»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.
AlexDaniel gfldex: Inf, but in this case the list is not infinite 00:00
00:00 vendethiel left
gfldex it's not a list, it's a Range 00:00
AlexDaniel right, it's an Array 00:01
m: my @a = <a b c d>; say WHAT @a
camelia rakudo-moar a5fe34: OUTPUT«(Array)␤»
gfldex that's not what I mean
tony-o_ the range you're using to select elems with is an Inf range 00:02
gfldex the array subscript [0..^*] will produce some Iterator that will iterate between 0 and Inf-1
the Iterator only looks at the lower and upper bound that is Range
AlexDaniel m: say 0..^∞
camelia rakudo-moar a5fe34: OUTPUT«0..^Inf␤»
AlexDaniel m: say 0..∞
camelia rakudo-moar a5fe34: OUTPUT«0..Inf␤»
00:03 dfcarpenterak joined
AlexDaniel gfldex: so you are saying that the index of last element is still < Inf-1 so that's why it is included? 00:04
gfldex yes
[Coke] perlpilot: S17-supply/syntax.t has been problematic since pre-xmas.
AlexDaniel kinda makes sense, but it doesn't dwim. If I wrote ^ there then perhaps I did that on purpose
Hotkeys what does the 'global' command for rakudobrew do 00:06
AlexDaniel gfldex: but I'm not going to insist that it should work :)
well, it does work. I mean that I'm not going to insist that it should work the way I want it to work :D 00:07
ooooooooooooooooooh 00:08
I think that I found something…
Hotkeys Doesn't 0..^Inf essentially == 0..Inf
it does in my head
AlexDaniel m: my @a = ‘a’..‘e’; say [@a[[0..$_],[$_^..*]] for ^(@a - 1)] 00:09
camelia rakudo-moar a5fe34: OUTPUT«[((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e))]␤»
AlexDaniel this is significantly shorter
ely-se Hotkeys: yes they should represent the same range
AlexDaniel m: my @a = ‘a’..‘e’; say (@a[(0..$_),($_^..*)] for ^(@a - 1)) 00:10
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
00:11 obarb is now known as brabo 00:12 choroba left
perlawhirl m: my @name = 'obarb'.comb; ~@name.rotate.join 00:13
camelia rakudo-moar a5fe34: OUTPUT«WARNINGS for /tmp/vpFb6C651P:␤Useless use of "~" in expression "~@name.rotate.join" in sink context (line 1)␤»
00:13 gist_nb joined
perlawhirl m: my @name = 'obarb'.comb; say @name.rotate.join 00:13
camelia rakudo-moar a5fe34: OUTPUT«barbo␤»
gfldex where in the docs do we doc List/Array subscripts?
perlawhirl almost :D
AlexDaniel m: my @a = ‘a’..‘e’; say [@a[^$_,$_..*] for 1..^@a]
camelia rakudo-moar a5fe34: OUTPUT«[((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e))]␤» 00:14
perlawhirl gfldex: doc.perl6.org/language/list
AlexDaniel avuserow: ↑ this is probably the shortest one
m: my @a = ‘a’..‘e’; say (@a[^$_,$_..*] for 1..^@a) 00:16
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
00:16 patrickz left
gfldex perlawhirl: that's where we _should_ doc Array slices :-> 00:16
i will close those gaps tomorrow 00:17
AlexDaniel m: ‘obarb’.flip.say 00:18
camelia rakudo-moar a5fe34: OUTPUT«brabo␤»
brabo AlexDaniel: hm? 00:19
AlexDaniel nvm :)
brabo tsk tsk 00:20
waking up random people at this time of day :p
sortiz m: say Inf - 1 == Inf, Inf + 1 == Inf # By definition
camelia rakudo-moar a5fe34: OUTPUT«TrueTrue␤»
brabo lizmat: ping! how was fosdem? 00:21
also hm, did i hear something about perl 6 release a while ago? 00:22
can it truly be? :)
00:22 firstdayonthejob left 00:23 SCHAAP137 left
Hotkeys it's up to spec as of 2015.12 00:23
unless you mean the 2016.01 release
brabo AlexDaniel: j/k btw, it is actually nice someone reminded me i was here ;)
Hotkeys: ye it will be the release i heard about
something about larry's bday or so? 00:24
