samcv \o/ i now have perl6 working on Solaris 00:02
MasterDuke nice 00:25
huh 02:09
where do you see that?
do you have your own openbsd install? 02:10
samcv yeah
it's when compiling nqp
Kaiepi just to be clear on the usage of nativecall expliticly-manage 02:11
samcv pretty unlikely my changes could cause that but i can double check on a master
Kaiepi it's meant for strings that the compiler shouldn't gc right?
samcv MasterDuke: cry.nu/p/7iwo 02:16
MasterDuke samcv: huh, don't even know where to start with that 02:19
samcv yeah. no issues on freebsd though 02:20
though openbsd does a lot of things different afaik 02:21
MasterDuke m: my ($a, $b); for (^1_000_000).kv -> $i, $v { $a = $i; $b = $v }; say "$a $b"; say now - INIT now 03:58
camelia 999999 999999
2.05613357
MasterDuke my ($a, $b); my $i; for ^1_000_000 -> $v { $a = $i++; $b = $v }; say "$a $b"; say now - INIT now
evalable6 999999 999999
0.3695138
MasterDuke why does the first call sink-all?
samcv github.com/MoarVM/MoarVM/commit/ef4e6fe945 this should help with problems with rakudobrew failing to update 04:06
it seems it only happens if there are stray files left over. and rakudobrew doesn't clean them up i suppose.
but it'll now show a message telling the user what folder to delete and to run it again after they delet it 04:07
nine Kaiepi: yes 06:39
[Tux] Rakudo version 2018.03-221-gaffeaa798 - MoarVM version 2018.03-105-gef4e6fe94
csv-ip5xs0.911 - 0.936
csv-ip5xs-209.054 - 9.071
csv-parser38.301 - 40.196
csv-test-xs-200.446 - 0.484
test9.366 - 9.552
test-t2.524 - 2.536
test-t --race1.046 - 1.127
test-t-2045.022 - 45.582
test-t-20 --race15.460 - 15.763
 07:16
AlexDaniel releasable6: next 07:26
releasable6 AlexDaniel, Next release in ā‰ˆ10 days and ā‰ˆ11 hours. 6 blockers. 0 out of 221 commits logged (āš  9 warnings)
AlexDaniel, Details: gist.github.com/1ff9ef5a27d3821109...3eab907c88
AlexDaniel Zoffix: re ā€œ(no changelog)ā€ in commit title, you can also include that in the description of the commit instead of a title, as I end up reading these anyway 07:28
Zoffix: alternatively you can just go to github.com/rakudo/rakudo/wiki/ChangeLog-Draft and include that commit in IGNORE&DELETE section 07:29
Zoffix AlexDaniel: "AlexDanielZoffix: re ā€œ(no changelog)ā€ in commit title" what are you talking about? 07:53
AlexDaniel Zoffix: github.com/rakudo/rakudo/commit/a0ee718d 07:54
Zoffix Ah 07:56
AlexDaniel undersightable6: check 08:19
undersightable6 AlexDaniel, OK! Working on itā€¦
AlexDaniel, 0 errors, 7 warnings: gist.github.com/3c0391b20261800071...e72b6a0398 08:21
AlexDaniel Geth:
undersightable6 you are wrong.
undersightable6: check
undersightable6 AlexDaniel, OK! Working on itā€¦
AlexDaniel, 0 errors, 7 warnings: gist.github.com/b22fb8401cb3e449bb...2d2e0c7b6c 08:22
AlexDaniel idiot. 08:23
there's no dalek and Geth 08:24
Zoffix synopsebot: help 08:31
synopsebot Zoffix, See: github.com/perl6/synopsebot
Zoffix weird that it's still running 08:32
lizmat Files=1239, Tests=76295, 318 wallclock secs (16.17 usr 5.55 sys + 2197.59 cusr 219.91 csys = 2439.22 CPU) 08:46
Geth Ā¦ rakudo: zoffixznet self-assigned When coercing to an unknown type, error is "Unable to parse expression in typename; couldn't find final ')'" github.com/rakudo/rakudo/issues/1723 08:51
pmurias TimToady: are 32bit implementations required to emulate int64 and uint64? the roast "spec" isn't clear on that 09:03
Zoffix ZOFFLOP: t/spec/S32-io/io-cathandle.t 09:11
Geth nqp: 4039aebbe3 | pmurias++ | src/vm/js/Compiler.nqp
[js] Fix uint16
 09:43
nqp: be655367c6 | pmurias++ | 5 files
[js] Add support for uint32
pmurias Zoffix: we have tests that storing 2 ** 33 in a uint8 doesn't throw 09:51
Zoffix: how do we handle that?
Zoffix: we also have fudged tests that it *does* throw 09:52
Zoffix You mean like my uint8 $ = 2**33? I'd expect that to throw
Well, not just I; that was discussed that `my native $foo = too-big => throw; and my native $foo = ok-sized; $foo++ => overflow` 09:53
lizmat yeah, assignment would throw, manipulating value in it would overflow / underflow silently 09:54
as that is what many algorithms expect 09:55
pmurias github.com/perl6/roast/blob/master...ive.t#L304
github.com/perl6/roast/blob/master...ive.t#L329 - more extreme assignment 09:56
lizmat: that semantics make sense, I'm not sure what we do about the current tests 09:57
lizmat perhaps creating an issue would help so the discussion gets concentrated ? 09:58
Zoffix m: my uint8 $u8; $u8 = 2 ** 32; dd $u8 09:59
camelia 0
Zoffix m: my uint8 $u8; $u8 = 2 ** 33+1; dd $u8
camelia 1
travis-ci NQP build passed. pmurias '[js] Add support for uint32' 10:01
travis-ci.org/perl6/nqp/builds/365029385 github.com/perl6/nqp/compare/a609b...655367c671
Zoffix m: my uint64 $u8 = 2**66; dd $u8
camelia Cannot unbox 67 bit wide bigint into native integer
in block <unit> at <tmp> line 1
Zoffix Makes sense to me to deem those tests wrong and toss them.
Geth rakudo/fix-R1723: fb7d5defbd | (Zoffix Znet)++ | 2 files
Stash partial fix for R#1723
 10:09
synopsebot R#1723 [open]: github.com/rakudo/rakudo/issues/1723 When coercing to an unknown type, error is "Unable to parse expression in typename; couldn't find final ')'"
pmurias Zoffix: what's a procedure for tossing wrong tests?
lizmat If we agree that they are wrong, they should also be changed in 10:10
Geth roast: 0d51cc6c18 | (Zoffix Znet)++ | S12-coercion/coercion-types.t
Fudged tests for good error in mistyped types in coercers

