buggable And the winning number is 7! Congratulations to alphah! You win a can of WD40! 00:00
Zoffix hahaha
buggable++
Wonder if WD40 is even a thing outside of North America :) 00:05
ugexe WD40 doesnt work on metric 00:06
geekosaur *eyeroll* 00:16
Zoffix ZofBot: You've rolled a six! 00:17
ZofBot Zoffix, ) You also can't make assignment metaoperators from diffy operators
Zoffix Since I object to smoking, I'm to name my ecosystem smoker "toaster" and will coin the term "toasting the ecosystem" :) 00:32
ugexe vapes the ecosystem 00:34
BenGoldberg If you forget to clean out the crumbs, your toaster may smoke. 00:58
geekosaur surely it should be 'roaster' 01:00
Zoffix I have a speech impediment. "roaster" is hard to say :P 01:02
BenGoldberg How does roasting differ from baking? 01:04
BenGoldberg suddenly thinks of simulated annealing, which is yet another erm, "hot" idea. 01:05
Zoffix hm, somehow I thought Proc has .pid :S 01:32
guess it's not really designed for async use with promises
Unhandled exception in code scheduled on thread 7 01:55
no such file or directory
greeaaaat
No line number or anything 01:56
Also notice run takes *@args ($, *@) but Proc.new takes $path, *@args, so if you change your run <zef --serial --debug install> to Proc::Async.new: <zef --serial --debug install> like I just did, it'll stop working -_- 01:57
ZofBot: complain all the things 01:58
ZofBot Zoffix, Getting information back from the external process's "STDOUT" or "STDERR", is done by a "Supply" that either gets characters or bytes
Zoffix This is Karma. After years of Perl folks making fun of PHP didn't go unpunished. 01:59
tnx.nl/php.html#names 02:00
"not enough memory" hm. I was starting 2 promises + 1 proc async for all the modules in the ecossystem on 24GB box 02:08
buggable: eco
buggable Zoffix, Out of 834 Ecosystem dists, 0 have warnings, 1 have errors, and 0 have no tags in META file. See modules.perl6.org/update.log for details
Zoffix m: say 834*3
camelia 2502
Zoffix tries throwing moar RAM at it
455GB... It better run now 02:10
MasterDuke how much does that cost per hour? 02:11
ugexe if you are installing to the same location for each module you might still be spending a bunch of time stuck waiting for locks 02:12
Zoffix boooo "Starting VM instance "toaster" failed. Error: The zone 'projects/perl6-build/zones/us-central1-b' does not have enough resources available to fulfill the request. Try a different zone, or try again later." 02:14
ugexe its possible it could be faster if you installed each one to a unique --install-to, and then programatically update PERL6LIB
Zoffix MasterDuke: $1.109
Well, speed isn't my issue ATM. Running it at all is :/
MasterDuke: with 24CPUs
MasterDuke that could start to add up if they ever let you run it 02:15
Zoffix Yeah. I don't plan on keeping that :D Just wanted to see how much RAM this program would take 02:16
(famous last words before "we forgot to turn off our VM for a month" stories) 02:17
Got it to boot with a 156GB RAM VM. Program is 70GB and climbing. 02:20
m: say 834*3*142
camelia 355284
Zoffix Guess it needs 355GB. There's like a billion of moar processes each using ~142MB
ugexe github.com/ugexe/zef/blob/55fb01b1...README.pod (grep $ zef -v --jobs=2 install CSV::Parser) zef used to do concurrency too, in 2015
===> Module count: 364 02:21
Zoffix :)
Well, it created all the procs and all the promises and started testing stuff, but RAM usage is still climbing 120GB now 02:22
aha, leved out at 124GB \o/ 02:23
ah, damn... that's cause the 5-minute job timeout I set kicked in and killed all the things :( 02:24
m: so try await start 42 02:32
camelia WARNINGS for <tmp>:
Useless use of "so " in expression "so try await start 42" in sink context (line 1)
Zoffix that's not useless yo
32CPU/210GB box. Much springier. With several-hour proc timeout. Hoping it'll complete :)O 02:35
geekosaur: toaster also starts with "t", just like "testing" BTW :) 02:36
yeah, seems to platau at ~124GB and 1240 "threads" (whatever htop calls threads) 02:38
144GB/1071 threads 02:45
didn't really move in the output of tested modules. Still finished with only 9
