Andrew Ho wrote:
Hello,
AH>So I think the next step is to figure out who can help look at that
latter
AH>problem. Is there a bug I or someone else should file somewhere?
MS>To make sure it isn't completely forgotten you can put it into the bug
MS>tracker with perlbug.
Very well, I have subm
Hello,
AH>So I think the next step is to figure out who can help look at that latter
AH>problem. Is there a bug I or someone else should file somewhere?
MS>To make sure it isn't completely forgotten you can put it into the bug
MS>tracker with perlbug.
Very well, I have submitted the bug and it
On Tue, Jul 26, 2005 at 03:27:17PM -0700, Andrew Ho wrote:
> So I think the next step is to figure out who can help look at that latter
> problem. Is there a bug I or someone else should file somewhere?
To make sure it isn't completely forgotten you can put it into the bug
tracker with perlbug.
Hello,
As the original instigator of this thread, maybe I can help continue its
usefulness. I think the current consensus is as follows:
* This all has nothing at all to do with with ExtUtils::MakeMaker
(so I have removed that list from the Cc list).
* There is nothing intrinsic
On Tue, Jul 26, 2005 at 10:57:08PM +0200, Gerrit P. Haase wrote:
> Andrew found relevant cahnges and you replied:
Oh yeah. Huh. Good question.
--
Michael G Schwern [EMAIL PROTECTED] http://www.pobox.com/~schwern
You are wicked and wrong to have broken inside and peeked at the
implemen
Michael G Schwern wrote:
On Tue, Jul 26, 2005 at 08:23:03PM +0200, Gerrit P. Haase wrote:
5.8.7 ships with Test::Harness 2.48 which doesn't have this code.
It fails to run the tests from the command line with the default
installation of perl-5.8.7 for Cygwin?
Huh?
From the initial repo
On Tue, Jul 26, 2005 at 08:23:03PM +0200, Gerrit P. Haase wrote:
> >5.8.7 ships with Test::Harness 2.48 which doesn't have this code.
>
> It fails to run the tests from the command line with the default
> installation of perl-5.8.7 for Cygwin?
Huh?
--
Michael G Schwern [EMAIL PROTECTED]
Michael G Schwern wrote:
On Tue, Jul 26, 2005 at 07:51:48PM +0200, Gerrit P. Haase wrote:
Yep, the example with the delete( $ENV ...) shows clearly that there
is a problem with forked/spawned processes which should be resolved.
In the first place I would be glad if someone (petdance?) could
On Tue, Jul 26, 2005 at 07:51:48PM +0200, Gerrit P. Haase wrote:
> Yep, the example with the delete( $ENV ...) shows clearly that there
> is a problem with forked/spawned processes which should be resolved.
>
> In the first place I would be glad if someone (petdance?) could fix TH
> for Cygwin s
Scott Bolte wrote:
On Mon, 25 Jul 2005 17:39:41 +0200, "Gerrit P. Haase" wrote:
Indeed, I see no unsetenv() in the perl sources.
Since it seems that the problem is independent of TH the problem should
be located in perl itself. I'll try with a recent devel version to see
if it fails here too
On Mon, 25 Jul 2005 17:39:41 +0200, "Gerrit P. Haase" wrote:
>
> Indeed, I see no unsetenv() in the perl sources.
>
> Since it seems that the problem is independent of TH the problem should
> be located in perl itself. I'll try with a recent devel version to see
> if it fails here too. Scott, h
Christopher Faylor wrote:
On Fri, Jul 22, 2005 at 03:16:29PM -0500, Scott Bolte wrote:
On Thu, 21 Jul 2005 00:14:51 -0700, Michael G Schwern wrote:
Its gotta be something about _default_inc(). Nothing looks wrong from
here. My only guess is that local $ENV{PERL5LIB} is ineffective. That
w
On Fri, Jul 22, 2005 at 03:16:29PM -0500, Scott Bolte wrote:
>On Thu, 21 Jul 2005 00:14:51 -0700, Michael G Schwern wrote:
>>
>> Its gotta be something about _default_inc(). Nothing looks wrong from
>> here. My only guess is that local $ENV{PERL5LIB} is ineffective. That
>> would account for t
On Thu, 21 Jul 2005 00:14:51 -0700, Michael G Schwern wrote:
>
> Its gotta be something about _default_inc(). Nothing looks wrong from
> here. My only guess is that local $ENV{PERL5LIB} is ineffective. That
> would account for the alternation.
