»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
jnthn Xliff: I've typically just put everything into a module (for example, if it's a CLI thing stuck the MAIN there too and marked it is export) and then the script is just "use My::Thing::CLI;" which is one statement to parse 00:00
And then all the rest is precompiled
Xliff jnthn++: Good idea. 00:03
00:05 khw joined
lizmat Xliff: since CompUnitRepo's are pluggable, you can write your own manager that would just manage pre-compiled versions 00:11
afk&
AlexDaniel [Coke]: fwiw I've been changing a lot of stuff to use Sake lately 00:15
yoleaux 18 Jan 2018 20:20Z <tbrowder> AlexDaniel: i was able to close the four tickets fixed by merge of rakudo PR #1339: RT #124281, RT #126742, RT #130477, and RT #132632
synopsebot RT#124281 [resolved]: rt.perl.org/Ticket/Display.html?id=124281 colonpairs in [POD] config options always produce strings
RT#126742 [resolved]: rt.perl.org/Ticket/Display.html?id=126742 [POD] config items should not include quotes for string values
RT#130477 [resolved]: rt.perl.org/Ticket/Display.html?id=130477 [BUG][POD] Pod config parses colopairs but simply stringifies whatever it matched
RT#132632 [resolved]: rt.perl.org/Ticket/Display.html?id=132632 [POD][NYI]. List and hash configuration value formats are not yet implemented (NYI)
AlexDaniel [Coke]: perhaps it does not have all the features, but that's some extremely pleasant dogfooding :) 00:16
00:21 lizmat left
AlexDaniel oh gosh… «»􏿽xE2􏿽x8C􏿽xA9 00:24
am I the only one thinking that this feature shouldn't exist really?
with all the nonsense it does, like shell-style quoting 00:25
the only times I used it was in code golf 00:26
because… well…
m: my $x = 42; say «60$x»
camelia (60 42)
tbrowder Zoffix: how can i get modules.perl6.org to update my module statuses. all of them had good tests when they were uploaded to CPAN. is there any way to update the cache without bumping the version on CPAN? do we need to add a tag in the .travis file for each version tag? 00:27
obviously i’m confused... 00:28
AlexDaniel tbrowder: “closed by merge of rakudo PR #1339”… and “, tests in roast PR #371” 00:31
tbrowder: because, technically, tickets are closed by commits with tests :)
that's for the next time 00:32
00:36 Zoffix joined
Zoffix tbrowder: you can push a commit like this github.com/perl6/modules.perl6.org...a41cc8e093 00:36
tbrowder: but it's not needed. As I've said, travis for CPAN dist source does not appear to be implemented 00:37
00:39 aborazmeh joined, aborazmeh left, aborazmeh joined
Zoffix And looks like it pointlessly tries to fetch it anyway: github.com/perl6/modules.perl6.org...pm#L23-L24 00:39
00:40 traxex_ is now known as traxex
Zoffix A somewhat rough fix would be to open the original META file packed inside the dist (those are not modified by PAUSE) and read the github source URL from it (well, try to see if it's there) somewhere here: github.com/perl6/modules.perl6.org...PAN.pm#L59 00:41
00:41 pmurias left
Zoffix And from what figure out travis credentials and pass them (via $dist->{_builder} hashref) to here github.com/perl6/modules.perl6.org...sCI.pm#L36 00:41
And as a bonus that'll also let us add a "repository" link people wanted. 00:42
And it'd prolly be the same for AppVeyor: github.com/perl6/modules.perl6.org...yor.pm#L39 00:43
+ with repo URL we can fetch stars and open Issues 00:44
Zoffix &
00:44 Zoffix left
AlexDaniel “From my point of view, Perl6 core developers should slow down their pace of improving the core language, and focus more on practical modules” 00:44
lol, we have too much effort put into improving the language? :D 00:45
00:46 shinobi-cl joined 00:47 lizmat joined
shinobi-cl r: subset Pos of Int where 1 .. *; class C { has @.arr; method new(@a) { return self.bless(arr => @a) }; method AT-POS(::?CLASS:D: Pos $row) {return @.arr[$row - 1];} }; my $c = C.new([1,2,3]); say $c[0]; 00:48
yoleaux 17 Jan 2018 04:45Z <lookatme> shinobi-cl: the type constraints should be `where * >= 1`
camelia Constraint type check failed in binding to parameter '$row'; expected Pos but got Int (0)
in method AT-POS at <tmp> line 1
in block <unit> at <tmp> line 1
Constraint type check failed for parameter '$row'
in method AT-POS at <tmp> line 1
in block <unit> at <tmp> line 1
00:49 lookatme joined
timotimo AlexDaniel: how dare we improve runtime performance 00:50
00:57 leont left
shinobi-cl r: subset Pos of Int where * >= 1 ; class C { has @.arr; method new(@a) { return self.bless(arr => @a) }; multi method AT-POS(::?CLASS:D: Pos $row) {return @.arr[$row - 1];} }; my $c = C.new([1,2,3]); say $c[0]; say $c[1]; 01:03
camelia C.new(arr => [1, 2, 3])
1
shinobi-cl so, i want my class to have indexes starting from 1. I guess i will create a special method to handle the element 0, to generate an exception, for example. Or maybe just leave it so it return the whole array? I'm not sure what should be the saner thing here... 01:05
lookatme add another candidate of AT-POS ? such as `multi method AT-POS($other) { fail "...." }` 01:07
timotimo add a proto to your class so no other AT-POS candidates get inherited from the parent class 01:11
01:14 shinobicl joined
lookatme r: subset Pos of Int where * >= 1 ; class C { has @.arr; method new(@a) { return self.bless(arr => @a) }; method AT-POS(::?CLASS:D: Pos $row) {return @.arr[$row - 1];} }; my $c = C.new([1,2,3]); say $c[0]; say $c[1]; 01:15
camelia Constraint type check failed in binding to parameter '$row'; expected Pos but got Int (0)
in method AT-POS at <tmp> line 1
in block <unit> at <tmp> line 1
Constraint type check failed for parameter '$row'
in method AT-POS at <tmp> line 1
in block <unit> at <tmp> line 1
01:17 Ven`` left, quotable6 left, quotable6 joined, ChanServ sets mode: +v quotable6 01:18 shinobi-cl left 01:29 markong left
TimToady m: say IntStr.new(3, "3").all ~~ Int 01:29
camelia True
mspo m: IntStr.WHAT 01:30
camelia ( no output )
tbrowder Zoffix: thnx, but i guess i have to bump the CPAN version to add the CPAN tag, yes? 01:33
01:34 troys is now known as troys_
AlexDaniel m: say +IntStr.new(3, ‘três’) 01:34
camelia 3
AlexDaniel m: say ~IntStr.new(3, ‘três’)
camelia três
01:39 Zoffix joined
Zoffix e: use lib <data/all-modules/github/zoffixznet/perl6-WhereList/lib>; use WhereList; sub s(*@a where all-items Int){ say 'here' }; s(1,2,3) 01:39
evalable6 here
Zoffix :)
tbrowder: yes. Dists are immutable, so if you want to mutate the META file you need to upload a new dist
lookatme: FWIW `where 1 .. *` is not wrong, although likely slower than `* >= 1`. The `where` clauses thunks and smartmatches, so `$foo where 1..*` is like `$foo ~~ 1..*` and smartmatch with numeric ranges checks the value is in range 01:41
m: say 42 ~~ 1..*; say -1 ~~ 1..* 01:42
camelia True
False
Zoffix m: { for ^100_000 { my $ = 42 ~~ 1..* }; say now - ENTER now }; { for ^100_000 { my $ = 42 >= 1 }; say now - ENTER now }
camelia 0.25361793
0.0078689
Zoffix mhm, like a billion times slower :) 01:43
lookatme Zoffix, oh, didn't know before :)
AlexDaniel m: my $x = 42; { for ^100_000 { my $ = $x ~~ 1..* }; say now - ENTER now }; { for ^100_000 { my $ = $x >= 1 }; say now - ENTER now } 01:45
camelia 0.22982925
0.0111522
Zoffix heh, right :) my version constant folded it 01:47
01:47 troys_ is now known as troys
Zoffix
.oO( zef plugin that generates stats during module testing on which routines are called the most so we'd know what has the most benefit optimizing )
01:50
And then shove it into toaster runs. 01:51
01:51 Zoffix left 01:53 khw left 01:56 itaylor57 joined
shinobicl lookatme: that's what i have, however, i want my class to be able to handle other things inside the [ ], like pairs or strings, so i have to use multimethods. 01:57
lookatme shinobicl, Okay, you can try proto, IDK whether it is help or not 02:01
02:06 khw joined
shinobicl r: class C { method AT-POS(::?CLASS:D: Str $x) { say "In AT-POS : $x"; } }; my $c = C.new(); say $c[1]; 02:08
camelia Type check failed in binding to parameter '$x'; expected Str but got Int (1)
in method AT-POS at <tmp> line 1
in block <unit> at <tmp> line 1
Type check failed in binding to parameter '$x'; expected Str but got Int (1)
in method AT-POS at <tmp> line 1
in block <unit> at <tmp> line 1
shinobicl r: class C { method AT-POS(::?CLASS:D: Str $x) { say "In AT-POS : $x"; } }; my $c = C.new(); say $c["1"];
camelia Type check failed in binding to parameter '$x'; expected Str but got Int (1)
in method AT-POS at <tmp> line 1
in block <unit> at <tmp> line 1
Type check failed in binding to parameter '$x'; expected Str but got Int (1)
in method AT-POS at <tmp> line 1
in block <unit> at <tmp> line 1
shinobicl r: class C { method AT-POS(::?CLASS:D: Str $x) { say "In AT-POS : $x"; } }; my $c = C.new(); say $c["A"];
camelia Earlier failure:
(HANDLED) Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5A' (indicated by ⏏)
in block <unit> at <tmp> line 1

Final error:
Type check failed in binding to parameter '$x';…
Earlier failure:
(HANDLED) Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5A' (indicated by ⏏)
in block <unit> at <tmp> line 1

Final error:
Type check failed in binding to parameter '$x'…
shinobicl hmmm. it wants a Int and only an Int. 02:09
r: class C { method AT-POS(::?CLASS:D: Int $x) { say "In AT-POS : $x"; } }; my $c = C.new(); say $c[1]; 02:10
camelia In AT-POS : 1
True
02:28 jeromelanteri joined 02:37 stmuk_ joined, syntaxman left 02:38 stmuk left 02:44 ilbot3 left 02:57 ilbot3 joined, ChanServ sets mode: +v ilbot3 02:58 Zoffix joined
Zoffix shinobicl: use "m:" trigger instead of "r:" You're duplicating all your evals by having it run on both backends 02:58
02:59 cdg joined
Zoffix shinobicl: you need to create your own postcircumfixes if you want to pass abitrary stuff 02:59
e: &postcircumfix:<[ ]>.candidates».signature».say
evalable6 (\SELF, Any:U $type, |c is raw)
(\SELF, int $pos)
(\SELF, int $pos, Mu \assignee)
(\SELF, …
Zoffix, Full output: gist.github.com/4c61ae692a2bd782da...9e939a10cc
Zoffix ^ 'cause those are the core ones and they coerce to ints
s: &postcircumfix:<[ ]>, \([], "x")
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/9024...ce.pm#L180
Zoffix m: class C { method AT-POS(::?CLASS:D: Str:D $x) { say "In AT-POS : $x"; } }; multi postcircumfix:<[ ]> (C \c, \v) { c.AT-POS: v }; my $c = C.new; say $c["A"]; 03:00
camelia In AT-POS : A
True
03:03 cdg left 03:06 troys is now known as troys_
shinobicl multi postcircumfix:<[ ]> (C \c, \v) ???? what kind of syntax is that? how should i search for it? 03:08
lookatme operator definition 03:11
Zoffix shinobicl: `multi` is a multi routine docs.perl6.org/syntax/multi `postcircumfix` is a category of ops and `:<[ ]>` is a longname specifying that you want to use `[` and `]` symbols for it. This defines a custom postcircumfix operator docs.perl6.org/language/functions#..._Operators then `(C \c, \v)` is the signature. It takes two args. `C` is the type constraint with your class C and
`\c` and `\v` are the parameters. `\c` is the same as `$c is raw` docs.perl6.org/type/Signature#inde...it__is_raw
lookatme great Zoffix ++
Zoffix s/longname/colonpair part of longname/;
like :foo<bar> except key is an empty string and value is a list of two strings 03:12
03:12 mempko left 03:14 syntaxman joined, syntaxman left, syntaxman joined 03:15 aborazmeh left 03:32 Zoffix left 03:36 edenc left 03:47 wmoxam joined 03:51 khw left 03:57 bloatable6 left, bloatable6 joined 03:58 unicodable6 left, unicodable6 joined, ChanServ sets mode: +v unicodable6 04:04 khw joined 04:11 Kaiepi left 04:12 Kaiepi joined
Xliff .tell lizmat Would that CompUnitRepo manager need to be written in nqp? 04:14
yoleaux Xliff: I'll pass your message to lizmat.
