gfldex m: say (-∞..∞).pick(5) 00:05
camelia ()
gfldex I would prefere that to throw an out of range exception
Zoffix agrees with reasoning in github.com/rakudo/rakudo/commit/5c...bf61982715 00:08
Geth rakudo: c91bcc2afc | (Zoffix Znet)++ | 2 files
Fix &infix:<eqv> with NaN-y Rationals

Phixes github.com/rakudo/rakudo/issues/1353
Check both numerator and denominators of both a and b are zero and return True. Otherwise, pass through `==` op.
We can't just `==` check numerators even if both denominators
  are zero, since ±Inf Rationals could have those differ, yet would
still be True `eqv`		 01:17
roast: 9b69d9787c | (Zoffix Znet)++ | S32-num/rat.t
Spec `eqv` of zero-denominator Rationals

Closes github.com/rakudo/rakudo/issues/1353 Rakudo fix: github.com/rakudo/rakudo/commit/c91bcc2afc
Zoffix Wondering if `eqv` is too loose with zero-denominator Rationals. For Nums, we care even about sign of zero, and I dug out comment: <TimToady> I would agree; eqv is not math, for which we have ==, but basically is about "would these serialize to the same thing" irclog.perlgeek.de/perl6-dev/2017-...i_13959621 01:30
But for Rationals, we `eqv` two zero-denominator Rationals if their numerators are both zero or are both of the same sign
m: say <42/0> eqv <1/0> 01:31
camelia True
Zoffix So should that be False, since it won't serialize the same?
For True: yes, they're equivalent anywhere math is involved
For False: no, they're not equivalent anywhere .numerator/.denominator are being worked on 01:32
m: (<43/0> + <42/0>).nude.say 01:35
camelia (85 0)
Zoffix Gonna leave it as `True`. If making it `False`, ^ the above would cause some uncertainty, since zero-denominator rats don't reduce 01:36
m: (<-2/0> + <4/0>).nude.say
camelia (2 0)
Zoffix crap 01:37
Maybe try always normalizing all zero-denominator Rats into <1/0>, <-1/0> and <0/0> 01:38
Zoffix puts it on the back burner, until .REDUCE-ME is removed 01:41
Geth roast: 86bf39fcdc | (Zoffix Znet)++ | S32-num/rat.t
Unspec zero-denominator Rationals leaving numerator untouched

This test was added earlier today and is not part of 6.c spec.
There are some math issues and we might want to solve them by normalizing newly-created zero-denomiator Rationals into <-1 0 1> numerators: R#1354 github.com/rakudo/rakudo/issues/1354		 01:55
synopsebot R#1354 [open]: github.com/rakudo/rakudo/issues/1354 Some operations with zero-denominator Rationals are broken
rakudo: e543c890ad | (Zoffix Znet)++ | 2 files
Improve USAGE message for subset/where params

Fixes github.com/rakudo/rakudo/issues/1346
  - Display `where` constraints as `where { ... }` instead of
   `-> ;; $_? is raw { #`(Block|83868104) ... }`. Looks like we
   did this awhile back too, but the detection of custom wheres got
   broken when .gist of blocks got changed
  - Display subset names instead of their nominal types in named params		 03:56
roast: b3d85987a7 | (Zoffix Znet)++ | S26-documentation/block-trailing.t
[v6.d REVIEW] slightly simplify code
 04:01
roast: 983aea724e | (Zoffix Znet)++ | S26-documentation/why-trailing.t
[v6.d REVIEW] slightly simplify code
 04:04
roast: 88b27431ec | (Zoffix Znet)++ | S32-exceptions/misc.t
[v6.d REVIEW] Formatting / more precise exceptions

  - Shorten long lines
  - Compress overly-vertical examples
  - Use actual exceptions instead of `Exception` in several
   throws-like tests		 04:13
roast: 1d71a97ac8 | (Zoffix Znet)++ | S32-list/seq.t
[v6.d REVIEW] Improve Seq.from-list tests

