AlexDaniel m: sub a(--> <0xf.8*:2[1,0]**0b11/0o200+2.0e2i>) { } 00:08
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed return value
at <tmp>:1
------> 3sub a(-->7⏏5 <0xf.8*:2[1,0]**0b11/0o200+2.0e2i>) { }
AlexDaniel what! 00:09
how is that malformed
m: dd <0xf.8*:2[1,0]**0b11/0o200+2.0e2i>
camelia ComplexStr.new(<0.96875+200i>, "0xf.8*:2[1,0]**0b11/0o200+2.0e2i")
AlexDaniel it doesn't like allomorphs? 00:10
m: sub a(--> <-3.1e-23-2.9e-23i>) { }; say +a 00:11
camelia -3.1e-23-2.9e-23i
AlexDaniel m: sub a(--> <0xf8+0b11i>) { }; say +a 00:16
camelia 248+3i
Geth rakudo: 86bac5ad59 | (Timo Paulssen)++ | src/Perl6/Grammar.nqp
heuristic for reduce metaop: don't get foiled by multiple [
 01:31
Geth roast: f59c3e7e84 | (Timo Paulssen)++ | S03-metaops/reduce.t
fix for R#2603: [[-1]] confuses rakudo parser
 01:53
linkable6 R#2603 [open]: github.com/rakudo/rakudo/issues/2603 [grammar and actions][parsing] Errors for "[[-" While Constructing Array
timotimo wait, were we going for a "only PRs get to go into master" thing just recently? 02:10
Geth ¦ problem-solving: fecundf assigned to jnthn Issue Introspection for operator precedence, associativity github.com/Raku/problem-solving/issues/222 04:54
lizmat timotimo: re "wait, were we going for a "only PRs get to go into master" thing just recently?" 07:16
no, I don't think there was an official decision, it's just that I decided that *I* will not commit anything directly to master anymore 07:17
Geth rakudo: 999680e657 | (Elizabeth Mattijsen)++ | src/core.c/Exception.pm6
Streamline X::Multi::Ambiguous listing

  - remove ":", it does not convey any information, clear it's a signature
  - add two spaces to set it off the first line
  - add "is default" status of the candidate when applicable		 08:07
rakudo: b8121cecb9 | (Elizabeth Mattijsen)++ | src/core.c/Exception.pm6
Lose the return constraint

It is not significant in dispatch, so should not be shown on a dispatch error.
rakudo: b005230615 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | src/core.c/Exception.pm6
Merge pull request #3857 from rakudo/ambiguous-is-default

Streamline X::Multi::Ambiguous listing
Geth rakudo/introducing-allomorph-class: 716cacdc4d | (Elizabeth Mattijsen)++ | 7 files
Introducing the allomorph class

  - a subclass of Str
  - IntStr/NumStr/RatStr/ComplexStr are now subclasses of allomorph
  - replaced several nqp::ifs by ternaries
  - streamlined many coercions, specifically RatStr.Numeric
  - simplified 16 infix:<eqv> candidates into a single one
... (8 more lines)		 08:57
rakudo: lizmat++ created pull request #3873:
Introducing the allomorph class
sena_kun releasable6, status 09:30
releasable6 sena_kun, Next release in ≈26 days and ≈9 hours. 1 blocker. Changelog for this release was not started yet
sena_kun, Details: gist.github.com/03ae9273799fe867a7...ffaf46b98e
MasterDuke ok, looks like there's a cycle in reify-at-least that's causing the 740k calls. AT_POS_SLOW calls reify-at-least 702 times. then reify-at-least (List:40) -> push-at-least (Iterator:49) -> push-exactly (Iterator:28 and Rakudo/Iterator:2028) -> pull-one (List:611) -> reify-at-least (List:40) 09:44
lizmat MasterDuke: I think you'll find them to be called on different objects 09:48
MasterDuke could be, but there doesn't need to be 740k calls in total 09:50
or you mean something is getting cloned that shouldn't?
MasterDuke m: my @a = ("a" .. "ya")[*]; say @a[(^@a).pick]; say @a.elems; 10:49
camelia r
25
MasterDuke m: my @a = ("a" .. "za")[*]; say @a[(^@a).pick]; say @a.elems;
camelia va
677
MasterDuke m: my @a = ("a" .. "yz")[*]; say @a[(^@a).pick]; say @a.elems; 10:50
camelia i
25
MasterDuke m: my @a = ("a" .. "yz")[]; say @a[(^@a).pick]; say @a.elems; 10:51
camelia p
25
MasterDuke m: my @a = ("a" .. "za")[]; say @a[(^@a).pick]; say @a.elems;
camelia tw
677
MasterDuke interesting jump. the range goes from 25 to 677 elems 10:58
m: say "yz".succ 10:59
camelia za
MasterDuke same with a sequence 11:01
m: my @a = ("a" ... "za")[*]; say @a[(^@a).pick]; say @a.elems; say now - INIT now 11:02
camelia mi
677
0.05097049
MasterDuke m: my @a = ("a" .. "za")[*]; say @a[(^@a).pick]; say @a.elems; say now - INIT now
camelia ls
677
5.22678342
lizmat MasterDuke: I meant that some of those calls are on the Reifier, and others on the List 11:29
MasterDuke sure 11:33
in reify-at-least, self is always a List::Reifier.new 12:41
Geth rakudo/negation-of-junctions-not-special: 893c475738 | (Elizabeth Mattijsen)++ | 2 files
Remove special handling of ne and != wrt Junctions