04:28 shinobicl left 04:31 shinobicl joined 04:32 wamba joined 04:35 Cabanossi left 04:36 Cabanossi joined 04:37 Kaiepi left, Kaiepi joined
shinobicl multi postcircumfix:<[ ]> (C \c, \v) ???? what kind of syntax is that? how should i search for it? 04:38
sorry, i posted that line again by mistake :(
AlexDaniel hehe 04:43
04:52 khw left
shinobicl Hi.. so i have this bidimensional array i need for a toy project of mine. It is immutable (hence the name)... So far the interface looks (more or less) like this... I want to upload it soon, but any ideas to improve the usability are welcome... pastebin.com/UXp0U18i 04:58
04:59 khw joined
shinobicl s/usability/usefulness/; 05:00
05:03 notbenh joined, troys_ is now known as troys 05:17 eliasr left
Xliff shinobicl: Why is @shape[1;2] 7 and not 5? 05:20
shinobicl sorry, it is actually 5.. it was a typo in the paste 05:23
05:24 khw left
Xliff kk 05:25
So the only time you can assign to the table is when it is initialized? What initializers do you have for it?
It currently looks like it takes 2 arrays. One for the header and one for the data. 05:26
Do you have other initializers? 05:30
shinobicl two. If you use an array as first param, this is the header. If you use a number, headers will be created in alpha order, like A, B, C, if you use 3 etc...An old version has also methods to get subsets by using regexes on columns. And to create indexes on any column. But it was a bit too much, so i took all this out and started over.
pastebin.com/CUVSdBWx <--- Current version
Xliff m: say (<A B C>,<D E F>,<G H I>),invert.say 05:34
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
invert used at line 1
Xliff m: say (<A B C>,<D E F>,<G H I>).invert.say
camelia Type check failed in invert; expected Pair but got List ($("A", "B", "C"))
in block <unit> at <tmp> line 1
05:35 mempko joined
Xliff m: ([Z] <X Y>).say 05:36
camelia ((X Y))
Xliff shinobicl: I am trying to think about how you would do a Hash initializer. 05:38
The first part is easy, headers = %h.keys
It's the values part that seems to be the problem.
I was hoping that Array.transpose existed, but it doesn't.
AlexDaniel what behavior would you expect from .transpose? 05:39
shinobicl pastebin.com/4YevcEmG <--- Current test suite 05:40
Xliff shinobicl: @shape[1;2] is still expecting 7 05:41
shinobicl yes, please ignore that... i'm writing more test right now :) not everythig passes, is just to show how it is used now :) 05:42
Xliff m: ([Z] (<X Y>, <a b>, <1 2>)).say
camelia ((X a 1) (Y b 2))
Xliff AlexDaniel: ^^ That, but working.
See the bottom of: docs.perl6.org/language/traps#___top 05:43
shinobicl: Fair enough, but I find that best practice is to fix bugs as you find them, less they fall through the cracks. ;)
AlexDaniel didn't take long for someone to point me to the trap that I've documented myself :D 05:44
Xliff LOL!
I guess I could write a transpose that special cases the one element bug. 05:45
05:50 committable6 left, squashable6 left, statisfiable6 left, committable6 joined, ChanServ sets mode: +v committable6, squashable6 joined, ChanServ sets mode: +v squashable6, statisfiable6 joined, ChanServ sets mode: +v statisfiable6
Geth doc: cfcb1f16bc | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Language/traps.pod6
Fix wording
05:52
synopsebot Link: doc.perl6.org/language/traps
AlexDaniel releasable6: next
releasable6 AlexDaniel, Next release in 1 day and ≈13 hours. Blockers: github.com/rakudo/rakudo/issues?q=...%9A%A0%22. Unknown changelog format
06:04 integral joined 06:05 wamba left 06:06 bwisti left 06:15 travis-ci joined
travis-ci Doc build passed. Aleks-Daniel Jakimenko-Aleksejev 'Fix wording' 06:15
travis-ci.org/perl6/doc/builds/330677639 github.com/perl6/doc/compare/006ec...cb1f16bcf4
06:15 travis-ci left 06:28 shinobicl left 06:38 troys left 07:00 Xliff_ joined 07:02 abraxxa joined 07:03 Xliff left 07:07 abraxxa left 07:11 darutoko joined 07:20 espadrine left 07:21 lowbro joined, lowbro left, lowbro joined, abraxxa joined 07:35 fluca1978 joined
Geth doc/master: 11 commits pushed by (Luca Ferrari)++
review: github.com/perl6/doc/compare/cfcb1...8d0df9cb5e
07:36
AlexDaniel fluca1978: “inspact” 07:39
also, “regexp”? Do we actually call it that? 07:40
quotable6: \bregexp\b
quotable6 AlexDaniel, OK, working on it! This may take up to three minutes (4482999 messages to process)
AlexDaniel, gist.github.com/922a8e6f5d82aecccd...4d7b0b959d
AlexDaniel quotable6: «regexp»
quotable6 AlexDaniel, OK, working on it! This may take up to three minutes (4482999 messages to process)
fluca1978 AlexDaniel: thanks, allow me to work on it 07:41
07:41 sena_kun joined
AlexDaniel no! Improving things is forbidden! :) 07:41
sure
quotable6 AlexDaniel, gist.github.com/b6a2a591652f4e63b2...5acd7b0b4d
AlexDaniel wow, that seems to be rather popular 07:42
Geth doc: 16c033c2e4 | (Luca Ferrari)++ | doc/Language/regexes.pod6
Change 'regexp' to 'regular expression'.

While the term 'regexp' is a short of 'regular expression'
  (see <irclog.perlgeek.de/perl6/2018-01-19#i_15709645) in this
context it could be better to call a regular expression with the extended form.
See <irclog.perlgeek.de/perl6/2018-01-1...709640>
07:51
synopsebot Link: doc.perl6.org/language/regexes
doc: d82dc3c4c7 | (Luca Ferrari)++ | doc/Language/regexes.pod6
Fix typo ('inspact' -> 'inspect').

See <irclog.perlgeek.de/perl6/2018-01-1...709637>
07:57 domidumont joined 07:59 abraxxa left 08:05 parv joined 08:09 domidumont left 08:10 domidumont joined
lizmat . 08:14
yoleaux 04:14Z <Xliff> lizmat: Would that CompUnitRepo manager need to be written in nqp?
lizmat .tell Xliff no, not as far as I know
yoleaux lizmat: I'll pass your message to Xliff.
08:15 stmuk joined 08:17 stmuk_ left 08:37 kst` joined 08:39 cgfbee left 08:41 kst left
lizmat WIP on P5 module to P6 porting guide: github.com/lizmat/P5-modules-to-P6...ting-guide 08:41
they're basically things I've run into while porting things like List::Util and friends 08:42
I will be on the road most of today and thus afk&
will check up on things about 14 hours from now
08:52 rindolf joined 08:53 cgfbee joined 09:07 lookatme left 09:08 parv left 09:10 leont joined 09:12 cpage left 09:26 ChoHag left 09:27 wamba joined 09:28 zakharyas joined 09:33 geekosaur joined 09:34 geekosaur left 09:35 geekosaur joined 09:38 leont left
El_Che the reddit thread is impossible to monitor for new messages. A good workaround is getting new messages through RSS: www.reddit.com/r/perl/comments/7r1...unity/.rss <-- lizmat 09:49
stmuk or block reddit with stayfocusd or similar :) 09:53
09:57 scimon joined
El_Che stmuk: ha, good point 09:58
TEttinger lizmat is on the road to the FaceSpaceReddDiggNewsBunker, a castle from ancient times with a room completely sealed off from social media
psch oh geez a reddit thread 09:59
El_Che TEttinger: she may be kidnapped first by angry reddit users 10:03
TEttinger that's what the castle part is for 10:04
if it fended off visigoths it can fend off visinerds
El_Che In the castle you have to fight the last boss: p5p 10:05
TEttinger weapon of choice, butterfly knives
El_Che img.etsystatic.com/il/bd8a45/88061...6_eu8b.jpg <--- butterfly ninja star 10:06
TEttinger etsy has some crazy stuff
El_Che it's from Into the Badlands tv-series 10:07
TEttinger ah
10:08 wamba left
TEttinger I wonder how they made it. laser cutting I don't know how it works on steel. I think 3D printing in metal uses selective laser sintering, which makes stuff like www.shapeways.com/product/F9HX968L...arketplace 10:10
I have one of those dice, it's everyone's favorite
then there's the "you aren't nerd enough to play this game" dice, with numbers in Tolkien's Tengwar script www.shapeways.com/product/QDVP8F2Q...ated-items 10:11
moritz if it's not in Unicode, I won't play with it :-) 10:19
10:26 wamba joined
scimon Why oh why oh why did I reopen this reddit thread>? 10:28
psch i kinda wonder about this whole "it's a new language" argument, and how many people using it have tried modern perl5 on a perl4 interpreter 10:29
or heck, 5.0 perl5
mind, i really have no idea how large the changes between perl4 and perl5 were, so it might well be that the argument holds perfectly well 10:30
geekosaur perl 5 introduced references, which ultimately led to a lot of changes
perl 3 -> perl 4 was more incremental
but as a *language* perl 5 was a lot like perl 4, and in fact there were very few things you could run into if you ran a perl 4 script on perl 5 10:31
stmuk I remember having to escape \@ in "" that was about it
psch ah, thanks geekosaur++ and stmuk++ 10:32
geekosaur yes, tjhat was the main one I wa thinking of --- but there were few rare others
psch so the argument as intended would mostly hold -- i.e. "write some bog-standard p5, run it with perl4, it mostly works" is trueish
geekosaur I would say not
scimon I luckily came in after the 4 -> 5 transition. 10:33
geekosaur perl 4 stuff ran in perl 5 mostly unmodified. perl 5 added enough stuff that was badly needed, that perl 5 code was usually not going to run in perl 4
I mean perl 4 didn;t even have "use"
scimon Having spent the 90's failing to take adventage of the internet craze by working in a bookshop.