grr, I think it's actually my buggy terminal that "froze" 02:49
oh it's asking for github pass... really :\
Well, this was a good 20 minutes of fun. 02:50
Zoffix deems the experiment a failure and goes to bed
Wonder how to go around the gitub pass thing. Is it asking for it just to pull from git://? I don't wanna give (or cache) any passwords on this box, which running 820+ pieces of software from complete strangers. 02:52
Geth roast: danielcliffordmiller++ created pull request #273:
Added test for RT 131381		 03:39
AlexDaniel \o/ 05:05
bisectable can now bisect from 2014.01 05:06
BenGoldberg bisectable6, printf "%.30f", pi 05:21
bisectable6 BenGoldberg, On both starting points (old=2015.12 new=8ec1810) the exit code is 0 and the output is identical as well
BenGoldberg, Output on both points: «3.141592653589790000000000000000»
travis-ci Rakudo build errored. Samantha McVey 'REPL: Don't show message to install line editor if using rlwrap 05:41
travis-ci.org/rakudo/rakudo/builds/238141912 github.com/rakudo/rakudo/compare/c...f3ea3a4a80
buggable [travis build above] ✓ All failures are due to timeout (0), missing build log (1), GitHub connectivity (0), or failed make test (0).
[Tux] This is Rakudo version 2017.05-320-g8ec181019 built on MoarVM version 2017.05-25-g62bc54e9 06:11
csv-ip5xs 2.914
test 13.457
test-t 4.534 - 4.560
csv-parser 13.670
as there is a perl5 installation running in the background (update to 5.26.0), that /might/ have small influence on timing 06:12
nine [Tux]: I do hope so. csv-ip5xs is moving into the wrong direction :/ 06:45
AlexDaniel bisectable6, old=2014.01 printf "%.30f", pi 07:35
bisectable6 AlexDaniel, Bisecting by output (old=2014.01 new=8ec1810) because on both starting points the exit code is 0
AlexDaniel, bisect log: gist.github.com/2c001dd45d6331cdd7...32c83dfd3c
AlexDaniel, There are 2 candidates for the first “new” revision. See the log for more details
AlexDaniel .tell BenGoldberg oh yeah, but it doesn't mean that it will default to 2014.01, because in most cases this is not what people are looking for. irclog.perlgeek.de/perl6-dev/2017-...i_14669729 07:36
yoleaux AlexDaniel: I'll pass your message to BenGoldberg.
Geth roast: eb8a73eea2 | (Elizabeth Mattijsen)++ | S03-operators/union.t
Initial set of thorough testing of (|) and friends
 08:12
rakudo/nom: a2add65e2e | (Elizabeth Mattijsen)++ | t/spectest.data
Make sure we run the new (|) tests
 08:18
roast: 9c6b1536db | (Elizabeth Mattijsen)++ | S03-operators/set.t
Remove union tests that are tested now in union.t
 08:21
pmurias what is translate_newlines when configuring the decoder supposed to do? 09:42
travis-ci Rakudo build canceled. Samantha McVey 'REPL: Don't show message to install line editor if using rlwrap
travis-ci.org/rakudo/rakudo/builds/238141912 github.com/rakudo/rakudo/compare/c...f3ea3a4a80
buggable [travis build above] ☠ Did not recognize some failures. Check results manually.
Geth nqp: 42fa5d7c9c | pmurias++ | 2 files
[js] Remove no longer used ops
 09:43
nqp: ec0e003589 | pmurias++ | 3 files
[js] Implement decode ops (NFG for them needs work)
pmurias the \r\n translation seems to be happening with it turned off :/ 09:45
sorry about the translate_newlines, I had a typo in the test 09:55
Geth nqp: 9ea137392e | pmurias++ | 2 files
[js] Support the translate_newlines option when configuring decoders
 10:03
nqp: 05e7227ada | pmurias++ | t/moar/05-decoder.t
Test decoders with the translate_newlines option
nqp: 4ab0db95dc | pmurias++ | src/vm/js/bin/run_tests.pl
[js] Reenable test 19
Zoffix Ah. Now I see why perl6.fail stopped updating. I upgraded perl5 but didn't install any of the modules 10:08
and damn cpanm is lightning fast compared to zef :( 10:09
timotimo we'll get there eventually ... 10:10
Geth roast: d4aea7748a | danielcliffordmiller++ (committed using GitHub Web editor) | S12-subset/subtypes.t
Added test for RT 131381
 10:13
roast: 5f5a9888b0 | danielcliffordmiller++ (committed using GitHub Web editor) | S12-subset/subtypes.t
Updated PR with requested changes
roast: a19c632aa6 | (Zoffix Znet)++ (committed using GitHub Web editor) | S12-subset/subtypes.t
Merge pull request #273 from danielcliffordmiller/master

