Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_logs/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by AlexDaniel on 12 June 2018.
00:04 entonian left
Geth rakudo/car-grant-constants: 60447c6267 | (Zoffix Znet)++ | src/core/Map.pm6
[CaR Grant] Make Map.list return a List instead of Seq

Partially fixes R#1344 github.com/rakudo/rakudo/issues/1344
Since (per discussion on the Issue) a Seq is not a "list", the
  .list/.cache methods need to return a Positional. This commit is
required for the grant to have proper coersion+typecheck on `@`-sigilled constants. The rest of the Issue still requires filling in the gaps in positional slurpy semantics
  ( R#1355 github.com/rakudo/rakudo/issues/1355 )		 00:43
synopsebot_ R#1344 [open]: github.com/rakudo/rakudo/issues/1344 [@LARRY] What is a "list"?
synopsebot_ R#1355 [open]: github.com/rakudo/rakudo/issues/1355 +@a/+a slurpies produce inconsistent results
roast/car-grant-constants: 1ccc16134e | (Zoffix Znet)++ | S02-types/hash.t
[CaR Grant] Spec Map/Hash.list/.cache return a List

R#1344 github.com/rakudo/rakudo/issues/1344 Rakudo fix: github.com/rakudo/rakudo/commit/60447c6267		 00:44
lizmat dinner& 00:51
Geth roast/car-grant-constants: 30ce60e070 | (Zoffix Znet)++ | S04-declarations/constant-6.d.t
[CaR Grant] Expand spec of @-sigilled constants

  - Spec throwage on attempts to parametarize
  - Spec coercion		 01:02
01:08 Kaiepi left 01:09 ufobat_ left 01:25 llfourn joined
Geth rakudo: jstuder-gh++ created pull request #1928:
Issue 1891: Add role to Signal for custom zero logic		 01:25
01:45 Kaiepi joined
hoelzro bisectable6: old=2014.09 my $foobar; say "$foobar"; CONTROL { exit 0 if $_ ~~ /foobar/; exit 1} 02:23
bisectable6 hoelzro, Bisecting by output (old=2014.09 new=b34d523) because on both starting points the exit code is 0
hoelzro, bisect log: gist.github.com/56395b053810b96629...f08692c47c
hoelzro, (2015-01-12) github.com/rakudo/rakudo/commit/0a...b59e4e1e8d
hoelzro bisectable6: old=2014.09 my $foobar; say "$foobar";
bisectable6 hoelzro, Bisecting by output (old=2014.09 new=b34d523) because on both starting points the exit code is 0
hoelzro, bisect log: gist.github.com/7f7ac288f3f6c2dc04...ad9c6c6b39 02:24
hoelzro, (2015-04-30) github.com/rakudo/rakudo/commit/64...241a0472bb
Zoffix /o\ 02:29
Any:U.hash returns an empty list
which means `constant %foo = class {}` would silently create `%foo` which is an empty `Map` 02:30
s/list/hash/;
Also, inconsistency: .list returns a List, while .hash returns a Hash. 02:31
(inconsistency being: when we have a choice to return whatever, we're returning an immutable ancestor in one case and mutable subclass in the other) 02:32
02:42 Kaiepi left 02:46 perlpilot left
Zoffix "@rustlang doc comment system is the best I have seen. Makes you look like you a professional or something." twitter.com/jackmott42/status/1008...5037304832 02:52
And looks like it's just a simple special comment where you type markdown. The real awesomeness lies in the HTML generator 02:53
I bet we could do a better thing. We already have supperiour markup, just need to make a better HTML generator for it :)
(and yeah, I said Markdown > Pod6 in the past, but in this case linkage the L<> can do is more valuable, giving Pod6 an edge) 02:54
crap meant to type that in #perl6 to snag a potential volunteer.... 02:55
Oh well :)
03:08 Zoffix left
Geth roast/car-grant-constants: ce969de416 | (Zoffix Znet)++ | S04-declarations/constant-6.d.t
[CaR Grant] Expand spec %-sigiled constants

  - Spec coersion
  - Spec throwage on parametarization		 03:12
03:17 Kaiepi joined 03:35 Kaiepi left 03:36 Kaiepi joined 04:32 ZofBot joined, ChanServ sets mode: +v ZofBot 04:39 ZofBot left, ZofBot joined, ChanServ sets mode: +v ZofBot
Geth roast/car-grant-constants: 49a6c5ebbe | (Zoffix Znet)++ | S04-declarations/constant-6.d.t
[CaR Grant] Fix ups for new constant tests

