Skip Menu |

This queue is for tickets about the GraphViz CPAN distribution.

Report information
The Basics
Id: 80709
Status: resolved
Priority: 0/
Queue: GraphViz

People
Owner: Nobody in particular
Requestors: corsepiu [...] fedoraproject.org
Cc:
AdminCc:

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



Subject: GraphViz-2.12 t/foo.t fails
Date: Thu, 08 Nov 2012 17:08:53 +0100
To: bug-GraphViz [...] rt.cpan.org
From: Ralf Corsépius <corsepiu [...] fedoraproject.org>
Hi Ron, GraphViz-2.12 t/foo.t fails with the error below on Fedora-16, 17 (perl-5.15.x), 18 and rawhide (perl-5.16.x): + make test PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t t/dumper.t .. ok # Failed test at t/foo.t line 98. # ' graph [ratio=fill]; # ' # doesn't match '(?^:^\s*graph\s*\[bb=.*)' # Failed test at t/foo.t line 99. # ' node [label="\N"]; # ' # doesn't match '(?^:^.+ratio=fill)' # Failed test at t/foo.t line 100. # ' node1 [label=London, pos="38,18", width="1.0604", height="0.5"]; # ' # doesn't match '(?^:^\s*node\s*\[\s*label\s*=\s*"\\N"\s*\];\s*)' Use of uninitialized value $this in pattern match (m//) at t/foo.t line 101. # Failed test at t/foo.t line 101. # undef # doesn't match '(?^:^.+label=London,)' # Failed test at t/foo.t line 98. # ' graph [ratio=fill]; # ' # doesn't match '(?^:^\s*graph\s*\[bb=.*)' # Failed test at t/foo.t line 99. # ' node [label="\N"]; # ' # doesn't match '(?^:^.+ratio=fill)' # Failed test at t/foo.t line 100. # ' node1 [label=London, pos="38,18", width="1.0604", height="0.5"]; # ' # doesn't match '(?^:^\s*node\s*\[\s*label\s*=\s*"\\N"\s*\];\s*)' Use of uninitialized value $this in pattern match (m//) at t/foo.t line 101. # Failed test at t/foo.t line 101. # undef # doesn't match '(?^:^.+label=London,)' # Failed test at t/foo.t line 98. # ' graph [ratio=fill]; # ' # doesn't match '(?^:^\s*graph\s*\[bb=.*)' # Failed test at t/foo.t line 99. # ' node [label="\N"]; # ' # doesn't match '(?^:^.+ratio=fill)' # Failed test at t/foo.t line 100. # ' node1 [label=London, pos="38,18", width="1.0604", height="0.5"]; # ' # doesn't match '(?^:^\s*node\s*\[\s*label\s*=\s*"\\N"\s*\];\s*)' Use of uninitialized value $this in pattern match (m//) at t/foo.t line 101. # Failed test at t/foo.t line 101. # undef # doesn't match '(?^:^.+label=London,)' # Failed test at t/foo.t line 98. # ' graph [ratio=fill];' # doesn't match '(?^:^\s*graph\s*\[bb=.*)' # Failed test at t/foo.t line 99. # ' node [label="\N"];' # doesn't match '(?^:^.+ratio=fill)' # Failed test at t/foo.t line 100. # ' node1 [label=London, pos="38,18", width="1.0604", height="0.5"];' # doesn't match '(?^:^\s*node\s*\[\s*label\s*=\s*"\\N"\s*\];\s*)' Use of uninitialized value $this in pattern match (m//) at t/foo.t line 101. # Failed test at t/foo.t line 101. # undef # doesn't match '(?^:^.+label=London,)' # Failed test at t/foo.t line 98. # ' graph [ratio=fill];' # doesn't match '(?^:^\s*graph\s*\[bb=.*)' # Failed test at t/foo.t line 99. # ' node [label="\N"];' # doesn't match '(?^:^.+ratio=fill)' # Failed test at t/foo.t line 100. # ' node1 [label=London, pos="38,18", width="1.0604", height="0.5"];' # doesn't match '(?^:^\s*node\s*\[\s*label\s*=\s*"\\N"\s*\];\s*)' Use of uninitialized value $this in pattern match (m//) at t/foo.t line 101. # Failed test at t/foo.t line 101. # undef # doesn't match '(?^:^.+label=London,)' # Failed test at t/foo.t line 98. # ' graph [ratio=fill];' # doesn't match '(?^:^\s*graph\s*\[bb=.*)' # Failed test at t/foo.t line 99. # ' node [label="\N"];' # doesn't match '(?^:^.+ratio=fill)' # Failed test at t/foo.t line 100. # ' node1 [label=London, pos="38,18", width="1.0604", height="0.5"];' # doesn't match '(?^:^\s*node\s*\[\s*label\s*=\s*"\\N"\s*\];\s*)' Use of uninitialized value $this in pattern match (m//) at t/foo.t line 101. # Failed test at t/foo.t line 101. # undef # doesn't match '(?^:^.+label=London,)' # Looks like you failed 24 tests of 30. t/foo.t ..... Dubious, test returned 24 (wstat 6144, 0x1800) Failed 24/30 subtests t/pod.t ..... ok t/simple.t .. ok Test Summary Report ------------------- t/foo.t (Wstat: 6144 Tests: 30 Failed: 24) Failed tests: 2-5, 7-10, 12-15, 17-20, 22-25, 27-30 Non-zero exit status: 24 Files=4, Tests=71, 0 wallclock secs ( 0.03 usr 0.01 sys + 0.31 cusr 0.06 csys = 0.41 CPU) Result: FAIL Failed 1/4 test programs. 24/71 subtests failed. For GraphViz-2.11 the testsuite works without any such errors. Ralf