Added test for RT 131381
roast: dcfb2db6ac | (Zoffix Znet)++ (committed using GitHub Web editor) | S12-subset/subtypes.t
Remove space in fudge #273
 10:15
Zoffix NeuralAnomaly: status 10:18
NeuralAnomaly Zoffix, [✘] Next release will be in 2 weeks and 2 days. Since last release, there are 35 new still-open tickets (32 unreviewed and 0 blockers) and 319 unreviewed commits. See perl6.fail/release/stats for details
Zoffix perl6.fail back in business
Geth rakudo/nom: 11fa6d944b | (Elizabeth Mattijsen)++ | src/core/set_operators.pm
Fix problem doing hash (|) object hash

Spotted while writing tests.		 10:20
roast: b5f9973e48 | (Elizabeth Mattijsen)++ | S03-operators/union.t
Add some more (|) test + some regrouping for clarity
 10:26
roast: c85b8f30de | (Elizabeth Mattijsen)++ | S03-operators/union.t
Add tests for list of pairs (|) mixed list
 10:36
lizmat afk again&
Zoffix m: say 50/0 cmp -100/0 10:46
camelia Same
Zoffix m: say sort 50/0, 100/0, -50/0 10:47
camelia Attempt to divide 50 by zero using div
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
Zoffix I thought this was fixed :|
c: all say sort 50/0, 100/0, -50/0 10:48
committable6 Zoffix, gist.github.com/ecb542ea294ed5ae74...c34e1cd819
Zoffix ah, duh
m: say map *.Num, sort 50/0, 100/0, -50/0 10:49
camelia (Inf Inf -Inf)
Zoffix
.oO( isn't that backwards? )
 10:50
m: say map *.Num, sort 50, 100, -50
camelia (-50 50 100)
Zoffix ah, prolly due to cmp issue
Ah hah. I see the problem :) 10:57
Will fix in a few hours.
actually, fix it some time next week. I wanna finish the toaster before the weekend, which I'll be taking off 11:36
|Tux| perl6 -e 'sub MAIN (List :$dirs=[]) { .say for @$dirs».split: /","/}' --dirs=d1 --dirs=d2 --dirs=d3,d4,d5 12:42
I expected that to give a single list, not (d1) (d2) (d3 d4 d5)
jnthn No, that won't flatten automatically (relatively little does) 12:47
.say for flat ...
Zoffix .say for flat $dirs.List».split: /","/ 12:52
.say for $dirs.join(",").split: /","/
m: (1, $((2, 3).Seq)).flat.perl.say; (1, $(2, 3)).flat.perl.say; 12:56
camelia (1, (2, 3).Seq).Seq
(1, $(2, 3)).Seq
Zoffix Looks like .perl of a Seq is forgetting to specially mark containerized seqs
s: (), 'perl', \()
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/11fa...st.pm#L810
Zoffix s: ().Seq, 'perl', \()
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/11fa...eq.pm#L157
|Tux| I played with flat too, but could not find something that "worked" 12:58
so it needs both .List *and* flat 13:00
Zoffix |Tux|: that's cause the items come from an Array and are containerized, so flat won't flatten them. The two examples I mentioned should work
|Tux| perl6 -e 'sub MAIN (List :$dirs=[]) { .say for flat @$dirs.List».split: /","/}' --dirs=d1 --dirs=d2 --dirs=d3,d4,d5
Zoffix Yeah
Or map a slip: .say for $dirs.map: |*.split: /","/ 13:01
Is there a point in (*@args ($, *@)) sig, rather than just (*@args) ? 13:07
like, literally that; unpacked to anons
Ah, yes: spec you want at least one item 13:09
m: sub foo (*@args ($, *@)) { @args }; sub bar(*@args) { @args }; dd [ foo($_), bar($_) ] with ()
camelia Too few positionals passed to 'foo'; expected at least 1 argument but got only 0 in sub-signature of parameter @args
in sub foo at <tmp> line 1
in block <unit> at <tmp> line 1
Zoffix tho where .so is fastar 13:14
4.61x faster 13:16
ZOFFLOP: t/spec/S10-packages/precompilation.tr 13:18
masak computer, any messages? 13:27
yoleaux 28 May 2017 22:26Z <MasterDuke> masak: re rt.perl.org/rt3/Public/Bug/Display...id=123572, what is the expected behavior of the :3[4, 22, 1] radix form? should digits greater than the base be allowed (e.g., the 4 in my example)? should values in the list greater than the base be allowed?
masak MasterDuke: the way I read the bug report is that exactly that should be disallowed: numbers greater than (or equal to) the base 13:28
Zoffix ZOFFLOP: t/spec/S11-modules/nested.t 13:33
Geth rakudo/nom: b22383fe62 | (Zoffix Znet)++ | src/core/Seq.pm
Fix Seq.Perl for containerized Seqs