Formatting/missing `constant` declarators		 04:54
04:54 Zoffix joined
Zoffix Yeah it was on mine. I remember you saying it 04:55
nm
05:45 Kaiepi left 05:49 PufferBot joined
Zoffix What was the nqp op to see if a thing is a native type? 06:36
nqp::objprimspec 06:37
07:23 [TuxCM] joined
[TuxCM] Rakudo version 2018.05-91-gb34d5231e - MoarVM version 2018.05-125-gbbf38a97b
csv-ip5xs0.881 - 0.893
csv-ip5xs-207.150 - 7.214
csv-parser23.596 - 23.726
csv-test-xs-200.442 - 0.460
test8.547 - 8.607
test-t2.302 - 2.388
test-t --race0.972 - 1.026
test-t-2040.094 - 40.642
test-t-20 --race13.699 - 13.890
 07:40
yoleaux 16 Jun 2018 21:08Z <nine> [TuxCM]: Even with Inline::Perl5 0.35?
zef update, so probably, yes 07:41
08:02 [TuxCM] left 08:37 [TuxCM] joined 08:48 [TuxCM] left 08:50 dct joined 09:24 ufobat_ joined 09:40 dct left 09:46 pmurias joined
pmurias jnthn: do you import nqp-j into IntelliJ when working on it? (I'm giving IntelliJ a try) 09:47
12:28 AlexDaniel left
tbrowder_ m: say “\x[a2]” 12:31
camelia ¢
tbrowder_ can anyone please explain what that character is being used as a variable (with a $ sigil) for in src/Perl6/*.nqp? also, 12:33
“that character” is the CENT SIGN 12:34
also, where is it defined? 12:35
thanks
12:48 [TuxCM] joined
timotimo it's short for $/.CURSOR 13:02
but since match and cursor have been merged, i think it's not necessary any more?
jnthn pmurias: No; the tweaks I've been doing of late are pretty tiny, thugh I probably would if I had to write any real amount of code. :) 13:09
13:38 dct joined 13:54 dct left 13:56 AlexDaniel joined 14:10 AlexDaniel left 14:39 lizmat left 14:48 lizmat joined 14:53 lizmat left, lizmat joined
tbrowder_ timotimo: thanks. 15:16
15:18 skids joined 15:33 dct joined 15:41 dct left 15:54 MasterDuke left 16:02 MasterDuke joined 16:13 AlexDaniel joined
AlexDaniel Is there any solution to R#1891 that is better than R#1928 ? 16:22
synopsebot_ R#1891 [open]: github.com/rakudo/rakudo/issues/1891 [regression][⚠ blocker ⚠] Signal(0) gives FreeBSD signals on Linux
R#1928 [open]: github.com/rakudo/rakudo/pull/1928 Issue 1891: Add role to Signal for custom zero logic
geekosaur if it works... 16:28
I could argue that "foreign" signals should be a different enum, which causes its elements to be (subclasses of?) Failure 16:29
Geth rakudo: 5603771fda | (Jeremy Studer)++ | 2 files
Add role to Signal for custom zero logic

We want to be able to take advantage of the powerful functionality of Enumerations, but also implement custom logic in the class. Enter roles.
Add a role to the Signal class (created via enum keyword earlier in the unit) to implement custom handling of zero values (to the type object acting as a sub). jnthn++ for role suggestion.		 16:37
rakudo: bfcfa5da2a | (Jeremy Studer)++ | src/core/signals.pm6
Allow types coercable to Int in Signally role

  AlexDaniel++ for pointing this out
rakudo: 5a6ff4073d | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | 2 files
Merge pull request #1928 from jstuder-gh/signally_role_zero_logic