It's not just the local scoping. There is
On Wed, Jul 20, 2005 at 06:54:13PM -0700, Andrew Ho wrote:
> Here are the diffs between 2.42 and 2.44, the next most recent version:
>
> http://www.zeuscat.com/tmp/test_harness_diff.txt
> http://www.zeuscat.com/tmp/test_harness_diff.html (colorized)
>
> I haven't peeked through the diff y
Hello,
MS>I would suggest working backwards though the CPAN versions of Test::Harness
MS>until it works. Then we can better spot the change which caused this bug.
MS>http://search.cpan.org/dist/Test-Harness/
I tested stepping down through versions of Test::Harness from CPAN.
Version 2.42 is th
On Tue, Jul 19, 2005 at 09:16:03PM -0500, Scott Bolte wrote:
> It is not a perl bug. It is something in the TH module.
>
> I tested with perl 5.8.7 and the previous 5.8.6. I also,
> through -I, tried both the old and new modules. It clearly
> is the new Test::Harness modu
On Tue, 19 Jul 2005 08:27:28 -0700, Michael G Schwern wrote:
> Try the same Test::Harness with an older perl. See if its a perl bug.
It is not a perl bug. It is something in the TH module.
I tested with perl 5.8.7 and the previous 5.8.6. I also,
through -I, tried both th
Try the same Test::Harness with an older perl. See if its a perl bug.
--
Michael G Schwern [EMAIL PROTECTED] http://www.pobox.com/~schwern
Insulting our readers is part of our business model.
http://somethingpositive.net/sp07122005.shtml
--
Unsubscribe info: http://cygwin.
On Mon, 18 Jul 2005 01:50:33 -0700, Michael G Schwern wrote:
>
> MakeMaker is not involved, just Test::Harness. All the action is in or
> called from Test::Harness::Straps->analyze_file.
I agree. TH is resetting INC and the directories passed
into test_harness() are lost after th
On Mon, Jul 18, 2005 at 12:31:36AM +0200, Gerrit P. Haase wrote:
> >Maybe a cygwin problem, I'll fetch a snapshot.
>
> With the latest snapshot I see the same behaviour.
>
> Are there other modules involved besides MM and TH?
MakeMaker is not involved, just Test::Harness. All the action is in o
On Mon, Jul 18, 2005 at 01:03:05AM +0200, Gerrit P. Haase wrote:
> >At this point I'd suspect there's something broken about %ENV. Try
> >replacing
> >this in THS->analyze_file:
> >
> >local $ENV{PERL5LIB} = $self->_INC2PERL5LIB;
> >if ( $Test::Harness::Debug ) {
> >local $^W=0; #
Michael G Schwern wrote:
On Mon, Jul 18, 2005 at 12:27:31AM +0200, Gerrit P. Haase wrote:
perl -MTest::Harness -Iblib/lib -Iblib/arch -wle '$Test::Harness::Debug =
1; $Test::Harness::Verbose = 1; runtests @ARGV' t/*.t
The lines we're looking for are "# PERL5LIB=..." from THS->analyze_file()
On Mon, Jul 18, 2005 at 12:38:51AM +0200, Gerrit P. Haase wrote:
> I have installed M/MS/MSCHWERN/ExtUtils-MakeMaker-6.30.tar.gz now,
> still the same, every second test fails because blib/lib isn't
> present in @INC (every second test).