If a Seq is containerized, it won't `flat`ten, so we need to preserve this property when roundtripping through .perl.		 13:39
roast: 3e60b77d8e | (Zoffix Znet)++ | S32-list/seq.t
Test Seq.perl.EVAL roundtrips containerized seqs

Rakudo fix: github.com/rakudo/rakudo/commit/b22383fe62		 13:40
rakudo/nom: 93524fb953 | (Zoffix Znet)++ | src/core/Proc.pm
Make invocation of Proc.spawn and &run 4.6x faster

The speedup is just for the routine calls themselves, achieved by changing the way to specify "at least one arg in @args" from unpacked double-slurpy to a `where` clause		 13:42
rakudo/nom: 92c187d2dc | (Zoffix Znet)++ | src/core/Proc/Async.pm
Fix Proc::Async.new not slurping first positional

  - Slurp all the args, including the first command
  - Makes it easy to pass stuff like `Proc::Async.new: <git log --oneline>`
   without having to extract `git` out of the rest of the list
  - Aligns interface with what we already do for &run and Proc.spawn		 13:58
roast: 4937efbc02 | (Zoffix Znet)++ | S17-procasync/basic.t
Test Proc::Async.new slurps all positionals
Zoffix ZofBot: it's crutical! Both crucial and critical! 14:00
ZofBot Zoffix, You can't put a combiner onto a control char
Zoffix leaves it in just for fun
Geth roast: 9c793637c5 | (Zoffix Znet)++ | S17-procasync/basic.t
Fix typo
 14:01
pmurias Zoffix: what's the bottle neck with zef? 14:05
Zoffix pmurias: no idea 14:06
It's just a million times slower. I couldn't even spot which modules were being installed by cpanm, it was installing them so fast 14:07
timotimo i wonder what a "perf record -g" would give us 14:11
like, is there an obvious hot spot for the whole process? 14:13
probably just a complete across-the-board slowness of rakudo + moarvm :\
ugexe well for installation you have to wait for precompile 14:19
timotimo p5 doesn't do that at all, right?
ugexe most everything else spawns a process to do work (wget, curl, whatever)
and probably the json/hash searching method
timotimo do you get faster zef execution when something like our UserAgent module is installed?
ugexe right p5 doesn't precompile things 14:20
I'm not sure. Net::HTTP used to be built into zef and I don't recall it being significantly faster
timotimo dang. 14:21
ugexe i think the fastest win would be not using json for the ecosystem listing 14:29
like something with some alphabetical ordering or something that could be stream parsed and short-circuit early 14:30
timotimo interesting, i was wondering if i should give JSON::Fast something "path filter" like
ugexe still, it has to search `provides` sections as well
timotimo the ecosystem api could transform the complete list to split it apart 14:31
ugexe well i dont use a json module in zef, and this is because it ensures the json read/generated matches what compunit::repository sees/can handle
timotimo i mean the web service
oh 14:32
ugexe yeah. i'm not sure what the best format would be, but thats probably what needs to be done. zef's local cache and compunit repository use \0 delimited lines to split an identity into parts for instance (but this doesn't allow nested data that would probably want to be searched) 14:33
er, \0 delmited fields, each dist on its own line^
timotimo so .split("\n")>>.split("\0")? 14:36
ugexe yeah, something like that
timotimo that's what you already have, right? 14:37
ugexe github.com/ugexe/zef/blob/master/l...e.pm6#L124
timotimo i wonder how much worth it'd be to have a lookup-table in front
or even in back
ah, that store re-writes the whole thing "from scratch" for reasons 14:38
but it's not actually sorted? so it could just be appending perhaps? 14:39
ugexe yeah it doesn't sort. and don't pay too much attention to the details as they are probably wrong
timotimo it's probably not a hot piece of code, but it would grow with number of installed dists
ugexe github.com/ugexe/zef/blob/master/l...he.pm6#L95 14:40
this is likely the slowest part
also Zef::Repository::* has that method
ha i see a comment on short-circuiting that used to exist too
timotimo how much more often do you search compared to storing? 14:41
ugexe oh yeah thats dumb 14:50
github.com/ugexe/zef/blob/master/l...he.pm6#L23 # *reads* what .store writes
but the localcache though thats probably not bad. let me point exactly at the more likely culprit 14:51
