Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_log/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by Zoffix on 27 July 2018.
AlexDaniel 💤 00:02
samcv does it make sense to change "Can not" to Cannot in one of our error messages? 00:05
Can not decode a utf-8 buffer as if it were $encoding
MasterDuke samcv: fwiw, in moarvm, there are way more "cannot" than "can not" 00:19
samcv yeah 00:21
Geth roast: 903d39546f | (Zoffix Znet)++ | 2 files
Add APPENDIX 3: Older Specs

For tests covering bug fixes and stuff in features that
  are deprecated in current version of the spec		 01:12
rakudo: 0a45ffbc67 | (Zoffix Znet)++ | t/spectest.data
Add APPENDICES/A03-older-specs/01-misc.t to list of test files
 01:13
roast: 1fefe0a255 | (Zoffix Znet)++ | 2 files
[v6.d REVIEW] Appendicise "-" magic open string cover tests

This feature is deprecated in 6.d
Orig: github.com/perl6/roast/commit/60a7a88e4		 01:15
roast: 56c2d19944 | (Zoffix Znet)++ | 2 files
[v6.d REVIEW] Appendicise slurp on '-' path

This feature is deprecated in 6.d
Orig: github.com/perl6/roast/commit/60a7a88e4		 01:16
travis-ci Rakudo build errored. Zoffix Znet 'Add APPENDICES/A03-older-specs/01-misc.t to list of test files' 01:45
travis-ci.org/rakudo/rakudo/builds/429377697 github.com/rakudo/rakudo/compare/a...45ffbc6771
buggable [travis build above] ✓ All failures are due to: timeout (1 failure). 01:45
[Tux] Rakudo version 2018.08-109-g0a45ffbc6 - MoarVM version 2018.08-85-gadce51fb8
csv-ip5xs0.910 - 0.929
csv-ip5xs-207.477 - 7.774
csv-parser24.519 - 24.677
csv-test-xs-200.426 - 0.432
test8.722 - 8.806
test-t2.059 - 2.102
test-t --race0.938 - 0.963
test-t-2035.860 - 37.292
test-t-20 --race12.490 - 12.606
 07:18
AlexDaniel weekly: reportable gist.github.com/2c1d0dbb2955715584...83c1090dce 08:35
notable6 AlexDaniel, Noted!
AlexDaniel weekly: zef too slow? Maybe it's possible to make it faster: github.com/ugexe/zef/issues/272 08:47
notable6 AlexDaniel, Noted!
AlexDaniel weekly: AlexDaniel worked on improving toaster to make things faster and more useful (failing modules are now bisected automatically). Still WIP, but we will see how it runs in the next weekly. In the process Whateverable got a noticeable refactor to make parts of it reusable. 08:51
notable6 AlexDaniel, Noted!
dogbert2 . 10:00
lizmat m: @a[999999999999999999999999999999] = 42; say @a.elems 10:37
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '@a' is not declared
at <tmp>:1
------> 3<BOL>7⏏5@a[999999999999999999999999999999] = 42;
lizmat use Array::Sparse; my @a is Array::Sparse; @a[999999999999999999999999999999] = 42; say @a.elems # 1000000000000000000000000000000 10:38
:-)
Geth rakudo: lizmat self-assigned Missing API functionality on Array github.com/rakudo/rakudo/issues/2288
6c93fde85d | (Elizabeth Mattijsen)++ | src/core/Rakudo/Iterator.pm6

