Geth roast: 5a3f3c68c5 | (Zoffix Znet)++ | S16-filehandles/filetest.t
Reword .IO.z test for directories

To have it pass on systems that return zero for directory sizes.
Part of addressing #320		 00:12
synopsebot SPEC#320 [open]: github.com/perl6/roast/issues/320 Windows roast failures just before 2017.09 release
Zoffix Some test is taking a hella long time on Windows... t\spec\S32-io\IO-Socket-Async.t, at a guess... 00:14
Ah, no that's the last passed one 00:16
travis-ci Rakudo build passed. Zoffix Znet 'Test overflow in .roll() 02:12
travis-ci.org/rakudo/rakudo/builds/314934244 github.com/rakudo/rakudo/compare/a...0be0f631cf
Geth roast: 37ff3ab58d | (Zoffix Znet)++ | S12-enums/basic.t
Test enums can be created via Seq of Pairs

RT#128017: rt.perl.org/Public/Bug/Display.html?id=128017		 02:24
synopsebot RT#128017 [new]: rt.perl.org/Ticket/Display.html?id=128017 enum treats a Seq of Pairs as a List of Str
Geth roast: 6e6eae6059 | (Zoffix Znet)++ | S12-enums/basic.t
Test can provide enum values via Pairs

RT#130041: rt.perl.org/Ticket/Display.html?id=130041		 02:30
synopsebot RT#130041 [open]: rt.perl.org/Ticket/Display.html?id=130041 [BUG] Pair in enum declaration should either DWIM or parsefail
Geth roast: 2d35dd4949 | (Zoffix Znet)++ | S02-types/whatever.t
No issues with //= and WhateverCode

RT#126415: rt.perl.org/Ticket/Display.html?id=126415		 02:38
synopsebot RT#126415 [open]: rt.perl.org/Ticket/Display.html?id=126415 [BUG] WhateverCode loses its mind inside //=
Geth roast: 0fb5a0116b | (Zoffix Znet)++ | S02-names/strict.t
`no strict` does not cause autovivification container issues

RT#125398: rt.perl.org/Ticket/Display.html?id=125398		 03:00
synopsebot RT#125398 [open]: rt.perl.org/Ticket/Display.html?id=125398 with no strict temp does not autcreate hashes and hash elements 03:01
Geth roast: 5cd8dff229 | (Zoffix Znet)++ | S03-binding/arrays.t
Binding Failure to Array throws useful error

RT#118397: rt.perl.org/Ticket/Display.html?id=118397		 03:06
synopsebot RT#118397 [open]: rt.perl.org/Ticket/Display.html?id=118397 [BUG] array binding causes failures to be thrown
Zoffix Segfault in t/spec/S07-hyperrace/hyper.t (after test 28, looks like) also consistent failure in t/spec/MISC/bug-coverage-stress.t (test 5) and t/spec/S12-construction/destruction.t looks to flop 4-5 tests as TODOs 04:19
travis-ci Rakudo build passed. Zoffix Znet 'Test overflow behaviour in .indent()
travis-ci.org/rakudo/rakudo/builds/314937049 github.com/rakudo/rakudo/compare/1...2c27f343a2
Geth rakudo: c0f99a393b | (Zoffix Znet)++ | src/Perl6/Grammar.nqp
Allow parse of Callable's signatures constraints without `:`

This now: `sub foo(&block (Dog --> Bool)) {...}` Means same as: `sub foo(&block:(Dog --> Bool)) {...}` Old behaviour was to complain about calling .Capture on a Callable
  - Per TimToady's nod in rt.perl.org/Ticket/Display.html?id...xn-1321791
  - Unifies syntax with Positional unpacking that allows both `:` and whitespace
   to indicate unpacking.		 04:22
roast: a31e4655e8 | (Zoffix Znet)++ | S06-signature/closure-parameters.t
Test unpacking Callable signatures

RT#123116: rt.perl.org/Ticket/Display.html?id=123116		 04:29
synopsebot RT#123116 [open]: rt.perl.org/Ticket/Display.html?id=123116 [BUG] Weird error for code parameters with subsignatures in Rakudo
Geth roast: 05cecddc21 | (Zoffix Znet)++ | S06-signature/closure-parameters.t
Test whitespace can be used to unpack Callable sig