psch ah, okay, i misinterpreted that then
geekosaur `do "getcwd.pl"`
scimon :D
stmuk people did mostly write perl 4 in perl 5 for several years .. maybe using CGI->new would be the only perl 5 10:34
10:34 ChoHag joined
stmuk I'm sure some did use perl 5 features from day one but probably a minority 10:34
scimon What gets me is all the people saying "You need to change the name" as if anyone has any control over it. Beyond Mr Wall. 10:35
stmuk I don't think it would make any difference whatever 10:37
scimon And... I really don't understand the If I have to stop using Perl5 I'm not moving to Perl6... it's weird. I *guess* it's an uncanny vallet thing.
geekosaur it's not
it's an activation energy thing. it takes a lot of energy to move everything to something incompatible
scimon That's kind of my point.
psch i'd agree that changing the name wouldn't really *do* much
geekosaur but once you've decided to put that energy in, you have lots of othher options on the table as well. perl 5 is not a captive market for perl 6 10:38
psch i mean, sure, maybe p5 can get hype from releasing a v7.0, but is that really enough to crush python and node, which everyone says are the big winners in the market
scimon Right... I really don't see the whole "Move Perl5 codebases to Perl6" working. Because most of the Perl5 codebases I know are massive. Scary scary massive.
geekosaur yes. I like things like Inline::Perl5 because it enables interoperability. perhaps we should focus on *that* 10:39
scimon I guess for me it comes down two out of the three (Python, Node and Perl6) Perl6 is the one that fits with the way I think more. 10:40
stmuk I find it a bit odd that the perl 5 community (used to perl 5 like DSLs and syntax extensions) are so scared of perl 6 especially when error messages are so perl 5 friendly
geekosaur because perl 5 and perl 6 being the same kind of *think* means that interop works better between them in subtle ways
scimon Python is too staid and reminds me of Java (shudder). Node has some lovely bits but a lot of cruft and the event model is annoying a lot of the time. 10:41
Anyhoo. Time to go poke about in a big pile of mason. At least it's not looking for a bug this time. 10:42
I looked at Inline::Perl5 one time but the first step about requiring a perl5 compiled to worked with it seemed like way too much work. 10:45
psch heh "the difference between perl5 the runtime and Perl 5 the language doesn't exist" says one reddit commentor, and yet "Only perl can parse Perl" used to be a popular saying..?
stmuk scimon: I have thought of bundling a "jumbo" Star-type distro with Inline::Perl5 and perl 5 10:47
or patching rakudobrew to support it
scimon Ok looking at the build process it doesn't acutally look at bad as I first thought. 10:49
stmuk scimon: BTW I notice on twitter you were playing elite .. is that the modern PC one? 10:50
scimon It is the only reason I have windows installed on my PC at home. 10:51
(And virtual box, turns out a gaming rig run dev VM's really well)
(That's a yes ;) )
psch okay, with "perl6 core devs should stop improving the core language and build useful modules instead" i'm closing the reddit thread 10:56
10:57 abraxxa joined 10:58 ufobat joined
jast yes, please don't improve the language, I hate improvements. especially optimizations, more docs and bugfixes... 10:58
10:59 TEttinger left
ilmari psch: how's that different from the original letter's "perl5 core devs should stop adding features and port the language to nqp"? 11:02
11:02 pmurias joined 11:03 abraxxa left
El_Che omg 11:04
perl 6 has a hard time finding its way into production and adding a 2nd unrelated runtime (and a hell to debug) is going to help that?
I love perl 6, but no way something like that would reach testing at $work (I would block it myself) 11:05
pmurias I guess Perl 5 users feeling someone wants to force them to move to Perl 6 before it's ready is breeding a lot of the resentment
stmuk I think there is a lot of pent up resentment about the decline of perl 5 and 6 is being scapegoated 11:06
El_Che a lot of it's not scapegoating
the version things is killing both projects
psch ilmari: i'm not sure i'm qualified to speak on that -- i have no idea how development of perl5 works or has worked for recent past
ilmari: but the parts that i'm interested in -- which is mostly r-j -- is kinda neglected and pretty broken 11:07
ilmari: i suppose i could write Inline::Java instead...
stmuk I don't think software project are killed by names or numbers .. they aren't that powerful
ilmari psch: the people who do it mostly don't care much about perl 6
jast I have to agree with stmuk here
El_Che people are real that don't touch perl6 because it's old perl and don't touch perl 5 because perl 6 has obsoleted it 11:08
they are called colleagues
ilmari but we do care about keeping perl5 alive and moving forward
El_Che and pretty much everywhere I go
jast heck, firefox survived two more-or-less forced renames
psch ilmari: ah, so we agree that "work on modules instead of core" is a bad suggestion? :)
ilmari psch: yes. and I was pointing out that at least one of the suggestions in the initial letter was equally bad 11:09
jnthn points out that many people who work on core stuff *also* have a bunch of modules
ilmari disclmaimer: I am actively involved in perl5 core development
Altreus it's so confusing that #perl and #perl6 have so many common conversers 11:12
jast my interpretation of the letter (we *are* talking about the one on perl.com, right?) isn't that it calls for people to stop working on perl5 core
ilmari «Place a moratorium on new features, with development confined to maintenance on the current runtime.» 11:13
psch ilmari: ah, i see. yes, i agree with that
ilmari «The Butterfly Perl 5 Project» section
jast there's the "idea" in it that porting perl5 to rakudo might be worthwhile, but the rest of the focus is on porting modules to perl6, mildly implying that maybe porting perl5 is a bit much
Altreus I just read this letter and I don't know whether I have an opinion so I'm going to borrow these ones 11:14
jast to me, the Butterfly Perl 5 section mainly served as contrast to the *other* plan (CPAN Butterfly)
the section concludes as, "Although I wholeheartedly would support a Butterfly Perl 5 Project, I’ve also come to the conclusion that it is no longer an itch I would want to scratch personally at this moment." 11:15
which to me is a reasonably clear indication that the actual plan is the next one :)
pmurias a "moratorium on new features seems 11:16
a "moratorium on new features" seems a lot like declaring it "dead and burried"
stmuk I think its a pity "v5" slang died
jast yeah, but that's what I just said, it's not what the letter actually calls for
tyil I think its a pity that the reddit threads are so hostile
jast it's just an option it's putting on the table
without actually pushing that option, instead pushing the other one
11:17 abraxxa joined
stmuk there haven't really been that many successful new perl 5 features TBH maybe // 11:17
jast notice that all the remaining sections focus on porting modules instead of changing anything about how perl 5 core is developed
tyil any suggestion that isnt "kill off perl 6 and long reign perl 5 only" gets downvoted and answered with unrelated remarks about hew perl 6 is not perl 5 and therefore the worst thing ever
stmuk jast: I doubt many "renamers" got past the first few paragraphs 11:18
jast I'm always wary of using new Perl 5 features because half of them get removed again later (at least subjectively)
plus the Perl job I worked before required compatibility with Perl 5.8...
psch tyil: to be fair, they're saying "call it something else" and not "kill it completely"
tyil psch: depends on which comment you're reading
jast personally I don't care what it's called and I don't think a different name would change anything 11:19
tyil but there is a huge hostility coming from most commenters
psch true
tyil jast: I agree with both parts
stmuk I think the "takeaway" is "address the wider programming community rather than the perl 5 one"
tyil I dont think I've ever taken or dropped a language because of the name it has
so long as it works for me, I'll use it 11:20
jast right, chances are everyone in the perl 5 community has already made up their mind
probably a bit of a waste of time to try and change it
at best you can change minds one at a time in a situation like this, and that's not what blog posts and comment threads are usually about 11:22
stmuk KBO 11:25
scimon Indeed. 11:30
11:30 astj left 11:31 astj joined 11:35 astj left 11:44 astj joined 11:45 markong joined 11:48 astj left, pmurias left 11:49 zakharyas left
El_Che there is a lot of hostility in the comments coming from perl 5 people, but what really make me sad is perl 6 people acting or believing that there isn't a problem at all. Perl 5 and Perl 6 both suffer from the current situation. 11:51
From the outside, the perception game, both risk to be seen as obsolete. The name is not a problem for Perl 5, but for Perl 6, seen as old when it was a new language. And Perl 5 seen as obsolete because there is a major new version already. 11:54
11:54 pmurias joined
El_Che Changing the name won't make millions flock to perl6 nor add tons of improvements to perl 5. But ot would help not loosing the game before the starting line 11:55
El_Che off: lunch
11:57 konsolebox_ left 11:58 konsolebox joined 12:02 fluca1978 left 12:04 vichib left
scimon So.... this conversation has been going on for a while right? With two sides entrenched in their beliefs? Oh god.... it's what the UK is going to be like for the new 20 years isn't it? 12:09
I can't escape. On the one hand Brexit and on the other Perl naming.... I'm going to go play some cards.
AlexDaniel I think that there is an increasing number of people agreeing that an alias would be beneficial 12:13
(as opposed to just leaving it as “Perl 6”)
I myself was against it for a long time for example 12:14
so IMO it's going on for a while not because we can't agree on anything, but because it takes time to accept this kind of revolutionary ideas :) 12:16
and it's totally understandable that people are against it. Without them we would be changing the name every day
for me personally the turning point was when I attempted to talk about Perl 6 with non-perl IT people 12:19
like, you have to go through the trouble of explaining why it's called Perl 6 if you want them to understand that you're talking about some brand new language and not perl5 12:21
12:21 pmurias left
AlexDaniel and for me that's totally a technical issue, not associated with any beliefs really. The name just doesn't work as well as it should, so fix it maybe 12:23
12:24 traxex_ joined
El_Che AlexDaniel: that's also my experience. 12:25
psch my experience has mostly been along the lines of "all perls are bad anyway" vOv 12:26
12:26 eliasr joined 12:27 traxex left, vike1 left, kaare__ joined, esh left, mingdao left
psch i like changing the Perl 6 brand because it does actually solve the biggest grievance perl5-exclusive people have 12:28
12:28 pmurias joined, vike joined
psch like it as an idea, that 12:28
is
12:28 syntaxman left
psch well, *perceived* as biggest from here :) 12:28
12:28 mingdao joined, esh joined, Ven`` joined 12:29 bloatable6 left, reportable6 left, bisectable6 left, syntaxman joined, bisectable6 joined, reportable6 joined, bloatable6 joined, benchable6 left, coverable6 left, releasable6 left, joy left, syntaxman is now known as Guest94921
AlexDaniel yea that's also a thing. Like, depending on your experience and the kind of people you talk to, you may have a different opinion on that topic 12:30
12:30 vichib joined, nativecallable6 left
AlexDaniel and it seems that “perl 6” works just fine for some people 12:30
good for them, I guess…
psch although i can already see the "oh now you change the name after blocking our next major version for 18 years and making us lose heaps of market share" argument... :) 12:32
but that's obviously very likely to be in bad faith anyway, sooo
AlexDaniel forgive and forget :) 12:33
.oO( we're not complaining about your language giving our brand new language a bad rep! :D )
12:34
12:34 joy joined 12:36 greppable6 left, reportable6 left, releasable6 joined, greppable6 joined, benchable6 joined, ChanServ sets mode: +v benchable6
AlexDaniel releasable6: next 12:36
releasable6 AlexDaniel, Next release in 1 day and ≈6 hours. Blockers: github.com/rakudo/rakudo/issues?q=...%9A%A0%22. Unknown changelog format
12:36 coverable6 joined
pmurias really dislikes the Perl 5 community vs Perl 6 community that the renaming talk seems to implicitly assume 12:36
12:36 reportable6 joined
Altreus especially since, as noted, the overlap is sufficient to keep confusing me 12:38
12:40 wamba left 12:41 pilne left 12:42 traxex_ is now known as traxex
pmurias Perl 6 being something that appeals to a community of experienced Perl programmers is at least for me a big selling point 12:44
jast okay, here's a new name idea, leeching off of the success of IPv6: PLv6 12:46
El_Che succes of IPv6? 12:48
12:49 grumpster joined
grumpster pmurias: how could perl6 *possibly* appeal to an *experienced* perl5 programmer? 12:49
literally every small detail is just sufficiently different to make any instinct go sideways
/\n/ matches "\r" 12:50
jast El_Che: hey, nice, you found my joke :}
grumpster there is significant whitespace
El_Che jast: lol
grumpster sigils differ
method invocations differ
even scoping seems not-exactly-the-same 12:51
so what's the appeal?
