Bug #27585 for Tie-EncryptedHash: reforgy fails in 1 out of ~30 test runs

Thu Jun 14 16:02:59 2007 ANDK [...] cpan.org - Ticket created

Subject:

reforgy fails in 1 out of ~30 test runs

You may have seen the testreport on http://www.nntp.perl.org/group/perl.cpan.testers/2006/11/msg374703.html This failure can be reproduced with these two lines of perl: $h{B}[4]{B} = { B => "B" }; $h{B}[4]{B}[4] = { B => [1..1] }; which stand for lines 33 and 38 of the test script. It just randomly happens not so often that these lines conflict. FWIW, my 321 smokes failed 10 times. Calculating the expected ratio on an infinite number of tests is left as an exercise to the reader:) Kind Regards,

Mon Nov 26 07:11:15 2007 cpan [...] pjedwards.co.uk - Correspondence added

From:

cpan [...] pjedwards.co.uk

I also experienced the same problem with t/reforgy.t in: Tie::EncryptedHash 1.21 Failure output is: 1..1 generating a large, random data structure... Not an ARRAY reference at t/reforgy.t line 38. Changed to avoid failure: ==== t/reforgy.t#1 - t/reforgy.t ==== 33c33 < $h{ralpha()}->[rnum()]->{ralpha()} = { ralpha() => ralpha() }; --- Show quoted text

> $h{ralpha()}->[rnum()]->{'HASH_'.ralpha()} = { ralpha() =>

ralpha() }; 38c38 < $h{ralpha()}->[rnum()]->{ralpha()}->[$_] = { ralpha() => [1..rnum()] }; --- Show quoted text

> $h{ralpha()}->[rnum()]->{'ARRAY_'.ralpha()}->[$_] = {

ralpha() => [1..rnum()] }; Cheers, Peter (Stig) Edwards

Mon Nov 26 07:11:18 2007 The RT System itself - Status changed from 'new' to 'open'

Tue Jul 15 05:23:32 2008 VIPUL [...] cpan.org - Correspondence added

Oops! Thanks for finding this rather hairy bug and for the patch. Applied in 1.23. cheers, vipul

Tue Jul 15 05:25:51 2008 VIPUL [...] cpan.org - Status changed from 'open' to 'resolved'