Rakudo impl: github.com/rakudo/rakudo/commit/c0f99a393b		 04:36
rakudo: 6061f0bcc5 | (Zoffix Znet)++ | src/core/Sequence.pm
Implement Sequence.Numeric

  - Gives proper .Numeric to RaceSeq and HyperSeq
  - Seq.Numeric overrides this with a variant that can handle .count-only
   optimization		 04:42
roast: d86c305731 | (Zoffix Znet)++ | 2 files
Test Hyper/Race Seqs have .Numeric implemented

Rakudo impl: github.com/rakudo/rakudo/commit/6061f0bcc5		 04:43
rakudo: 2f955d55e3 | (Zoffix Znet)++ | src/Perl6/Grammar.nqp
Revert "Allow parse of Callable's signatures constraints without `:`"

This reverts commit c0f99a393b9f42ce9ff996ce82e0099ccba668f7.
Just noticed the original syntax with `:` is actually a Signature literal appended to the parameter, so it makes a lot of sense, while adding `<.ws> ()` unpacking version makes less sense and there's not much reason to have more than one syntaxes for this feature.		 04:45
roast: db549b09e5 | (Zoffix Znet)++ | S06-signature/closure-parameters.t
Revert tests for <.ws> Callable sig unpacking

Test were added today and not part of 6.c-errata.
Feature got canceled: github.com/rakudo/rakudo/commit/2f955d55e3		 04:50
Zoffix (writing docs)++ # makes you realize your idea is crap
ZofBot: SLEEP! 04:51
ZofBot Zoffix, oO( yeah, should've applied that bashrc thingyā€¦ ) hehe hmmm ok now I see an issue greppable6: rakudobrew AlexDaniel, gist
Geth rakudo: 9fd5042b84 | (Zoffix Znet)++ | src/core/Exception.pm
Make X::Cannot::Capture PDGier