over say golang
traxex what if someone were to point out that there are experienced perl5 programmers on this channel who find perl6 appealing
jast I am an experienced perl5 programmer, and perl6 appeals to me
traxex wow
jast because (a) perl5 has enough slightly annoying quirks and (b) perl6 has more than enough awesome new things to work with 12:52
stmuk Well I'm an experienced perl5 programmer who prefers it ... but I have to admit most I meet are either indifferent or hostile
I think the "renamers" just want the name changed because they don't like the idea of perl 6 being a sort of perl
jast I'm a sucker for expressiveness, and if there's anything perl6 definitely is, it's more expressive
Juerd Next year I will have 20 years experience with Perl 5. In all those years, many things appealed to me even though they were not called Perl.
jast I think it's very clearly a perl, but I'm also completely indifferent to renaming efforts 12:53
Juerd In fact, this year already.
grumpster when core Perl5 toolchain people come in and say: irclog.perlgeek.de/perl6/2016-01-05#i_11830221
you have a clear problem
jast for me it's about 15 yearas of experience with Perl 5
grumpster it boggles my mind in the utmost degree how this can be brushed aside as "hostile" 12:54
Juerd stmuk: I'm a strong proponent of renaming Perl 6, because I love both and would like them to have a chance at succeeding.
Ven`` learned Perl 5 after learning Perl 6...
jast that comment you linked to doesn't read hostile at all, just frustrated
Ven`` Reddit isn't very good for such discussion :\. There are 50 new comments on that thread, but I have no idea how to see them... 12:55
jast which I guess is fair :) there's definitely a lot of room for improvement in a project as young (all things considered) as perl6
stmuk I don't see the correct answer to the lack of "p6doc" which is to install it in that log either
:)
jast Ven``: you know, to get a basic minimum of functionality all you have to do is buy reddit gold....... (or use a well-known browser extension) 12:56
Ven`` jast: I don't have a reddit account
and I don't plan on getting one..
jast who knows, the extension might work anyway
El_Che !last rss
oops
Ven`` RHS probably works anyway, yes
El_Che www.reddit.com/r/perl/comments/7r1...unity/.rss
Ven`` ...why is everything htmlescaped there..? 12:57
stmuk perl5 doesn't even have non-experimental signatures or p5-mop in it yet .. the idea that changing the perl 6 name would help with their basic issues is crazy
and what about the "given/when" issues
Ven`` whereso and whereis are no more at least ;-). 12:58
El_Che it's not crazy at all
AlexDaniel pmurias: that's a selling point for a portion of perl5 devs, but how many perl5 devs do we have out there nowadays? Like, shouldn't we focus on everyone else instead?
El_Che renaming is to no create the false impression that you're working with an obsolete version
grumpster the issues for most perl5 programmers are mostly "I can not disassociate myself with this other thing I do not like, and am not considering using. I can not stop my managers, dev-friends, random folks in a coffee shop that saw my laptop sticker to keep bugging me about Perl6, because from their perspective the two are related"
pmurias grumpster: the knowledge of details doesn't translate over, but (most of the) good stuff has Perl 6 equivalents 12:59
grumpster give perl5 that - and all hostility will go away, as *most* ( not all, most ) want to simply forget perl6 exists
stmuk El_Che: its too late .. shutting the stable door after the horse bolted
grumpster if it works out as a language - great, and we wish it luck
El_Che: most perl5-ers I know do not care about "this is obsolete" - they *like* it that way 13:00
the main point is that the conversation keeps resurfacing "when are you upgrading to P6?"
El_Che grumpster: it's a about new blood 13:01
grumpster this is what most "hostile people" are trying to escape in vain
sure - it might be
jast by the same logic, IPv6 should have been renamed long ago
El_Che ipv6 is an update
grumpster but it's not the "new blood" that writes not-so-nice reddit replies
El_Che perl 6 wasn't
jast ipv6 is not backward compatible in essentially the same way that perl6 is
grumpster El_Che: the new blood is indifferent, except it sees the tension and goes "hmmm... I'll go do some rust instead" 13:02
stmuk the "renaming" talk is bikeshedding of the worst sort
jast throw ipv4 packets at ipv6 and they get thrown away :) except if you do weird hacks (which is part of what the letter is about)
grumpster stmuk: bikeshedding will continue until clarity improves 13:03
jast people who choose rust for something probably have reasons that would make perl6, or any similar languages, a questionable choice in the first place
13:03 grumpster left 13:05 pmurias left
jast in any case, the blurb on perl6.org is pretty clear on what perl6 is and isn't 13:05
if people don't even read the first sentence on a project's website, how is *anything* going to get through to them? 13:06
okay, second sentence :)
stmuk the question about p6doc is covered in the faq BTW 13:10
13:10 pmurias joined
pmurias grumpster: would a rename even solve that problem? If people where asking you about the "Wicked" (or whatever) language that they heard Perl was renamed to would it be better? 13:13
13:14 vichib left
stmuk pmurias: no it would be more confusing. It would also be an incredible amount of effort to change every ref to "perl 6" by something else and some would be missed. And we would be the laughing stock of reddit at the end of it 13:15
[Coke] scimon++ # cards
Juerd jast: The problem is not getting interested people, who read the website, to know what things are. 13:17
jast: The problem is getting people interested, and something seemingly superficial such as the name is utterly important for that, especially because most people will decide whether to look into something, based on only the most superficial information.
jast if all they have is a completely new name, that's not going to convince them to look at the language, either 13:18
to convince people that it's worth looking at, you have to make it easy for them to be exposed to something that appeals to them 13:19
Juerd jast: While I do think the name clash is fundamentally wrong, that's just a matter of principle and doesn't actually hurt any project. The way the name Perl hurts both Perls isn't one that can be fixed within the community because it has nothing to do with the community.
jast I'm not saying the name is free of problems, but the more fundamental problem is exposing people to what perl6 is about in the first place
Juerd jast: Do you have a hackerspace nearby?
Go there. Ask fellow geeks about Perl. Then ask them why they haven't tried Perl 6 yet. 13:20
jast not exactly
but I don't think that's the right way to approach this issue, anyway 13:21
Juerd I will predict the following answers: 1. Perl is dead or should die, 2. Because it's Perl. These are not my opinions, but there does almost seem to be concensus over these things.
jast eliciting people's gut reactions doesn't tell us anything new. I *know* there are lots of associations with the name "Perl"
13:21 aborazmeh joined, aborazmeh left, aborazmeh joined
Juerd Gut reactions are way more important than you think. 13:21
jast if you want to do something *now*, get more cool perl6 code out there, in small and easily-digested portions
stmuk virtualsue spoke to a room of over 100 go devs and asked for hands up who had heard of "perl6" 13:22
about 3
jast yes, which is why it's more important to focus on *changing* them, instead of arguing about things that aren't going to happen any time soon, anyway
stmuk that's 3% and 100% had heard of it by the end of the evening
jast btw I'm going to do my part by doing a perl6 talk here at work at some point in the near-to-mid future :) 13:23
Juerd jast: Most of the time I agree. But whenever the discussion comes up, and it does so quite often, I keep hoping that TimToady will change his mind and agree to renaming Perl 6.
stmuk and if you change the name that's back to 0%
jast discuss that with him and not with me ;)
Juerd Personally, I think Camelia and Rakudo would be great candidates for the name of my favourite programming language. 13:24
stmuk: I think they may have heard about Perl 6 but had filtered the information because they're not interested in Perl.
stmuk: At least that's a pattern that I keep seeing with people around me who aren't Perl programmers.
[Coke] I spent 5 minutes once explaining the difference between 5 & 6 to someone here who asked a very nice question about it, and it's never been a concern since.
jast I'm willing to concede that the name could use discussing, but there are more immediately worthwhile things that can be done to raise awareness of perl5 vs perl6 (and why perl6 is great ;)) 13:25
Juerd [Coke]: I spent those 5 minutes at least a hundred times already and will continue doing so.
[Coke] where here == $DAYJOB
Juerd: well done, thank you 13:26
Juerd And even if Perl 6 is renamed to something better, some explanation will still be necessary
I love to talk about the language and if anything, the name is a good conversation starter. 13:27
[Coke] Agreed.
13:28 pmurias left
[Coke] I'm a little inured to the "your language is bad and you should feel bad" argument because I backed into doing work with ColdFusion over a decade ago and never still have to support developers using it as their primary language for developing web apps 13:29
Juerd www.linuxjournal.com/article/2070 " One of the ideas I keep stressing in the design of Perl is that things that are different should look different."