Issue 1891: Add role to Signal for custom zero logic
16:48 AlexDaniel left, AlexDaniel joined 17:02 MasterDuke left 17:11 MasterDuke joined 17:39 dct joined
Zoffix m: my int constant PICK_CHAIN_BY_NAME = 0; multi foo (int) { say "in" }; foo PICK_CHAIN_BY_NAME 18:06
camelia Cannot resolve caller foo(Int); none of these signatures match:
(int)
in block <unit> at <tmp> line 1
Zoffix Looks like some perf improvement in PseudoStash will occur soon, 'cause it got a bunch of `my int constant` but we currently ignore the type so those ain't native ATM 18:07
m: my $a := 4; my $b := 8; { for ^10000_000 { my $ = $a +| $b; Nil; }; say now - ENTER now } 18:08
camelia 1.2266776
Zoffix m: my int $a = 4; my int $b = 8; { for ^10000_000 { my $ = $a +| $b; Nil; }; say now - ENTER now }
camelia 0.69476671
Zoffix 10M... guess no observable improvement will be seen :) 18:09
Geth rakudo: zoffixznet self-assigned Native-typed constants NYI github.com/rakudo/rakudo/issues/1932
jstuder-gh++ created pull request #1933: Issue 1891 Continued: Add special case for Inf in Signally CALL-ME
 18:11
18:18 dct left
timotimo just made a random SDL2_raw example 2.5x faster by replacing a "state $counter" inside a routine with a "my int $counter" outside of it 18:20
Zoffix cool... but that's no longer `state`? 18:22
timotimo it's equivalent in this case 18:23
because the outside is actually the program's mainline
timotimo another 2x improvement in another script 18:32
Zoffix stares at github.com/perl6/roast/blob/6.c-er...#L149-L150 18:33
It clearly *doesn't* work :P 18:34
m: my Num constant baz = 42;
camelia ( no output )
Zoffix $ ./perl6 -e 'my Num constant baz = 42;'
===SORRY!=== Error while compiling -e
Type check failed in constant declaration of baz; expected Num but got Int (Int)
Geth roast/car-grant-constants: ae2eb4a16e | (Zoffix Znet)++ | S04-declarations/constant.t
Remove traling whitespace
 18:35
roast/car-grant-constants: df4acccf94 | (Zoffix Znet)++ | S04-declarations/constant.t
[CaR Grant] Fix incorrect typed constant test

The test declares a constant of type `Num` but tries to assign a value of type `Int` to it. The only reason this test didn't explode in the past was because type constraints on constants weren't plugged into anything and were entirely ignored.
Fix test by using correct type of the value. Extra 6.d tests in S04-declarations/constant-6.d.t (master branch) also test the typecheck failure when types do not match.		 18:38
Zoffix |7d cherry-pick to 6.c-errata: github.com/perl6/roast/commit/df4acccf94
ZofBot Zoffix, Will remind you on 2018-06-24T14:38:31.544734-04:00 about cherry-pick to 6.c-errata: github.com/perl6/roast/commit/df4acccf94
Geth rakudo/home-precomp-with-fix: 4e782f6b75 | (Rob Hoelz)++ | src/core/CompUnit/RepositoryRegistry.pm6
Prepend inst# to home when setting up repositories

The keys to the $repos hash are in the form of inst#/path/to/repo (among others), but $home was just the path before this change, causing all nqp::atkey($repos, $home) calls to come back as null		 18:39
rakudo: hoelzro++ created pull request #1934:
Prepend inst# to home when setting up repositories
rakudo/car-grant-constants: 0e2a7d93cb | (Zoffix Znet)++ | src/Perl6/World.nqp
Make safely_stringify routine safer

Right now it doesn't check whether the Int/Str/List objects
  are actually instances before trying to unbox them or peek at their
attributes, which causes explosions if the exception being stringified has one of those types objects as a value.
Fix by printing '(Int)'/'(Str)'/'(List)' if the things are type objects and aren't instances.		 18:42
samcv REPL seems broken on jvm 18:44
Zoffix What's the error?
samcv unless more recent changes have happened since a few days ago at least 18:45
no error
i type: "my $foo = 10" press enter and then it does nothing
Zoffix I think I pushed a change like 1 or 2 days ago
Oh, that went into post-rlease branch tho 18:46
AlexDaniel m: multi sub foo(Int() $x) { say ‘hmmm’ }; say foo Inf 18:48
camelia hmmm
True
AlexDaniel m: multi sub foo(Int() $x) { say $x }; say foo Inf
camelia Cannot convert Inf to Int:
in sub foo at <tmp> line 1
in block <unit> at <tmp> line 1
AlexDaniel is that right?
Zoffix m: say Inf.Int.^name 18:49
camelia Failure
geekosaur Inf is a floating point thing, isn't it? out of range for Int
AlexDaniel yes but it goes into that candidate 18:50
Zoffix AlexDaniel: more or less yeah. It's just we currently don't check the coerced value is of the right type, so it fails to detect that it got a Failure and not an Int
Geth nqp: 2ac63b49f2 | (Samantha McVey)++ | tools/build/MOAR_REVISION
Bump MoarVM to get native int lexpad fixes