i guess congrats are in order.. if i understood it correctly it has been a long time coming :)
[Coke] brabo: _the_ release was on hristmas day. 00:27
brabo aha okay 00:28
ye
those weeks were a bit dazed and confused at my end ^^ 00:29
Hotkeys m: say ('a'..'f').rotor(2=>-1) 00:31
camelia rakudo-moar a5fe34: OUTPUT«((a b) (b c) (c d) (d e) (e f))␤»
00:31 spider-mario left
Hotkeys this is fun 00:31
m: say ('a'..'f').rotor(3=>-2)
camelia rakudo-moar a5fe34: OUTPUT«((a b c) (b c d) (c d e) (d e f))␤»
AlexDaniel ooh, maybe it could be solved with rotor 00:32
Hotkeys that's what I was trying to do
I think it'd be longer than your solution up there
AlexDaniel m: my @a = ‘a’..‘e’; say (@a.rotor($_, Inf, :partial) for 1..^+@a) 00:33
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
AlexDaniel m: my @a = ‘a’..‘e’; say (@a[^$_,$_..*] for 1..^@a) 00:34
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
Hotkeys well you're much smarter than I
AlexDaniel Hotkeys: it's not too bad
let's see if we can golf it down…
avuserow: ↑ 00:35
Hotkeys: oh, maybe it could be done with rotor but without for loop…
Hotkeys does this count as golfing 00:36
AlexDaniel yes
Hotkeys m: my @a = ‘a’..‘e’; say (@a.rotor($_, ∞, :partial) for 1..^+@a)
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
Hotkeys :p
AlexDaniel yes :)
00:36 Guest7721 left
Hotkeys just barely shorter 00:38
m: my @a = ‘a’..‘e’; (1..^+@a).map({@a.rotor($_, ∞, :partial)})
camelia ( no output )
Hotkeys oh
forgot the say
stupid repl
m: my @a = ‘a’..‘e’; (1..^+@a).map({@a.rotor($_, ∞, :partial)}).say
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
AlexDaniel it's not shorter this way? :)
Hotkeys not with say
couple characters longer
I guess if we're golfing we can remove spaces though 00:39
m: my @a=‘a’..‘e’;(1..^+@a).map({@a.rotor($_,∞,:partial)}).say
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
AlexDaniel I'm confident that you can just make it with rotor alone
one second
Hotkeys maybe
00:40 lostinfog left
AlexDaniel Hotkeys: I'll show you what I mean 00:41
m: my @a = ‘a’..‘e’; say @a.rotor(1, 4 => -5, 2, 3 => -5, 3, 2)
camelia rakudo-moar a5fe34: OUTPUT«((a) (b c d e) (a b) (c d e) (a b c) (d e))␤»
AlexDaniel now all you have to do is just generate these params
Hotkeys oh 00:42
I was thinking you wanted them grouped together like you had
AlexDaniel Hotkeys: yes but that can be done later, I think…
revhippie like a sequence of splices
Hotkeys sure
AlexDaniel Hotkeys: with another rotor! 00:43
00:43 lichtkind left
AlexDaniel m: my @a = ‘a’..‘e’; say @a.rotor(1, 4 => -5, 2, 3 => -5, 3, 2).rotor(2) 00:43
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)))␤»
00:44 vendethiel joined
AlexDaniel how do I pass an array as arguments? 00:46
00:47 wamba left
Hotkeys how do you mean? 00:48
AlexDaniel rotor(@args)
? 00:49
Hotkeys you slap a | on it
iirc
rotor(|@args) 00:50
AlexDaniel right
00:51 ely-se left
AlexDaniel m: my @a = ‘a’..‘e’; say @a.rotor(flat ($_, 5-$_ => -5 for 1..^5)).rotor: 2 00:52
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)) ((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)) ((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)) ((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e))…»
AlexDaniel m: my @a = ‘a’..‘e’; say @a.rotor(flat ($_, @a-$_ => -@a for 1..^+@a)).rotor: 2
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)) ((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)) ((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)) ((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e))…»
AlexDaniel how do I limit it?
