00:00 lucasb left 00:11 sena_kun joined 00:12 Altai-man_ left 00:36 pmurias left 00:38 pmurias joined 00:43 pmurias left 02:09 Altai-man_ joined 02:12 sena_kun left 02:25 sena_kun joined 02:27 squashable6 left, Altai-man_ left 02:29 squashable6 joined 03:09 Altai-man_ joined 03:12 sena_kun left 05:10 sena_kun joined 05:11 Altai-man_ left 07:09 Altai-man_ joined 07:12 sena_kun left 09:10 sena_kun joined 09:12 Altai-man_ left
nine Xliff: the temporary name would not prevent multiple processes from wasting time, precompiling the same file. 09:45
lizmat Files=1296, Tests=109719, 210 wallclock secs (28.32 usr 8.24 sys + 2947.30 cusr 274.94 csys = 3258.80 CPU) 10:27
10:36 robertle joined
Geth rakudo: 7b07518843 | (Elizabeth Mattijsen)++ | 2 files
Move ValueObjAt definition to the bootstrap

We need it earlier than the core setting.
11:01
11:09 Altai-man_ joined 11:12 sena_kun left
lizmat m: CORE::.raku # was trying to fix this, but this proved to be more gnarly than I hoped it would be 11:30
camelia No such method 'WHICH' for invocant of type 'ContainerDescriptor'
in block <unit> at <tmp> line 1
Geth roast: aa898a3e9d | (Elizabeth Mattijsen)++ | 13 files
Batch #16 of roast RT -> GH ticket migration
11:51
|Tux| Rakudo version 2019.11-465-g7b0751884 - MoarVM version 2020.01-15-g1fa9764bf
csv-ip5xs0.725 - 0.756
csv-ip5xs-206.417 - 6.444
csv-parser23.215 - 23.384
csv-test-xs-200.431 - 0.431
test7.177 - 7.291
test-t1.769 - 1.787
test-t --race0.791 - 0.816
test-t-2029.722 - 30.448
test-t-20 --race8.547 - 8.582
12:00
Geth rakudo: 3dd9edffc6 | (Elizabeth Mattijsen)++ | src/core.c/Supply-factories.pm6
Implement Supply.comb(N)
13:02
roast: a30f436ffb | (Elizabeth Mattijsen)++ | S17-supply/comb.t
Add tests for Supply.comb(N)
13:03
13:10 sena_kun joined 13:12 Altai-man_ left 13:55 benchable6 left 13:56 greppable6 left, shareable6 left, nativecallable6 left, unicodable6 left, evalable6 left, tellable6 left, notable6 left 13:57 reportable6 left, quotable6 left 13:58 squashable6 left, bisectable6 left, releasable6 left, committable6 left, sourceable6 left 13:59 statisfiable6 left, coverable6 left, bloatable6 left 14:11 lucasb joined 14:20 shareable6 joined, greppable6 joined, nativecallable6 joined, benchable6 joined 14:37 quotable6 joined 14:38 tbrowder joined, unicodable6 joined, reportable6 joined, notable6 joined 14:40 evalable6 joined 14:41 tellable6 joined 14:46 Xliff left 14:56 bloatable6 joined, coverable6 joined 14:57 statisfiable6 joined, bisectable6 joined, sourceable6 joined 14:58 releasable6 joined, committable6 joined, squashable6 joined 15:09 Altai-man_ joined 15:12 sena_kun left
Altai-man_ a non-offensive remainder the release is blocked by two regressions 16:13
tellable6 2020-01-13T12:22:32Z #raku-dev <lizmat> Altai-man_: I have *no* idea what is causing the error: all calls to `return` in that section of code, have been wrapped into a `try` already :-(
AlexDaniel Altai-man_: tickets? 16:17
releasable6: status
releasable6 AlexDaniel, Next release will happen when it's ready. 1 blocker. 0 out of 466 commits logged
AlexDaniel, Details: gist.github.com/e02a06b0e27a09cbfb...43b1aa8c30
AlexDaniel Altai-man_: things will go a bit faster if these are filed separately
Altai-man_ ADT and overload::constant 16:18
AlexDaniel too much noise in that thread :)
Altai-man_ will file tickets after $dayjo
16:35 domidumont joined 17:10 sena_kun joined 17:11 Altai-man_ left
AlexDaniel timotimo: did you look at ADT? 18:17
I don't understand what's going on there
18:23 sena_kun left 18:24 MasterDuke joined 18:29 SmokeMachine left 18:31 kawaii left 18:33 sena_kun joined, Geth left 18:34 kawaii joined 19:11 sena_kun left 19:26 sena_kun joined
lizmat sena_kun: I've reverted the commits that caused overload::constant breakage 19:37
sena_kun lizmat: and it works now? (I'll check myself, but just in case) 19:41
lizmat sena_kun: I haven't checked
I have checked: and no it doesn't fix the issue 19:42
Non ast passed to WANTED: List
===SORRY!===
Cannot find method 'returns' on object of type List
So bisect is wrong, it seems :-(
19:51 domidumont left
lizmat what is a native buf ? 19:54
was that ever specced ?