Changes: 2018.05-124-g4eda279b5..2018.05-131-g9fb02c950 44f17cea3 Fix a comment in MVMContext 1085a51e0 Allow MVMContext atkey to work on all sizes of native ints 9c48a675f Add support for all native int types to getlexref_ni d8b729d8e Add new atkey_u op ccb84367d Add support for all native int's in lexprimspec bbf38a97b Flesh out compare-oplist-interp-order.sh a bit more
nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...g9fb02c950
0d7af14e99 | (Samantha McVey)++ | 12 files

Implement the new atkey_u op so we can properly get these native types from the lexical pad.
Also update stage3 .moarvm files.
nqp: ec508f9548 | (Samantha McVey)++ | docs/ops.markdown
Add documentation on atkey_u newly added op
 18:53
rakudo: 8921cea954 | (Samantha McVey)++ | tools/build/NQP_REVISION
Bump NQP/Moar to get native int lexpad fixes

Changes: 2018.05-83-gc8519b79d..2018.05-86-gec508f954 ec508f954 Add documentation on atkey_u newly added op 0d7af14e9 Fix Unhandled lexical type error w/ native int's and Rakudo REPL 2ac63b49f Bump MoarVM to get native int lexpad fixes		 18:54
¦ rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....gec508f954
samcv so now native int's work properly in repl without giving Unhandled lexical type 'int8' in lexprimspec for '$foo' 18:55
Zoffix samcv++
samcv though for uint's it now gives a new error
No such method 'perl' for invocant of type 'Perl6::Compiler'
doing: `my uint $foo = 10` then after that just typing `$foo` and pressing enter 18:56
Zoffix AlexDaniel: RT#132980
synopsebot_ RT#132980 [open]: rt.perl.org/Ticket/Display.html?id=132980 Coercion type apparently does not check the actual type of the coerced value
samcv not sure where to go about trying to find that issue though 18:58
how can i trace where that exception comes from?
and i think it must be handled in the repl somewhere i guess?
Geth roast/car-grant-constants: 943b0dce43 | (Zoffix Znet)++ | S04-declarations/constant-6.d.t
[CaR Grant] Remove X::ParametricConstant dummy
 19:11
samcv ah looks like things are perfectly fixed but they are much better for int's 19:12
Geth roast/car-grant-constants: 50772172f5 | (Zoffix Znet)++ | S04-declarations/constant-6.d.t
[CaR Grant] Fudge newly failing new tests

Typecheck is now implement, but coercion isn't so tests fail to compile due to typecheck failure
samcv doing 'my int8 $t = -4', '$t', '$t = 4' i get Cannot find method 'qast' on object of type NQPMu
19:14 travis-ci joined
travis-ci NQP build passed. Samantha McVey 'Add documentation on atkey_u newly added op' 19:14
travis-ci.org/perl6/nqp/builds/393373899 github.com/perl6/nqp/compare/0d7af...508f9548ea
19:14 travis-ci left
[Coke] waves from the lobby of the TPC hotel. 19:23
yoleaux 02:40Z <Zoffix> [Coke]: FYI it appears TPF's grant commenting page is losing some comments and it's been happening for possibly ~3m colabti.org/irclogger/irclogger_log...-06-17#l54
samcv waves back 19:25
19:35 dct joined
samcv ok the problem now seems to be it loses the value 19:43
`my uint $foo = 10`, `say $foo` => 0 19:44
(on repl)
Geth roast/car-grant-constants: fd20de7ae2 | (Zoffix Znet)++ | S04-declarations/constant-6.d.t
[CaR Grant] Fix up test class name clashes

  colabti.org/irclogger/irclogger_log...06-17#l633		 19:49
20:01 dct left
Geth roast/car-grant-constants: f403cf56ec | (Zoffix Znet)++ | S04-declarations/constant-6.d.t
[CaR Grant] Remove all fudges / fix plans
 20:18
rakudo/car-grant-constants: b6cd61228b | (Zoffix Znet)++ | src/core/Exception.pm6
[CaR Grant] Implement X::ParametricConstant exception

Thrown when user attempts to parametarize %/@/& constants using `my Bar constant @foo` syntax. As discussed[^1] in the Work Proposal, allowing this syntax creates a lot of varying behaviour such as `constant @foo .= new: 1, 2, 3` creating a `List`, yet `my Int constant @foo .= new: 1,2, 3` would have ... (6 more lines)		 20:22
20:25 ufobat_ left
samcv maybe jnthn would know more. getting Cannot find method 'qast' on object of type NQPMu too 20:30
20:39 pmurias left
Zoffix mf 20:44
Geth rakudo: 7e5a862e54 | (Jeremy Studer)++ | src/core/signals.pm6
Add special case for Inf in Signal CALL-ME
rakudo: cd8d0445ef | (Jeremy Studer)++ | src/core/signals.pm6
Rm Inf candidate and alter Int() candidate logic