Mention user may have meant to use `&callable:(Sig)` form		 05:00
roast: cd4dba6faf | (Zoffix Znet)++ | S32-list/seq.t
Test Seq.Numeric makes use of .count-only
 05:02
roast: ae7f3bf610 | (Zoffix Znet)++ | S32-list/seq.t
Fix infiniloop in test when in failure path
 05:04
travis-ci Rakudo build passed. Zoffix Znet 'Rename Test::Helper's is-run `$status` to `$exitcode` 05:34
travis-ci.org/rakudo/rakudo/builds/314995168 github.com/rakudo/rakudo/compare/0...7c771cd765
Rakudo build passed. Zoffix Znet 'Test Failure.new(Exception.new) does not segfault 06:43
travis-ci.org/rakudo/rakudo/builds/314995477 github.com/rakudo/rakudo/compare/e...01fea1832c
Rakudo build passed. Zoffix Znet 'Test List.ASSIGN-POS gives PDG error 07:36
travis-ci.org/rakudo/rakudo/builds/315001295 github.com/rakudo/rakudo/compare/1...d56be258b5
[Tux] Rakudo version 2017.11-96-g9fd5042b8 - MoarVM version 2017.11-28-gba1fb9542
csv-ip5xs1.091 - 1.096
csv-ip5xs-2013.082 - 13.109
csv-parser12.405 - 12.459
csv-test-xs-200.448 - 0.457
test11.688 - 11.826
test-t2.894 - 2.905
test-t --race1.222 - 1.236
test-t-2051.647 - 52.968
test-t-20 --race18.118 - 18.151
 08:10
lizmat Files=1231, Tests=75913, 315 wallclock secs (14.62 usr 5.23 sys + 2177.16 cusr 213.00 csys = 2410.01 CPU) 08:53
Geth nqp: 499d3f0560 | (Elizabeth Mattijsen)++ | tools/build/MOAR_REVISION
Bump Moar for the latest fixes
 09:22
Ā¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...3-g93a77dc
rakudo: 83eac0e423 | (Elizabeth Mattijsen)++ | tools/build/NQP_REVISION
Bump NQP for the latest goodies
 09:38
Ā¦ rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2017....9-g499d3f0
lizmat m: my @matrix[2;2] = [1,2],[3,4]; @matrix[<0>;0] for ^1000000
camelia ( no output )
lizmat m: my @matrix[2;2] = [1,2],[3,4]; @matrix[<0>;0] for ^1000000
camelia ( no output )
lizmat m: my @matrix[2;2] = [1,2],[3,4]; @matrix[<0>;0] for ^1000000
camelia ( no output )
lizmat m: my @matrix[2;2] = [1,2],[3,4]; @matrix[<0>;0] for ^1000000
camelia ( no output )
Geth roast: c9d4633758 | pmurias++ | S05-metasyntax/unicode-property-pair.t
Add a test for <:!NumericValue>
 10:47
nqp: 87f0e9f176 | pmurias++ | src/vm/js/nqp-runtime/unicode-props.js
[js] Fix wrong identifier
 10:48
nqp: 9e418e899e | pmurias++ | src/vm/js/nqp-runtime/unicode-props.js
[js] Refactor unicode properties that take an arg
nqp: 7460b33e29 | pmurias++ | src/vm/js/nqp-runtime/unicode-props.js
[js] Implement :Name unicode property
tbrowder AlexDaniel: request review of PR #1300 which should help with work on table and other pod formatting code 11:12
AlexDaniel R#1300
synopsebot R#1300 [open]: github.com/rakudo/rakudo/pull/1300 Untwine Pod
AlexDaniel tbrowder: it says ā€œThe changes to roast (see below) are still required.ā€, what does that mean exactly? 11:13
tbrowder hm, not sure, will check... 11:14
AlexDaniel tbrowder: I'm just skimming over the text, but it says ā€œThis commit increases the amount of memory used during compilation from 700Mb => 1200Mbā€ 11:15
not sure what *exactly* this implies, but that's probably unacceptable
(given that we have a bunch of people trying to use perlĀ 6 on rpi and other small linux boards) 11:16
tbrowder yes, but that was with a very old rakudo. can you add thouā€™s PR as a branch so others can play with it? 11:17
AlexDaniel ah, it's for the synthetic test 11:21
tbrowder also, the changes to roast referenced in PR have been merged already.
ā€œothersā€ => ā€œmoiā€ 11:23
AlexDaniel yeah I've left a comment 11:24
.tell lizmat github.com/rakudo/rakudo/wiki/Ticket-updates 11:50
yoleaux AlexDaniel: I'll pass your message to lizmat.
AlexDaniel (please feel free to beautify the page any way you like)
lizmat . 11:54
yoleaux 11:50Z <AlexDaniel> lizmat: github.com/rakudo/rakudo/wiki/Ticket-updates
lizmat looks ok to me so far :-)
AlexDaniel reportable6: 2017-11-19T14:04:30Z 2017-11-26T22:21:12Z 11:56
reportable6 AlexDaniel, OK, working on it! This may take up to 40 seconds
AlexDaniel one link is wrong :)
reportable6 AlexDaniel, gist.github.com/a1f860270afb3df84c...6daf395082 11:57
AlexDaniel reportable6: 2017-11-05T12:53:21Z 2017-11-13T22:37:40Z 12:02
reportable6 AlexDaniel, OK, working on it! This may take up to 40 seconds
AlexDaniel ā€¦ and one has a SPAM ticket in itā€¦
reportable6 AlexDaniel, gist.github.com/44d5028345ec26cd43...fb1d297670 12:03
tbrowder AlexDaniel: when working the pod config quote issue, i think the only legal quoting and other entries ought to be compatible with the table in synopsis S26 (config, delimited blocks). that makes the solution much easier since only a small set of config value entry forms are defined there. then we can warn or fail if anything else is input. 12:15
travis-ci Rakudo build failed. Zoffix Znet 'Allow parse of Callable's signatures constraints without `:` 12:17
travis-ci.org/rakudo/rakudo/builds/315170716 github.com/rakudo/rakudo/compare/4...f99a393b9f
buggable [travis build above] ā˜  All failures are due to: missing build log (3 failures), failed make test (3 failures). Across all jobs, 2 unique test files failed.
AlexDaniel tbrowder: so what exactly do you want to forbid? 12:19
tbrowder things like :caption(Q|blah|) 12:20
and other quote types. itā€™s clear that pod doesnā€™t allow unicode except inside a formatting code block so the double-chevron quotes shouldnā€™t be allowed either. if the user doesnā€™t use one of the prescribed formats in the table then either warn or ignore is my vote (not fail unless the parser fails). 12:25
