AlexDaniel releasable6: next 00:36
releasable6 AlexDaniel, Next release in 7 days and ≈18 hours. Blockers: github.com/rakudo/rakudo/issues?q=...%9A%A0%22. Changelog for this release was not started yet
AlexDaniel, Details: gist.github.com/77d3cab5f775686253...27d7e04e0f
AlexDaniel oh wow, time flies 00:37
Geth roast/rework-dot-equals-opt: 2e5fb5b7f5 | (Zoffix Znet)++ | S03-operators/inplace.t
Localize class used in a test
 00:48
roast/rework-dot-equals-opt: c932edc5a1 | (Zoffix Znet)++ | S03-operators/inplace.t
Test .= inside andthen and relatives

Closes github.com/rakudo/rakudo/issues/1506 Rakudo fix: github.com/rakudo/rakudo/commit/2daad12f5897835		 01:05
roast/rework-dot-equals-opt: cfa9debfd5 | (Zoffix Znet)++ | S03-operators/inplace.t
More coverage for .= calls
 01:49
rakudo/rework-dot-equals-opt: 01237782fe | (Zoffix Znet)++ | t/packages/Test/Helpers/QAST.pm6
Implement `qast-contains-callmethod` helper routine
 01:53
roast/rework-dot-equals-opt: 07da836c33 | (Zoffix Znet)++ | S03-operators/inplace.t
More coverage for .= calls
 02:00
roast/rework-dot-equals-opt: 804308186a | (Zoffix Znet)++ | S03-operators/inplace.t
Improve test routines used

By using `isa-ok` the tests would pass even if `.=` portions never ran. Test the variable actually have type objects in 'em now.		 02:01
rakudo/rework-dot-equals-opt: abea32429b | (Zoffix Znet)++ | 2 files
Restore .dispatch:<.=> optimization

The original attempt[^1] tried to get rid of `dispatch:<.=>` altogether by generated optimized QAST right away, in the Actions. That proved to be a flawed approach, as `.=` variants
  (dotty methodcall and infixish op), as well as named arg variants
  (normal and fake-infix adverbs), along with various places where
... (8 more lines)		 02:13
rakudo: zoffixznet++ created pull request #1509:
Rework dot equals opt
rakudo/master: 4 commits pushed by (Zoffix Znet)++ 02:14
roast: zoffixznet++ created pull request #388:
Rework dot equals opt
roast/master: 5 commits pushed by (Zoffix Znet)++
rakudo: 2ba7634ca8 | (Zoffix Znet)++ | t/packages/Test/Helpers/QAST.pm6
Fix bugglet in `qast-contains-callmethod` recursion
 02:21
rakudo: 7793f420e4 | (Zoffix Znet)++ | t/08-performance/02-qast-rewrites.t
Test full QAST for .= opt

Some of the .dispatch:<.=>s can hide in QAST::Blocks declarations portion.		 02:22
rakudo: FCO++ created pull request #1510:
Fix Parameter.usage-name, if the parameter name's second char is a * …		 02:27
rakudo: 3c73099ce8 | (Fernando Correa de Oliveira)++ | src/core/Parameter.pm
Fix Parameter.usage-name, if the parameter name's second char is a * or a ! it ignores it
 02:34
rakudo: 07748c4994 | (Zoffix Znet)++ (committed using GitHub Web editor) | src/core/Parameter.pm
Merge pull request #1510 from FCO/fix-usage-name

Fix Parameter.usage-name, if the parameter name's second char is a * …
travis-ci Rakudo build passed. Zoffix Znet 'Merge pull request #1509 from rakudo/rework-dot-equals-opt 02:59
travis-ci.org/rakudo/rakudo/builds/339735333 github.com/rakudo/rakudo/compare/1...0f2f7439e1
Geth rakudo: 2cc7b6317e | (Zoffix Znet)++ | src/core/Any-iterable-methods.pm
Fix Any.tail(Callable) violating Iterator protocol

Fixes RT#131617: rt.perl.org/Ticket/Display.html?id=131617
When we .skip-at-least on our iterator, it's possible we skipped all the values already. When we then use that same iterator to make a new Seq, we violate protocol when that Seq is consumed.
Fix by checking return value of .skip-at-least and returning an empty Seq when we skipped all the values.		 03:30
synopsebot RT#131617 [new]: rt.perl.org/Ticket/Display.html?id=131617 [BUG] Any.tail(Callable) violates Iterator protocol
roast: 3e60598fea | (Zoffix Znet)++ | S32-list/tail.t
Test .tail(Callable) does not violate Iterator protocol