The Inf was blowing up due to returning Failure on attempted coercion and then using it in a comparison.
  AlexDaniel++ for the explanation and for suggestion on how to alter the
method. See comments
  [here](github.com/rakudo/rakudo/pull/1933...397898977)
rakudo: 61878c3e07 | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | src/core/signals.pm6
Merge pull request #1933 from jstuder-gh/signally_role_zero_logic

Issue 1891 Continued: Add special case for Inf in Signally CALL-ME
Zoffix there's one 6.c-errata test that blocks the coercive behaviour of Associative constants... sorta
m: use Test; throws-like 'constant %hash = "nothash"', X::TypeCheck
camelia 1..2
ok 1 - 'constant %hash = "nothash"' died
ok 2 - right exception type (X::TypeCheck)
ok 1 - did we throws-like X::TypeCheck?
Zoffix $ ./perl6 -e 'use Test; throws-like ｢constant %hash = "nothash"｣, X::TypeCheck'
# Got: X::Hash::Store::OddNumber
m: try EVAL '"notahash".Map'; say $!.^name 20:46
camelia X::Hash::Store::OddNumber
Zoffix AlexDaniel: what do you think? Can this be changed? The changed behaviour in that particular test is just the type of the exception, but the larger implication of the behaviour change is that `constant %foo = :42foo, :100bar` used to throw but now it works 20:47
And the sucky implication is this: 20:48
m: class Foo {}; constant %foo = Foo
camelia 5===SORRY!5=== Error while compiling <tmp>
Type check failed in constant declaration of %foo; expected Associative but got Foo (Foo)
at <tmp>:1
------> 3class Foo {}; constant %foo = Foo7⏏5<EOL>
Zoffix Now that silently succeeds and %foo is an empty Map, because Any:U.Map returns that 20:49
m: class Foo { method Map { Map.new: (:42foo, :70bar) } }; constant %foo = Foo 20:50
camelia 5===SORRY!5=== Error while compiling <tmp>
Type check failed in constant declaration of %foo; expected Associative but got Foo (Foo)
at <tmp>:1
------> 3:42foo, :70bar) } }; constant %foo = Foo7⏏5<EOL>
Zoffix This works now tho
And another pro-change argument is that before CaR grant, the @-sigilled constants actually always called .cache, so if we're blocking the change, then we have this inconsistency that @-sigils coerce but %-sigils don't 20:51
And another pro-change argument is that Constants Work Proposal and CaR Grant both discussed the coercive behaviour and in ~4 months no one objected. 20:53
20:54 robertle left
AlexDaniel Zoffix: well, with v6.d in mind, is there any possibility to have both behaviors depending on which language version is used? 20:56
Zoffix should be 20:57
Obligatory reminder about R#1289 21:00
synopsebot_ R#1289 [open]: github.com/rakudo/rakudo/issues/1289 [6.d][severe] [6.d BLOCKER] Implement a Way to Know Caller's Language
AlexDaniel Zoffix: I mean, it makes sense, but we should really try not to make changes like this, I think 21:05
previous behavior wasn't blatantly wrong, and there's a spectest… 21:06
so for me it's a bit hard to justify the change 21:07
withing v6.c I mean, for v6.d go crazy with it :)
within*
Zoffix mhm 21:08
21:10 perlpilot joined
Geth rakudo/car-grant-constants: 4dcf5929d3 | (Zoffix Znet)++ | 2 files
Implement $*W.lang-ver-before method

Tells you whether the current language version is before the given version letter. For use where language-specific behaviour needs to be implemented.
   $*W.lang-ver-before('d')
   ?? nqp::say('Bruh, your lang is too old')
   !! nqp::say('6.d or above. You get a new feature!')		 21:53
Zoffix AlexDaniel: what about generic stuff like this: 21:58
m: my Int(Any) constant z = "42"
camelia ( no output )
Zoffix There ain't no spectests covering it, but now it'll be fatal 21:59
m: my Int(Any) $z = "42"
camelia 5===SORRY!5=== Error while compiling <tmp>
Coercion Int(Any) is insufficiently type-like to qualify a variable
at <tmp>:1
------> 3my Int(Any) $z7⏏5 = "42"
expecting any of:
constraint
Zoffix Like that ^
AlexDaniel: well, actually since it's a constant it'll probably be possible to just coerce the stuff, but still, in the past all of this type stuff was entirely ignored 22:00
m: my Int(Date) constant z = "42"; dd z
camelia "42"
Geth rakudo/car-grant-constants: a1f544ad42 | (Zoffix Znet)++ | src/Perl6/Actions.nqp
[CaR Grant] Implement basic typecheck and coercers on constants

