Perl 6 language and compiler development 2.0 | Logs at irclog.perlgeek.de/perl6-dev/today | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by moderator on 22 May 2018.
00:11 lizmat joined 00:29 Kaiepi joined 00:30 Kaiepi joined 00:32 Kaiepi joined 00:43 Kaiepi joined
Geth rakudo: skids++ created pull request #1858:
Colon lists		 01:09
01:58 ilbot3 joined
moderator Perl 6 language and compiler development 2.0 | Logs at irclog.perlgeek.de/perl6-dev/today | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
02:00 Kaiepi joined
Geth roast/car-grant-unreduce: 746566091d | (Zoffix Znet)++ | S32-num/rat.t
Revert "[CaR Grant] Change spec of zero-denominator rationals normalization"

This reverts commit 798b571f5f293ec8906aad2f7256ccf53ef40656.
There are a couple more tests that expect lack of normalization and normalizing ZDRs leads to poorer error reporting.
An attempt will be made to avoid ZDR normalization and instead to extract ZDR's functionality into a separate role, and then we can be a bit more liberal with on-demand, internal-only normalization.		 02:42
02:51 Zoffix joined
Zoffix m: class Foo does Rational[Int,Int] {}; role Bar {}; my $x := Foo.new.^mixin: Bar; multi z(Numeric) { say "n" }; multi z(Rational) is default { say "here"}; multi z(Bar) is default { say "there"}; z $x 02:51
camelia Ambiguous call to 'z(Foo+{Bar})'; these signatures all match:
:(Numeric $)
:(Rational $)
:(Bar $)
in block <unit> at <tmp> line 1
Zoffix I think the ZeroDenominator role is dead from the startline. We already even use `is default` for <cmp> op for Rational. If we add yet another role to have to differentiate with, it'd be even worse :( 02:52
I'm starting to regret this grant :P
Though I recall regretting and even wanting to quit IO grant too, but I learned a lot from it.
Plan C: blow everything up and re-design from scratch as a learning tool to see how our impl differs. 02:57
Plan D: dial down all of the ZDR stuff and just blow up more instead of doing IEEE stuff.
Plan E: make stuff slower 03:02
Plan F: FFFFFFFFFFFUUUUUUUUU
^_^
Gonna do with Plan C and should have @results by end of Monday.
03:02 Zoffix left 04:19 shareable6 joined
Geth nqp: mryan++ created pull request #451:
Mryan ast debug		 05:27
bartolin I've seen failing tests in S26-documentation/block-leading.t and S26-documentation/why-leading.t (rakudo commit 26f0033427): gist.github.com/usev6/1164d70c77a6...9b73c1389b 05:41
yoleaux 19 May 2018 04:25Z <yy6p6> bartolin: is there any way to compile a file into an executable jar?
bartolin has to leave for $work now, but wanted to report it.
Kaiepi PufferBot, all 05:56
PufferBot [openbsd] Running complete Rakudo build and tests (this will take a while)...
[openbsd] Failed to build Rakudo and run all tests... See the output at fpaste.scsys.co.uk/577516 05:57
Kaiepi bah, have to make realclena 05:58
should probably add --gen-nqp to its config as well 05:59
06:01 shareable6 joined 06:04 PufferBot joined
Kaiepi PufferBot, all 06:04
PufferBot [openbsd] Running complete Rakudo build and tests (this will take a while)...
[openbsd] Failed to build Rakudo and run all tests... See the output at fpaste.scsys.co.uk/577517 06:05
Kaiepi hm 06:06
nqp's not recognizing slice as a moar op
06:09 PufferBot joined
Kaiepi works fine when i build it myself though... 06:09
BeastieBot, all 06:11
BeastieBot [freebsd] Running complete Rakudo build and tests (this will take a while)...
06:22 PufferBot joined 06:27 BeastieBot_ joined, BeastieBot joined
Kaiepi alright, one more time now that i updated them 06:28
PufferBot, all
PufferBot [openbsd] Running complete Rakudo build and tests (this will take a while)...
[openbsd] Failed to build Rakudo and run all tests... See the output at fpaste.scsys.co.uk/577518 06:29
Kaiepi BeastieBot, all 06:30
BeastieBot [freebsd] Running complete Rakudo build and tests (this will take a while)...
[freebsd] Failed to build Rakudo and run all tests... See the output at fpaste.scsys.co.uk/577519
06:33 BeastieBot joined
Kaiepi BeastieBot, all 06:33
BeastieBot [freebsd] Running complete Rakudo build and tests (this will take a while)...
[freebsd] Failed to build Rakudo and run all tests... See the output at fpaste.scsys.co.uk/577520
06:35 BeastieBot joined
Kaiepi BeastieBot, all 06:35
BeastieBot [freebsd] Running complete Rakudo build and tests (this will take a while)...
Kaiepi alright there we go, building rakudo now 06:36
06:58 robertle joined
BeastieBot [freebsd] Failed to build Rakudo and run all tests... See the output at fpaste.scsys.co.uk/577521 07:00
07:10 brrt joined
AlexDaniel Kaiepi: ok, good point… 07:46
Kaiepi: I somewhat forgot about freebsd
Kaiepi: are any of these failures new?
I have never seen a clean run to be honest
07:47 shareable6 joined
Kaiepi i haven't checked in a while so most likely 07:47
07:55 PufferBot joined
Kaiepi i'll test one more time with pufferbot in case any other tests break 07:55
PufferBot, all
PufferBot [openbsd] Running complete Rakudo build and tests (this will take a while)... 07:56
[openbsd] Failed to build Rakudo and run all tests... See the output at fpaste.scsys.co.uk/577522 07:57
Kaiepi nope, still the same nqp error
PufferBot, test 07:58
PufferBot [openbsd] Running tests...
[openbsd] Failed tests... See the output at fpaste.scsys.co.uk/577523
lizmat Files=1239, Tests=76372, 320 wallclock secs (15.51 usr 5.38 sys + 2209.74 cusr 220.94 csys = 2451.57 CPU) 07:59
Kaiepi weird, i can build and run tests fine using a different copy of the repo 08:04
regular tests passed, doing stresstest now 08:07
same tests fail hastebin.com/anesuzixaq.bash 08:28
(ignore the IO::Socket::Async ones, those have to do with not having ipv6 configured)
AlexDaniel Kaiepi: can you ticket everything that is failing? 08:43
or do we have a ticket somewhere already?
09:05 brrt joined
Geth rakudo: fc57f9f055 | skids++ | src/Perl6/Actions.nqp
Execute code in elided duplicate argument pair values
 09:08
rakudo: 732976007a | skids++ | src/Perl6/Actions.nqp
Just an :op<list> will suffice here... jnthn++ for feedback
rakudo: 4bbb87f554 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | src/Perl6/Actions.nqp
Merge pull request #1856 from skids/elided_pairs

Execute code in elided duplicate argument pair values
lizmat FWIW, I'm seeing a quite a few documentation related spectest fails 09:09
something not merged yet ?
AlexDaniel could be so, but fwiw that doesn't affect the upcoming release in any way 09:14
.oO( master branch – go crazy with it! )
 09:18
Geth rakudo: cb66dfcccb | (Tom Browder)++ | src/Perl6/Pod.nqp
clean up code, expand space chars considered when normalizing text
 09:23
rakudo: 6f1b1cb66f | (Tom Browder)++ | src/Perl6/Pod.nqp
correct unicode numbers for the breaking vs. non-breaking space chars
rakudo: 27e8dab9f8 | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | src/Perl6/Pod.nqp
Merge pull request #1857 from tbrowder/nbsp-improv

clean up code, expand space chars considered when normalizing text
roast: 3bceb85476 | (Tom Browder)++ | S26-documentation/12-non-breaking-space.t
expand tests for non-breaking space handling
roast: a5de222edf | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | S26-documentation/12-non-breaking-space.t
Merge pull request #430 from tbrowder/expand-nbsp-tests

expand tests for non-breaking space handling
lizmat hmmm... still the same spectest errors :-( 09:47
Geth nqp: mryan++ created pull request #452:
Fix stringification of Sub when using --target=ast		 10:03
10:04 travis-ci joined
travis-ci Rakudo build failed. Elizabeth Mattijsen 'Merge pull request #1856 from skids/elided_pairs 10:04
travis-ci.org/rakudo/rakudo/builds/382569740 github.com/rakudo/rakudo/compare/d...bb87f554dc
10:04 travis-ci left
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 10:04
Kaiepi AlexDaniel, i'll write them up tomorrow 10:16
it's getting pretty late here
Geth nqp: ba0367278d | (Martin Ryan)++ | src/QAST/Children.nqp
Fix stringification of Sub when using --target=ast
 10:45
nqp: 748eb890cb | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | src/QAST/Children.nqp
Merge pull request #452 from mryan/mryan-ast-debug

Fix stringification of Sub when using --target=ast
nqp: d8ee853cd9 | (Nick Logan)++ | 2 files
[js] Add getppid op

  `getppid` was added in nodejs v9.2.0		 11:19
nqp: 687e2f12c8 | (Paweł Murias)++ (committed using GitHub Web editor) | 2 files
Merge pull request #441 from ugexe/js-getppid

  [js] Add getppid op
tbrowder_ lizmat: what spectest errors? 11:34
lizmat t/spec/S26-documentation/block-leading.t Failed tests: 1, 3, 5, 7-8, 10, 12, 14, 22, 24, 26, 28 11:35
43, 45, 47, 49
tbrowder_ arg! i must admit i didn’t check all pod tests as i usually do. checking now... 11:38
Geth nqp: mryan++ created pull request #453:
Mryan ast debug		 12:03
tbrowder_ lizmat: spec failure is in newline handling: after the nbws changes, some pod lines are retaining newlines unexpectedly...investigating. 12:13
lizmat afk for a bit& 12:15
12:18 ExtraCrispy joined
tbrowder_ lizmat: looks like a fix is eminent. nbsp treatment changed \\s to horizontal spaces thus newlines got left out. fix is something like making a pass with \\v+ and substitute with ‘ ‘. 12:31
12:38 shareable6 joined
|Tux| Rakudo versionRakudo version 2018.04.1-128-g27e8dab9f - MoarVM version 2018.04.1-123-g9b9aa8c58
csv-ip5xs0.848 - 0.857
csv-ip5xs-208.707 - 9.033
csv-parser24.802 - 25.346
csv-test-xs-200.421 - 0.467
test8.788 - 8.850
test-t2.303 - 2.319
test-t --race0.989 - 1.126
test-t-2041.785 - 41.965
test-t-20 --race14.452 - 15.884
 12:55
2018.04.1-128-g27e8dab9f - MoarVM version 2018.04.1-123-g9b9aa8c58
brrt [Tux] - can you do a benchmark on moarvm branch 'jit-stack-walker' 12:56
.ask lizmat can you do a bechmark with moarvm branch 'jit-stack-walker'
yoleaux brrt: I'll pass your message to lizmat.
|Tux| brrt, not right now. Busy getting a replacement rx2800 up and running with HP-UX 11.31 12:58
tbrowder_ hm, going to add specific unicode vert space chars instead of using \\v and see how that works... 13:00
13:59 lucasb joined
lucasb Just to conclude the subject of :nth, I don't see a conflict with :nth being 1-based and using *-1 to mean the last occurrence. That's two different ends of the list of match occurrence, the left and right. It's two different orientations, no need to keep them consistent with each other. The design could be aligned in two ways, either be consistent with the well established convention of using *-1 to 14:08
mean the last element (like in array indexing), or be consistent with the fact that :nth is 1-based: since the left end is incremented by 1, also increment the right end by 1 too, and so *-2 becomes *-1 and *-1 becomes just "*".
Just to conclude the subject of :nth, I don't see a conflict with :nth being 1-based and using *-1 to mean the last occurrence. That's two different ends of the list of match occurrence, the left and right. It's two different orientations, no need to keep them consistent with each other. The design could be aligned in two ways, either be consistent with the well established convention of using *-1 to
mean the last element (like in array indexing), or be consistent with the fact that :nth is 1-based: since the left end is incremented by 1, also increment the right end by 1 too, and so *-2 becomes *-1 and *-1 becomes just "*".
oooh, sorry. f*ck mouse pasting :)
To me, the later is the wrong path. masak and colomon [1] first suggested and implemented :nth(*-1), but it was later changed without discussion. But anyway, this unfortunate detail has been reported, and if it's deemed intended, then fine. 14:09
[1] irclog.perlgeek.de/perl6/2014-11-23#i_9707222
timotimo lucasb: i find it troubling to see *-based indexing as "from the end" tbh 14:14
i've used *-based indexing for percentile sampling, for example
m: my @data = (^100 .roll(1000)); say @data[* div 100, * div 20, * div 2, * div 20 * 19, * div 100 * 99] 14:15
camelia (71 86 50 79 10)