[Coke] s/never//
jkramer Isn't the cro tool installed with the cro package/module? I got the modules but can't find cro in path
timotimo did you add the path that zef installs stuff to? 13:30
like /home/timo/perl6/install/bin
stmuk ${ROOT}/share/perl6/site/bin/cro
jkramer I don't have ~/perl6 o_O
timotimo no problem
doesn't zef output "installed one script to /foo/bar/baz"? 13:31
sena_kun `.perl6` maybe
timotimo oh, the perl6 folder is my own creation
it's where i have my rakudo, moarvm, and nqp git clones
jkramer Ich have ~/.perl6 but it only contains a "precomp" directory with a bunch of files with hashes as names 13:32
I'll re-install cro and see if zef says anything about its location 13:33
Ahh it's ~/.rakudobrew/moar-nom/install/share/perl6/site/bin 13:34
timotimo oh rakudobrew 13:35
jkramer Shouldn't eval "$(rakdubrew init -)" in my zshrc add that to PATH?
timotimo no
you need to "rakudobrew rehash" every time scripts are installed
moritz or, like, don't use rakudobrew :-)
jkramer Ah
timotimo also, rakudobrew isn't a good fit for people who don't hack on the rakudo/moar core every day
huggable: rakudobrew
huggable timotimo, nothing found
timotimo huggable: r
huggable timotimo, nothing found
timotimo huggable: p6 13:36
huggable timotimo, nothing found
timotimo what's teh factoid called that gives you zoffix' build tool ...
jkramer What else would you recommmend?
moritz jkramer: are you developing rakudo, or only using it?
timotimo github.com/zoffixznet/r
moritz if just using, github.com/nxadm/rakudo-pkg
jkramer moritz: Just using it, but I actually like rakudobrew. I thought it's basically perlbrew for P6 :) 13:37
timotimo it installs forwarder scripts for everything zef installs so that you can switch between multiple installed versions of rakudo in an instant
a user would normally only ever have one version installed, thus these forwarders are only making things more difficult 13:38
(because you have to regenerate them every time something is installed)
AlexDaniel huggable: r6
huggable AlexDaniel, nothing found
AlexDaniel huggable: r
huggable AlexDaniel, nothing found
AlexDaniel huggable: z
huggable AlexDaniel, Helper script for Rakudo Perl 6 core development: github.com/zoffixznet/z
AlexDaniel oh, here it is
timotimo no, z is the wrong one
oyu want r 13:39
13:39 Ven`` left
AlexDaniel hmmm… 13:39
timotimo what's with my typing today
AlexDaniel huggable: r :is: github.com/zoffixznet/r
huggable AlexDaniel, Added r as github.com/zoffixznet/r
13:41 vichib joined
stmuk so rakudobrew is bad because its aimed at devs and the recommended replacement is also aimed at devs building the latest development commit? 13:41
13:42 Zoffix joined
Zoffix stmuk: it builds latest tag 13:42
(I forget what rakudo.org page says on that; it might need updating)
stmuk (git describe --abbrev=0 --tags)
ah
s/latest version/latest release/ maybe? 13:44
Zoffix changed 13:45
stmuk you could just add a --make-install to the Configure.pl line if you wanted to golf it down a line 13:47
and can't help thinking it would be easier as a shell script too 13:48
scimon I thought Rakudo Star was the recommended install. 13:50
stmuk scimon: it depends who you speak to
Zoffix :)
scimon perl6.org/downloads/ ? 13:51
13:51 zakharyas joined
scimon Personally I use these ones. github.com/nxadm/rakudo-pkg/releases 13:52
Zoffix scimon: yeah, you heard right. But many users wish to have more bleeding-edge releases so they use compiler releases or even build off development commits. There's a tool to achieve the latter called "rakudobrew" and it has some issues, so the recommendation *for bleed users* is to not use rakudobrew and to just use the alias.
Yeah, packages are good too
scimon Oh I know I used Rakudobrew for a while.
But as you say. Issues.
Zoffix Rakudo Star is basically our most-thoroughly-tested release.
Juerd Yep, I want to get bleeding edge rakudo and use rakudobrew for that.
stmuk is using a forked version of rakudobrew which installs the lastest release and is modified so the test target actually works 13:53
scimon The monthly release pacakages are a nice why to stay bleeding-ish edge but aheas of star.
Juerd And whenever that borks I just rm -r ~/.rakudobrew and start over :)
scimon :D
Juerd scimon: I'd probably use monthly releases if installing them was a single command. I'm lazy. 13:54
stmuk I should pinch some of r into my forked 'brew to support Inline::Perl5
:>
Juerd Basically I upgrade rakudo ever other week and hope the performance is good enough for my production needs yet 13:55
scimon There is a chap who has a PPA for unbuntu of the monthly release but it's a bit behind. (Currently on 2017-11)
Ahhhhh.
Juerd Most importantly, memory use. It's improved a lot but not there yet.
scimon What are you trying to do?
jkramer Ok I dumped the rakudobrew install and re-installed with that r-thing, but it looks like it doesn't bring zef
Zoffix Juerd: the command is just `ugrade-perl6` with my "r" instructions: github.com/zoffixznet/r And you don't need to `rakudobrew rehash` when installing binaries. They Just Work
Juerd scimon: Well, I'd like to use Perl 6 on cheap virtual server instances with only 512 MB of memory.
jkramer Ah nevermind, that are install instructions for zef too
Juerd Zoffix: Thanks, I hadn't looked into r yet. 13:56
El_Che Juerd: I have 32-bit ubuntu pkgs for that reason
scimon Oh... yeah. I've not tried that yet.
13:56 aborazmeh left
Zoffix jkramer: right :) it's separate because you only need to install zef once, not after every upgrade of rakudo 13:56
Juerd github.com/Juerd/p6-mqtt/blob/master/foo.pl is my canary 13:57
Once that runs without major memory usage (and memory leaks), Perl 6 is probably ready to replace my Perl 5 scripts. 13:58
stmuk ha perlbrew supports cperl now 14:04
timotimo m: react { whenever Signal(SIGINT) { say "got signal $_" } }; # today's fun mistake
camelia got signal SIGINT
stmuk although last time I tried cperl it didn't even compile on linux only os x 14:05
scimon Juerd: Interesting. 14:06
stmuk wonders if r doesn't exist as a script to avoid support issues :) 14:07
Zoffix I shove everything into aliases. That's the only reason. 14:10
Plus saves the endless debates on the safety of `\curl -L blah.blah | bash`
nige yes - short aliases are good - "go" is a great runtime name, short and verby for running a computer program 14:13
it is totally pants however from a Google-ability point of view
14:14 bisectable6 left, bisectable6 joined
nige short, and distinctive - easy to type and say 14:14
having watched this channel over the years and its latent punning power I'm sure members could come up with a short list for TimToady++ to consider 14:15
Zoffix FWIW we were talking about bash aliases for compiling from source :P 14:16
nige sorry Zoffix missed the context there 14:17
timotimo Juerd: what rate of memory growth are you observing in foo.pl? 14:18
huh. after a while it stops giving new messages apparently 14:21
El_Che GoFurtherLang 14:26
nige "gfl" 14:27
three letter acronyms are not a bad place to look for trade markable names - they sometimes collide with stock market codes though 14:29
ICQ - "I Seek You" was a an inherently strong brand name based 14:30
Zoffix New blog post: "Long Live Perl 5!": rakudo.party/post/Long-Live-Perl-5
nige on three letters - google-friendly etc
stmuk 4f865671685c6d53b53dfacf4beb9e5faf8aecf8 Ensure Configure.pl uses gmake for builds on BSDs 14:32
why?
Zoffix stmuk: there's some comments on the PR: github.com/rakudo/rakudo/pull/1335 14:33
timotimo i do see it slowly grow. maybe a megabyte every two minutes? 14:35
stmuk I can't make any sense out of that .. I thought all the makefiles didn't use GNU make features? I've never seen any problems installing on BSD with its native make
"BSD's make isn't entirely compatible with GNU's" from the PR
it doesn't have to be
14:35 trnh joined 14:36 vichib left, bwisti joined
nige Zoffix++ +1 to the "Elephant in the Room" section in rakudo.party/post/Long-Live-Perl-5 - just 2 more cents - it would be great if the alias could be trade marked and ideally google-friendly 14:38
Zoffix stmuk: IIRC that user was trying to fix compilation on BSDs. They committed fixes to MoarVM as well: github.com/MoarVM/MoarVM/commit/28...2609e8338b 14:39
14:39 giraffe left
nige Zoffix do you see the alias as being the runtime name? 14:40
stmuk has he added a Makefile with GNUisms I wonder? 14:41
14:41 grumpster joined
Zoffix nige: in Perl 6 we have much stronger distinction between language and implementation than Perl 5. The alias concerns the name of the language, not what any particular implementation decides to call its executables. 14:42
grumpster Zoffix++ # excellent interpretation
stmuk I think if a dependency on gmake has been added it should be documented in the Moar docs .. its likely to break existing BSD ports as well
jast Zoffix++ # +1 on the entire post
Zoffix stmuk: as far as I recall those were the only commits in rakudo/moarvm from that person
grumpster however there remains a problem: the open-letter was written by one of the most ( if currently not *the* most ) prolific contributors
until the author stands up and says that she wholehartedly supports "Since Perl 5 is actively maintained, its users should not feel compelled to migrate to anything else, and should continue using what works for them" 14:43
the hurt is not going to go away I am afraid
14:44 grumpster left
tbrowder Zoffix: maybe i read the post too rapidly, but what about the issue of 6 blocking 5 to 5+N? 14:44
jast she did mention specific reasons why people would migrate: "because they can, they want to try, or just to see how Perl 6 will work out for them" 14:45
nige Zoffix - yes - I've got your point there - someone could alias the alias to the particular runtime they might want use
jast clearly if none of these apply to someone, that person wouldn't migrate
Zoffix tbrowder: I think that's covered by The Naming Issue
stmuk Zoffix: OK I'll do some BSD make testing and probably email Ben 14:46
Zoffix & 14:47
14:47 Zoffix left
nige a commercial company creating their own version of Perl 6 - could call it <their-name> implementing Perl <Alias>&#174; 14:50
stmuk the FreeBSD build is broken even if I install gmake 14:51
nige mico$ixy implementing Perl Rak
where Perl is a master brand and Rak is the alias/sub-brand 14:53
Juerd timotimo: That depends on the number of messages received, but for typical work loads, around 250 kB per minute. 14:56
timotimo do you also sometimes get it just not receiving stuff any more?
Juerd timotimo: Yeah, that it stops giving new messages is a new thing. Didn't happen before, but today it keeps happening. Could just be the server disconnecting for whatever reason; I haven't debugged it yet.
timotimo now it's just sending keep-alives
Juerd timotimo: That shouldn't happen; I haven't seen that happen yet. 14:57
By the way, if you want to get a truck load of data per second, subscribe to the topic "#", which is the mqtt wildcard. 14:58
timotimo uh oh :)
AlexDaniel stmuk: can you create a ticket?
Juerd e.g. whenever $m.subscribe("#") { say .<message>.decode("utf8-c8") } 14:59
timotimo From argument to subbuf out of range. Is: 39, should be in 0..37
huh, fun.
is it invalid to ask those for a topic?
Juerd Hmmm
timotimo no, it also happenns with just .<message>.decode 15:00
Juerd I usually keep to off by 1 bugs. Two is not my style ;)
Hm, is this the server spewing invalid data, or a bug in my library. The latter is much more likely but I'm hoping for the former. 15:01
timotimo if i just put a try around the while parse loop it keeps going
how does the parse function deal with incomplete data?
Juerd It should return False
stmuk AlexDaniel: yeah sure ... it looks like support for a non-standard BSD compiler has broken the standard one (at least on FreeBSD). I suspect the added GNU Make dependency isn't needed either 15:02
Juerd In any case, an $offset of 39 is weird 15:03
15:03 vichib joined 15:04 vichib left
Juerd It's been a while since I wrote this and the variable length message length encoding loop weirds me out. I'll need a minute to try to understand this :) 15:04
timotimo OK :)
15:05 vichib joined
Juerd Oh, it's the second subbuf that fails, not the first one. 15:06
15:08 trnh left
Juerd Bug found 15:08
- return False if $length > $buf.elems + $offset; 15:09
+ return False if $length + $offset > $buf.elems;
timotimo lizmat: the Telemetry doc page links to Telemetry::Instrument::AdHoc, but that isn't part of the docs :(
15:11 nige left
timotimo is someone good with vim regexes and highlighting? it looks like the vim highlighter for perl6 chokes on comments that have many - in them 15:12
15:12 nige joined, giraffe joined
Juerd Bug fix pushed to p6-mqtt github 15:13
Damn, the amount of binary garbage published to test.mosquitto.org is staggering. 15:14
15:15 rindolf left
mspo never heard of that 15:20
15:22 rindolf joined, tony-o joined
Juerd MQTT is a publish/subscribe protocol, Mosquitto is an open source message broker (mqtt server), test.mosquitto.org is their public server and probably the best real world test case ever :D 15:22
15:22 nige left
Juerd I know at least 2 people who never bothered to install their own mqtt server because test.mosquitto.org is surprisingly stable 15:23
stmuk the build is also broken on OpenBSD
15:23 nige joined 15:24 zakharyas left 15:25 zakharyas joined
mspo stmuk: which build? 15:25
stmuk rakudo 15:26
mspo nqp and moar build okay? 15:30
stmuk: I'm just speculating that the netbsd build is probably also broken
stmuk I meant in MoarVM .. its recent changes to the FreeBSD and OpenBSD config 15:31
I'd guess NetBSD isn't affected
RT#132741
synopsebot RT#132741 [new]: rt.perl.org/Ticket/Display.html?id=132741 [BUILD] Broken on FreeBSD and OpenBSD
mspo rt could use a L&F update :) 15:33
stmuk L&F?
mspo look and feel 15:34
shouldn't that build be using $CC from the environment?
