ugexe Concurrent::Stack installs fine for me 01:08
japhb Apparently your local stress is not stressful enough .... 01:11
[Coke] I think I ratcheted down the flapper protection to Nil on blin, should probably bump it back up slightly. (I hate needing it. :| ) 01:41
... or maybe that was on the ake release scripts? 01:42
definitely don't want flapper protection on the release.
05:16 [Coke] left 05:20 [Coke] joined 09:57 rakkable left 09:58 rakkable joined 12:08 [Coke] left, [Coke]_ joined 12:22 [Coke]_ left 12:36 [Coke] joined
[Coke] 118 failures so far on ast blin run 12:38
looks like the test might be hanging around 915 done.
91%
ah, no, there is some progress still 12:39
91.17%
Killed some stalled tests. 12:59
(3 or 4 that were running >30m CPU time)
Probably need to add a Blin failure condition if the module's been running for x minutes to kill it and mark it as "too slow". 13:00
126 fails 13:56
releasable6 Next release in ā‰ˆ4 days and ā‰ˆ3 hours. There are no known blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 15:00
ugexe it might be time to hack in a rakuast skip list for blin 15:09
[Coke] I could also just retest the known failures from last tim. 15:12
*e
that's a much easier lift, probably.
133 fails, we're at 98.5% 15:13
(several of those may just show an end of output because I was killing things that were long running and making no progress/using no cpu)
(retest) and for now skip the moarvm ones. 15:15
ugexe what i mean is many of the failures are never going to pass
like the $*W ones 15:16
[Coke] No, I get it: 1) start with the list of all the previous failures, 2) remove all the ones we know will never pass. 3) test what's left. 15:57
and we can occasionally do a full run until ast is the default.
16:06 liztormato joined
liztormato rakkable: eco-provides $*W 16:06
meh 16:07
16:07 liztormato left
timo rakkable: eco-provides $*W 16:25
rakkable timo, Running: eco-provides $*W, please be patient!
timo, Found 110 lines in 38 files (30 distributions):
timo, gist.github.com/f921e83015f3affedc...e3f6c603c9
[Coke] ast run done, final results pushed. 16:50
17:03 [Coke] left 17:04 [Coke] joined
lizmat timo: thanks, so there's 30 dists that use $*W 17:10
ugexe some false positives in there 17:22
like Cro::HTTP it hit my $WRONG-METHOD = False;
lizmat hmmm... I though the search was verbatim hmmm... 17:23
ugexe it might be... and the * isn't rendering right sometimes or some such 17:25
like in Cro::HTTP its actually called $*WRONG-METHOD
lizmat right... but most of the matches appear to be legit 17:27
ugexe regarding that `while 0 { LAST ...; }` bug i mentioned a couple of days ago... what do we think about fixing that to not fire without revision gating it? or at least have it that way in rakuast? 17:34
lizmat I'd say it qualifies as a bug, and should therefore be fixed 17:36
lizmat recalls needing to go through hoops to make .map *not* run the LAST phaser if it didn't run at all 17:37
ugexe yeah i'm having a similar issue with lazy loops 17:59
m: my @a = (while 1 { LAST say 24; say 666; }); say 42 18:00
camelia 24
42
ugexe like that
m: my @a = (while 1 { LAST say 24; say 666; }); say 42; say $_ for @a; say 44 18:02
camelia 24
No such method '!capture_phasers' for invocant of type 'Code'
in block <unit> at <tmp> line 1

42
ugexe works with --optimize=off
right now i've got 10 branches im done with, but many of them are blocked by ones that will come before them (and thus it doesnt make sense to push them until the prereq branch gets merged) 20:22
i'll also be out of country for a week starting next weekend 20:23
Geth rakudo: ugexe++ created pull request #6466:
RakuAST: don't compose a class to resolve an attributive parameter		 21:46
rakudo: ugexe++ created pull request #6467:
RakuAST: stop reporting an orphaned declarator doc as a missing declarand		 21:47
rakudo: ugexe++ created pull request #6468:
RakuAST: allow a $!attr parameter in a sub nested in a method
rakudo: ugexe++ created pull request #6469:
RakuAST: don't fatalize the // left side of a dynamic-variable read		 22:07
rakudo/main: 6 commits pushed by (Nick Logan)++ 22:16
rakudo: ugexe++ created pull request #6470:
RakuAST: branch on a native-int condition directly		 22:17
rakudo: ugexe++ created pull request #6471:
RakuAST: don't run LAST in a loop that never runs its body		 22:28
ugexe jfc that LAST stuff really was convoluted and i really tried to find a way to simplify it but alas 22:29

Powered by Cro and the RakuĀ® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!