R#1723 github.com/rakudo/rakudo/issues/1723
lizmat / fudged in 6.c-errata
GH R#1724 # Inconsistent handling of Cool -> Str migration 10:11
synopsebot R#1724 [open]: github.com/rakudo/rakudo/issues/1724 [@LARRY] Inconsistent handling of Cool -> Str migration
Zoffix pmurias: no written procedures yet. If there's good agreement the tests are wrong, then just change them in master and cherry pick the change to 6.c-errata. Otherwise, PR the change and ping @LARRY to review whether the change is acceptable
Zoffix &
pmurias github.com/rakudo/rakudo/issues/1725 - issue for our integer problem 10:56
do we need the EVAL: github.com/perl6/roast/blob/master...uint.t#L10 ? 15:19
Geth roast: 5a11a77618 | usev6++ | S32-io/IO-Socket-INET.t
[JVM] Fudge newly added test
 15:23
lizmat pmurias: the alternative would be some kind of array that lists supported int types in core? 18:39
int1 int2 int4 are specced, but NYI
Geth rakudo: MasterDuke17++ created pull request #1727:
Use explicit native types in some places		 20:32
rakudo: 9d52444d8e | (Daniel Green)++ | 4 files
Use explicit native types in some places
 20:35
rakudo: 72c7c5bf46 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | 4 files
Merge pull request #1727 from MasterDuke17/use_explicit_native_types_in_some_places

Use explicit native types in some places
MasterDuke lizmat: re github.com/rakudo/rakudo/pull/1722 do you think any part of it is good? 21:24
lizmat MasterDuke: may I get back to you on that tomorrow?
MasterDuke sure, no pressing need 21:25
lizmat is pretty tired from a lot of stuff today and is only capable of menial work atm (or zonk out in front of the TV)
please ping me in case I forget :-)
good night, #perl6-dev!
MasterDuke k

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!