OK 00:53
m: my @a = ‘a’..‘e’; say @a.rotor(|($_, @a-$_ => -@a for 1..^+@a), ∞).rotor: 2
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
AlexDaniel there we go!
avuserow: ↑ 00:54
Hotkeys: you can actually omit + in ^+@a 00:55
Hotkeys ah
AlexDaniel m: my @a = ‘a’..‘e’; say (@a.rotor($_, ∞, :partial) for 1..^@a) 00:56
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
00:57 tmtowtdi left
AlexDaniel Hotkeys: now the question is how to get rid of 「for」 and do the same thing with Z 00:57
m: my @a = ‘a’..‘e’; say @a.rotor(|(1..^@a Z ((+@a-1 … 1) Z=> (-@a, -@a … *))), ∞).rotor: 2 01:03
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
AlexDaniel Hotkeys: easy!
well, I'm sure that there is an easier way…
like, one of the hundred of other suggested alternatives :)
01:05 vendethiel left
AlexDaniel Hotkeys: thanks for mentioning rotor, I didn't even think about it 01:05
Hotkeys: we can also do it without :partial 01:10
m: my @a = ‘a’..‘e’; say (@a.rotor: $_, @a-$_ for 1..^@a).rotor: 2
camelia rakudo-moar a5fe34: OUTPUT«((((a) (b c d e)) ((a b) (c d e))) (((a b c) (d e)) ((a b c d) (e))))␤»
AlexDaniel Hotkeys: saves a character or two
anyway, enough of this golfing spam :) 01:11
Hotkeys lol
shortened it a bit a think 01:14
the non-for one
m: my @a = ‘a’..‘e’; say @a.rotor(|(1..^@a Z (@a-1...0) »=>» -@a), ∞).rotor: 2
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
Hotkeys AlexDaniel: 01:15
AlexDaniel oh, how interesting
Hotkeys m: my @a = ‘a’..‘e’; say @a.rotor(|(1..^@a Z (@a-1…0) »=>» -@a), ∞).rotor: 2
camelia rakudo-moar a5fe34: OUTPUT«(((a) (b c d e)) ((a b) (c d e)) ((a b c) (d e)) ((a b c d) (e)))␤»
Hotkeys there got that sequence char in there
AlexDaniel not too bad
okay I think that I have to make a gist with all the solutions 01:16
Hotkeys lol
what is this for anyway
AlexDaniel Hotkeys: someone asked a question
Hotkeys ah
AlexDaniel and we answered it… a couple of times :)
Hotkeys I must've missed it
perlawhirl haha, yes... i assume whatever reason avuserow needed it for has passed. 01:19
but it's a fun problem to try to solve many ways with perl6
AlexDaniel yeah, I enjoyed that 01:20
01:25 dfcarpenterak left, addison_ left 01:26 sevvie joined 01:27 vendethiel joined 01:35 yeahnoob joined 01:38 sno left 01:40 ELBeavers left 01:50 vendethiel left 01:54 flo__ left
Hotkeys I do stackexchange codegolf 01:55
sometimes
with perl 6
I meant to say that all in one line whoops
02:01 leont left 02:14 cuonglm joined
cuonglm Anyone has faced problem with building rakudo from latest master 02:14
I got "Stage mast : No registered operation handler for 'stat_time'" 02:15
japhb cuonglm: master? not nom? 02:18
Or are you talking about MoarVM's branch? 02:19
(Normally you let Rakudo's configure choose the proper commit to build NQP, which then chooses the proper MoarVM commit.) 02:20
cuonglm japhb: it's master
I simply cd to rakudo source
git pull 02:21
perl Configure.pl --gen-moar --gen-nqp --backends=moar --prefix=/usr/local/stow/perl6
make
then it failed in mast stage
You can see full error here paste.debian.net/378252/ 02:23
japhb cuonglm: What does `git describe --all` output?