15:35 konsolebox left, astj joined
stmuk yes I'm sure I've used other compilers successfully via environmental config only .. compiler flags on command line 15:35
mspo netbsd has clang too 15:36
15:38 trnh joined 15:39 astj left 15:40 vichib left, vichib joined 15:41 konsolebox joined 15:45 st_elmo joined 15:50 domidumont left 15:51 domidumont joined 15:52 kaare__ is now known as kaare_ 15:56 nige left, konsolebox left 15:57 domidumont left, nige joined, lowbro left 16:03 st_elmo left, konsolebox joined, st_elmo joined 16:10 wamba joined 16:12 zakharyas left, zakharyas joined
jkramer m: my Array[Int where * < 5] $foo; $foo.push($_) for ^5; 16:12
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in typename; couldn't find final ']' (corresponding starter was at line 1)
at <tmp>:1
------> 3my Array[Int7⏏5 where * < 5] $foo; $foo.push($_) for ^5
expectin…
jkramer m: my Array[subset SmallNum of Int where * < 5] $foo; $foo.push($_) for ^5; 16:13
camelia ( no output )
jkramer m: my Array[subset SmallNum of Int where * < 5] $foo; $foo.push($_) for ^10;
camelia Type check failed in assignment to ; expected SmallNum but got Int (5)
in block <unit> at <tmp> line 1
jkramer Shouldn't the first version work too? 16:15
16:15 nige left
jkramer m: sub foo(Int $x where * > 5) { say "yeah $x" }; foo(7); foo(2); 16:16
camelia yeah 7
Constraint type check failed in binding to parameter '$x'; expected anonymous constraint to be met but got Int (2)
in sub foo at <tmp> line 1
in block <unit> at <tmp> line 1
16:18 scovit joined 16:21 nige joined 16:22 bdmatatu joined 16:23 sjoshi joined
scovit hello, is it normal that a fresh install of zef from github on rakudo.2017-12 gives me 'No such method 'subst' for invocant of type 'Any'' 16:24
?
timotimo can you get a stack trace for that? maybe with --ll-exception even?
scovit sure: gist.github.com/scovit/4038354b9e5...d1082644de 16:27
[Coke] what does 'zef info zef' say? 16:28
(Identity is probably the important bit) 16:29
timotimo unfortunately we can't see from the stacktrace what provided the result for the "ver" call
scovit zef info zef says this :gist.github.com/scovit/412c4add9a6...fffe227afa
[Coke] seems recent (I'm running 0.1.32 here, need to upgrade!) 16:31
scovit let's say that I modify the sources at /pasteur/homes/vscolari/share/perl6/site/sources/50B47A043364E01112770F7D1C590E62CECFF648
how do I tell perl6 to take the modifications?
timotimo not a good idea
i'd rather git clone zef, change the code there, then zef install --force-install . 16:32
scovit mmh let's try
timotimo or just -I. to pull the code in
scovit Ok, I reinstalled zef 16:35
now back to zef install File::Temp
Str $v = "0.0.2"
Any $v = Any
No such method 'subst' for invocant of type 'Any'
16:35 nige left
timotimo where does the first $v come from? 16:35
scovit before there are also other Str $v = "0.0.x" kind of stuff
timotimo oh, there's just one dd in the entirety, and it gets called more often. sure. 16:36
Geth doc: ec806e6df9 | (Alex Chen)++ | doc/Language/faq.pod6
Update Language/faq.pod6
doc: 8f4b7c8791 | (Alex Chen)++ | doc/Language/variables.pod6
Document naming conventions
synopsebot Link: doc.perl6.org/language/faq
Link: doc.perl6.org/language/variables
doc: 93b9bcea6c | (Alex Chen)++ | 2 files
Add myself on CREDITS
16:37 wander joined
scovit yep, I putted the dd there 16:37
but I do not know what are those $v
and why it ends with Any
when zef installs zef it works fine
timotimo it's probably the Distribution::DEPRECATED class that gets the ver method called on it
can you put a "dd self" in there, too? 16:38
scovit it seems that doing that triggers an infiniloop 16:40
timotimo oh, haha
of course
let's put in an evil hack
scovit O can try with self.WHAT
timotimo m: sub test { my $*FOO = rand; say $*FOO; say CALLERS::<$*FOO>; test unless $++ > 5 } 16:41
camelia ( no output )
timotimo m: sub test { my $*FOO = rand; say $*FOO; say CALLERS::<$*FOO>; test unless $++ > 5 }; test
camelia 0.364240484362533
Nil
0.954991139474574
0.364240484362533
0.954318373402591
0.954991139474574
0.789046088766774
0.954318373402591
0.708910210756786
0.789046088766774
0.647764382083217
0.708910210756786
0.522312396438158
0.647…
timotimo cool, that works
scovit: put my $*FOO = 1; dd self without CALLERS::<$*FOO>; at the beginning or end of that method 16:42
scovit ok ill try
16:42 nige joined
scovit I will copy you just the last line of output 16:43
16:44 wander left
scovit here: gist.github.com/scovit/b5ad6cfed67...74b9bc797b 16:45
updated now 16:46
timotimo is the dd before or after the subst?
scovit it is beforew
timotimo strange. the ver method it uses should have used $!version which is set 16:47
scovit ok, let me check
16:48 kraih joined
scovit there is this extra error: Use of uninitialized value of type Any in string context. 16:48
timotimo right. i wonder where that comes from, exactly
oh
scovit line 32 16:49
timotimo no, not from our dynamic var
hm, not sure what calls Str there, but it's likely the ver method giving an undefined value again 16:50
16:50 konsolebox left
timotimo what are you trying to install here? maybe i can try to reproduce 16:50
scovit File::Temp 16:51
timotimo is that the one that explodes?
scovit yes
timotimo it installs just fine here, but i'm on latest rakudo master; you're on 2017.12, right?
scovit right 16:52
I can downgrade
to rakudo star
timotimo what, downgrade zef?
oh
16:53 khw joined
kraih lizmat: it would appear that you're being quite dishonest regarding the renaming issue irclog.perlgeek.de/perl6/2018-01-17#i_15702313 vs www.youtube.com/watch?v=olILAdlWFZk&t=488 16:56
16:57 konsolebox joined
kraih honestly, at this point i just want to see Perl 6 die 16:58
tyil that sounds a bit aggressive
geekosaur we noticed, troll.
16:58 ChanServ sets mode: +o geekosaur
kraih lizmat is right, lets leave the sister language solution behind 16:58
stmuk /ignore kraih is probably best at this point
scovit :) 16:59
kraih i'll wear that ban as a badge of honor
tyil whether or not you're here complaining, we like Perl 6, and we'll keep on using it
16:59 setty1 joined
geekosaur no, god himself has just spoken to you, can you not tell? 16:59
tyil oh
sorry :(
I still like Perl 6 tho 17:00
timotimo i find it sensible to point out that inconsistency in liz' opinion fwiw
tyil I'd just wish for the p5 ond p6 communities to be friendly to eachother so we can benefit from one another
kraih but think about it, i went from designing the butterfly cover of the first Perl 6 book, to wanting to see Perl 6 die
geekosaur not when trolls want to feed on it
17:00 st_elmo left
scimon I do feel it's a real shame you feel that way kraih. 17:00
geekosaur trolls must feed, you know, they are the true gods
tyil timotimo: it might've been, but not if he says he wants to see the lang die instead of remain sensible
kraih: sounds like an issue on your part 17:01
have you considered professional help?
mst kraih: while the inconsistency was worth pointing out, your other comments are not constructive and merely laying additional damage atop that caused by liz' divisive article
tyil: not cool.
geekosaur just like hate is the only truth
timotimo not sure what changed between the presentation and the 6.c release date
geekosaur just like destruction is the only creation 17:02
tyil I didnt find liz's article divisive tbh
stmuk rakudo.party/post/Long-Live-Perl-5 is positive
mst tyil: it came across to a lot of people as a giant fuck you to the perl5 community; zoffix's post is much more balanced and much more constructive 17:03
17:03 trnh left 17:04 geekosaur sets mode: -o geekosaur
mst also calling the sister language thing a 'necessary fiction' was *incredibly* counterproductive, since that was absolutely not how any of the people involved in that actually felt 17:04
tyil mst: but it wasnt, people who want to be angry for no reason read it as something divisive just to be justify an angry mob
mst tyil: if you're unwilling to even try to empathise with the people who were upset, then you're part of the problem and should just keep quiet about the subject 17:05
tyil they'll be angry no matter what we do or propose
mst a small number of people, yes, but this upset lots more people than that
you might like to read strangelyconsistent.org/blog/how-pe...ill-us-all 17:06
kraih mst: i think the issue needs to escalate
mst I think that coming into #perl6 to tell the perl6 devs you want perl6 to die is just as terrible an idea as writing an article on perl.com telling the perl5 devs you want perl5 to die 17:07
tyil she never said anything like that 17:08
ilmari the article _did_ suggest stopping perl5 feature development and port it to nqp
tyil telling people they have no right to speak because they dont empathize with an angry mob is much worse than the open letter
mst tyil: having given a lightning talk called 'perl5 is the new perl4' that did say that, her remarks are inevitably read with the context - and 'stopping feature development' is effectively killing a project
stmuk perl6.org/ still says "little sister" but people can express their own opinions
ilmari even if it was presented as a "radical" idea and discarded
stmuk I'm a "sisterist" myself
tyil ilmari: that wouldnt be killing perl 5, that would be securing a future
17:09 zakharyas left
mst tyil: "stopping feature development" is not "securing a future" 17:09
ilmari tyil: except for the people who _want_ to keep perl5 moving and adding new features to it
mst but, again, you're not even trying to empathise, just dismissing the people who're upset as 'an angry mob'
ilmari the giant freeze between 5.10 and 5.12 was bad enough, nobody wants a repeat of that
mst so I don't believe you're actually trying to be constructive
tyil the people who are upset are calling for killing off perl 6, a language I like a lot, so yeah, excuse for not empathizing with this idea 17:10
mst ...