Hi Yes, you are correct. V 2.13 has been uploaded to CPAN. CHANGES: - Re-package distro because users get errors during testing. See RT#80709. Since I had this same error during my testing, I assume the uploaded version contains un-patched code. The errors are not in GraphViz, they are in the test code which has hard-coded line numbers where it looks for strings in the output. The output has been reformatted recently, and no longer matches those assumptions. See t/foo.t for details. Note: I did not write those tests :-).
Subject: Re: [rt.cpan.org #80709] GraphViz-2.12 t/foo.t fails
Date: Fri, 09 Nov 2012 02:31:46 +0100
To: bug-GraphViz [...] rt.cpan.org
From: Ralf Corsépius <corsepiu [...] fedoraproject.org>
On 11/08/2012 10:46 PM, RSAVAGE via RT wrote: Show quoted text
> <URL: https://rt.cpan.org/Ticket/Display.html?id=80709 > > > Hi > > Yes, you are correct. V 2.13 has been uploaded to CPAN.
GraphViz-2.13/t/foo.t exhibits exactly the same breakdown. Ralf
Hi Ralf On Thu Nov 08 20:34:59 2012, corsepiu@fedoraproject.org wrote: Show quoted text
> On 11/08/2012 10:46 PM, RSAVAGE via RT wrote:
> > <URL: https://rt.cpan.org/Ticket/Display.html?id=80709 > > > > > Hi > > > > Yes, you are correct. V 2.13 has been uploaded to CPAN.
> > GraphViz-2.13/t/foo.t exhibits exactly the same breakdown.
Not on my machine. I develop under Debian. Which OS are you using? I can't say the OS matters. I'm just trying to determine which factors are relevant. TIA.
Subject: Re: [rt.cpan.org #80709] GraphViz-2.12 t/foo.t fails
Date: Fri, 09 Nov 2012 05:39:28 +0100
To: bug-GraphViz [...] rt.cpan.org
From: Ralf Corsépius <corsepiu [...] fedoraproject.org>
Hi Ron, On 11/09/2012 05:06 AM, RSAVAGE via RT wrote: Show quoted text
> <URL: https://rt.cpan.org/Ticket/Display.html?id=80709 > > > Hi Ralf > > On Thu Nov 08 20:34:59 2012, corsepiu@fedoraproject.org wrote:
>> On 11/08/2012 10:46 PM, RSAVAGE via RT wrote:
>>> <URL: https://rt.cpan.org/Ticket/Display.html?id=80709 > >>> >>> Hi >>> >>> Yes, you are correct. V 2.13 has been uploaded to CPAN.
>> >> GraphViz-2.13/t/foo.t exhibits exactly the same breakdown.
> > Not on my machine. > > I develop under Debian. Which OS are you using?
I am the Fedora maintainer of GraphViz and am building it in a variety of environments/OSes in preparation to integrate it into Fedora. - Locally/native: Fedora 17/x86_64 - In local mock environments (Fedora/RH's variant of packaging build chroots): I.e. I am building in chroots for Fedora 16/17/18/rawhide x86_64, occasionally also for i386. - Inside of Fedora's official build system (@Redhat/fedora aka. koji). Cf. http://koji.fedoraproject.org/koji/taskinfo?taskID=4670363 for the build logs (check build.log) one of the breakdowns (This built is for rawhide - Fedora/Red Hat's development version). In all cases, the breakdown happens deterministically for me. Show quoted text
> I can't say the OS matters. I'm just trying to determine which factors > are relevant.
I am wondering about 2.13, esp. because the only changes I can spot between 2.12 and 2.13 seem to have been the version numbers. Typically, testsuite breakdowns like these on RH's/Fedora's are related to - missing package dependencies (In perl-terms: Missing other perl dists). - networking (koji/mock/the chroots do not properly support networking). - Actual bugs ;) Ralf
Subject: Re: [rt.cpan.org #80709] GraphViz-2.12 t/foo.t fails
Date: Fri, 09 Nov 2012 16:06:04 +1100
To: bug-GraphViz [...] rt.cpan.org
From: Ron Savage <ron [...] savage.net.au>
Hi Ralf On 09/11/12 15:42, Ralf Corsépius via RT wrote: Show quoted text
> Typically, testsuite breakdowns like these on RH's/Fedora's are related to > - missing package dependencies (In perl-terms: Missing other perl dists). > - networking (koji/mock/the chroots do not properly support networking). > - Actual bugs ;)
OK. I can see a problem with the line-spacing assumptions in the tests. I can fixed that and release a new version immediately. -- Ron Savage http://savage.net.au/ Ph: 0421 920 622
Hi Ralf V 2.14 is on CPAN.
Hi Ralf Fixed in V 2.14. Just closing the ticket.