AlexDaniel m: my $x is default(Q|abc|); say $x 12:26
camelia abc
jnthn . 12:27
yoleaux 11 Dec 2017 23:56Z <Zoffix> jnthn: I moved those inside the locks. If that's incorrect let me know: github.com/rakudo/rakudo/commit/8df8ce09ce
11 Dec 2017 23:56Z <Zoffix> jnthn: that was RE this (that bot didn't pass): `jnthn how come `$!signals-by-Str-setup` and `$!signals-setup` are set to `True` *outside* the lock? Isn't there a race that as one thread just enters the lock another one just exited it but hasn't yet set those to True, so the other thread would check it and find it false and duplicate the work?
11 Dec 2017 23:57Z <Zoffix> jnthn: github.com/rakudo/rakudo/commit/1f...bebea8374d
jnthn .tell Zoffix seems reasonable
yoleaux jnthn: I'll pass your message to Zoffix.
tbrowder i added the the table of config formats to the docs
AlexDaniel tbrowder: I see. Well, maybe, I'm not entirely sure 12:29
ilmari would it make sense to have a thread-safe version of 'once'? 12:30
lizmat that i
*is* the long term goal, afaik
and/or making state variables thread-safe 12:31
because that's really the root of the problem
tbrowder the config quote issue is that the original enclosing quote pair is not getting stripped. that should probably be taken care of in the grammar, but it can be done in nqp land.
ilmari lizmat: that sound a lot harder than wrapping up 'unless $foo-done { $foo-lock.protect { unless $foo-done { ...; $foo-done = True } } }' in some nicer syntax 12:33
lizmat but how to prevent the race condition on creating the lock? or would you do that at init time ?
tbrowder i can extend the acceptable quotes if you want, but i would like to see a list. note that allowing Q|| also means allowing Q[] and other bracketing forms.... 12:34
AlexDaniel tbrowder: playing a bit more with it right now, it seems like it thinks that everything is a stringā€¦
tbrowder also q, qq, qqw, ... 12:35
lizmat ilmari: also, you don't really need a protect block there: just a flag that it's running/has run
tbrowder yee
yes, all is stringified, but only stuff in <> shows without the quotes, all else shows the quotes as well as the contents. 12:37
so the set of chars to be stripped can be large if all are allowed. but that can be handled. 12:38
bottom line imho is: the first cut solution may not find all things a user might try, so the problem remains open-ended unless the set allowed is clearly defined. 12:41
ilmari lizmat: I guess an unless cas($done, False, True) { ... } would be enough? 12:42
AlexDaniel just tried :foo[5, 8] and is a bit disappointed
tbrowder i donā€™t think that is unreasonable, and it appears Damian thought so, too.
AlexDaniel tbrowder: well, try it, we'll see how it goes 12:43
tbrowder: but generally I dislike the idea of stripping the quotes afterwards
lizmat ilmari: to wrap the setting of the state variable? that could work, I guess ?
AlexDaniel it should know the insides without stripping anythingā€¦
ilmari lizmat: instead of the unless/protect/unless dance
jnthn Trying to make state variable "threadsafe" won't go anywhere good
lizmat :-( 12:44
jnthn The clue's in the name
"state"
What happens if we mutate state from two threads at once?
Boom.
timotimo hm. should we implement once in terms of atomic ops perhaps somehow? 12:45
jnthn Well, once is menat to be "once per closure clone", not "once ever"
I think what a lot of those problems want *is* once ever
timotimo hm, true
jnthn It's kinda like a Promise that triggers its work on the first time you await it 12:46
AlexDaniel ( related trap: docs.perl6.org/language/traps#Usin...once_block )
jnthn Promise::Deferred or some such
Or just Deferred or something 12:47
tbrowder i do too, but thatā€™s all i can see in nqp landā€”grammar is another ball game which can maybe be handled by someone interested. i canā€™t do that now, but can start in that direction. i can do the nqp thing and note that it is an incremental solution. actually, World.nqp may have the answer, too. my goal is to cure the RT asap, but not neccesarily have it be a final solution. where do you want my effort put? 12:50
AlexDaniel tbrowder: well, you decide. But to me it feels like all pod-related code had enough incremental solutions, and it's time to do stuff properly 12:51
tbrowder well, iā€™m a bit more pragmatic than you i guess :) 12:52
AlexDaniel tbrowder: we will have to take it anyway as long as it makes things better :) 12:53
tbrowder ok, holiday stuff is taking lots of time, but iā€™ll look into doing it right higher up the chain if i can. 12:55
AlexDaniel tbrowder++
DrForr Speaking of pod-related code... 12:57
github.com/drforr/perl6-Pod-To-BlogspotHTML - The walker should be rewritten to an iterator (probably), the test suite is being rewritten using roast, and as it stands it's probably not properly subclassable, but those are all issues I'll work out before Xmas break. 13:00
I'll also likely include an internal Pod::To::HTMLFragment that does the heavy lifting, so BlogspotHTML can override just a few methods and get something that looks nicer probably with style sheets. 13:01
Geth roast: a894326bfe | (Zoffix Znet)++ | S12-construction/named-params-in-BUILD.t
Can bind to native type attributes in build methods