well i guess i don't know where exactly, but probably github.com/ugexe/zef/blob/master/l...ystems.pm6 (which has a possible benchmarkable code snippet here github.com/ugexe/zef/blob/master/x...ory.t#L45) 14:56
AlexDaniel m: BEGIN { @*ARGS=‘--dirs=d1,d2’, ‘--dirs=d3,d4’ }; sub MAIN (List :$dirs=[]) { .say for flat @$dirs.List».split: /","/ } 15:27
camelia d1
d2
d3
d4
AlexDaniel mc: BEGIN { @*ARGS=‘--dirs=d1,d2’, ‘--dirs=d3,d4’ }; sub MAIN (List :$dirs=[]) { .say for flat @$dirs.List».split: /","/ }
committable6 AlexDaniel, ¦2015.12: «d1␤d2␤d3␤d4»
AlexDaniel m: BEGIN { @*ARGS=‘--dirs=d1,d2’ }; sub MAIN (List :$dirs=[]) { .say for flat @$dirs.List».split: /","/ }
camelia Usage:
<tmp> [--dirs=<List>]
AlexDaniel mc: BEGIN { @*ARGS=‘--dirs=d1,d2’ }; sub MAIN (List :$dirs=[]) { .say for flat @$dirs.List».split: /","/ }
ffffffffffffffffffffffffffffffffuuuuuuuuuuuuuuuuuuuuuuuuu
this is really annoying… 15:28
mc: BEGIN { @*ARGS=‘--dirs=d1,d2’ }; sub MAIN (List :$dirs=[]) { .say for flat @$dirs.List».split: /","/ }
committable6 AlexDaniel, ¦2015.12: «Usage:␤ /tmp/qAkip8Jc9Z [--dirs=<List>] «exit code = 2»» 15:29
AlexDaniel okay 15:30
Zoffix dam., I can't spell for shit 15:39
[Coke] we have a bot for that. 15:41
also: "damn" 15:42
;)
Zoffix :)
We have a bot for spelling? What? 15:43
[Coke] well, it's me running the spelling tests. 15:44
Zoffix Ah :)
AlexDaniel cmomittble6: say 42 16:25
committable6 AlexDaniel, Seems like you forgot to specify a revision (will use “v6.c” instead of “say”)
AlexDaniel Zoffix: ♥
committable6 AlexDaniel, ¦v6.c (19 commits): «42»
Zoffix AlexDaniel: ?
AlexDaniel Zoffix: these bots are for those who can't spell and can't learn the command syntax :) 16:26
Zoffix Ah :D
Coll
I mean Cool
AlexDaniel yre wellcoume
Zoffix Hahaha. twitter.com/kraih/status/870313476518010880 16:50
For all the Perl 6 hate, he sure loves our features :P 16:51
nine is just happy to see features still flowing 17:00
Zoffix is amused Perl 5 allows whitespace as heredoc delimiters 17:01
nine A feature dearly missed in Perl 6 ... 17:06
Zoffix Really? Feels to be of a "you think it's cute now..." type of ideas. 17:07
say q:to/ /;␤ Warning! This command is bad and you should feel bad!␤ qx/rm -fr ~/␤Don't run it, bruh␤ 17:09
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/XBO12Dblt4
Ending delimiter not found
at /tmp/XBO12Dblt4:5
------> 03 08⏏04<EOL>
expecting any of:
whitespace
Zoffix
.oO( hm, maybe I should put spacing around the command to meke it more visible )
say q:to/ /;␤ Warning! This command is bad and you should feel bad!␤ ␤ qx/rm -fr ~/␤␤Don't run it, bruh␤
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/qQUilf8Of_
Ending delimiter not found
at /tmp/qQUilf8Of_:7
------> 03 08⏏04<EOL>
expecting any of:
whitespace
Zoffix Oh mah gawd! My system!
evalable6: you're dumb
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/rTGy48MUrj
Undeclared routines:
dumb used at line 1. Did you mean 'DUMP'?
you're used at line 1
nine Zoffix: I was being quite sarcastic :) 17:28
Zoffix Ah :) 17:29
Looks like there's some confusion among users using rakudobrew thinking they're using R*: perl6maven.com/getting-started-with...3327453415 17:57
Also, `2017.04.2-2-ge1c086b`; it's one of the lemons. People don't realize they use crap commits.
AlexDaniel Zoffix: wait, why is evalable6 dumb? 18:04
Zoffix AlexDaniel: 'cause I didn't want to eval any code 18:05
AlexDaniel well, um, maybe evalable knows better? 18:06
Zoffix In fact, its replies interrupted my acting out a fatal programming error and if I don't get an oscar, I might just sue 18:07
AlexDaniel is happy it didn't actually run
Zoffix m: say qx/cal/
camelia qx, qqx is disallowed in restricted setting
in sub restricted at src/RESTRICTED.setting line 1
in sub QX at src/RESTRICTED.setting line 11
in block <unit> at <tmp> line 1
Zoffix it's restricted anywa'
AlexDaniel ye right… last time I was doing regressionable stuff, I found a bunch of shitfiles scattered everywhere 18:08
Zoffix XD
I tend to put all my files under /tmp
AlexDaniel restricted setting does not help, especially if you're using 2 year old commits to test stuff…
Zoffix m: use NativeCall; sub system(Str) is native {}; say system 'cal' 18:09
camelia June 2017
Su Mo Tu We Th Fr Sa
1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30