RT#131617: rt.perl.org/Ticket/Display.html?id=131617 Rakudo fix: github.com/rakudo/rakudo/commit/2cc7b6317e
travis-ci Rakudo build passed. Zoffix Znet 'Test full QAST for .= opt 03:32
travis-ci.org/rakudo/rakudo/builds/339736696 github.com/rakudo/rakudo/compare/2...93f420e4b0
Geth rakudo: b2a21fa965 | (Zoffix Znet)++ | src/core/Failure.pm
Fix .perl of handled Failures

Fixes RT#131496: rt.perl.org/Ticket/Display.html?id=131496
Simple put them through the `orelse` call to set the `handled` flag, if we need to set it.		 04:10
synopsebot RT#131496 [new]: rt.perl.org/Ticket/Display.html?id=131496 [BUG] Failure.perl doesn't roundrip `handled` flag
Geth roast: 9625f78f3e | (Zoffix Znet)++ | S04-exceptions/fail.t
Test .perl of handled Failures

RT#131496: rt.perl.org/Ticket/Display.html?id=131496 Rakudo fix: github.com/rakudo/rakudo/commit/b2a21fa965		 04:11
travis-ci Rakudo build passed. Zoffix Znet 'Merge pull request #1510 from FCO/fix-usage-name 04:17
travis-ci.org/rakudo/rakudo/builds/339738662 github.com/rakudo/rakudo/compare/7...748c4994a6
Rakudo build passed. Zoffix Znet 'Fix Any.tail(Callable) violating Iterator protocol 04:49
travis-ci.org/rakudo/rakudo/builds/339747194 github.com/rakudo/rakudo/compare/0...c7b6317e98
lizmat S32-str/utf8-c8.t , test 56 is failing reliably for me (on MacOS 04:59
# at t/spec/S32-str/utf8-c8.t line 195 05:00
# expected: $(6, 171)
# got: $(6, 37, 65, 66)
Geth rakudo: 311ef07fd8 | (Elizabeth Mattijsen)++ | src/core/List.pm
Unify List.roll/pick invocant specification

One could argue that a List:U should be considered a one-element List though.		 05:15
lizmat sleep& 05:18
Geth rakudo: ac87ea2ac8 | (Stefan Seifert)++ | src/core/CompUnit/PrecompilationStore/File.pm
Fix multi-threaded precompilation

The lock file is sufficient to protect the precompilation store from access by multiple processes. However if multiple threads trigger precompilation they will use the same store object which thinks its already sufficiently locked. So multiple writers within the same process will access the store concurrently. We cannot rely on the lock file in ... (9 more lines)		 13:07
lizmat Files=1234, Tests=76203, 305 wallclock secs (14.47 usr 5.34 sys + 2088.72 cusr 209.17 csys = 2317.70 CPU) 13:38
Geth rakudo: f559c6d8bf | (Zoffix Znet)++ | src/Perl6/Actions.nqp
Fix issues in constant type declarations