RT#127845: rt.perl.org/Ticket/Display.html?id=127845 Rakudo fix: github.com/rakudo/rakudo/commit/77...2828eb36da		 13:18
synopsebot RT#127845 [open]: rt.perl.org/Ticket/Display.html?id=127845 [BUG] Making a parameter of a BUILD submethod a uint8 triggers a bytecode validation error in Rakudo
Geth roast: 77a081fb22 | (Zoffix Znet)++ | S02-names/is_default.t
Reformat tests for better readability
 13:25
roast: 785446408e | (Zoffix Znet)++ | S02-names/is_default.t
Test is default() respects type constraint

RT#126107: rt.perl.org/Ticket/Display.html?id=126107		 13:31
synopsebot RT#126107 [open]: rt.perl.org/Ticket/Display.html?id=126107 [BUG] "of" type constraint inconsistently applied
lizmat commute& 13:33
Geth roast: 0d79eda3dc | (Zoffix Znet)++ | S02-names/is_default.t
Test default `is default()` gets adjusted to type constraint

RT#126107: rt.perl.org/Ticket/Display.html?id=126107		 13:35
dogbert2 m: my Int @a[2; 2] is default(0) 13:38
camelia 5===SORRY!5=== Error while compiling <tmp>
is default on shaped Array[Int] not yet implemented. Sorry.
at <tmp>:1
------> 3my Int @a[2; 2] is default(0)7ā5<EOL>
expecting any of:
constraint
Geth roast: 0431615286 | (Zoffix Znet)++ | S02-types/baggy.t
Test creating setty/baggy types with lazy iterables throws

RT#125580: rt.perl.org/Ticket/Display.html?id=125580		 13:43
synopsebot RT#125580 [open]: rt.perl.org/Ticket/Display.html?id=125580 [BUG] bag/set related issues 13:44
dogbert2 Zoffix++, you're on a roll
AlexDaniel Zoffix++ indeed 13:47
it's like if there's a Zoffathon that nobody told us about :)
Geth rakudo: ced4af662f | (Zoffix Znet)++ | t/05-messages/02-errors.t
Test use of wrong sigil suggests right var

RT#125277: rt.perl.org/Ticket/Display.html?id=125277		 13:52
synopsebot RT#125277 [open]: rt.perl.org/Ticket/Display.html?id=125277 [LTA] error message when using the wrong sigil with an @array or %hash
Geth rakudo: 21555412b1 | (Zoffix Znet)++ | t/05-messages/02-errors.t
Remove extraneous comma in test description
 13:55
dogbert2 zofbot hello 14:11
zofbot: hello
ZofBot: are you alive? 14:13
ZofBot dogbert2, If I chewed master's shoe, for instance, he used to kick me; but if I chewed Fred's, Fred would tickle me under the ear
AlexDaniel good bot 14:18
ZofBot: say something racist
ZofBot AlexDaniel, So, maybe we need to add a correcsponding special case somewhere
Geth roast: 70f09dd721 | (Zoffix Znet)++ | S02-names/pseudo.t
Small formatting adjustment
 14:20
rakudo: a82e0e7da5 | (Zoffix Znet)++ | src/core/PseudoStash.pm
Fix gut spillage in pseudopackages

When going too high up with CALLER::, CALLERS::, or SETTING::
Fixes RT #119521: rt.perl.org/Ticket/Display.html?id=119521		 14:24
synopsebot RT#119521 [open]: rt.perl.org/Ticket/Display.html?id=119521 [BUG] ā€œctxlexpad needs an MVMContextā€ when reaching too far out to get to the truth in Rakudo
Geth roast: 7f55679952 | (Zoffix Znet)++ | S02-names/pseudo.t
no guts spillage when going too high up scope in pseudopackages