Fixes R#2286		 10:54
AlexDaniel m: my @a[999999999999999999999999999999] = 42; say @a.elems 11:21
camelia Illegal dimension in shape: 999999999999999999999999999999. All dimensions must be integers bigger than 0
in block <unit> at <tmp> line 1
Zoffix .tell jnthn FWIW there are 4 @LARRY 6.d Issues that could use a comment: github.com/rakudo/rakudo/issues?q=...3A%40LARRY (also there's a question for you on github.com/rakudo/rakudo/issues/2286 about $!reified) 11:23
yoleaux Zoffix: I'll pass your message to jnthn.
lizmat m: my @a; @a[999999999999999999999999999999] = 42; say @a.elems # AlexDaniel 11:26
camelia Cannot unbox 100 bit wide bigint into native integer
in block <unit> at <tmp> line 1
lizmat jnthn: do you know a HLL way to find out if something is a container? 11:36
Zoffix huggable: HLL iscont 11:46
huggable Zoffix, say $v.VAR ~~ Scalar|Proxy ?? "conted" !! "deconted"
Zoffix lizmat: that's one way that I know of ^ 11:47
lizmat thought of "it.VAR.^name ne it.^name" 11:52
this also handles IntLexRef 11:54
m: my int $a; sub a(\a) { dd $a.VAR.^name }; a $a
camelia "IntLexRef"
Zoffix ah, cool 12:06
m: sub is-cont(\a) { so a.VAR.^name eq a.^name }; my $a; dd is-cont $a 12:07
camelia Bool::False
Zoffix doesn't seem to work here
oh, is my conditional upsidedown
m: use nqp; my int $a; dd nqp::iscont($a) 12:08
camelia 1
Zoffix m: sub is-cont(\a) { so a.VAR.^name ne a.^name }; my $a = 42; dd is-cont $a
camelia Bool::True
Zoffix m: sub is-cont(\a) { so a.VAR.^name ne a.^name }; my int $a = 42; dd is-cont $a
camelia Bool::False
Zoffix lizmat: doesn't seem to match what `nqp::iscont` does for natives
huggable: HLL iscont :is: say $v.VAR ~~ Scalar|Proxy ?? "conted" !! "deconted" or sub is-cont(\a) { so a.VAR.^name ne a.^name }; 12:09
huggable Zoffix, Added HLL iscont as say $v.VAR ~~ Scalar|Proxy ?? "conted" !! "deconted" or sub is-cont(\a) { so a.VAR.^name ne a.^name };
lizmat well, for my purpose I needed to know if I can assign to it having its own container 12:09
synopsebot R#2286 [open]: github.com/rakudo/rakudo/issues/2286 [fudged tests committed] `(1,2) X+ ()` crashes
Geth roast: c44518fb8e | (Elizabeth Mattijsen)++ | S03-metaops/cross.t
Unfudge test for R#2286
travis-ci Rakudo build passed. Elizabeth Mattijsen 'Make sure X handles empty lists on RHS 19:55
travis-ci.org/rakudo/rakudo/builds/429723048 github.com/rakudo/rakudo/compare/0...93fde85da0
lizmat notable6: weekly 20:45
notable6 lizmat, 13 notes: gist.github.com/97248d50de4529864d...4c172c4848
lizmat notable6: reset weekly 21:01
notable6 lizmat, Moved existing notes to “weekly_2018-09-17T21:01:48Z”
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2018/09/17/...ersus-six/ 21:46
samcv lizmat++ 21:51
sjn lizmat++ #weekly #mustread 21:53
samcv finally all the tests i wrote for utf16le and utf16be and utf16 and detecting bom with decode stream are passing \o/ 21:58
made sure that .readchars(1) makes sure things don't get unaligned
AlexDaniel lizmat++ 22:15
timotimo lizmat++ 22:20
samcv so on my utf16 testing branch, `utf16` detects encoding based on BOM, if BOM doesn't exist it uses the machine endianess. If set to utf16be and it sees a big endian BOM, the BOM is ommited. likewise for little endian but for LE BOM 22:27
i still really don't like the byte order marker... i think this is a good way to do it. 22:30
Perl 5 will tell me that a character is illegal if i try to decode the BOM in the wrong endianess (which results in a non-character) 22:31
ah looks like perl 5 will follow the spec and not skip the BOM. i kind of like that way since you are able to store all characters in it 22:33
timotimo i wonder if :keepbom should be a little like :!chomp 22:38
a program written by someone who doesn't expect to get utf16 (say, maybe they use a heuristic to figure out the right encoding), they might be surprised about the garbage at the beginning of the file 22:39
samcv i need to see if the noncharacter which is the BOM with endianess reversed is valid in utf16. i mean. you probably shouldn't have it. but if it's not valid period 22:42
though i suppose dying on reversed BOM character is not as bad as utf8 i feel. because there *are* illegal characters in utf16. namely you can't represent the low surrogates or high surrogates in anyway 22:49
TimToady it was intentionally designed to be invalid when reversed, iirc 22:56
and I think has been generalized to every plane, such that the *fffe and *ffff are considered invalid 22:57
samcv yeah it has TimToady
this whole thing is so messy though!
TimToady an aBOMination :) 22:58
samcv 10/10 pun
AlexDaniel samcv: should've started that message with ZERO WIDTH NO-BREAK SPACE 23:12
foobar 23:13
bar
cool, that goes all the way to colabti without issues :) 23:14
diakopter aBOM ☃ 23:23
MasterDuke timotimo: did you see my comments in #moarvm last night? i figured out what that error i was getting was from 23:32
timotimo oh, was it a typo'd class name? 23:33
MasterDuke QAST::Something() vs QAST::Something.new() 23:34
timotimo i should have been able to spot that :\ 23:35
MasterDuke and then i also had a QAST::WVar instead of QAST::WVal
the sad thing is it's not the first or even second time i've done that and gone on a long goose chase before realizing what the real problem was
timotimo yeah :| 23:36
i've very briefly looked into making that a compile-time error
i didn't get it working, and it's been so long ago that i don't know any more what was up
MasterDuke i know there's less interest in nqp errors, but wow, i've lost some time to them 23:38

Powered by Cro and the Raku® Programming Language.

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

Thank you!