4cd5ba 100644 --- a/t/deparse_cmp.t +++ b/t/deparse_cmp.t @@ -115,7 +115,7 @@ sub deparse { my $fh = shift; local $ENV{PERL_HASH_SEED} = 0; # GRRR!!!! otherwise deparse output will not be consistent - IPC::Open3::open3(\*WRITE, \*READ, \*ERR, "perl", "-MO=Deparse") or die $!; # should be fatal (if it fails, or just because it's plain wrong). + IPC::Open3::open3(\*WRITE, \*READ, \*ERR, $^X, "-MO=Deparse") or die $!; # should be fatal (if it fails, or just because it's plain wrong). # out various handles my $w = IO::Select->new(\*WRITE);