Partially addresses github.com/rakudo/rakudo/issues/1504
  - Fixes crash with types with `::` in the name
  - Fixes crash with parametarized types like `Array[Numeric]`
  - Changes default type to `Mu` instead of `Any`, to match what we
... (5 more lines)		 13:48
teatime announces (expecting little interest/relevence to y'all, but just in case I'm wrong) his intention to volunteer w/ the Debian Rakudo packaging team.
DrForr teatime++ 13:50
AlexDaniel [Coke]: maybe you can find a use for this: github.com/perl6/doc/issues/1765#i...-364660372 14:57
Zoffix m: my Numeric constant @foo = 1, 2, 3; 16:03
camelia ( no output )
Zoffix Having trouble imagining good behaviour for that ^. One one hand, it could typecheck the assigned value is a Positional[Numeric]. However, we allow `.=` op, so then `my Numeric constant @foo = 1, 2, 3` would need to be List[Numeric], but Lists cannot be parametarized (making an Array[Numeric] in this case is too surpising IMO, 'cause it ain't no constant). 16:04
I think I'll just make this case throw something along the lines of "Parametarization of {$sigil}-sigiled constants is not supported" 16:05
* "so then..." I meant "so then `my Numeric constant @foo .= 1, 2, 3` would need to be" (with the .= op) 16:06
AlexDaniel m: for 1..1000 { $^i %% $_ && put "$_ " for ^$i } 16:21
camelia (signal SEGV)1
1
1
2
1
1
2
3
1
1
2
4
1
3
1
2
5
1
1
2
3
4
6
1
1
2
7
1
3
5
1
2
4
8
1
1
2
3
6
9
1
1
2
4
5
10
1
3 …
Zoffix m: my constant %h = :{(True) => 42}; %h<foo> = 42; dd %h 16:24
camelia :{Bool::True => 42, :foo(42)}
Zoffix hmm
c: 2017.12 my constant %h = %(:70bar, :42foo); %h<foo> = 42; dd %h 16:25
committable6 Zoffix, ¦2017.12: «Hash % = {:bar(70), :foo(42)}␤»
Zoffix I thought we did some sort of decontainerization or something
m: my constant @h = :70bar, :42foo; dd @h 16:26
camelia (:bar(70), :foo(42))
Zoffix m: my constant @h = [:70bar, :42foo]; dd @h
camelia [:bar(70), :foo(42)]
Zoffix Ah, k, I see: twitter.com/zoffix/status/853669556178563072 16:33
I assigned a single pair and thought it ended up an immutable Hash, but it just remained as Pair 16:34
Then, I guess the paramaterization thing does make sense. 16:35
Except that a `constant` with % or @ sigil could be not a constant at all.
Need to write down the spec for this stuff before implementing anything. 16:36
Zoffix &
Geth rakudo: 5bd4e27948 | (Zoffix Znet)++ | docs/archive/constants-type-constraints-proposal-2018-02-10.md
Add constants type constraints proposal
 18:52
rakudo: a66af37f88 | (Zoffix Znet)++ | docs/archive/constants-type-constraints-proposal-2018-02-10.md
Add TOC
 18:53
Zoffix .tell jnthn Wrote down a proposal for how I imagine type constraints on constants working, if you wanted to comment. This is basically clarification to 6.c spec rather than anything new: github.com/rakudo/rakudo/blob/mast...8-02-10.md 18:54
yoleaux Zoffix: I'll pass your message to jnthn.
Zoffix .tell TimToady Wrote down a proposal for how I imagine type constraints on constants working, if you wanted to comment. This is basically clarification to 6.c spec rather than anything new: github.com/rakudo/rakudo/blob/mast...8-02-10.md
yoleaux Zoffix: I'll pass your message to TimToady.
Zoffix oopps copy-pasta fail 18:55
Geth rakudo: c059563742 | (Zoffix Znet)++ | docs/archive/constants-type-constraints-proposal-2018-02-10.md
Fix editing glitch
 18:57
rakudo: 467d8945b1 | (Zoffix Znet)++ | docs/archive/constants-type-constraints-proposal-2018-02-10.md
Clarify reasoning for X::ParametricConstant
 19:21
rakudo: 801b1fa8f2 | (Samantha McVey)++ | 8 files
Fix Unicode prop tests on Windows

Filenames that contain `:` are not allowed on Windows, so use `;` instead.		 19:39
roast: 85dab99cee | (Zoffix Znet)++ | S03-operators/inplace.t
Test .= with constants

Part of github.com/rakudo/rakudo/issues/1504
The fake-infix adverbs don't yet work. Couldn't get the fudger to fudge them so I just commented them out (will work on them very soon anyway)		 19:52
travis-ci Rakudo build passed. Samantha McVey 'Fix Unicode prop tests on Windows 20:53
travis-ci.org/rakudo/rakudo/builds/339929257 github.com/rakudo/rakudo/compare/4...1b1fa8f213
Geth nqp: 8e7201eb64 | (Zoffix Znet)++ | src/vm/moar/QAST/QASTOperationsMAST.nqp
Fix slurpies in `if` ops; jnthn++ for the tip

Fixes RT#105872: rt.perl.org/Ticket/Display.html?id=105872
When deciding whether the block needs the args, we check the arity. However, for slurpies, arity is 0, so the code think we don't need no args passed.
Fix by looking at .count if .arity is zero.		 20:54
synopsebot RT#105872 [open]: rt.perl.org/Ticket/Display.html?id=105872 [BUG] 42, 44, 22 -> *@a { say @a.perl }
rakudo: dfb6d951d7 | (Zoffix Znet)++ | tools/build/NQP_REVISION
[NQP Bump] 8e7201e Fix slurpies in `if` ops; j […]

NQP bump brought: github.com/perl6/nqp/compare/2018....6-g8e7201e		 20:57
rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....6-g8e7201e
ef1d22f4c1 | (Zoffix Znet)++ | src/Perl6/Actions.nqp

We use it to figure out whether we need to pass args to
  `if` ops: github.com/perl6/nqp/commit/8e7201eb64
roast: aa90737df1 | (Zoffix Znet)++ | S04-statements/if.t
Test `if` with slurpies on its block

RT#105872: rt.perl.org/Ticket/Display.html?id=105872
Rakudo fix: github.com/perl6/nqp/commit/8e7201eb64
   github.com/rakudo/rakudo/commit/dfb6d951d7
   github.com/rakudo/rakudo/commit/ef1d22f4c1		 21:03
dogbert17 m: say "ab" ~~ /^ a ?% (b) $/ 21:27
camelia ===SORRY!===
QAST::Block with cuid 1 has not appeared

Powered by Cro and the Raku® Programming Language.

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

Thank you!