  - Use more appropriate test routines
  - Use more elaborate &body to test it generates proper things
Orig: github.com/perl6/roast/commit/1461...d7c1eb9961		 04:35
travis-ci Rakudo build failed. Zoffix Znet 'Improve USAGE message for subset/where params 04:45
travis-ci.org/rakudo/rakudo/builds/323485321 github.com/rakudo/rakudo/compare/c...43c890ad58 04:46
buggable [travis build above] ☠ All failures are due to: failed make test (6 failures). Across all jobs, only t/05-messages/01-errors.t test file failed.
Geth rakudo: 1d772dc002 | (Zoffix Znet)++ | src/core/List.pm
Fix &list(Seq:D) to produce a List

Per github.com/rakudo/rakudo/issues/1344 , a Seq is not a "list", because it does not do a Positional role.		 04:50
roast: 6636645ac9 | (Zoffix Znet)++ | S03-operators/context-forcers.t
Change incorrect &list(Seq) tests

Per github.com/rakudo/rakudo/issues/1344 , a Seq is not a "list", because it does not do a Positional role.
Looks like the original tests[^2] did start off testing for List and
  because they were added alongside `cache` listop, they tested for
... (6 more lines)		 04:56
roast/6.c-errata: 4a3719602f | (Zoffix Znet)++ | S03-operators/context-forcers.t
Change incorrect &list(Seq) tests

Per github.com/rakudo/rakudo/issues/1344 , a Seq is not a "list", because it does not do a Positional role.
Looks like the original tests[^2] did start off testing for List and
  because they were added alongside `cache` listop, they tested for
... (6 more lines)		 04:57
Zoffix .ask MasterDuke how come github.com/rakudo/rakudo/pull/1064/files converts `'` in literals to `'"'"'` Makes it kinda hard to read in usage messages. I suspect this was to automatically add shell quoting, but that quoting doesn't apply to Windows 05:26
yoleaux Zoffix: I'll pass your message to MasterDuke.
travis-ci Rakudo build failed. Zoffix Znet 'Fix &list(Seq:D) to produce a List 05:40
travis-ci.org/rakudo/rakudo/builds/323490576 github.com/rakudo/rakudo/compare/e...772dc002e2
buggable [travis build above] ☠ All failures are due to: failed make test (6 failures). Across all jobs, only t/05-messages/01-errors.t test file failed. 05:41
Geth roast: cd327665b8 | (Zoffix Znet)++ | S32-exceptions/misc.t
Add missing semicolon
 06:14
rakudo: 27fbd7ab77 | (Zoffix Znet)++ | 2 files
More MAIN USAGE fixes

  - Fix string literal breakage caused by original fix[^1] to #1346
  - Make quote post-processing only apply when we're dealing with
   a string literal, not with variable names (another reason why
   I think we shouldn't do it:
   irclog.perlgeek.de/perl6-dev/2017-...i_15636320 )
  [1] github.com/rakudo/rakudo/commit/e5...c707f73403		 06:17
synopsebot RAKUDO#1346 [closed]: github.com/rakudo/rakudo/issues/1346 type info for named MAIN args ignores subset names
roast: 44602bd7e4 | usev6++ | S32-io/out-buffering.t
[jvm] Skip newly added test
 06:41
Zoffix Got some warts with `&list(Seq:D)` all signs point to the right answer for `list (1, 2, 3).Seq` is `((1, 2, 3).Seq)` 06:42
m: dd list $(1, 2, 3) 06:43
camelia ($(1, 2, 3),)
Zoffix like we do here.
signs == &list and .list are not equivalent. There's no equivalence with, say, QuantHashes. Same as there's no equivalency between &hash and .hash 06:44
and HEAD~10's behaviour of &list(Seq:D) returning a Seq:D is really a quirk of `+@` slurpy with PositionalBindBlahBlah going on 06:46
m: sub (+l){ dd l }((1, 2, 3).Seq)
camelia (1, 2, 3).Seq
Zoffix m: sub (+l){ dd l }(42)
camelia (42,)
Zoffix ^ so &list(Seq:D) should work like the 42 version there and return a 1-item list.
I'm just gonna revert all I did in this area today: (a) 6.c had test that &list(Seq:D) was no-op; we judged Seq being a list as wrong and changed it, but it can be argued that Seq is *sometimes* a list due to PositionalBindFailover and in this case &list is a no-op precisely because of the failover; (b) from user's perspective, IMO &list(Seq:D) giving a Seq is less surprising and more DWIMMI than 06:53
&list(Seq:D) giving a 1-item list with that Seq in it; (c) I don't want to leave special Seq:D -> List handling in &list, since we don't do anything special for containerazied Positionals or QuantHashes
oh, I didn't even commit anything to rakudo 06:54
Geth roast: 51d2569718 | (Zoffix Znet)++ | S03-operators/context-forcers.t
Revert "Change incorrect &list(Seq) tests"