cuonglm Opps, it output "heads/nom" 02:25
japhb That's actually a good thing. I was worried you'd at some point in the past manually put yourself on master. :-) 02:26
I remember something about stat_time being undone on the Moar side, but I thought that had been fixed. Don't remember the details,
but you could check the MoarVM commit log for the last couple weeks
japhb must run
02:27 yurivish joined 02:28 molaf__ left
skids counglm: try removing moar and nqp binaries from the install path maybe? 02:29
02:31 kid51 left
cuonglm skids: I'm trying, it seems to work 02:32
it gets latest moar and nqp then building them again 02:33
02:33 labster left 02:34 labster joined 02:38 sevvie left
cuonglm Thanks all, work perfect now 02:39
skids np enjoy :-) 02:40
02:49 cpage_ left 02:53 cuonglm left, johan_ joined 02:54 johan left, gist_nb left, cpage_ joined 02:57 maybekoo left 03:00 johan_ left 03:02 soundobstacle left 03:05 vendethiel joined, johan_ joined 03:08 Actualeyes left 03:11 yqt left, jojotus left
AlexDaniel ok, there we go: gist.github.com/AlexDaniel/1e2d1c50963d37c5d43a 03:12
Hotkeys: ↑
03:12 jojotus joined
AlexDaniel I've attempted to make them all as short as possible, but there are probably some ways to golfify some of those 03:12
and yeah, I did not remove the spaces around operators. I think that it's more fair this way 03:13
(i.e. you're not going to use obfuscated code)
03:16 alpha123 left 03:20 sno joined 03:26 vendethiel left 03:30 alpha123 joined 03:32 noganex_ joined 03:35 noganex left 03:37 Snowdog joined
Snowdog Is anyone here? 03:38
colomon yes?
Snowdog I noticed that today, a new version of Perl6 came out.
colomon right
usually monthly release
…. errr, though it’s not the usual date for it, now that I think about it. 03:39
colomon has been a bit out of the loop
Snowdog However, when I build it with Rakudobrew, it sets it up as a separate installation from the 2015-12 version, and now all my libraries have to be reinstalled.
That's not really a viable option. Is there any way I can report this?
colomon Snowdog: github.com/tadzik/rakudobrew/issues 03:40
Snowdog An upgrade shouldn't require a new installation.
Thanks.
colomon I’m not quite sure what you did? what command did you use to build it? 03:41
Snowdog I think everyone is still in testing mode, but if the software is going to be used, then it has to be trusted.
I can build the new installation in2 ways:
jdv79 isn't that how perlbrew works? perl and libs are seperated on perl ver?
Snowdog 1) rakudobrew build moar or rakudobrew build moar 2016.01.1
revhippie i think that's kind of the point, yeah 03:42
Snowdog As instructed on the perl6 page: www.perl6.org
colomon if you built rakudobrew build moar before and rakudobrew build moar again today, the libraries should get reinstalled automatically.
jdv79 its not an "upgrade". its a new install.
Snowdog Unfortunately, no.
Right, it's a new install.
Will 2015-12 ever be upgraded? 03:43
My only point is that it can't be used in a production environment if I have to reinstall and test all the libraries each time it upgrades.
That's all. I'll report it. 03:44
jdv79 that's how perlbrew and rakudobrew work though.
colomon it would be handy to have a way of bundling the standardl libraries you install. I’m not sure that should be rakudobrew’s responsbility.
Snowdog Really? Each new build requires a total reconfiguration?
colomon Snowdog: each new installation does
jdv79 perhaps in the future it'll be made more smooth 03:45
colomon the whole point is to let you easily switch between different installs of rakudo.
Snowdog OK, thanks. That's just frustrating. I know it's a new language, but I thought it was ready for release.
jdv79 that sort of stuff is very immature at the moment. it can only get better over time.
geekosaur while there has been an official release, things are still changing fairly rapidly
especially in the area of packaging/distribution 03:46
Snowdog I think it's a cool language. I'm looking forward to using it.
jdv79 :( its "ready"; not battle tested. sorry.
colomon but I wouldn’t expect rakudobrew to change in the way Snowdog wants? isn’t the current behavior a feature, not a bug?
maybe there should be a “release” tag (or something like that) for rakudobrew which treats the latest releases over time as a single install? 03:48
jdv79 perhaps a feature could be added to "mirror/port/re-install" one perl6's libs to another but its working as expected right now afaik
Snowdog Aye!
03:49 xtreak joined, azawawi joined
xtreak Hi I am a perl newbie. I was learning grammars. They seem very powerful. I was trying to use it to parse my toy language. 03:50
AlexDaniel xtreak: welcome! 03:51
xtreak How can I use it throw errors when there is an unescaped quote in the string like 'hello' i am good'
Is it possible or am I doing it wrong. Sorry if its too naive. 03:52
jdv79 no idea. maybe look at the Perl6 grammar itself or wait til someone who knows wakes up to see your question and if nobody answers in a few hours ask again. 03:54
xtreak Thanks thats a gud idea. Is the perl6 parser written in Perl6 or Perl5? I don't have much experience on C. 03:55
03:57 ZoffixWin joined, ZoffixWin left, ZoffixWin joined
skids xtreak: The parser is in Perl 6 03:57
ZoffixWin .tell [Coke] The announcement is great news, but it really should include a download link :/ RE: blogs.perl.org/users/coke/2016/02/r...eased.html
yoleaux ZoffixWin: I'll pass your message to [Coke].