(Mu)
Zoffix hehe
AlexDaniel luckily we don't have many retarded vandals here, so I can get away with not implementing secure eval for a little bit more… even when you give them clear instructions on how to trick the restricted setting 18:11
Zoffix Oh noes! You've used the prohibited word!
AlexDaniel which one? “eval”? 18:12
use MONKEY-SEE-NO-EVAL;
Zoffix AlexDaniel: the one that was arbitrarily chosen to be offending to mentally disabled, despite no one actually using that word to describe the actual mentally disabled, just like no one uses "idiot" to describe the actual mentally disabled 18:13
I got shit for using it last time, and I'll be damned if I let you get away with using it! 18:14
Apologize!
AlexDaniel Zoffix: why talk at all if not to offend those who always want to be offended? 18:15
but still, yes, I'll pick a different word next time 18:16
Zoffix The argument's natural conclusion is all insults must be removed from use.
AlexDaniel: out of curiosity, which word will you plan on using as a substitute? 18:17
AlexDaniel Any suggestions? I was thinking, like, “faggots”, but then realized… 18:18
to be honest, I've never really thought about the origin of insults 18:19
especially those that are not in my native language… 18:20
like, insults are just insults, no? I guess not…
Zoffix AlexDaniel: no, never found a good replacement for retarded. All the related words don't quite have the same strength to them, and well, they all equate the insulted person or object to those of lower mental abilities, so if you're not going to use retarded for that reason, you can't use things like "idiot" either, and from then it all spirals down to all insults being bad.
AlexDaniel Zoffix: “brainless”? 18:21
Zoffix AlexDaniel: I think insults are a lot about context. There's nothing negative about being a baby, a child, or a girl; it's the context alone that give those word insult-like qualities when they're used as insults. 18:22
AlexDaniel: strength isn't there :) Even before the cultural shift to avoid "retarded" it was a bit too vulgar to use in, say, an office meeting, but "brainless" is much milder. 18:23
I guess modifiers could bring up to equal strength, something like "god damn brainless" or "fucking brainless", though the latter overshoots a bit 18:24
AlexDaniel
.oO( braińļeşs͞ ?)
 18:26
Zoffix hah 18:27
AlexDaniel oh wait, modifiers, not combiners!
u: MODIFIER
unicodable6 AlexDaniel, U+02B0 MODIFIER LETTER SMALL H [Lm] (ʰ)
AlexDaniel, U+02B1 MODIFIER LETTER SMALL H WITH HOOK [Lm] (ʱ)
AlexDaniel, 249 characters in total: gist.github.com/5e8636fe3f8543165b...0544ee00fe
AlexDaniel ᴮᴿᴬᴵᴺᴸᴱᶳᶳ ? 18:28
yes, no S 18:29
[Coke] If you can't find a better insult, perhaps you have an issue yourself.
Zoffix Is that a bad thing? :P 18:30
AlexDaniel [Coke]: that's a good insult
Zoffix haha
[Coke] I'm not going to call you names about it, if that's what you're asking. :)
AlexDaniel Zoffix: I like the idea of replacing all possibly offending words with a random sequence of unicode characters 18:34
Zoffix :D
AlexDaniel then anybody can make anything he wants out of it
they want, I mean
Zoffix :)
AlexDaniel ok I'll just shut up not to make a bigger fool out of myself :X 18:35
Zoffix Don't you mean a bigger ʷ˞ːˍ˳˳˧? 18:37
masak that looks like Principia Mathematica 18:38
[Coke] f̡͙̈́̒̏ő̴̧̹͢o͇̍ͨ͗͝l̸̨̻̳̖ 18:43
m: "f̡͙̈́̒̏ő̴̧̹͢o͇̍ͨ͗͝l̸̨̻̳̖".comb>>.NFD.map(*[0].chr).join.say 18:50
camelia fool
AlexDaniel too complicated 18:51
[Coke] wow, that looks horrible on the irc logs: irclog.perlgeek.de/perl6-dev/2017-...i_14672419
AlexDaniel m: say "f̡͙̈́̒̏ő̴̧̹͢o͇̍ͨ͗͝l̸̨̻̳̖".samemark(“dogs”)
camelia fool
[Coke] even shorter: samemark('a') 18:52
AlexDaniel m: say "f̡͙̈́̒̏ő̴̧̹͢o͇̍ͨ͗͝l̸̨̻̳̖".samemark(‘’) 18:53
camelia f̡͙̈́̒̏ő̴̧̹͢o͇̍ͨ͗͝l̸̨̻̳̖
AlexDaniel didn't work…
[Coke] m: say "f̡͙̈́ő̴̧̹o͇̍ͨ͗l̸̨̻̳".samemark(‘a’) 18:56
camelia fool
Zoffix m: say "f̡͙̈́ő̴̧̹͢o͇̍ͨ͗l̸̨̻̳".samemark(‘⁣’) 18:57
camelia fool
Zoffix :}
Geth rakudo: jsimonet++ created pull request #1094:
Fix for RT #131201 (error message when proto has no implementation)		 19:17
Zoffix ZOFVM: Files=1240, Tests=138233, 125 wallclock secs (23.90 usr 3.20 sys + 2561.17 cusr 151.78 csys = 2740.05 CPU) 19:44
Geth rakudo/nom: 252dbf3a91 | (Zoffix Znet)++ | src/core/Rakudo/Internals.pm
Fix hang in deepmap with Iterable type objects