This reverts commit 6636645ac9c0fbe96d7a9af27fe6b1af07c50429.
They're right if you squint hard enough and alternatives are worse:
  irclog.perlgeek.de/perl6-dev/2017-...i_15636448		 06:55
roast/6.c-errata: 85ff3ca337 | (Zoffix Znet)++ | S03-operators/context-forcers.t
Revert "Change incorrect &list(Seq) tests"

This reverts commit 6636645ac9c0fbe96d7a9af27fe6b1af07c50429.
They're right if you squint hard enough and alternatives are worse:
  irclog.perlgeek.de/perl6-dev/2017-...i_15636448		 06:56
roast: fd95618356 | (Zoffix Znet)++ | S03-operators/context-forcers.t
Add comment for future core devs
 06:58
travis-ci Rakudo build passed. Zoffix Znet 'More MAIN USAGE fixes 07:09
travis-ci.org/rakudo/rakudo/builds/323500240 github.com/rakudo/rakudo/compare/1...fbd7ab7737
yoleaux AlexDaniel`: RT/GH weekly 07:19
Geth rakudo: d78297fcf4 | (Zoffix Znet)++ | src/core/List.pm
Revert "Fix &list(Seq:D) to produce a List"

This reverts commit 1d772dc002e2740ffb744d5e210d5e15e04a1c1f.
This was eventually deemed not the right course of action:
  irclog.perlgeek.de/perl6-dev/2017-...i_15636448		 07:36
roast: 79803f6322 | (Zoffix Znet)++ | S03-operators/context-forcers.t
[v6.d REVIEW] Expand/clarify spec for numeric coercers