ZoffixWin Or how to get it... 03:58
jdv79 xtreak: iirc the parser is probably src/Perl6/Grammar.nqp 03:59
also see Actions.nqp
nqp is just a Perl6 subset (Not Quite Perl) 04:00
xtreak Sure thanks. I hope it does have instructions to compile it again so that I can change and see the effects.
ZoffixWin At the risk of sounding like a grumpy old man.. I have no clue at all how to get the freshly released Rakudo. I see no tag in the repo and rakudo.org/ only has leads to 2015.11 R*
jdv79 xtreak: docs aren't great yet. best to try something and just ask in here when it doesn't work out. 04:01
ZoffixWin And I hang out here and use Perl 6 and have contributed to the Ecosystem... What would a new person make out of this?
jdv79 ZoffixWin: still keepin it real. nice. 04:02
perhaps a pr person wouldn't be a bad idea. what does the tpf do again?
:)
perlawhirl AlexDaniel: just saw you gist... nice work 04:08
AlexDaniel :)
Hotkeys ++ 04:10
AlexDaniel actually, I really liked the last one 04:11
I haven't seen [\+] thing earlier
Hotkeys ZoffixWin: github.com/rakudo/rakudo/releases
they're there
AlexDaniel how is it called? What does it do exactly?
xtreak I think [\+] prints the intermediate results in the reduction process 04:13
Hotkeys m: [\+] 1..10
camelia ( no output )
Hotkeys m: say [\+] 1..10
camelia rakudo-moar a5fe34: OUTPUT«(1 3 6 10 15 21 28 36 45 55)␤»
Hotkeys seems like you're right xtreak
or rather not print but 04:14
creates a list of all the intermediary results
xtreak Yes :)
I think most of the reduction operators can be made like this by having a \ prefixed to them 04:15
Hotkeys "By default reduction meta operators are eager. To lazily generate values, prefix the operator with a \. If the non-meta part contains a \ already, quote it with [] (e.g. [\[\x]])."
m: my $lazy := [\+] 1..*; say $lazy[^10] 04:16
camelia rakudo-moar a5fe34: OUTPUT«(1 3 6 10 15 21 28 36 45 55)␤»
Hotkeys m: my $lazy := [\+] 1..*; say $lazy[^20]
camelia rakudo-moar a5fe34: OUTPUT«(1 3 6 10 15 21 28 36 45 55 66 78 91 105 120 136 153 171 190 210)␤»
Hotkeys pretty neat
xtreak friedo.com/blog/2016/01/exploring-p...-operators 04:17
p6: my $lazy := [\[\+]] 1..*; say $lazy[^20] 04:18
camelia rakudo-moar a5fe34: OUTPUT«5===SORRY!5=== Error while compiling /tmp/tmpfile␤Two terms in a row␤at /tmp/tmpfile:1␤------> 3my $lazy := [\[\+]]7⏏5 1..*; say $lazy[^20]␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ statemen…»
xtreak Any example of how to use the operations that already contain a / at the beginning? 04:19
Sorry \ 04:20
04:20 samb1 left
ZoffixWin Hotkeys, thanks. Do I still get to grumble about the release process or did I use up all of my allocated points for today? 04:21
jdv79, not sure whether that was sarcasm, but being more organized about the release process and at least faking excitement about new releases would help Perl 6's adoption greatly IMO. 04:22
jdv79 no sarcasm here. couldn't the tpf or someone else do that for us? 04:25
ZoffixWin Do we have anyone from TPF in the channel? If not, I think they're too far away to do this job efficiently. 04:26
AlexDaniel .tell avuserow gist.github.com/AlexDaniel/1e2d1c50963d37c5d43a
yoleaux AlexDaniel: I'll pass your message to avuserow.
