Bug #100489 for Test-HexDifferences: Test failure with DCANTRELL/Test-Differences-0.63.tar.gz

Fri Nov 21 02:47:30 2014 SREZIC [...] cpan.org - Ticket created

Subject:

Test failure with perl 5.21.6

12_dumped_eq_dump_or_diff.t fails with perl 5.21.6. With perl 5.21.5 the test suite runs fine, so this may or may not be a regression in perl itself: # Failed test 'subtest 'expected undef' of '' compare diag' # at t/12_dumped_eq_dump_or_diff.t line 37. # Got diag (282 bytes): # '+---+-------------------------+---+----------+' # '| Ln|Got | Ln|Expected |' # '+---+-------------------------+---+----------+' # '* 1|'0000 : 31 : 1 * 1|undef *' # '* 2|' * | |' # '+---+-------------------------+---+----------+' # Expected diag (294 bytes): # '+---+---------------------------+---+----------+' # '| Ln|Got | Ln|Expected |' # '+---+---------------------------+---+----------+' # '* 1|'0000 : 31 : 1\n * 1|undef *' # '* 2|' * | |' # '+---+---------------------------+---+----------+' # Looks like you failed 1 test of 30. t/12_dumped_eq_dump_or_diff.t .. Dubious, test returned 1 (wstat 256, 0x100) Failed 1/30 subtests

Sat Nov 22 20:05:56 2014 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Correspondence added

On Fri Nov 21 02:47:30 2014, SREZIC wrote: Show quoted text

> 12_dumped_eq_dump_or_diff.t fails with perl 5.21.6. With perl 5.21.5 > the test suite runs fine, so this may or may not be a regression in > perl itself: > > # Failed test 'subtest 'expected undef' of '' compare diag' > # at t/12_dumped_eq_dump_or_diff.t line 37. > # Got diag (282 bytes): > # '+---+-------------------------+---+----------+' > # '| Ln|Got | Ln|Expected |' > # '+---+-------------------------+---+----------+' > # '* 1|'0000 : 31 : 1 * 1|undef *' > # '* 2|' * | |' > # '+---+-------------------------+---+----------+' > # Expected diag (294 bytes): > # '+---+---------------------------+---+----------+' > # '| Ln|Got | Ln|Expected |' > # '+---+---------------------------+---+----------+' > # '* 1|'0000 : 31 : 1\n * 1|undef *' > # '* 2|' * | |' > # '+---+---------------------------+---+----------+' > # Looks like you failed 1 test of 30. > t/12_dumped_eq_dump_or_diff.t .. > Dubious, test returned 1 (wstat 256, 0x100) > Failed 1/30 subtests

This broke some time after v5.21.5-539-g6dfba0a. A binary search would help.

Sat Nov 22 20:05:57 2014 The RT System itself - Status changed from 'new' to 'open'

Sun Nov 23 13:10:54 2014 ANDK [...] cpan.org - Correspondence added

Is this not just due to DCANTRELL/Test-Differences-0.63.tar.gz ?

Sun Nov 23 13:59:00 2014 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Correspondence added

On Sun Nov 23 13:10:54 2014, ANDK wrote: Show quoted text

> Is this not just due to DCANTRELL/Test-Differences-0.63.tar.gz ?

How insightful! Indeed, the tests pass if I use DCANTRELL/Test-Differences-0.62.tar.gz.

Sun Nov 23 14:00:45 2014 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Reference by ticket #100515 added

Mon Nov 24 10:07:23 2014 dcantrell [...] cpan.org - Correspondence added

I'm inclined to say that this is Test::HexDifferences being too fussy. I don't think it's appropriate to rely so exactly on the diagnostic output from testing code. That output (and the change in Test::Differences 0.63) is designed to make the output useful to humans, not to computers. Test::HexDifferences should probably be using Text::Diff directly instead of wrapping Test::Differences.

Sun Dec 21 17:23:41 2014 SREZIC [...] cpan.org - Subject changed from 'Test failure with perl 5.21.6' to 'Test failure with DCANTRELL/Test-Differences-0.63.tar.gz'

Sun Jul 05 10:23:20 2015 STEFFENW [...] cpan.org - Correspondence added

Thank you for the good hints, what's the problem and how to fix. Solved in version 0.009 Renewed with Test::Diff and without Test::Differences in 1.000

Sun Jul 05 10:23:23 2015 STEFFENW [...] cpan.org - Status changed from 'open' to 'resolved'

Sun Jul 05 10:23:25 2015 STEFFENW [...] cpan.org - Fixed in 0.009 added