This is a proof of concept. It breaks the following tests:
t/spec/S03-junctions/boolean-context.t (Wstat: 256 Tests: 74 Failed: 1)
   Failed test: 72
t/spec/S03-junctions/autothreading.t (Wstat: 768 Tests: 106 Failed: 3) ... (9 more lines)		 13:42
rakudo: lizmat++ created pull request #3874:
Remove special handling of ne and != wrt Junctions		 13:43
lizmat notable6: weekly 14:32
notable6 lizmat, 1 note: 2020-08-17T17:28:54Z <codesections>: bsdimp.blogspot.com/2020/08/a-35-ye...nd-in.html
lizmat notable6: weekly reset 14:35
notable6 lizmat, Moved existing notes to “weekly_2020-08-24T14:35:54Z”
timotimo ooh i can make it into a weekly again \o/ 14:52
nwc10 this week! 14:53
it's like the good old days
mmm, the worm has turned 14:54
timotimo well, mentioned, not writing it
[Tux] Rakudo version 2020.08.1-5-gb00523061 - MoarVM version 2020.08-4-g0e7417add
csv-ip5xs0.826 - 0.852
csv-ip5xs-208.140 - 8.377
csv-parser27.502 - 27.515
csv-test-xs-200.396 - 0.402
test7.795 - 8.167
test-t1.947 - 1.962
test-t --race0.839 - 0.858
test-t-2032.899 - 33.455
test-t-20 --race9.492 - 9.943
 15:38
Geth rakudo/rakuast: fcec447693 | (Jonathan Worthington)++ | 4 files
Basic RakuAST resolution of multi-part type names

Meaning that things like `say Proc::Async` now compile correctly.		 16:22
rakudo/rakuast: be43baea3b | (Jonathan Worthington)++ | src/Raku/Grammar.nqp
Add Pod grammar rules into RakuAST-based grammar

No actions to go with them for now, but this lets us parse a bunch more spectests, since Pod is used widely throughout them.
rakudo/rakuast: b6af8b2d79 | (Jonathan Worthington)++ | src/Raku/ast/call.rakumod
Add missing visit-children methods on call nodes
rakudo/rakuast: e5c74e067e | (Jonathan Worthington)++ | 3 files
Do some missing Str => RakuAST::Name bits
 17:30
rakudo/rakuast: 2feebc64ce | (Jonathan Worthington)++ | src/Raku/ast/scoping.rakumod
Fix code declaration walking

When we have an immediate block user, we still need to walk the condition part of the AST to find code objects to emit the code for.
lizmat And yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2020/08/24/2020-...rvey-time/ 18:25
samcv lizmat++ 19:18
moritz lizmat++ 19:55
lucs lizmat++ jjmerelo++ 20:17
patrickb Oh! A new weekly! lizmat++ 20:22
Geth rakudo: 038fbc44a1 | (Patrick Böker)++ | tools/build/create-jvm-runner.pl
Improve the JVM relocatability situation

The logic to turn the third party jars into relative paths was broken.
Also utilize the ${NQP_HOME} prefix in the non-relocatable case.		 20:34
rakudo: c990124684 | (Patrick Böker)++ | tools/build/create-jvm-runner.pl
Implement $PERL6_HOME fallback in the JVM runner

The fallback for the $PERL6_HOME env var was missing in the JVM runner script.
rakudo: c1784bda13 | (Patrick Böker)++ (committed using GitHub Web editor) | tools/build/create-jvm-runner.pl
Merge pull request #3862 from patrickbkr/jvm-fixes

Make JVM honor $PERL6_HOME env variable
rakudo/rakuast: 42ce04337d | (Jonathan Worthington)++ | src/Raku/ast/scoping.rakumod
Fix typo; MasterDuke++
 20:39
[Coke] ugexe++ 21:16

Powered by Cro and the Raku® Programming Language.

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

Thank you!