  - Test more variants of coersion
  - Remove fudge for RT#124540 ; the prefix:<-> coersion first
   coerses and THEN negates result. The latter is deemed "using"
   a value, and hence it's right that it explodes the Failure
   created by the coersion. I double checked the speculations
   referenced by the two test sections and don't see anything
   saying it must preserve Failures (and it does coroborate the
   "coerces, THEN negates" explanatiuon)		 07:40
synopsebot RT#124540 [new]: rt.perl.org/Ticket/Display.html?id=124540 Roast rakudo skip/todo test:./S03-operators/context-forcers.t line:96 reason: 'failure modes of Str.Numeric'
roast: a3518f85c8 | (Zoffix Znet)++ | S03-operators/context-forcers.t
Make extra sure potential Failures don't get exploded
 07:43
Zoffix Found one more LTA-ness with &list: it converts Arrays to Lists 08:01
And really, it's the +@ slurpy that's got its stuff all over the place 08:02
m: sub (+@l is raw){ dd @l }([1, (2, 3)])
camelia (1, $(2, 3))
Zoffix Why did it turn that Array into a List?
m: sub (+@l){ dd @l }((1, (2, 3)))
camelia [1, (2, 3)]
Zoffix here it turns a List into an Array 08:03
m: sub (+l){ dd @l }([1, (2, 3)])
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '@l' is not declared
at <tmp>:1
------> 3sub (+l){ dd 7⏏5@l }([1, (2, 3)])
Zoffix and that one is same as `raw`
Opened as R#1355 and gonna put &list/.list spec on hold for now 08:25
synopsebot R#1355 [open]: github.com/rakudo/rakudo/issues/1355 +@a/+a slurpies produce inconsistent results
[Tux] Rakudo version 2017.12-76-gd78297fcf - MoarVM version 2017.12-14-g4541cf6f6
csv-ip5xs1.141 - 1.303
csv-ip5xs-2013.450 - 14.077
csv-parser12.532 - 14.391
csv-test-xs-200.428 - 0.475
test11.726 - 11.882
test-t2.882 - 2.933
test-t --race1.205 - 1.260
test-t-2051.524 - 51.675
test-t-20 --race18.666 - 19.257
 09:41
gfldex m: say so "½" ~~ /(\d+)/; say $0 11:21
camelia False
Nil
gfldex is that intentional? (If yes, it's a ENODOC trap.) 11:22
tbrowder hi, #perl6-dev, anyone watching? 11:33
gfldex I'm watching in waiting for answers.
tbrowder: do you want to join me? :-> 11:34
tbrowder gfldex: hi!
i have an nqp challenge for the heavy hitters here: check my translation of sub parse_line found in p5 module Text::ParseWords (on cpan). 11:35
i'm close, but as they say in the carnival games, "no cigar" 11:37
my attempt so far is at gist gist.github.com/tbrowder/4a0b64e6c...20aef47afd 11:38
gfldex m: say "½".Rat; 11:48
camelia Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5½' (indicated by ⏏)
in block <unit> at <tmp> line 1
gfldex is that intentional?
ilmari m: say "½".unival 11:49
camelia 0.5
ilmari m: say "⅓".unival
camelia 0.333333
ilmari m: say "⅓".unival.Rat
camelia 0.333333
ilmari m: dd "⅓".unival.Rat 11:50
camelia <1/3>
ilmari m: dd "⅓".unival
camelia <1/3>
lizmat [Tux]: am I seeing this correctly? is test-t --race now faster than csv-ip5cs ? 12:33
ip5xs rather
?
[Tux] ja en nee. de snelste run is sneller, maar de tweede run is trager 12:34
sorry, nl instead of en: the window is wider: the fastest run is faster, but the slowest run is slower 12:35
lizmat still :-)
within noise then :-)
[Tux] yes, sure
gfldex lizmat: given that next next year will start with a monday, a weekly with a little outlook of the next Perl 6-year would be nice. 13:14
lizmat gfldex: not sure if Monday 1st would be the best time to look forward, not after having all that booze :-) 13:19
both for the writer as well as the reader :-) 13:20
El_Che so no one else have failing tests on the latest rakudo? 13:31
I wonder if it's docker related or not
lizmat www.learningperl6.com/2017/12/31/p...algorithm/ # I think brian could use some pointers 13:35
like you also can use a sub in a sequence ?
lizmat is going afk in a mo
timotimo m: my @foo = 1, 2, 3; my @bar = [ @foo ]; @foo[1] = 99; say @foo; say @bar 14:51
camelia [1 99 3]
[1 2 3]
timotimo oh, now i get it
the , in there was the important bit 14:52
