Skip Menu |

This queue is for tickets about the Test-Mimic-Recorder CPAN distribution.

Report information
The Basics
Id: 117499
Status: open
Priority: 0/
Queue: Test-Mimic-Recorder

People
Owner: Nobody in particular
Requestors: SREZIC [...] cpan.org
Cc:
AdminCc:

Bug Information
Severity: (no value)
Broken in: 0.012005
Fixed in: (no value)



Subject: Tests fail with perl 5.19.11 and later
RT-Send-CC: ANDK [...] cpan.org
On Sun Sep 04 15:24:35 2016, SREZIC wrote: Show quoted text
Bisect?
Show quoted text
> Bisect?
commit 2484f8dbbb584bb7bc7107e23ebd49f694eb3ea2 Author: Father Chrysostomos <sprout@cpan.org> Date: Sun Jun 23 07:08:24 2013 -0700 Stop folding of ops from changing mutability
On Mon Sep 05 14:55:17 2016, ANDK wrote: Show quoted text
> > Bisect?
> > commit 2484f8dbbb584bb7bc7107e23ebd49f694eb3ea2 > Author: Father Chrysostomos <sprout@cpan.org> > Date: Sun Jun 23 07:08:24 2013 -0700 > > Stop folding of ops from changing mutability
I was planning to write a patch for this module, but I have neither the time nor the energy for it, as it is proving to be more work than I bargained for. I can, however, explain what perl commit 2484f8d is doing. The return value of 3+2 (for example) is no longer a constant 5 if you use it in lvaluish context such as \(3+2) or foreach(3+2). Each time that you reference it you will get a difference scalar. I suspect the module is depending on \(3+2) (or whatever) returning a reference to the same constant every time. Note also that changes in 5.22 may break this module further. In 5.22 a sub ref may be stored directly in the stash, without there having to be a glob. In general, if you have code to handle special symbol table entries other than typeglobs, you should check that the entry is on a type that your code supports, and vivify the typeglob (\*{$name}) otherwise, to make the code future-proof.