If an Iterable type object is given to deepmap, .list gets called on it and the resultant 1-el iterable with the type object inside of it gets used to map over. However, the logic that descends into iterables forgets to check if they're type objects, so we end up infinitely recursing by descending into the type object iterable, ... (5 more lines)		 19:46
roast: 6b95eaacc2 | (Zoffix Znet)++ | S32-list/deepmap.t
Test deepmap and hypering does not hang on type objects

Bug find: irclog.perlgeek.de/perl6/2017-06-01#i_14672468 Rakudo fix: github.com/rakudo/rakudo/commit/252dbf3a91		 19:47
roast: d8261eaa79 | (Zoffix Znet)++ (committed using GitHub Web editor) | S32-list/deepmap.t
Add ticket ref to test
 21:12
dogbert17 m: sub grab(+@a) { "grab $_".say for @a }; grab(flat $(1, 2)) 21:21
camelia grab 1
grab 2
dogbert17 commit: 2016.01 sub grab(+@a) { "grab $_".say for @a }; grab(flat $(1, 2)) 21:22
committable6 dogbert17, ¦2016.01: «grab 1 2»
dogbert17 commit: 2016.07 sub grab(+@a) { "grab $_".say for @a }; grab(flat $(1, 2))
wake up bot
commit: 2016.08 sub grab(+@a) { "grab $_".say for @a }; grab(flat $(1, 2)) 21:23
bisect: sub grab(+@a) { "grab $_".say for @a }; grab(flat $(1, 2)) 21:31
bisectable6 dogbert17, Bisecting by output (old=2015.12 new=252dbf3) because on both starting points the exit code is 0
dogbert17, bisect log: gist.github.com/6e030f4b457143ddd9...a9a6334eda
dogbert17, (2017-02-18) github.com/rakudo/rakudo/commit/ca...8c5da329c6
dogbert17 .ask Zoffix should 'sub grab(+@a) { "grab $_".say for @a }; grab(flat $(1, 2))' return «grab 1 2» or «grab 1␤grab 2␤» 21:33
yoleaux dogbert17: I'll pass your message to Zoffix.
AlexDaniel slaps committable6 21:50
by the way 21:53
c: 348886d9b^,348886d9b my Array[Int] $x; say $x>>.Str;
github.com/rakudo/rakudo/commit/34...3254e4d709
committable6 AlexDaniel, ¦348886d9b^: «» ¦348886d: ««timed out after 10 seconds» «exit signal = SIGHUP (1)»»
AlexDaniel not that it makes any difference, but still
Zoffix dogbert17: 1, 2. The flat or containerization is irrelevant; the (1, 2) list becomes the [1, 2] array 22:16
yoleaux 21:33Z <dogbert17> Zoffix: should 'sub grab(+@a) { "grab $_".say for @a }; grab(flat $(1, 2))' return «grab 1 2» or «grab 1␤grab 2␤»
Zoffix Good grief. The social media is unusable this week. Everyone thinks they're witty and clever, parroting the same tired lines about American politics. 22:35
Pretty amusing.
ZofBot: "Mein Covfefe"! Please dispense your approval by mashing the "Like" button. Thank you. 22:36
ZofBot Zoffix, 365 test-t 5
geekosaur makes about as much sense as anything else being said on the subject... 22:37
Geth rakudo/nom: d09bf7fbb2 | (Julien Simonet)++ | src/core/Exception.pm
Fix for RT #131201 (error message when proto has no implementation)

Check if the array is empty or not.		 22:39
rakudo/nom: b4f1f43221 | (Julien Simonet)++ (committed using GitHub Web editor) | src/core/Exception.pm
Update Exception.pm