  - Implement typechecking on &- and $- sigilled, sigilles, and
   backslash-sigilles constants
  - Throw X::ParametricConstant on attempts to specify parametarized
   @-, %-, and &- constants (using `my Int constant @foo` and
   similar syntax only; `constant @foo = Array[Int].new` is still
... (8 more lines)		 22:01
roast/car-grant-constants: 86e2e20e07 | (Zoffix Znet)++ | S04-declarations/constant-6.d.t
[CaR Grant] 6.d-fy relevant parts

Coercers on %-sigilled constants are actually blocked by one spectest and so need to go into 6.d language.		 22:02
AlexDaniel Zoffix: yeah :( not enough spectests 22:03
it depends 22:04
I guess we can try changing it, and see what pops out of toaster
Zoffix ok
Geth rakudo/post-release-2018.06: 10 commits pushed by (Aleks-Daniel Jakimenko-Aleksejev)++, (Jeremy Studer)++, (Samantha McVey)++, (Zoffix Znet)++
review: github.com/rakudo/rakudo/compare/2...71cd9e1a41		 22:05
rakudo/car-grant-constants: 17 commits pushed by (Jeremy Studer)++, (Zoffix Znet)++, (Aleks-Daniel Jakimenko-Aleksejev)++, (Elizabeth Mattijsen)++, (Samantha McVey)++
review: github.com/rakudo/rakudo/compare/a...bf22e9c02e		 22:06
22:06 Kaiepi joined
Geth roast/car-grant-constants: 5 commits pushed by (JJ Merelo)++, (Zoffix Znet)++, (Tom Browder)++, (Jeremy Studer)++ 22:06
22:14 Kaiepi left
Zoffix ZOFVM: Files=1295, Tests=153418, 155 wallclock secs (21.78 usr 3.46 sys + 3285.70 cusr 178.72 csys = 3489.66 CPU) 22:14
22:15 Kaiepi joined
Geth rakudo: zoffixznet++ created pull request #1935:
Implement typecheck and auto-coercers on constants		 22:15
rakudo/post-release-2018.06: 9 commits pushed by (Zoffix Znet)++
roast: zoffixznet++ created pull request #437:
Spec typecheck and auto-coercive behaviour for constants
roast/post-release-2018.06: 20 commits pushed by (Zoffix Znet)++
review: github.com/perl6/roast/compare/9dc...12b2b6ecaa
22:27 perlpilot left
[Coke] Zoffix: are you saying comments are not making it through, or comments are showing up and then disappearing. 22:49
Zoffix [Coke]: not making it through, it seems 22:50
[Coke] I don't see anything in the moderation queue.
Most of the spam-moderated stuff is chinese text selling nikes. 22:51
very rarely a human comment will get stuck there. 22:52
Zoffix ¯\_(ツ)_/¯
[Coke] I don't know if there's a "so spammy it's deleted" variant of moderation; can't imagine that's true if I still hae to blacklist 50 comments a month selling sneakers.
so, I believe you! but I can't see it. 22:53
Zoffix ok :)
Kaiepi would being able to get the file descriptor for IO::Socket::Async connections be desirable? 23:00
also, trying to get it for IO::Socket::INET connections on my system gives me -1 when it probably shouldn't be 23:02
Zoffix [Coke]: ok, try number 2 :) "Your comment has been received and held for approval by the blog owner." 23:13
[Coke]: the page you see after submitting a comment looks messed up BTW: i.imgur.com/hnbCMiT.png 23:14
[Coke] Yes. that is a known bug that will eventually be fixed by getting rid of mt
Sorry. :(
ok, checking to see the queue. 23:15
Zoffix k
[Coke] it was there, should be published. (takes a few minutes to propogate)
Zoffix cooll
/j #perl6 23:19
23:29 lizmat left
MasterDuke [Coke]: fyi, i also just submitted a comment 23:34
23:46 lizmat joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!