New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
perl 5.27.1 breaks Test2::Plugin::IOSync #16069
Comments
From @exodistExample reports: This module passes on perls 5.8.8 -> 5.26 so long as Test-Simple and Strangely 5.8.1 fails with the exact same problems as 5.27.1, maybe a From my own debugging I see 2 problems, they may be related: 1) writes appear to happen out of order (that is we have multiple clones of 2) PerlIO layers seem to be applied differently than they were before, Please note that there is a module all these load called All these problems are related to PerlIO layers, the module itself uses I tried to break this down into a minimal repro case, but was not able to -Chad Perl Info
|
From @jkeenanOn Fri, 07 Jul 2017 20:25:57 GMT, exodist7@gmail.com wrote:
Here is the commit to blead which appears to have broken this CPAN module: ##### [perl #131221] improve duplication of :via handles At the immediately preceding commit: ##### At the breaking commit: ##### t/IOEvents.t (Wstat: 256 Tests: 1 Failed: 1) But note that in the test run above I got failures in 8 test files. The number of files with failures in at least one of the CPANtesters reports was only 3. I haven't kept up with Test2 in the past two years, so I'm unfamiliar with the style of testing done in the failing files and cannot further diagnose the problem. Thank you very much. |
The RT System itself - Status changed from 'new' to 'open' |
From @exodistWould it be possible to make it not diplicate the via layer when the getarg On Jul 7, 2017 4:26 PM, "James E Keenan via RT" <perlbug-followup@perl.org>
|
From @exodistI have found ways to mitigate all the breakages that do not involve changes to perl itself, and they do not break older versions. If nothing else is done for this task I should be ok. That said, I want to propose that a via layer without a GETARGS should simply not be applied during cloning. Or at the very least make it so that a GETARGS that is present, but returns undef can implement that behavior. |
From @jkeenanOn Sat, 08 Jul 2017 03:25:53 GMT, exodist7@gmail.com wrote:
Resolving ticket.
Chad, if you would like to get discussion going about this, I'd suggest sending email to p5p with fresh discussion about it. Thank you very much. |
@jkeenan - Status changed from 'open' to 'rejected' |
Migrated from rt.perl.org#131716 (status was 'rejected')
Searchable as RT131716$
The text was updated successfully, but these errors were encountered: