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 13 May 2018.
01:01 AlexDani` joined 01:13 cog_ joined
Geth roast/car-grant-midrat: 08973cf72e | (Zoffix Znet)++ | S32-num/midrat.t
[CaR Grant] Test MidRat->Num degradation
 01:26
roast/car-grant-midrat: 96717e3677 | (Zoffix Znet)++ | S32-num/midrat.t
[CaR Grant] Clarify magic numbers used
 01:29
01:34 |Tux| joined 01:57 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
MasterDuke has anybody seen this? Unicode's recommendations for regex engines: unicode.org/reports/tr18/ 02:33
samcv: ^^^ 02:34
AlexDani` reportable6: 2018-05-07T00:00:00Z 2018-05-14T00:00:00Z 03:22
reportable6 AlexDani`, OK, working on it! This may take up to 40 seconds
AlexDani`, gist.github.com/61ec075bdfab44d1b1...2c150f4d94 03:23
AlexDaniel weekly: reportable: gist.github.com/61ec075bdfab44d1b1...2c150f4d94 03:39
notable6 AlexDaniel, Noted!
AlexDaniel perl6.org and other stuff is down 03:53
Zoffix moritz timotimo
don't know who I should ping actually
04:31 Tison joined 04:32 cog_ joined
Tison help wanted: github.com/perl6/nqp/issues/443 04:42
besides, why do we treat intentionally defined subrules as 'non-declarative'? It looks hard to me even if i redefine token ws as { <!ww> \\s* } it still be a non-declarative token because of <!ww> 04:44
05:05 ufobat joined
Tison m: my $A="44.rc0"; if $A ~~ m/( \\d**1..4 "." <["a" . "b" . "rc" ]> +$ ) / {say "$0";} else {say "no match";} 05:17
camelia Potential difficulties:
Repeated character (.) unexpectedly found in character class
at <tmp>:1
------> 3$A="44.rc0"; if $A ~~ m/( \\d**1..4 "." <7⏏5["a" . "b" . "rc" ]> +$ ) / {say "$0";}
Quotes are not metacharacters in…
Tison m: my $A="44.rc0"; if $A ~~ m/( \\d**1..4 "." <["a" "b" "rc" ]> +$ ) / {say "$0";} else {say "no match";} 05:18
camelia Potential difficulties:
Quotes are not metacharacters in character classes
at <tmp>:1
------> 3$A="44.rc0"; if $A ~~ m/( \\d**1..4 "." <7⏏5["a" "b" "rc" ]> +$ ) / {say "$0";} els
Repeated character (") unexpectedly found …
Tison m: my $A="44.rc0"; if $A ~~ m/( \\d**1..4 "." <[abrc]> +$ ) / {say "$0";} else {say "no match";}
camelia no match
Tison oh sorry
I think here is camelia
06:50 Ven`` joined 07:18 robertle joined 07:52 cog_ joined
jnthn .tell Tison It's precisely the <!ww> that makes the default `ws` non-declarative, and that's because declarative things are restricted to those things that are regular (in the sense of "regular language"). 09:04
yoleaux jnthn: I'll pass your message to Tison.
09:05 cog_ joined 09:15 Tison joined
Tison . 09:16
yoleaux 09:04Z <jnthn> Tison: It's precisely the <!ww> that makes the default `ws` non-declarative, and that's because declarative things are restricted to those things that are regular (in the sense of "regular language").
Tison ok 09:21
09:54 Ven`` joined 10:05 AlexDaniel joined 10:16 cog_ joined
|Tux| Rakudo version 2018.04.1-76-g9b915f09d - MoarVM version 2018.04.1-101-gd5830c853
csv-ip5xs0.846 - 0.857
csv-ip5xs-208.679 - 8.716
csv-parser35.829 - 36.264
csv-test-xs-200.453 - 0.458
test8.689 - 8.752
test-t2.350 - 2.353
test-t --race0.966 - 0.970
test-t-2041.849 - 41.959
test-t-20 --race14.489 - 14.517
 10:37
10:58 Zoffix joined
Zoffix AlexDaniel: it's back up now, right? 10:58
AlexDaniel: as for whom to ping, admins of www box: github.com/perl6/infrastructure-do...nistration
AlexDaniel Zoffix: yeah 11:35
huggable: infrastructure 11:36
huggable AlexDaniel, nothing found
AlexDaniel huggable: infrastructure :is: github.com/perl6/infrastructure-doc
huggable AlexDaniel, Added infrastructure as github.com/perl6/infrastructure-doc
AlexDaniel undersightable6: status 11:43
undersightable6 AlexDaniel, OK! Working on it…
AlexDaniel, 0 errors, 5 warnings: gist.github.com/5e08dabce41592f4fe...e1a787dd20 11:45
AlexDaniel Zoffix: rakudo.org is still down ↑
11:56 Ven`` joined 12:03 Ven`` joined
AlexDaniel Zoffix: nevermind, moritz++ 12:37
Geth rakudo: tbrowder++ created pull request #1826:
Fix for GH #1821: unexpected table failure		 12:45
synopsebot RAKUDO#1821 [open]: github.com/rakudo/rakudo/issues/1821 Can't get Pod table formatted right
lizmat . 12:48
Geth roast: tbrowder++ created pull request #427:
Add tests for fix for GH #1821: unexpected table failure
13:03 Ven`` joined
Geth nqp: W4anD0eR96 assigned to pmurias Issue If `token ws` defined by user, we should treat it as declarative github.com/perl6/nqp/issues/443
W4anD0eR96 assigned to ugexe Issue If `token ws` defined by user, we should treat it as declarative github.com/perl6/nqp/issues/443

Bump Moar to get the latest JIT improvements		 13:31
13:31 lucasb joined 13:37 skids joined 13:40 Zoffix left
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...6-g37bbb65
rakudo: fb86392c92 | (Elizabeth Mattijsen)++ | tools/build/NQP_REVISION
Bump NQP to get the latest JIT improvements
 14:06
¦ rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....9-gc1ecee8
14:12 Ven`` joined 14:21 cog_ joined 14:38 Ven`` joined
AlexDaniel heh, “Remove 4to6 error messages” – amazing title for an upcoming commit 16:16
yo 16:33
'$|' => ':autoflush on open',
no such thing, do we have an alternative? 16:34
[Coke] No global alternative available. TTY handles are unbuffered by default, for others, set out-buffer to zero or use :!out-buffer with open on a specific IO::Handle. 16:37
from docs.perl6.org/language/5to6-perlvar
timotimo 4to6? we have warnings for perl4 users, too? :) 16:38
AlexDaniel timotimo: yes 16:39
m: say $#
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of $# variable; in Perl 6 please use .fmt
at <tmp>:1
------> 3say $#7⏏5<EOL>
AlexDaniel “Deprecated in Perl 5. Removed in Perl v5.10.0.”
timotimo nice. 16:42
Geth rakudo/master: 4 commits pushed by (Aleks-Daniel Jakimenko-Aleksejev)++ 17:58
roast: a672d539db | (Aleks-Daniel Jakimenko-Aleksejev)++ | S32-exceptions/misc2.t
Remove some tests for 4to6 and 5to6 error messages