RT #119521: rt.perl.org/Ticket/Display.html?id=119521		 14:25
rakudo: d4ad6fa953 | (Zoffix Znet)++ | t/05-messages/02-errors.t
Test RO assignment indicates value of the thing

RT#131362: rt.perl.org/Ticket/Display.html?id=131362		 15:37
synopsebot RT#131362 [new]: rt.perl.org/Ticket/Display.html?id=131362 [LTA] Can ā€œCannot modify an immutable Xā€ error message also include the value? (42 = $x)
Geth rakudo: 653017561f | (Zoffix Znet)++ | t/05-messages/02-errors.t
Test declaring enum with uninitialized hash warns

RT#130446: rt.perl.org/Ticket/Display.html?id=130446		 15:48
synopsebot RT#130446 [new]: rt.perl.org/Ticket/Display.html?id=130446 [REGRESSION] [LTA] Creating an enum from a Hash does not work but no longer warns (enum Bits (%thing))
Geth rakudo: fc52143bee | (Zoffix Znet)++ | 2 files
Fix crash when using allomorphs as enum values

Mixing both NumericEnumeration and StringyEnumeration causes a conflict with Str method resolution.
Fix by adding NumericStringyEnumeration role and using it for allomorphs.		 16:21
roast: a2cec00d1c | (Zoffix Znet)++ | S12-enums/basic.t
Test can create enum with a Hash

RT#130446: rt.perl.org/Ticket/Display.html?id=130446		 16:22
synopsebot RT#130446 [new]: rt.perl.org/Ticket/Display.html?id=130446 [REGRESSION] [LTA] Creating an enum from a Hash does not work but no longer warns (enum Bits (%thing))
Geth roast: ba6740c170 | (Zoffix Znet)++ | S12-enums/basic.t
Test "NumericStringyEnumeration" uses key as .Str value

Rakudo fix: github.com/rakudo/rakudo/commit/fc52143bee		 16:24
rakudo: 1c7d15d73d | (Zoffix Znet)++ | src/core/PseudoStash.pm
Fix gut spillage in pseudo packages (2)

Fixes cases missed by first pass in
  github.com/rakudo/rakudo/commit/a82e0e7da5		 16:58
roast: 9fbde14b7a | (Zoffix Znet)++ | S02-names/pseudo.t
Test more gut-throwing pseudopackage combinations

  - Also, indicate what mixed combination failed when test fails
  - Rakudo fix: github.com/rakudo/rakudo/commit/1c7d15d73d		 16:59
dogbert17 m: my Set $s = <spam eggs bacon>.Set; say $s.elems; say "There are $s.elems in the set" # is this intended, i.e. that all elements are written out ? 17:31
camelia 3
There are eggs bacon spam.elems in the set
lizmat dogbert17: that feels wrong 17:32
geekosaur dogbert17, if you say $s,.elems() instead it will dtrt
er $s.elems()
otherwise it assumes the .elems is meant literally
dogbert17 geekosaur++
lizmat ah, yes :-) 17:33
geekosaur interpolation of expressions without explicit {} is always going to be a tricky heuristic thing "every DWIM has a WAT"
timotimo aye, we only interpolate method call chains and non-$-sigiled things if there's some kind of postcircumfix at the end
Geth roast: 69d9d04601 | (Zoffix Znet)++ | S04-declarations/my.t
Test binding/assignment to grouped sigilless vars