MasterDuke Zoffix: maybe we need to add an `if $*OS.is-win` path? the reason for that patch (if i remember correctly) was so someone could copy and paste the output of the USAGE message and it would work
yoleaux 05:26Z <Zoffix> MasterDuke: how come github.com/rakudo/rakudo/pull/1064/files converts `'` in literals to `'"'"'` Makes it kinda hard to read in usage messages. I suspect this was to automatically add shell quoting, but that quoting doesn't apply to Windows
timotimo gfldex: ½ doesn't match \d because it's not a digit, i'd say. that should go in the docs, though 14:53
m: sub pretty-print(Int $seconds) { say $seconds.polymod(60*60*24, 60*60, 60, 1) Z~ " " Z~ <Days Hours Minutes Seconds> }; pretty-print(5 * 60 + 3); pretty-print(7 * 60*60*24 + 22 * 60*60) 14:58
camelia (303 Days)
(79200 Days)
timotimo haha, oh no
gfldex i'm not gonna wait 79200 days! 14:59
timotimo m: sub pretty-print(Int $seconds) { say $seconds.polymod(24, 60, 60, 1) Z~ " " Z~ <Days Hours Minutes Seconds> }; pretty-print(5 * 60 + 3); pretty-print(7 * 60*60*24 + 22 * 60*60)
camelia (15 Days)
(0 Days)
gfldex in fact, I plan to be dead by then :->
timotimo docs.perl6.org/routine/polymod#class_Int - anyway, the examples here are already for dividing stuff into seconds, minutes, days, hours 15:00
but i used it the wrong way around
m: sub pretty-print(Int $seconds) { say $seconds.polymod(60, 60, 24) Z~ " " Z~ <Seconds Minutes Hours Days> }; pretty-print(5 * 60 + 3); pretty-print(7 * 60*60*24 + 22 * 60*60)
camelia (3 Seconds)
(0 Seconds)
timotimo but i'm not sure why it doesn't iterate over the result with the Z
oh, the " " in the middle, haha
m: sub pretty-print(Int $seconds) { say $seconds.polymod(60, 60, 24) Z~ (" " xx *) Z~ <Seconds Minutes Hours Days> }; pretty-print(5 * 60 + 3); pretty-print(7 * 60*60*24 + 22 * 60*60) 15:01
camelia (3 Seconds 5 Minutes 0 Hours 0 Days)
(0 Seconds 0 Minutes 22 Hours 7 Days)
timotimo m: sub pretty-print(Int $seconds) { say ($seconds.polymod(60, 60, 24) Z~ (" " xx *) Z~ <Seconds Minutes Hours Days>).grep(none(*.starts-with("0 ")) }; pretty-print(5 * 60 + 3); pretty-print(7 * 60*60*24 + 22 * 60*60)
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in argument list; couldn't find final ')' (corresponding starter was at line 1)
at <tmp>:1
------> 3s Days>).grep(none(*.starts-with("0 ")) 7⏏5}; pretty-print(5 * 60 …
timotimo m: sub pretty-print(Int $seconds) { say ($seconds.polymod(60, 60, 24) Z~ (" " xx *) Z~ <Seconds Minutes Hours Days>).grep(none(*.starts-with("0 "))) }; pretty-print(5 * 60 + 3); pretty-print(7 * 60*60*24 + 22 * 60*60)
camelia (3 Seconds 5 Minutes)
(22 Hours 7 Days)
timotimo m: sub pretty-print(Int $seconds) { say ($seconds.polymod(60, 60, 24) Z~ (" " xx *) Z~ <Seconds Minutes Hours Days>).grep(none(*.starts-with("0 "))).reverse }; pretty-print(5 * 60 + 3); pretty-print(7 * 60*60*24 + 22 * 60*60)
camelia (5 Minutes 3 Seconds)
(7 Days 22 Hours)
timotimo m: sub pretty-print(Int $seconds) { say ($seconds.polymod(60, 60, 24) X~ " " Z~ <Seconds Minutes Hours Days>).grep(none(*.starts-with("0 "))).reverse }; pretty-print(5 * 60 + 3); pretty-print(7 * 60*60*24 + 22 * 60*60) 15:02
camelia 5===SORRY!5=== Error while compiling <tmp>
Only identical operators may be list associative; since 'X~' and 'Z~' differ, they are non-associative and you need to clarify with parentheses
at <tmp>:1
------> 3y ($seconds.polymod(60, 60, 24) X…
timotimo m: sub pretty-print(Int $seconds) { say (($seconds.polymod(60, 60, 24) X~ " ") Z~ <Seconds Minutes Hours Days>).grep(none(*.starts-with("0 "))).reverse }; pretty-print(5 * 60 + 3); pretty-print(7 * 60*60*24 + 22 * 60*60)
camelia (5 Minutes 3 Seconds)
(7 Days 22 Hours)
timotimo good post, gfldex++ 15:03
Geth roast: jstuder-gh++ created pull request #373:
Test Issue 1204 (Strip slip in quotewords)		 17:57
Zoffix gfldex: yes "½".Rat is intentional. The ruling was the user needs to use unival() for those and only Nd were allowed. I image regex is the same 18:10
MasterDuke: yeah, adding $*DIST.is-win sounds like a plan. 18:11
Geth roast: 4788582437 | (Jeremy Studer)++ | 3 files
Test Issue 1204 (Strip slip in quotewords)