jdv79 idk
04:27 nchambers is now known as ncampers
ZoffixWin TBH, I'm more picturing a documented release process that will include all the stuff that needs to be done: include download links, post on blogs.perl.org, Twitter, Facebook, Reddit, HackerNews, other popular sites. Include a detailed-enough changelog to make it easier for a handful of bloggers to do a quick write up and spread the news. 04:27
jdv79 we don't have that? what do they do now for a release? 04:28
ZoffixWin is unsure.
jdv79 i know there's a script or scripts somwehere that are followed 04:29
at least one of moritz or [Coke] or FROGGS[mobile] knows about that
ZoffixWin My feedback is only that this looks more like something I'd email my boss when he asked me what I've been doing for the last month: blogs.perl.org/users/coke/2016/02/r...eased.html 04:30
Anyway. Night all :)
04:31 AlexDaniel left
jdv79 peace 04:31
04:32 samb1 joined
jdv79 neat that [Coke] mentioned the nyc meeting 04:32
revhippie it just occurred to me that in that gist from earlier ( gist.github.com/AlexDaniel/1e2d1c50963d37c5d43a ) that 'for' in several of them is part of the expression within the 'say' 04:33
04:38 vendethiel joined 04:41 MadcapJake joined
MadcapJake Great news everyone! github.com/github/linguist/pull/2832 04:41
04:44 xtreak left
skids Oh yay! 04:44
04:46 xtreak joined 04:48 ncampers is now known as nchambers, nchambers is now known as dtscode 04:52 vendethiel left 04:54 vendethiel joined, Actualeyes joined 04:56 skids left 04:58 revhippie left 05:00 vendethiel left 05:02 dtscode left, dtscode joined 05:03 dtscode left, dtscode joined 05:09 DarthGandalf left
timotimo MadcapJake: while great in theory, now the real work begins. waiting for the PR to get merged. likely takes about 1-2 years 05:09
05:09 xinming_ left 05:11 DarthGandalf joined
MadcapJake hahaha 05:11
timotimo seriously, though 05:13
MadcapJake timotimo: issuestats.com/github/github/linguist
timotimo the last time we tried to get better perl6 highlighting ...
MadcapJake oh in language-perl ?
timotimo i don't think so 05:14
MadcapJake Yeah when I started language-perl6fe, I knew there was one in language-perl, but I didn't want to be hindered by waiting for PRs to merge xD
timotimo github.com/github/linguist/pull/900
MadcapJake woah 05:15
timotimo github.com/github/linguist/pull/750 - not as bad
so yeah. we've been burned in the past 05:16
i have no intent on saying "i'm sure they'll merge it today!" for the rest of my life. OP will surely deliver, and all that. 05:17
MadcapJake wow, well the nice part of the PR I ref'd, it's just highlighting, only adds a submodule and a few reference changes to the selectors that I use
timotimo "only adds a submodule" sounds pretty fucking bad :)
MadcapJake lol
05:17 khw left
MadcapJake well that appears to be there modus operandi for language support 05:18
timotimo what, wait for a few years to see if it's worth it to click the "merge PR" button?
MadcapJake no submodules :P 05:19
github.com/github/linguist/tree/ma...r/grammars
speaking of... 05:20
05:21 laz78 joined, BenGoldberg left 05:22 azawawi left
MadcapJake had to make sure they merge the latest 1.5.0 highlighter (which I pushed after he made his PR). Oh lovely submodules! 05:22
05:24 xtreak left
MadcapJake timotimo: just figured out that Slack snippets uses CodeMirror for highlighting, gonna take a crack at writing a Perl 6 "mode" 05:24
timotimo i saw that, yeah :)
we don't have a codemirror mode yet? 05:25
MadcapJake not in the main project, i believe
looks like azawawi has one 05:26
timotimo nice
MadcapJake oops, dead link
timotimo uh oh :(
MadcapJake found it! 05:27
05:28 xtreak joined
MadcapJake github.com/azawawi/farabi6/blob/1d...l6-mode.js 05:28
timotimo oh, of course! farabi!