Removing tests to match current rakudo behavior.
Part of rakudo/rakudo#1356.
synopsebot RAKUDO#1356 [open]: github.com/rakudo/rakudo/issues/1356 [RFC] Let's remove most (if not all) perl5-oriented error messages for variables
18:40 cog_ joined
bartolin I'm looking at some I/O tests that fail on the jvm backend. I've found this comment that states that "[i]t's allowed for implementations to pass the first print unbuffed, to test if the handle is writable": github.com/perl6/roast/blob/a672d5...ring.t#L11 18:43
yoleaux 5 May 2018 14:48Z <Zoffix> bartolin: RE strings not recognized as numbers/unicode minus... in nqp::numify. If those were recognized before the numify change, then the code to make that stuff work should be nqp::radix. Otherwise, I think only the Unicode minus might need to be added to JVM's numify to make any new failures pass.
bartolin but if I'm not mistaken, we *require* the first write to be passed unbuffed with some other tests, e.g. github.com/perl6/roast/blob/a672d5...open.t#L67 18:44
18:44 Zoffix joined
Zoffix bartolin: yeah, they need a flush in the test routine. Don't think there's reason to *require* first write to be unbuffered 18:46
flush before this line, I mean: github.com/perl6/roast/blob/a672d5...open.t#L46 18:47
bartolin Zoffix: I see. thanks for looking. I'll adjust 'spurt-slurp' accordingly (after testing). 18:48
Zoffix bartolin++ 18:49
releasable6 Next release in ≈4 days and ≈23 hours. 2 blockers. Please log your changes in the ChangeLog 19:00
Geth roast: 0d7781fd46 | usev6++ | S32-io/open.t
Write buffered data to file before reading

On MoarVM we don't buffer the first write. The additional call to 'flush' makes the test routine independent of that implementation detail. Cmp. irclog.perlgeek.de/perl6-dev/2018-...i_16164630		 19:06
19:07 b2gills joined 20:15 committable6 joined, squashable6 joined
lizmat notable6: weekly 20:59
notable6 lizmat, 3 notes: gist.github.com/39a5bf5d77eaeb4789...72b717dde8
lizmat notable6: weekly reset 21:26
notable6 lizmat, Moved existing notes to “weekly_2018-05-14T21:26:58Z”
Geth nqp: 26809a4115 | (Elizabeth Mattijsen)++ | tools/build/MOAR_REVISION
Bump MoarVM to get the latest fixes.
 21:27
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...8-geebc462
21:36 cognominal joined 22:18 dct joined
lizmat And another Perl 6 Weekly hits The Net: p6weekly.wordpress.com/2018/05/14/...ough-time/ 22:23
22:32 greppable6 joined
Geth rakudo: 20e2f1d205 | (Elizabeth Mattijsen)++ | tools/build/NQP_REVISION
Bump NQP to get the latest MoarVM improvements
 22:35
¦ rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....0-g26809a4
nqp/W4anD0eR96-patch-1: cfaa4ef0c7 | 陈梓立++ (committed using GitHub Web editor) | src/QRegex/NFA.nqp
User-defined `ws` should be declarative
 22:36
nqp: W4anD0eR96++ created pull request #444:
User-defined `ws` should be declarative		 22:38
¦ rakudo: zoffixznet self-assigned `$`-sigiled constants don't get constant folded github.com/rakudo/rakudo/issues/1829 22:48
23:02 skids joined
Geth roast: 771a2bbeb1 | thundergnat++ (committed using GitHub Web editor) | S32-num/fatrat.t
Fix bogus TODO test. RT #124815

Was checking if Str type was a Str instance instead of if a FatRat type was a FatRat. The Str type just happened to be a stringified gist of a FatRat type and the Str instance was '(FatRat)', but it was still wrong because an instance isn't a type.		 23:08
synopsebot RT#124815 [new]: rt.perl.org/Ticket/Display.html?id=124815 Roast rakudo skip/todo test:./S32-num/fatrat.t line:191 reason: 'unknown'
23:09 entonian joined 23:15 entonian joined 23:16 entonian joined
Geth roast: 14f5138f07 | thundergnat++ (committed using GitHub Web editor) | S32-num/fatrat.t
Fix bogus FatRat test #124814

Bogus test condition. Was trying to assign a Rat to a FatRat constrained container. Of course it failed, just like it does if you try to assign a non-Int to an Int constrained container or a non-Rat to a Rat constrained container.		 23:16
23:37 entonian joined
Geth roast: 6a99fa7406 | (Zoffix Znet)++ (committed using GitHub Web editor) | S32-num/fatrat.t
Remove useless comment
 23:45

Powered by Cro and the Raku® Programming Language.

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

Thank you!