Change message to the one proposed by @zoffixznet.
  "Routine does not have any candidates. Is only the proto defined?"
rakudo/nom: 02614f6477 | (Zoffix Znet)++ (committed using GitHub Web editor) | src/core/Exception.pm
Merge pull request #1094 from jsimonet/nom

Fix for RT #131201 (error message when proto has no implementation)
rakudo/nom: 3524c2b79d | (Zoffix Znet)++ | t/05-messages/01-errors.t
Fix broken test
 22:47
rakudo/nom: 8e0d000fbb | (Zoffix Znet)++ | t/05-messages/01-errors.t
Test only-proto error is sane

RT#131201: rt.perl.org/Ticket/Display.html?id=131201 Rakudo fix: github.com/rakudo/rakudo/commit/02614f6477
Zoffix cpan@perlbuild4~/CPANPRC/rakudo (nom)$ ./perl6 -e 'Blob.split' 22:48
Cannot resolve caller split(Blob); Routine does not have any candidates. Is only the proto defined?
in block <unit> at -e line 1
I guess that fix fixes rt.perl.org/Ticket/Display.html?id...et-history too
Geth rakudo/nom: b667e818c5 | (Zoffix Znet)++ | t/05-messages/01-errors.t
Fix Blob.split/Blob.splice errors are sane

RT#131367: rt.perl.org/Ticket/Display.html?id=131367 Rakudo fix: github.com/rakudo/rakudo/commit/02614f6477		 22:52
Zoffix (cause IIRC the proto being being placed into Mu was for something to do with shaped array candidates) 22:54
AlexDaniel Zoffix: your favorite type of errors: irclog.perlgeek.de/perl6/2017-06-01#i_14673283 22:55
Zoffix haha 22:56
I guess that error should now be extended to mention IO::CatHandle's &.on-switch 22:57
AlexDaniel honestly, when I came to perl 6 with some perl 5 knowledge, these errors were extremely helpful 23:04
now that I think perl 6, I hate them with passion
I guess you can't get both 23:05
timotimo :D 23:06
Zoffix Sure you can. You pack them into "TeachingAid::Errors" module and remove them from core.
timotimo hm. then people coming from perl5 would need to "use TeachingAid::Errors" or something in all their scripts 23:07
that'll be the new "use strict"? :P
Zoffix Considering these errors interfere with normal Perl 6 programs and you'll be a "Perl 6 programmer writing Perl 6" infinitely longer than "Perl 5 programmer writing Perl 6", they don't pay for themselves. 23:08
timotimo that's true
Zoffix Why would that need be new strict? If you feel these suggestions would help you, you'd alias `perl6` to `perl6 -MTrainingAir::Errors` and be done with it. I can't recall these errors ever being helpful to me in my switch from perl 5 to perl6. I may have mistyped .elems with .length, but "no such method" would get me straight equally well 23:09
timotimo hm, if you alias it, that wouldn't be picked up by hashbangs, but you could no longer run "perl6 somescript.p6" without it sometimes breaking 23:10
Zoffix hates those errors enough not to care about that :D 23:11
timotimo :)
AlexDaniel Zoffix: why not just fix all cases where they interfere 23:12
timotimo they were put in as a feature
Zoffix AlexDaniel: from my POV? Because it's a useless feature and spending human hours fixing it is a waste. 23:13
AlexDaniel Zoffix: but if you were an ✓×⋅²≈ trying to shove perl 5 code into rakudo, would these errors actually be “useless”? 23:14
I don't think so 23:15
I loved these error messages so much
Zoffix AlexDaniel: how much worse would've been generic errors?
"no such method .length on Array"
m: -> \new { say new if $++ }(42)
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of C++ constructor syntax; in Perl 6 please use method call syntax
at <tmp>:1
------> 3-> \new { say new if 7⏏5$++ }(42)
Zoffix Ah, yes, I remember my days as a C++ like it were NEVER 23:16
as a C++ programmer
Zoffix leaves to hack on toaster
AlexDaniel well, I've never thought about it this way :)
what I know for sure is that generic messages don't have any love in them 23:17
AlexDaniel shrugs 23:18
maybe you are right
travis-ci Rakudo build failed. Zoffix Znet 'Merge pull request #1094 from jsimonet/nom 23:38
travis-ci.org/rakudo/rakudo/builds/238539028 github.com/rakudo/rakudo/compare/2...614f64772e
buggable [travis build above] ☠ Did not recognize some failures. Check results manually.
Zoffix buggable: recognize harder! 23:44

Powered by Cro and the Raku® Programming Language.

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

Thank you!