Re: Is this a release blocker for 5.43.9?
[email protected] (Eric Herman) Fri, 20 Mar 2026 14:08:28 +0100
| Newsgroups | perl.perl5.porters |
|---|---|
| Message-ID | <CAE3QHemguZpcK1GErocztwaw0FMDzi-Oiu5+bVGyoXOG-ueuTw@mail.gmail.com> |
Thank you for confirming and creating https://github.com/Perl/perl5/issues/24302 Unless someone advises otherwise in the next short while, I shall continue with the release as-is. Cheers, Eric On Fri, 20 Mar 2026 at 12:37, James E Keenan <[email protected]> wrote: > On 3/20/26 06:54, Eric Herman wrote: > > While testing 5.43.9 tarballs prior to release, I encountered a > > YAML::PP failure related to Tie::IxHash. > > > > We bisected the commits between 5.43.8 and 5.43.9 (not pushed) and found > > > > # first bad commit: [b3f68b479c9876437620503fb7d54e14bceadcee] Convert > > OP_CONST hash keys in key+val lists to HEKs > > > > Should this block the 5.43.9 release? > > > > IMO, it should not block 5.43.9 -- though we should, of course, try to > fix it before then if possible. > > We generally reserve the term "release-blocker" for something that must > be fixed before the annual *production* release can be issued. To me, > at least, this implies a stricter standard than we would use for a > monthly dev release. > > At the moment I'm confirming your report with our regular > Porting/bisect.pl program. When that is done I'll open a bug ticket and > label it BBC. That should be the place where we discuss the specific > Tie-IxHash/YAML-PP problem. If this problem is severe enough, we can > always slap a "release blocker" label on the ticket. >