This tests for bugs exhibited in
  [Issue 1204](github.com/rakudo/rakudo/issues/1204).
Corresponds with development in Rakudo:
   Commit ad684de383bb0c0ff051d7b0901583a9b5c2e398
   [PR 1350](github.com/rakudo/rakudo/pull/1350)		 18:14
roast: bb44171de7 | (Zoffix Znet)++ (committed using GitHub Web editor) | 3 files
Merge pull request #373 from jstuder-gh/issue_1204

Test Issue 1204 (Strip slip in quotewords)
roast: 22f00c7613 | (Zoffix Znet)++ | 2 files
[v6.d REVIEW] Use more appropriate test routine
 18:16
roast: 37133cb65a | (Zoffix Znet)++ | S03-operators/buf.t
[v6.d REVIEW] Expand infix:<~> with Blob tests

  - Test both return and result of meta assign
  - Test other forms of the op
Origs:
  github.com/perl6/roast/commit/a85c...c972f0bf97
  github.com/perl6/roast/commit/7073...990eb6d12e		 18:53
roast: 3589f5f4c0 | (Zoffix Znet)++ | S32-list/minmax.t
[v6.d REVIEW] remove trailing whitespace
 19:15
roast: ecf8c6d7fc | (Zoffix Znet)++ | S32-list/minmax.t
[v6.d REVIEW] Unfudge min/max flattening tests

The skip-fudge expects no flattening to happen with single-arg
  `[min]`/`[max]` with Range arg, but I don't see what would be a
viable answer in such a case, other than min/maxing the Range's endpoints, which would merely be *an optimization*, not a functional difference, and IMO should not be specced. (N.B.: don't see ... (6 more lines)		 19:29
rakudo: 4909430c73 | usev6++ | src/vm/jvm/runtime/org/perl6/rakudo/Binder.java
[jvm] Adjust logic for handling anonymous params

This makes the code more similiar to what we do for MoarVM in BOOTSTRAP.nqp
  (following commit 88acdbb745).		 19:36
Zoffix .ask lizmat in this commit, suggestion for `$#` is `.fmt` do you remember which Perl 5 feature that covers? The only one I know is $#array which in Perl 6 would be @array.end. Wondering if I should just replace the .fmt with .end or have both of them because there's also some .fmt feature: github.com/rakudo/rakudo/commit/bd...6f069c4e01 19:48
yoleaux Zoffix: I'll pass your message to lizmat.
lizmat metacpan.org/pod/perlvar#Deprecate...-variables 19:52
yoleaux 19:48Z <Zoffix> lizmat: in this commit, suggestion for `$#` is `.fmt` do you remember which Perl 5 feature that covers? The only one I know is $#array which in Perl 6 would be @array.end. Wondering if I should just replace the .fmt with .end or have both of them because there's also some .fmt feature: github.com/rakudo/rakudo/commit/bd...6f069c4e01
lizmat "$# was a variable that could be used to format printed numbers."
Zoffix lizmat++ thanks
lizmat afk& 19:53
travis-ci Rakudo build errored. usev6 '[jvm] Adjust logic for handling anonymous params 20:52
travis-ci.org/rakudo/rakudo/builds/323654712 github.com/rakudo/rakudo/compare/d...09430c737d
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. Across all jobs, only t/02-rakudo/repl.t test file failed.
Geth roast: 1e22663ad3 | (Zoffix Znet)++ | S32-exceptions/misc.t
[v6.d REVIEW] Remove conflicting $# var test