tyil I'll use perl 5 or perl 6 depending on what I feel works best
I don't see *any* need to be this angry at either side
17:10 ChanServ sets mode: +o mst
mst tyil: enough. 17:10
you're calling people who feel like there was a call to kill off perl5 an angry mob and then claiming that they're bad for talking about wishing perl6 would die. this is hypocritical, counterproductive and prima facie stupid. the conversation is now over. I will enforce it with quiets if anybody attempts to continue this divisive bullshit. 17:11
17:12 mst sets mode: -o mst 17:13 scimon left
scovit how does one wipe the zef cache/database ? 17:20
jnthn zef nuke
17:20 MilkmanDan left
jnthn oh, I think :) 17:20
timotimo well, zef nuke has multiple modes 17:21
jnthn Depends what you want to be rid of
timotimo at least one of them is probably going to nuke stuff you don't want nuked
scovit luckily this time zef update did the trick
17:21 MilkmanDan joined
scovit but still I think there might be a bug 17:21
17:23 kraih left
stmuk zef caching does seem sometimes over agressive 17:25
17:31 ChoHag left
El_Che good evening 17:34
sjn o/ 17:35
El_Che hi salve 17:36
long time no see
did the UN sent you here to keep the peace? (Even Trump loves Norway) 17:49
tadzik because they grow his hair? 17:50
17:51 astj joined
El_Che tadzik: 17:51
:)
17:55 astj left
sjn is lurking full-time these days 17:56
tadzik so, who's going to fosdem? 17:57
18:01 mempko left 18:10 khw left
El_Che hopefully everyone. I plan to lock all the angry perl 5 and 6 people in a room. They can only come out if they agree on a path forward (or those who survive) 18:12
sjn is going to fosdem o/ 18:15
18:16 Kaiepi left, domidumont joined
timotimo i'm not going to fosdem 18:18
18:18 Kaiepi joined 18:22 khw joined 18:25 nige left 18:26 nige joined, sjoshi left
DrForr sjn++ # yay! 18:29
18:33 darutoko left
[Coke] I am not going to fosdem. 18:34
18:36 st_elmo joined 18:38 wamba left 18:39 tyil joined 18:41 nativecallable6 joined 18:42 Scimon joined
Scimon So. I think I'm going to step back from IRC for a few weeks. After 15 years of being a perl developer and hearing how nice and welcoming the perl community is I'm feeling very down. 18:45
I was hoping to go to my first perl conference this year and give a talk on Perl6. But I don't have the mental energy to cope with this sort of environment. 18:46
So see you all in a while.
18:46 Scimon left 18:51 traxex left 19:02 traxex joined
tyil tadzik: I'll be at fosdem :> 19:03
DrForr I'll be there as well, natch. 19:04
Sorry about Scimon (just got in) but I can definitely understand how he feels. 19:06
tyil ^ 19:07
El_Che he left 19:10
he has happier things to think of, his upcoming marriage
DrForr That'll do it, good on him! (and spouse-to-be FTM) 19:12
El_Che happy news aside, we're not showing our best side either
so it should be a lesson for us all 19:13
DrForr cranks /I Love Rock-n-Roll/ while waiting for things to pick up. 19:15
Though the drama is apparently playing out on reddit, not here.
19:17 setty1 left
mspo you need another dime? 19:20
drama?
DrForr The mp3 jukebox doesn't require dimes, lucky for me - I'd have to find an exchange office. 19:22
El_Che mp3? You commoner
flac is da shit
DrForr Ogg FTW.
El_Che lossless or die 19:23
DrForr moves on from her rock-n-roll to rocky-horror. 19:25
masak sjn: \o 19:27
mspo hot patootie 19:29
19:29 mempko joined
mspo I think eddie listens to the radio, not a jukebox 19:29
DrForr Just remember, no matter how you slice it, it's still Meatloaf. 19:30
19:31 raschipi joined
Xliff_ I just watched a show with Meatloaf in it <_< 19:33
yoleaux 8 Oct 2017 16:51Z <thundergnat> Xliff_: Ah. Oh well, it was worth checking since I was getting that exact same behavior with an older version.
raschipi Meatloaf has multiple versions? <_< 19:34
Xliff_ 2017, yoleaux? Really?
raschipi yoleaux is a bot that delivers messages
Xliff_ I know. That message is old
raschipi Oh, ok.
Xliff_ I think I was supposed to get one from lizmat.
raschipi It's because now you have the _ at the end. 19:35
19:36 quotable6 left, quotable6 joined, ChanServ sets mode: +v quotable6 19:37 nige left 19:40 raschipi left 19:43 nige joined
[Coke] cannot remember the last time he ended up caring what format a music file was in. 19:49
sena_kun well, with a great pair of headphones, a good record and a good ears you can actually notice a subtle difference if you know to what exactly you should listen to. :) 19:52
but this difference is indeed subtle. 19:53
El_Che there is quality of course, but there is also future-proofing your music collection
[Coke] I'm sure I couldn't.
El_Che in case you need to convert to an other format
sena_kun [Coke], did you tried?
El_Che you don't want a lossy source
psch isn't playback of records destructive? 19:54
sena_kun El_Che, sure. I did not implied that "MAH SOUND" is the only reason to keep flac. :)
psch i.e. you get used to your own slightly-deteriorated-from-lots-of-playing-it sound
El_Che psch: you can convert your new LP's right away to flac :)
psch or was that only casettes?
El_Che psch: everything physical has wear 19:55
psch i'm pretty those definitely get worse over time
+sure
El_Che casette > LP > CD
psch El_Che: well casette playback mumble magnetism mumble mumble
:s
El_Che yeah, I was a snobby kids buying good cassettes for my walkman 19:56
the dirt cheap ones were crap
[Coke] sena_kun: yes
sena_kun [Coke], ah, okay then. 19:57
19:58 ChoHag joined 19:59 st_elmo left
psch the biggest difference in playback quality for me was still from an audio interface upgrade 20:05
like, even 128 bit mp3s sounded... i guess "more detailed" is the best phrase 20:06
but i strongly believe in diminishing returns there :p 20:07
20:10 khw left 20:12 domidumont left
moritz for me, the biggest upgrade was from tape to CD :-) 20:15
[Coke] moritz: agreed.
[Coke] listens to a presentation on us tax reform 20:16
psch i barely experienced that, and definitely didn't pay enough attention to notice the upgrade :)
mspo [Coke]: in 192bit sound? 20:20
20:22 khw joined 20:23 setty1 joined
stmuk you can take the FLAC from my cold dead fingers 20:27
20:29 rindolf left 20:30 zacts left
timotimo opus ftw 20:33
20:34 zacts joined 20:46 rindolf joined
Xliff_ .tell raschipi Oh. LOL. Missed the '_' 20:47
yoleaux Xliff_: I'll pass your message to raschipi.
Xliff_ Sorry. FLAC is too heavy for my available storage. 20:49
Lossy or not, I get more bang for my megabyte-buck from MP3
FLAC = great taste. MP3 = less filling. :P 20:50
20:50 evalable6 left 20:51 evalable6 joined, ChanServ sets mode: +v evalable6
stmuk losers :P 20:51
sena_kun stmuk, just ooc, what headphones/dac/amp are you using? :) 20:53
timotimo i currently listen to music and stuff using a 60 euros soundbar attached to the headphone output of an asus screen that's fed via displayport :)
timotimo can hear the cringe from hundreds of miles away 20:54
psch timotimo: i used to listen on 40€ headphone via onboard sound
stmuk sena_kun: I can't remember but its from China and has loads of bits
psch timotimo: i could hear (and record) the usb port interfering with the DAC, because the noise changed pitch when i lifted my mouse of my desk 20:55
*off
sena_kun stmuk, ah, ok. Good luck with your flac. :) (actually like 70-80% of my records are in flac)
timotimo psch: the headphone jacks on my laptops all tended to break only a year or so into use ;(
timotimo AFK
stmuk FiiO E7 Portable Headphone DAC Amplifier 20:57
21:06 mempko left, bdmatatu left 21:08 pilne joined 21:16 colomon joined, colomon left 21:20 comborico1611 joined
comborico1611 Anyone ever use a straight razor? 21:38
yoleaux 18 Jan 2018 22:05Z <sena_kun> comborico1611: also see github.com/emacs-tw/awesome-emacs
El_Che Xliff_: you can keep flacs and convert to mp3 for devices with less space 21:43
Xliff_: 320kbits/s is fiiiine :)
21:50 TEttinger joined
timotimo so with languages like haskell, why don't we rename perl6 to "Wall" or "Larry"? 21:51
comborico1611 I'm for that. 21:52
Would you choose Wall or Larry?
21:52 S007 left
timotimo i'm not sure how many programming languages are named after people immediately involved in their development 21:52
like, Ada wasn't actually developed by ada lovelace
and i think haskell curry also didn't have a hand in haskell? not sure if they were even still alive when haskell was started in ~1850 21:53
moritz proposal for anybody who proposes a language rename: create an IRC channel with that name, and see how many people follow
timotimo but that enforces the limitations of irc channel names on the language name :( 21:54
like, what if i want to say "perl6 is like C# but without the C and more of the #", so i'd propose to call it ##?
comborico1611 moritz: Ew. Touchy, touchy. You sound grumpy. 21:55
sena_kun while the name is the topic, by the way, just wanted to give an opinion(no, not a holywar one, more like 6.d one). 21:56
timotimo i'm hoping my silly suggestion got a tiny smirk out of moritz
psch ##++
/o\
AlexDaniel joined #6lang :D
timotimo rather than just unhappyness
comborico1611 timotimo, you never answered me question!
sena_kun about github.com/perl6/6.d-prep/tree/mas...lease-name
timotimo excuse me, which one was that?
comborico1611 Would you choose Wall or Larry? 21:57
timotimo it's not for me to decide :P
comborico1611 I hate this Dismorphia. 21:58
El_Che I am pretty sure someone already proposed Gloria
AlexDaniel comborico1611: if we're choosing between these two, then Gloria
comborico1611 You just didn't want to answer!
sena_kun of course, I don't have a years of experience to say something, but still I think that "dismorphia" is 1)sounds like some disease(I know it's not, but still); and 2)it's harder to type. As for another options, I can say that "rakudo"(I know it's the implementation name) is better in my opinion, yup. Just my 5 cents.
Xliff_ El_Che: OOC, now...but the device without much space is my hard drive.
comborico1611 AlexDaniel: that's not very programmer-ish of you.
sena_kun ugh, my grammar is awful as always.
comborico1611 I suggest 6lerp. 21:59
Pronounced Slurp.
Xliff_ On topic: If the name has to be changed, then 6lang works. I still think this is Perl, though.
El_Che Xliff_: I don't keep my music on my laptop :)
AlexDaniel comborico1611: I guess that's from my engineer-ish part :)
Xliff_ There is enough Perl in Rakudo for it to still be Perl.
El_Che: This is my desktop. I have a lot of MP3s. 22:00
El_Che Rakudo is a very cool name
moritz thinks he has worked too long on Perl 6 to accept a changed name
Xliff_ moritz++
sena_kun No.
I don't want a name change.
El_Che moritz: you are allowed to still call it Perl 6
you and sena_kun :)
sena_kun I have a t-shirt with "Perl 6" already. :)
El_Che and probably stmuk
stmuk ha
comborico1611 Remember zoffix's key distinction. It's not a name change, it's just an alias.
El_Che moritz: I would by the 2nd edition of your book, the Rakudo one :) 22:01
moritz yet people still talk about renaming
sena_kun sure
Xliff_ BTW... another OOC.
comborico1611 El_Che: lol
Xliff_ Is there a .transpose equivalent for Arrays?
comborico1611 They talk because they haven't refined their minds enough to promote the alias argument.
