develooper Front page | perl.perl5.porters | Postings from May 2019

It's time for Release Candidates for 5.30

From:
Karl Williamson
Date:
May 8, 2019 18:14
Subject:
It's time for Release Candidates for 5.30
Message ID:
9f7866b4-efe0-eeba-5df2-7b88a5578743@khwilliamson.com
Here are all the pending issues.

First blockers,

a) https://rt.perl.org/Ticket/Display.html?id=133958
Tony has a patch that he's planning to apply today

b) https://rt.perl.org/Ticket/Display.html?id=134031
This turns out to caused by a bug in pow() on Free BSD.  The distro knew 
something was wrong and wrote a work around that depended on faulty 
string to number conversion that we have now fixed.  There is some less 
than ideal code in the distro that could be easily changed that 
immunizes it against this bug.  The original author of the distro told 
me it shouldn't be a blocker.  I have filed a ticket against Free BSD
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=237800
and the distro itself
https://rt.cpan.org/Ticket/Display.html?id=129486&results=611425c71d12a91c868bbf130a310979
In any event, there's nothing for perl to do, unless we want to write 
our own pow() for integer inputs, which doesn't feel right to me.  So 
this ticket should be removed from the blockers list

c) https://rt.perl.org/Ticket/Display.html?id=134044
I have a patch that could be applied.  But these undefined behaviors are 
not new in 5.29.  We either apply the patch (just the first one of two, 
I think) or we remove it from the blockers list.

d) https://rt.perl.org/Ticket/Display.html?id=128972
Again I have a patch, that I've tested on Tony Cook's machine, but it 
doesn't fail there even without the patch.  This ticket has been 
deferred before, so it can be deferred again.  Either apply the patch or 
remove from the blockers list.

Thus all four blockers have been dealt with, or could be dealt with 
immediately.

Then the other potential issues.

Unicode 12.1 was released yesterday with no changes from what we have in 
blead.

cpan/Win32.c has a fix to #include a missing hdr which prevents it from 
compiling on cygwin with g++.  We could take this, or defer it.
http://nntp.perl.org/group/perl.perl5.porters/254714

There's a missing cast in locale.c which causes it not to compile on 
early gcc.
https://rt.perl.org/Ticket/Display.html?id=134080
This is trivial to fix, or it could wait

I believe the project was harmed last year by the late release of 5.28. 
And I don't want that to happen again.  We're at the point where we 
should be making a couple final decisions and then making RCs.



nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About