lucasb timotimo: do you mean that in general (including arrays subscripts) or just this case of :nth?
timotimo m: my @data = (^100 .roll(1000)).sort; say @data[* div 100, * div 20, * div 2, * div 20 * 19, * div 100 * 99] 14:16
camelia (0 4 47 95 98)
timotimo in general
lucasb so... you don't like @array[*-1] to mean the last? 14:17
timotimo oh, no i do
lucasb ah :)
timotimo am i being inconsistent? hmm
lucasb is inconsistent all the time
14:22 shareable6 joined 14:35 brrt joined 15:02 skids joined
Geth rakudo: tbrowder++ created pull request #1859:
fix nbsp-fix problem by adding vertical and other missing breaking an…		 15:41
roast: 641c6277f6 | skids++ | S02-types/capture.t
Test duplicate named argument elimination
 15:51
16:04 Kaiepi joined 16:08 shareable6 joined 16:53 robertle joined
tbrowder_ n: my $r := / < [\\x[0009] ] > /; 17:30
nqp: 17:31
nqp: 17:32
m:
evalable6
tbrowder_ m: use nqp; say(“foo”) 17:35
camelia foo
tbrowder_ nqp: say(“foo”)
camelia Confused at line 2, near "say(“foo”)"
at gen/moar/stage2/NQPHLL.nqp:701 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/NQPHLL.moarvm:panic)
from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/nqp.moarvm:comp_unit)
from ge…
tbrowder_ nqp: say(“foo”); 17:36
camelia Confused at line 2, near "say(“foo”)"
at gen/moar/stage2/NQPHLL.nqp:701 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/NQPHLL.moarvm:panic)
from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/nqp.moarvm:comp_unit)
from ge…
tbrowder_ nqp: nqp:say(“foo”);
camelia Confused at line 2, near "nqp:say(“f"
at gen/moar/stage2/NQPHLL.nqp:701 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/NQPHLL.moarvm:panic)
from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/nqp.moarvm:comp_unit)
from gen/…
tbrowder_ nqp: nqp::say(“foo”); 17:37
camelia Confused at line 2, near "nqp::say(“"
at gen/moar/stage2/NQPHLL.nqp:701 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/NQPHLL.moarvm:panic)
from gen/moar/stage2/NQP.nqp:921 (/home/camelia/rakudo-m-inst-1/share/nqp/lib/nqp.moarvm:comp_unit)
from gen/…
tbrowder_ m: my $r = /< :Zs >/; 17:40
camelia ( no output )
lucasb ah, fancy quotes
nqp doesn't support it
jdv79 are grammars faster yet?
tbrowder_ m: my $r=/< :Zs >/; say “ “ ~~ $r 17:42
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in curly double quotes; couldn't find final '”' (corresponding starter was at line 1)
at <tmp>:1
------> 3my $r=/< :Zs >/; say “ “ ~~ $r7⏏5<EOL>
expecti…
tbrowder_ m: my $r=/< :Zs >/; say(“ “ ~~ $r); 17:44
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in curly double quotes; couldn't find final '”' (corresponding starter was at line 1)
at <tmp>:1
------> 3my $r=/< :Zs >/; say(“ “ ~~ $r);7⏏5<EOL>
expec…
tbrowder_ m: my $r=/<:Zs>/; say “ “ ~~ $r; 17:46
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in curly double quotes; couldn't find final '”' (corresponding starter was at line 1)
at <tmp>:1
------> 3my $r=/<:Zs>/; say “ “ ~~ $r;7⏏5<EOL>
expectin…
tbrowder_ m: my $r=/<[ ]>/; say “ “ ~~ $r; 17:49
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in curly double quotes; couldn't find final '”' (corresponding starter was at line 1)
at <tmp>:1
------> 3my $r=/<[ ]>/; say “ “ ~~ $r;7⏏5<EOL>
expectin…
17:53 ggoebel joined 17:57 shareable6 joined 18:30 ggoebel joined 19:00 robertle joined 19:38 shareable6 joined 20:10 Ven`` joined 20:16 Ven`` joined 20:40 Ven`` joined 21:12 Kaiepi joined 21:43 Ven`` joined 22:52 ggoebel joined
Geth rakudo: skids++ created pull request #1861:
Workaround Nil value for &DEPRECATED in phaser from trait (R#1219)		 23:47
synopsebot R#1219 [open]: github.com/rakudo/rakudo/issues/1219 is DEPRECATED doesn't work on Routines in precomped modules

Powered by Cro and the Raku® Programming Language.

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

Thank you!