sena_kun :S 19:59
then this is not a culprit...
sena_kun re-runs blin on ADT 20:00
AlexDaniel it doesn't depend on anything so I think you can just gist the whole thing 20:03
let's see 20:04
c: HEAD gist.github.com/AlexDaniel/2119ef7...6bbc011f30
committable6 AlexDaniel, Using file “01-tree.p6” as a main file, other files are placed in “./sandbox”
AlexDaniel, ¦HEAD(0f0947b): ««timed out after 10 seconds» «exit signal = SIGHUP (1)»»
AlexDaniel c: HEAD gist.github.com/AlexDaniel/2119ef7...6bbc011f30
committable6 AlexDaniel, Using file “01-tree.p6” as a main file, other files are placed in “./sandbox”
AlexDaniel, ¦HEAD(0f0947b): ««timed out after 10 seconds» «exit signal = SIGHUP (1)»»
AlexDaniel oh come on?
committable6: timeout=30
committable6 AlexDaniel, timeout is now set to “30” (default value is “10”)
AlexDaniel c: HEAD gist.github.com/AlexDaniel/2119ef7...6bbc011f30
committable6 AlexDaniel, Using file “01-tree.p6” as a main file, other files are placed in “./sandbox”
AlexDaniel, ¦HEAD(0f0947b): ««timed out after 10 seconds» «exit signal = SIGHUP (1)»» 20:05
AlexDaniel c: HEAD,HEAD,HEAD,HEAD gist.github.com/AlexDaniel/2119ef7...6bbc011f30
committable6 AlexDaniel, Using file “01-tree.p6” as a main file, other files are placed in “./sandbox”
AlexDaniel, ¦HEAD,HEAD,HEAD,HEAD (4 commits): ««timed out after 10 seconds» «exit signal = SIGHUP (1)»»
AlexDaniel ok there was too much cruft in there 20:06
sena_kun Blin points on add2ec0d491218e59e6c7a517e37458e1a6c3daf
AlexDaniel bisect: old=2019.11 gist.github.com/AlexDaniel/2119ef7...6bbc011f30 20:07
bisectable6 AlexDaniel, Using file “01-tree.p6” as a main file, other files are placed in “./sandbox”
AlexDaniel, Bisecting by exit code (old=2019.11 new=0f0947b). Old exit code: 0
AlexDaniel, bisect log: gist.github.com/de0beac95a553ba4f1...0a8f0287c6
AlexDaniel, (2020-01-03) github.com/rakudo/rakudo/commit/ad...8e1a6c3daf
AlexDaniel sena_kun: yeah :)
c: 2019.11,add2ec0d49^,add2ec0d49,HEAD gist.github.com/AlexDaniel/2119ef7...6bbc011f30
committable6 AlexDaniel, Using file “01-tree.p6” as a main file, other files are placed in “./sandbox”
AlexDaniel, gist.github.com/0b3e4e7832c13d43ca...bd341f7a39 20:08
AlexDaniel bisect: old=add2ec0d49 gist.github.com/AlexDaniel/2119ef7...6bbc011f30 20:09
bisectable6 AlexDaniel, Using file “01-tree.p6” as a main file, other files are placed in “./sandbox”
AlexDaniel, Bisecting by exit code (old=add2ec0 new=0f0947b). Old exit code: 3
AlexDaniel, bisect log: gist.github.com/5dda407dc9c23702f2...9302d4a2e7
AlexDaniel, (2020-01-14) github.com/rakudo/rakudo/commit/e7...7da154d255
AlexDaniel huh? 20:10
c: e76f8e4d3e^,e76f8e4d3e gist.github.com/AlexDaniel/2119ef7...6bbc011f30
committable6 AlexDaniel, Using file “01-tree.p6” as a main file, other files are placed in “./sandbox”
AlexDaniel, gist.github.com/70baea0a88486f0157...b5c60ecacd
AlexDaniel see, that commit is also problematic somehow
lizmat: ↑ somehow this recent commit affects ADT module (2020-01-14) github.com/rakudo/rakudo/commit/e7...7da154d255 20:11
lizmat AlexDaniel: but the breakage of ADT predates that, no ? 20:12
I mean, that one was from yesterday
AlexDaniel lizmat: yes, and then it breaks a bit more :)
lizmat wow 20:13
AlexDaniel “# You failed 3 tests of 7” vs “# You planned 7 tests, but ran 6 # You failed 5 tests of 6”
lizmat I see... the first test used to return some bogus structure, it now returns Nil 20:14
hmmm
AlexDaniel lizmat: the .raku part breaks because it does `add_method` for “perl” and “gist” 20:15
I'm not sure how that can be fixed on our side… 20:16
like, the module strongly assumes that the method name is .perl
lizmat addiing a "raku" to that list, does *not* fix the problem
as long as the module inherits from Mu, Mu::raku will call module.perl when needed
m: class A { method perl() { say "perl" } }; A.raku 20:17
camelia ( no output )
lizmat hmmm.. apparently not ?