MakeMaker isn't involved. Its something inside
Test::Harne
On Mon, Jul 18, 2005 at 12:27:31AM +0200, Gerrit P. Haase wrote:
> >perl -MTest::Harness -Iblib/lib -Iblib/arch -wle '$Test::Harness::Debug =
> >1; $Test::Harness::Verbose = 1; runtests @ARGV' t/*.t
> >The lines we're looking for are "# PERL5LIB=..." from THS->analyze_file()
> >telling us how TH
I have installed M/MS/MSCHWERN/ExtUtils-MakeMaker-6.30.tar.gz now,
still the same, every second test fails because blib/lib isn't
present in @INC (every second test).
Gerrit
--
=^..^=
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/pro
Gerrit P. Haase wrote:
Maybe a cygwin problem, I'll fetch a snapshot.
With the latest snapshot I see the same behaviour.
Are there other modules involved besides MM and TH?
Gerrit
--
=^..^=
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cyg
Michael G Schwern wrote:
On Sun, Jul 17, 2005 at 11:43:39PM +0200, Gerrit P. Haase wrote:
Nope, Teat::Harness now at version 2.52:
$ make test
/usr/bin/perl.exe "-MExtUtils::Command::MM" "-e" "test_harness(0,
'blib/lib', 'blib/arch')" t/*.t
Ok, see if MakeMaker has anything to do with thi
On Sun, Jul 17, 2005 at 11:43:39PM +0200, Gerrit P. Haase wrote:
> Nope, Teat::Harness now at version 2.52:
>
> $ make test
> /usr/bin/perl.exe "-MExtUtils::Command::MM" "-e" "test_harness(0,
> 'blib/lib', 'blib/arch')" t/*.t
Ok, see if MakeMaker has anything to do with this by running:
perl -MT
Michael G Schwern wrote:
On Sun, Jul 17, 2005 at 10:16:40PM +0200, Gerrit P. Haase wrote:
I'm going to guess somethign wacky is going on inside Test::Harness. What
version is being used?
The dafault Test::Harness which is included with perl-5.8.7:
$ perl -e 'use Test::Harness; print "$Test:
On Sun, Jul 17, 2005 at 10:16:40PM +0200, Gerrit P. Haase wrote:
> >I'm going to guess somethign wacky is going on inside Test::Harness. What
> >version is being used?
>
> The dafault Test::Harness which is included with perl-5.8.7:
> $ perl -e 'use Test::Harness; print "$Test::Harness::VERSION\n
On Sun, Jul 17, 2005 at 10:39:05PM +0200, Gerrit P. Haase wrote:
> >Hmm. That's very strange. Particularly the oscilating nature. The reason
> >its failing only when in lib/ is because '.' is in @INC so it can find
> >the module. If you run the tests under taint mode (which removes the .)
> >th
Michael G Schwern wrote:
On Sun, Jul 17, 2005 at 12:57:17PM +0200, Gerrit P. Haase wrote:
we are on Cygwin: 1.5.18(0.132/4/2) 2005-07-02 20:30, perl is the latest
stable release: This is perl, v5.8.7 built for cygwin-thread-multi-64int
We have a problem building modules, i.e. make test fails
Hi Michael,
Michael G Schwern wrote:
On Sun, Jul 17, 2005 at 12:57:17PM +0200, Gerrit P. Haase wrote:
we are on Cygwin: 1.5.18(0.132/4/2) 2005-07-02 20:30, perl is the latest
stable release: This is perl, v5.8.7 built for cygwin-thread-multi-64int
We have a problem building modules, i.e. ma
Hello,
we are on Cygwin: 1.5.18(0.132/4/2) 2005-07-02 20:30, perl is the latest
stable release: This is perl, v5.8.7 built for cygwin-thread-multi-64int
We have a problem building modules, i.e. make test fails if a .pm file
in the source package is located below the root in a subdirectory called
35 matches
Mail list logo