Xliff_ Rather, is there a .transpose for Arrays.
moritz Xliff_: [Z] is kinda like transpose 22:02
AlexDaniel sena_kun: but Dismorphia is just a name for the release
Xliff_ moritz: Has a key flaw. I was talking about this with AlexDaniel, last night
moritz m: my @a = [1, 2, 3], ['a', 'b', 'z']; say ([Z] @a).perl
camelia ((1, "a"), (2, "b"), (3, "z")).Seq
Xliff_ m: ([Z] <a b c>).say 22:03
camelia ((a b c))
AlexDaniel sena_kun: soo… why is that a problem? Some time later we'll move to 6.e and it will be Ear-Infection, and everyone will forget about Dismorphia :)
sena_kun AlexDaniel, if it's not an "alias", then I don't mind(though still dislike it for the reasons above). 22:04
Xliff_ moritz: ^^ should be ((a) (b) (c))
stmuk dysmorphia, is a mental health condition where a person spends a lot of time worrying about flaws in their appearance
moritz Xliff_: yes, that sucks 22:05
AlexDaniel sena_kun: that's *just* a language release name. 6.c was *C*hristmas, 6.d is *D*ismorphia
sena_kun though I probably should to address it to Zoffix. because of "Collate all the feedback on the matter in presentable".
stmuk is there one for worrying about names of things? :>
Juerd Was diwali (or however that's spelled) dropped?
AlexDaniel sena_kun: the feedback presentable is about the language name/alias/whatever, not about the 6.d release
moritz let's name it D&D, shall we?
AlexDaniel Juerd: well, not really, but some explanation here: github.com/perl6/6.d-prep/tree/mas...lease-name 22:06
Xliff_ moritz: I think there should be a real Array.transpose that addresses this.
m: ([Z] <a <b c d> ce).say
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in quote words; couldn't find final '>' (corresponding starter was at line 1)
at <tmp>:1
------> 3([Z] <a <b c d> ce).say7⏏5<EOL>
expecting any of:
a…
Xliff_ m: ([Z] <a <b c d> e>).say
camelia ((a <b c d> e))
Juerd AlexDaniel: Thanks for the link 22:07
Xliff_ m: ([Z] ('a', <b c d>, 'e').say
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in parenthesized expression; couldn't find final ')' (corresponding starter was at line 1)
at <tmp>:1
------> 3([Z] ('a', <b c d>, 'e').say7⏏5<EOL>
expecting a…
Xliff_ m: ([Z] ('a', <b c d>, 'e')).say
camelia ((a b e))
sena_kun AlexDaniel, you lost me. :) So what can(should?) I do with my tiny powers to reduce chances that one day I'll wake up and write "Dismorphia" with a marker on my t-shirt?
Xliff_ m: ('a', <b c d>, 'e').say
camelia (a (b c d) e)
sena_kun like, pm someone or do nothing. 22:08
AlexDaniel sena_kun: you can always write v6.d instead
but the fact that dysmorphia is an actual mental disorder… is probably a good argument 22:09
not to even name a release like that, I mean
Juerd AlexDaniel: dysmorphia ne dismorphia though :)
AlexDaniel sure, I see that, but it doesn't help much 22:10
Juerd But it's hardly a significant difference :)
El_Che I like Zoffix' BrokenChicken alias. It could be a good name
Juerd AlexDaniel: Agreed
22:11 khw left
El_Che "I wrote the new app in BrokenChicken, boss" 22:11
Xliff_ s/Broken/Robot/ and it's good.
Might get sued, though
AlexDaniel sena_kun: so “A new spec of 6lang was just released – v6.d “Dismorphia””
version I mean
but I hope you get the point :)
sena_kun e.g. when I duckduckgo "dismorphia" I have "Body dysmorphic disorder - Wikipedia" as a first link. And butterflies are like 4th.
moritz feels like in a Kafka novel 22:12
sena_kun AlexDaniel, I get the point, just my 5 cents.
ok, back to working~
AlexDaniel back to being ill
sena_kun o/
El_Che moritz: which one, mr Samsa?
moritz dunno, long time since I read one 22:13
psch vor dem gericht /o\ 22:14
comborico1611 moritz: "
moritz: "This is pure Kafka. Who's Kafka!? Tell me!" What is this quote from?
AlexDaniel .tell Zoffix Oh well. So turns out that “dysmorphia” (with y instead of i) is actually something bad. Perhaps that argument alone is good enough for getting the Dismorphia suggestion out of v6.d prep docs… irclog.perlgeek.de/perl6/2018-01-19#i_15713129 22:15
yoleaux AlexDaniel: I'll pass your message to Zoffix.
AlexDaniel .tell Zoffix or maybe it's actually good if we give it a positive spin :)
yoleaux AlexDaniel: I'll pass your message to Zoffix.
comborico1611 It's from the 1995 film Congo. 22:17
TEttinger kafka wrote Metamorphosis 22:18
comborico1611 Going to eat supper now. It's Friday! 22:19
timotimo you haven't been embarassed until you claimed during a lightning talk that kant wrote metamorphosis
TEttinger kant stop me now
El_Che timotimo++
comborico1611 lol
22:19 comborico1611 left
TEttinger 6.d Ditto 22:20
ditto, the transforming pokemon
stmuk Dwalin? 22:21
timotimo Dwayne (The Rock) Programming Language
stmuk Dent?
timotimo arthur? 22:22
TEttinger dweomer
stmuk Dent, Arthur, Dent
22:22 abraxxa left, khw joined
Xliff_ Determinant 22:28
timotimo deuteronomy
stmuk Demon Seed 22:29
timotimo Dark Souls
geekosaur deuterostomy
Xliff_ Dance Dance Revolution
TEttinger dinnertime 22:30
dudette
timotimo dinner time is great
sena_kun # debug
TEttinger haha
done. 22:31
stmuk Donald
TEttinger duck.
geekosaur (he hates that.)
TEttinger he hates wearing pants, yes 22:32
22:32 Kaiepi left, pmurias joined
stmuk Dionysus 22:32
TEttinger damocles 22:33
I'd be really surprised if no one has made a reference to the sword of damocles as analogous to potential Russian blackmail of Trump 22:34
stmuk weezil 22:35
Dweezil
El_Che FrankieJr 22:36
22:36 Guest94921 left 22:37 syntaxman joined, syntaxman left, syntaxman joined
timotimo how did we reach F already 22:37
pmurias TEttinger: you mean with not enough recent drama around we need to mix in US politics to liven things up? 22:38
github.com/fglock/Perlito/ # click baiting a alternative Perl 5 implementation that could be grown into a Butterfly Perl 6 22:40
TEttinger neat 22:42
pmurias Perlito5 is interesting in that it's a Perl 5 implementation that has origins as a Perl 6 one 22:46
22:47 Zoffix joined
Zoffix . 22:47
yoleaux 22:15Z <AlexDaniel> Zoffix: Oh well. So turns out that “dysmorphia” (with y instead of i) is actually something bad. Perhaps that argument alone is good enough for getting the Dismorphia suggestion out of v6.d prep docs… irclog.perlgeek.de/perl6/2018-01-19#i_15713129
22:15Z <AlexDaniel> Zoffix: or maybe it's actually good if we give it a positive spin :)
Zoffix AlexDaniel: don't care much about the release name. There's a commit in 6.d-prep somewhere (might've been commit that added Dismorphia) with lots of comments dissing Dismorphia. I bailed out of that discussion but perhaps it had some conclusions 22:48
22:50 espadrine joined
Zoffix is amused that out of all the things in 6.d-prep repo, the only comments Zoffix ever seen is about naming and not any of the techincal issues :P 22:50
timotimo i regret to have to say i'm also not weighing in on the 6.d prep stuff at all :( 22:51
Zoffix The last two are the most contentious I'd imagine: github.com/perl6/6.d-prep/blob/mas...ually-safe 22:53
And github.com/perl6/6.d-prep/blob/mas...s-not-set-
The safecall one might be a bad idea due to perf penalty that'd entail 22:54
Yeah. I'm -1 on it now (if Nil methods are fixed not to try to call Cool candidates, the safecall will be much more usable)
jnthn It's possible we might be able to find a performant way 22:55
Zoffix \o/ 22:57
jnthn I'm in two minds about it from a language design perspective
22:57 cdg joined
jnthn On the one hand we've said the ?, +, and * are about the MRO 22:57
22:57 sortiz joined
jnthn On the other, I agree .? isn't really DWIM in the way anyone would really expect at the moment 22:58
ooh, "Make start blocks in sink context attach an error handler" should be on my todo list probably 22:59
Zoffix :) 23:00
jnthn I consider the non-blocking await work for 6.d largely complete
Supplies have been using the guts that back that internally, even in 6.c 23:01
So the code paths are already getting a good workout
23:03 trnh joined 23:05 Zoffix left
Altreus I wonder whether anyone can suggest some real-world examples of grammars so I can glean some techniques and a bit more knowledge than a contrived tutorial 23:12
I'm hoping it'll answer some of my immediate questions so I don't have to pester :)
stmuk Altreus: parsing any config file or structured data whatever 23:13
Altreus slightly surprised the yaml module I found isn't using a grammar 23:15
which is sort of why I asked for suggestions - I don't know whether the examples I find are good ones or bad ones 23:16
ah yamlish uses one
jnthn Docker::File (I think) uses one
eco Docker::File 23:17
eco: Docker::File
Aww
buggable jnthn, Docker::File 'Parsing and generation of Dockerfiles.': github.com/jnthn/p6-docker-file
jnthn Oh :)
timotimo i'd assume our template languages are implemented with grammars, though they could just as well be search&replace
jnthn Yeah, I did call it Docker::File :)
iirc it's a medium-complexity grammar
Altreus dank 23:20
jnthn bitt
Altreus the yamlish parser uses a dot in <.document-prefix> but yours and the JSON one don't, in <image> for example 23:21
jnthn The . just means "don't capture"
Altreus ah, that's another question I had :)
I was going to write a grammar for a "format" that seems to be a bit random. Sometimes people use proper XML, and sometimes people use something XML-like but apparently with an amount of guesswork on the part of the parser 23:23
Would it make sense for TOP to try XML-like and then just pass the input string to the other grammar to try?
wait
I mean try XML, and then pass it on to the XMLish parser
probably with a warning
23:27 cdg left 23:28 pecastro joined
jnthn Could try something like that 23:29
That or just try writing a very liberal XML parser... :) 23:30
Altreus honestly I don't even know if I can. I think I might have to write format-specific logic to know whether a tag is nested or ended 23:32
jnthn *nod* 23:33
Note that it's possible to embed code conditions and stuff into a grammar
Altreus I don't grok the term "code conditions" 23:36
Do you mean conditions as in requirements or conditions as in contextual situation? 23:37
psch m: my token foo { "a" <?{ rand < .5 ?? note() !! 1 }> "b" }; say ("ab" ~~ &foo) xx 2 23:39
camelia Noted
(「ab」 「ab」)
Noted
psch m: my token foo { "a" <?{ rand < .5 ?? note() !! 1 }> "b" }; say ("ab" ~~ &foo) xx 2
camelia Noted
(「ab」 「ab」)
psch inside the <?{ }> you have $/ iirc?
which is "match so far"
23:41 pilne left 23:44 trnh left, pmurias left 23:45 sena_kun left 23:48 cdg joined 23:49 leont joined 23:52 cdg left 23:53 yoleaux left 23:57 rindolf left