Bug #52724 for Parse-Marpa: Undefined subroutine die_with_parse

Sun Dec 13 01:55:30 2009 grandpa [...] cpan.org - Ticket created

Subject:

Undefined subroutine die_with_parse_failure

use strict; The code below generates the run time error 'Undefined subroutine &Parse::Marpa::Grammar::die_with_parse_failure called at C:/Perl/site/lib/Parse/Marpa.pm line 172'. use warnings; use Parse::Marpa; say ${Parse::Marpa::mdl ( (do {local $/ = undef; my $source = <DATA>; \$source;}), \('2') ) }; __DATA__ semantics are perl5. version is 1.004000. start symbol is Expression. Expression: /\d+/, /[+]/, /\d+/. q{$_[0] + $_[2]}. It seems likely all occurrences of Parse::Marpa::Grammar::die_with_parse_failure should be replaced by 'Parse::Marpa::Internal::Grammar::die_with_parse_failure' in Marpa.pm Are there other similar changes that need to be made?

Sun Dec 13 12:06:04 2009 jkegl [...] cpan.org - Taken

Sun Dec 13 12:10:35 2009 jkegl [...] cpan.org - Correspondence added

Thanks. I feel a bit embarrassed about this one. This looks like exactly the same bug as 45136. It was a simple fix, but obviously I screwed it up.

Sun Dec 13 12:10:36 2009 The RT System itself - Status changed from 'new' to 'open'

Sun Dec 13 13:02:11 2009 jkegl [...] cpan.org - Correspondence added

I just uploaded Parse::Marpa 1.005_000 to CPAN. This time I've added a regression test. This is a developer's release. If all looks well in a few days, I'll make it an official release as 1.006000.

Sun Dec 13 13:03:02 2009 jkegl [...] cpan.org - Broken in 1.004000 added

Mon Dec 14 22:18:59 2009 jkegl [...] cpan.org - Correspondence added

Just uploaded 1.005_001. The new regression test didn't make it into the MANIFEST in the previous release. I'd like to see what cpantesters has to say about this fix before making the official release. On Sun Dec 13 13:02:11 2009, JKEGL wrote: Show quoted text

> I just uploaded Parse::Marpa 1.005_000 to CPAN. This time I've added a > regression test. > > This is a developer's release. If all looks well in a few days, I'll > make it an official release as 1.006000.

Wed Dec 16 13:01:03 2009 jkegl [...] cpan.org - Correspondence added

Just uploaded 1.005_002. Fixing a portability issue in the regression test. The regression test is a test of an error message, and what with character counting, different file names, etc., error messages can be difficult to compare across systems. The fix itself looks fine, but I won't make the official release until the testing is in order. On Mon Dec 14 22:18:59 2009, JKEGL wrote: Show quoted text

> Just uploaded 1.005_001. The new regression test didn't make it into > the MANIFEST in the previous release. I'd like to see what cpantesters > has to say about this fix before making the official release. > > On Sun Dec 13 13:02:11 2009, JKEGL wrote:

> > I just uploaded Parse::Marpa 1.005_000 to CPAN. This time I've added a > > regression test. > > > > This is a developer's release. If all looks well in a few days, I'll > > make it an official release as 1.006000.

>

Thu Dec 17 22:02:47 2009 jkegl [...] cpan.org - Correspondence added

I have a decision to make on the current release candidate, which I'll describe for anyone interested. The new regression test fails on exactly one platform, due to an extra space in the output. I strongly suspect a bug in the test environment, as other tests on similar platforms pass. The decision: 1.) Ignore the one cpantesters FAIL as spurious. Go straight to an official release. 2.) Work around it. Change the test so the spacing issue is ignored, and create another release candidate.

Fri Dec 18 18:42:29 2009 jkegl [...] cpan.org - Correspondence added

Other cpantester FAIL's have come in, so my theory about that one FAIL being due to an alpha particle won't fly. :-) Fixing the one bug (abend on attempt to report parsing errors) uncovered glitches in the routine for reporting parsing errors. I've now fixed these newest glitches, and the fixes have passed one round of testing. The fixes will go into another release candidate, which I will shortly put onto CPAN.

Sat Dec 19 23:11:28 2009 jkegl [...] cpan.org - Correspondence added

I've uploaded release candidate 1.005_003. I'll wait to see what the cpantesters say. On Fri Dec 18 18:42:29 2009, JKEGL wrote: Show quoted text

> Other cpantester FAIL's have come in, so my theory about that one FAIL > being due to an alpha particle won't fly. :-) > > Fixing the one bug (abend on attempt to report parsing errors) uncovered > glitches in the routine for reporting parsing errors. I've now fixed > these newest glitches, and the fixes have passed one round of testing. > > The fixes will go into another release candidate, which I will shortly > put onto CPAN. > >

Sun Dec 27 21:05:26 2009 jkegl [...] cpan.org - Correspondence added

cpantesters liked 1.005_003, so I've made it official release 1.006000 just uploaded to CPAN. If nothing untoward happens over the next few days, I will close this bug as resolved.

On Sat Dec 19 23:11:28 2009, JKEGL wrote:
Show quoted text

> I've uploaded release candidate 1.005_003. I'll wait to see what the
> cpantesters say.
>
> On Fri Dec 18 18:42:29 2009, JKEGL wrote:

> > Other cpantester FAIL's have come in, so my theory about that one FAIL
> > being due to an alpha particle won't fly. :-)
> >
> > Fixing the one bug (abend on attempt to report parsing errors) uncovered
> > glitches in the routine for reporting parsing errors. I've now fixed
> > these newest glitches, and the fixes have passed one round of testing.
> >
> > The fixes will go into another release candidate, which I will shortly
> > put onto CPAN.
> >
> >

>
>

Thu Dec 31 01:13:20 2009 jkegl [...] cpan.org - Status changed from 'open' to 'resolved'

Thu Dec 31 01:13:21 2009 jkegl [...] cpan.org - Fixed in 1.006000 added

Bug #52724 for Parse-Marpa: Undefined subroutine die_with_parse_failure