It's already tested to throw ::Perl5Var exception; don't test for ::Obsolete		 21:08
rakudo: 0444a2c33f | (Zoffix Znet)++ | 2 files
Awesomify `$#` var error

  - Throw more appropriate X::Syntax::Perl5Var instead of X::Obsolete
  - Suggest .fmt if it's just the `$#` var, which is an old Perl 5
   variable for formatting numbers:
   metacpan.org/pod/perlvar#Deprecate...-variables
  - Suggest @foo.end if it's the `$#var` var; include identifier name
   in the error.		 21:09
roast: 4967ec515b | (Zoffix Znet)++ | S32-exceptions/misc.t
Unfudge `$#` var tests

Rakudo fix: github.com/rakudo/rakudo/commit/0444a2c33f		 21:11
Zoffix .ask AlexDaniel perhaps the todo-fudged versions should be removed from the spec? The `$/` is a proper Perl 6 var. The rest... I dunno... the Perl5-ism errors only ever hurt me not helped me; I think the fewer of them we have the better. If I write `$+` it usually means I meant `$++` and I don't want it suggesting some crap from some other language... 21:25
yoleaux Zoffix: I'll pass your message to AlexDaniel.
Zoffix github.com/perl6/roast/commit/0a1d...dddfe0b869
.ask AlexDaniel github.com/perl6/roast/commit/0a1d...dddfe0b869
yoleaux Zoffix: I'll pass your message to AlexDaniel.
Zoffix m: my $x = new class Foo {}: 21:26
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of C++ constructor syntax; in Perl 6 please use method call syntax
at <tmp>:1
------> 3my $x = new class 7⏏5Foo {}:
Zoffix This is the worst offender IMO, because it blocks a much more visible way to write a `.new` on a class's definition. 21:27
m: my $x = .new without class Foo {} 21:29
camelia ( no output )
Geth rakudo: 6f36b02137 | (Zoffix Znet)++ | 2 files
Smartify harness automatic job number calculation

  - Use TEST_JOBS if we were given any
  - Instead of using default 6 jobs, calculate using the number of CPUs
   on the box. Multiplying number by 1.3 is the optimal number of
   test jobs I found a while back.
  - On harness6 old default was 1 by mistake; it was supposed to be 6
  - Default to 6 cores if our cpu-core computation fails for some reason		 21:48
roast: 7145799cce | (Zoffix Znet)++ | 2 files
[v6.d REVIEW] Test for actual types instead of their .^names

Orig: github.com/perl6/roast/commit/34ce...bae08c6e26		 22:06
roast: 201865ab6f | (Zoffix Znet)++ | integration/weird-errors.t
[v6.d REVIEW] Localize test's vars and subs

So they don't affect other stuff in the file Orig: github.com/perl6/roast/commit/834a...b0f1dbcd83		 22:11
roast: 309c992bb5 | (Zoffix Znet)++ | S10-packages/precompilation.t
[v6.d REVIEW] Fix typo in comment

Orig: github.com/perl6/roast/commit/716b...124cb41e59		 22:13
roast: 438f4155ca | (Zoffix Znet)++ | S12-construction/BUILD.t
[v6.d REVIEW] Remove trailing whitespace
 22:16
roast: 80600333cf | (Zoffix Znet)++ | S12-construction/BUILD.t
[v6.d REVIEW] Use more appropriate test routines