RT#124210: rt.perl.org/Ticket/Display.html?id=124210
synopsebot RT#124210 [open]: rt.perl.org/Ticket/Display.html?id=124210 [BUG] Assignment to several declared SSA-style \x variables fails in Rakudo
timotimo that's also the rule that has you write @foo[] or similar to get an array interpolated
dogbert17 the array one I knew, guess I should have figured out the () case as well 17:34
timotimo m: sub postcircumfix:<^ ^>(|) { }; my @test; say "this is test: @test^^" 17:35
camelia this is test: @test^^
timotimo hm, i wonder why that would be
geekosaur think it's related to the way/possibly time that the perl6 parser gets braided with the string parser? 17:36
there's at least one bug on file about oddities in how braiding works iirc 17:37
timotimo m: sub bleh(&infix:<+>) { say 1 + 2 }; test(* * *) 17:38
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
test used at line 1
geekosaur alternately the ^^ is just being misparsed because it conflicts with some other rule, but I can't think of one it'd conflict with offhand (of course that proves ~nothing)
timotimo m: sub test(&infix:<+>) { say 9 + 9 }; test(* * *)
camelia 81
timotimo ^- this is fun
Geth roast: softmoth++ created pull request #367:
Formatting Codes don't produce spurious ""		 18:14
synopsebot SPEC#367 [open]: github.com/perl6/roast/pull/367 Formatting Codes don't produce spurious ""
Geth roast: 84e2f22389 | (Brad Gilbert)++ (committed using GitHub Web editor) | S02-types/native.t
fix assigning typo from 0d871001ff
 18:53
lizmat so I was looking at stackoverflow.com/questions/477804...trld-input 20:16
and thought simplistically that this should work: while with prompt("> ") { $^a.say }
m: while with prompt("> ") { $^a.say } # alas, it doesn't
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
with used at line 1
lizmat so I added a "with" sub: sub with(\a) { defined(a) ?? a !! Nil } 20:17
m: sub with(\a) { defined(a) ?? a !! Nil }; while with prompt("> ") { $^a.say }
camelia > Ā»Wann treffen wir drei wieder zusamm?Ā«
> Ā»Um die siebente Stundā€˜, am BrĆ¼ckendamm.Ā«
> Ā»Am Mittelpfeiler.Ā«
> Ā»Ich lƶsche die Flamm.Ā«
> Ā»Ich mitĀ«
>
lizmat which does work
so I was wondering: why don't we add this "with" sub to the core ? 20:18
Zoffix -1 on the name: causing confusion with the statement modifier + posing the question of why there isn't a &without. -1 on adding a sub, since you can just write while prompt('> ') // Nil { } to get the same effect, no? 20:56
yoleaux 12:27Z <jnthn> Zoffix: seems reasonable
Zoffix :)
Doesn't prompt return Nil in the first place? 21:03
geekosaur I would hope it returns the Str it read
lizmat Zoffix: it shouldn't return on an empty string 21:05
Zoffix Ah right. Then // Nil is what I'd use
lizmat alas, "" is also false and exits the while 21:06
Zoffix On the topic of adding things: there's an old ticket to add TR/// that I plan to do soon, if anyone wanted to object 21:07
lizmat not me :-)
Zoffix loop { say prompt('> ') // last } 21:11
m: @ = $*IN.lines; dd prompt 21:13
camelia Nil
Zoffix geekosaur: ^ I meant when it reached EOF 21:14
Zoffix goes back to lurking 21:15
lizmat decommute& 21:17
Geth roast: ba86027f3e | usev6++ | 5 files
[jvm] Unskip now passing tests

Fixed with nqp commit 7eaebf5abd		 22:03
travis-ci Rakudo build passed. Zoffix Znet 'Test use of wrong sigil suggests right var 22:07
travis-ci.org/rakudo/rakudo/builds/315368886 github.com/rakudo/rakudo/compare/8...d4af662fa9
Geth roast: ronaldxs++ created pull request #368:
`no strict` autovivification 2 level hash		 22:46
synopsebot SPEC#368 [open]: github.com/perl6/roast/pull/368 `no strict` autovivification 2 level hash
Geth roast: aa5691d102 | (Ronald Schmidt)++ | S02-names/strict.t
`no strict` autovivification 2 level hash

RT#125183: rt.perl.org/Ticket/Display.html?id=125183		 22:56
roast: 9abb9ae6e7 | (Zoffix Znet)++ (committed using GitHub Web editor) | S02-names/strict.t
Merge pull request #368 from ronaldxs/no-strict-autovivify-2-level-hash

  `no strict` autovivification 2 level hash
synopsebot RT#125183 [new]: rt.perl.org/Ticket/Display.html?id=125183 [BUG] Crash on no strict and multi-dimensional hash
synopsebot SPEC#368 [closed]: github.com/perl6/roast/pull/368 `no strict` autovivification 2 level hash
rakudo: softmoth++ created pull request #1307:
Drop extraneous spaces in code and table blocks		 23:41
synopsebot RAKUDO#1307 [open]: github.com/rakudo/rakudo/pull/1307 Drop extraneous spaces in code and table blocks

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!