testing a fix 20:28
ok, the fix appears to fix the overload::constant one 20:37
20:37 Geth joined
Geth rakudo: b1e986c282 | (Elizabeth Mattijsen)++ | src/core.c/Mu.pm6
Make sure that .raku will call .perl on a class that has that

Apparently I thought I had implemented that already, but obviously I hadn't. This means that:
   class A { method perl() { say "perl" } }
   A.raku; # perl
   A.new.raku; # perl
Also took the opportunity to rewrite Mu.raku using ternaries.
20:43
lizmat AlexDaniel sena_kun ^^
sena_kun lizmat: is it a fix for ADT? :) 20:44
AlexDaniel lizmat++
sena_kun: for the first part, it seems
sena_kun awesome!
lizmat++
lizmat it is a fix for overload::constant
sena_kun that's already great. :] 20:45
lizmat now reverting the earlier reverts to see if overload::constant is still fixed
nope, still needs the THROW revert 20:46
sena_kun lizmat: which ones should I cherry-pick? 20:52
or, hmm, Blin can't run branches. :S
so cherry-picking won't really help for another round of blin...
lizmat b1e986c282adcdd86529f in any case
sena_kun ouch, cherry-picking it has a conflict... 20:54
releasable6: status
releasable6 sena_kun, Next release will happen when it's ready. 3 blockers. 0 out of 470 commits logged
sena_kun, Details: gist.github.com/fb80c151997883ed71...cbaf3486e9
20:55 AlexDaniel left 21:05 MasterDuke left
lizmat aaaaahhh... it looks like "when" codegens as a succeed 21:10
and overload::constant uses "when"
so that at least ties those things together 21:11
21:12 sena_kun left 21:27 sena_kun joined
sena_kun lizmat: does it mean you can write a fix? 21:30
lizmat sena_kun: the fix is in the two commits I reverted
I don't understand *why* that fixes it 21:31
but overload::constant uses "when" inside the slang
and "when" codegens to throw a succeeds
so that at least is a bit of a smoking gun, but still not sure who supplied the bullet or who pulled the trigger 21:32
so HEAD is now clean on overload::constant
sena_kun roger that 21:33
as for cherry-picking: the problem is that the fix includes parts from `is_built` commits and I, apparently, can't smoothly pick it. 21:34
at this level I wonder if we should just move the release commit to HEAD again, that'd add some pain in writing changelog, but still
lizmat before we do that, let me look some more at ADT
sena_kun ok, I'll write a bit with running Blin on master
that'd be super great to address ADT today to get Blin results tomorrow's morning. 21:35
21:38 lucasb left
lizmat ok, confirmed that add2ec0d491218e59e6c7a is what breaks ADR 21:47
*ADT
sena_kun \o/ 21:50
22:01 AlexDaniel joined, AlexDaniel left, AlexDaniel joined 22:03 Xliff joined 22:24 Kaiepi left 22:25 Kaiepi joined 22:40 sena_kun left 22:56 sena_kun joined
lizmat ok, some ADT progress: 23:00
adding "raku" to the list of methods to be generated, makes it pass the raku / perl rename
I'll write it on the issue :-) 23:01
sena_kun lizmat++
sena_kun runs blin for HEAD and goes to sleep 23:02
Xliff Hah! Thought that might be the case. 23:18
Geth rakudo: f1f690e045 | (Elizabeth Mattijsen)++ | src/Perl6/bootstrap.c/BOOTSTRAP.nqp
Make sure that Attribute.new.is_built defaults correctly

When the "is built" trait was added, all calls in the bootstrap and the setting that created attributes, had the new "is_built" attribute also specified. That was in fact wrong. What *should* have been done, is to make sure the new "is_built" attribute would default to the value (implicitely) given by the "has_accessor" attribute. Only that would ensure that Attribute's contract would not be broken.
This solves the ADT blocker (#3427)
23:22
Xliff m: my $s = Supplier.new; my $sup = $s.Supply; $s.^lookup('tap').^name.say 23:23
camelia Mu
Xliff m: my $s = Supplier.new; my $sup = $s.Supply; $s.^lookup('tap')
camelia ( no output )
Xliff m: my $s = Supplier.new; my $sup = $s.Supply; $sup.^lookup('tap').^name.say
camelia Method
Geth rakudo: ada7c36548 | (Elizabeth Mattijsen)++ | 2 files
Remove now needless specification of "is_built"
23:31
lizmat also calls it a day& 23:32