Orig: github.com/perl6/roast/commit/f210...f232e153a3
travis-ci Rakudo build errored. Zoffix Znet 'Awesomify `$#` var error 22:20
travis-ci.org/rakudo/rakudo/builds/323668099 github.com/rakudo/rakudo/compare/4...44a2c33fb5
buggable [travis build above] ✓ All failures are due to: timeout (3 failures).
Geth roast: 0147de8705 | (Zoffix Znet)++ | S16-filehandles/mode.t
[v6.d REVIEW] Improve IO.mode tests

  - Use our standard temp file routines for temp files
  - Shorten test descriptions
  - Use more appropriate test routines
  - Toss full-file Windows fudge. Any failures here will be dealt
   separately, as part of resolving #320		 22:27
synopsebot SPEC#320 [open]: github.com/perl6/roast/issues/320 Windows roast failures just before 2017.09 release
roast: 3fafce04be | (Zoffix Znet)++ | S11-modules/nested.t
[v6.d REVIEW] Remove ded comment

  - It seems to refer[^1] to a now-removed code
  - There's no such RT ticket
  [1] github.com/perl6/roast/commit/8181...4ef68befed		 22:34
roast: 899cf5e3fa | (Zoffix Znet)++ | S32-str/split-simple.t
[v6.d REVIEW] Add missing `plan` in subtest

Orig: github.com/perl6/roast/commit/c3f4...2aecd1d267		 23:09
roast: 0ee6a4eb27 | (Zoffix Znet)++ | S05-transliteration/trans.t
[v6.d REVIEW] Use more appropriate test routines

Orig: github.com/perl6/roast/commit/2010...e654f120d6		 23:11
roast: 86787bf61a | (Zoffix Znet)++ | S06-signature/types.t
[v6.d REVIEW] Use more appropriate test routines

Orig: github.com/perl6/roast/commit/3fe0...484ec089ce		 23:16
roast: 1f54bd963c | (Zoffix Znet)++ | S32-str/split.t
[v6.d REVIEW] Use more appropriate test routines

Orig: github.com/perl6/roast/commit/0d86...950fad0e9f		 23:19
travis-ci Rakudo build errored. Zoffix Znet 'Smartify harness automatic job number calculation 23:25
travis-ci.org/rakudo/rakudo/builds/323673079 github.com/rakudo/rakudo/compare/0...36b02137fe
buggable [travis build above] ☠ All failures are due to: timeout (2 failures), failed make test (1 failure). Across all jobs, 2 unique test files failed.
Geth roast: 500cb47fb8 | (Zoffix Znet)++ | S12-enums/basic.t
[v6.d REVIEW] Localize symbols in the test file

To avoid conflicts in other parts of the file		 23:29
rakudo: e5b4f37f18 | (Zoffix Znet)++ | t/05-messages/02-errors.t
Test .substr errors has right suggestsions

RT#128039: rt.perl.org/Ticket/Display.html?id=128039		 23:33
synopsebot RT#128039 [resolved]: rt.perl.org/Ticket/Display.html?id=128039 [BUG] Misleading suggestion in .substr error message for positive index values in Rakudo
Geth roast: 55c3087001 | (Zoffix Znet)++ | S32-str/substr.t
[v6.d REVIEW] DSETEM in .substr error

DSETEM = "Do Not Spec Exact Text of Error Messages". These proved of little value and were interfering with readability changes in 6.c. Keep these in Rakudo's tests for now. If other impls need it, we can always do an Appendix set of tests to the spec that impls can optionally follow.
Moved to rakudo in github.com/rakudo/rakudo/commit/e5b4f37f18		 23:34

Powered by Cro and the Raku® Programming Language.

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

Thank you!