MadcapJake oops, why does github sometimes bring you to a commit instead of master?
timotimo good question
MadcapJake only happens if you click a filename in search 05:29
timotimo ah, that could be
MadcapJake is there an estabalished mime type for perl6? «text/x-perl6» ? 05:35
05:37 johan joined 05:38 johan_ left
MadcapJake i'm just gonna assume that's it, only other possibility i could see is «text/x-perl-6» 05:38
05:38 Cabanossi left 05:40 Cabanossi joined
timotimo i'd be for "this is perl6 program code" 05:42
actually, make that a space between perl and 6, and make it a " "
05:43 dtscode is now known as 64MAAYGZ0
MadcapJake mime types can have spaces? 05:43
05:44 64MAAYGZ0 is now known as mniip
timotimo no 05:44
timotimo bbl 05:45
05:45 mniip left 05:46 johan left, nchambers joined
MadcapJake timotimo: haha it was just merged, it will go live later this week github.com/github/linguist/pull/2832 05:49
05:51 johan joined 05:55 jferrero joined 05:58 nchambers is now known as I, I is now known as Guest72381 05:59 _Gustaf_ joined, Guest72381 is now known as nchambers 06:02 xtreak left
sammers p6: use JSON::Tiny; my %h = a => '1', c => '3', b => '2'; to-json(%h).say; to-json(%h.sort).say; 06:08
camelia rakudo-moar a5fe34: OUTPUT«===SORRY!===␤Could not find JSON::Tiny in:␤ /home/camelia/.perl6/2015.12-232-ga5fe347␤ /home/camelia/rakudo-m-inst-1/share/perl6/site␤ /home/camelia/rakudo-m-inst-1/share/perl6/vendor␤ /home/camelia/rakudo-m-inst-1/share/perl6␤ C…»
sammers hmm
having an issue with this. after I sort a hash JSON::* is nesting the sorted hash inside a new [];
like this, pre-sort: { "a" : "1", "c" : "3", "b" : "2" } 06:09
after sort: [ { "a" : "1" }, { "b" : "2" }, { "c" : "3" } ]
it is also grouping each pair into its own object 06:10
06:12 nightfrog left
sammers p6: my %h = a => '1', c => '3', b => '2'; %h.say; %h.sort.say; 06:12
camelia rakudo-moar a5fe34: OUTPUT«a => 1, b => 2, c => 3␤(a => 1 b => 2 c => 3)␤»
06:12 rodarmor left
sammers hmm 06:12
_Gustaf_ o/ 06:16
sammers p6: my %h = a => '1', c => '3', b => '2'; %h.say; %h.sort.hash.say;
camelia rakudo-moar a5fe34: OUTPUT«a => 1, b => 2, c => 3␤a => 1, b => 2, c => 3␤»
06:19 rodarmor joined 06:28 cpage_ left 06:29 sjoshi joined 06:36 psy_ left 06:38 sno left 06:44 geraud left 06:47 Sgeo left, nightfrog joined 06:54 CurtisOvidPoe_ joined 06:57 vendethiel joined 07:00 domidumont joined 07:03 xinming joined, Sgeo joined 07:04 alpha123 left 07:05 domidumont left, domidumont joined 07:11 psy_ joined 07:14 perlawhirl left 07:15 zakharyas joined 07:16 CurtisOvidPoe_ left 07:17 CIAvash joined 07:19 FROGGS[mobile]2 joined 07:20 vendethiel left, flaviusb left 07:23 FROGGS[mobile] left
stmuk_ hmmm I think we should stop promoting rakudobrew as a way of n00bz getting rakudo once R* is out 07:23
07:25 vendethiel joined 07:30 nakiro joined 07:40 flaviusb joined 07:41 jake__ joined 07:42 darutoko joined 07:45 firstdayonthejob joined, jake__ left 07:46 vendethiel left 07:53 vendethiel joined 07:56 FROGGS joined 07:57 Actualeyes left 08:00 virtualsue joined, sno joined 08:01 [Tux] left 08:02 [Tux] joined
timotimo maybe 08:09
especially since we practically dragged it kicking and screaming into the spotlight to begin with :)
08:10 Ven joined 08:11 fireartist joined 08:12 digidog joined
stmuk_ true 08:17
08:19 zakharyas1 joined, zakharyas left
timotimo we probably ought to update perl6.org's top-of-the-page thingie with the latest release, too 08:22
stmuk_ I wondered about that but it might be less confusing for outsiders to just put the R* accouncement there 08:25
only R* announcements went on rakudo.org/ 08:26
its all very confusing
moritz right, let's wait for R* 08:27
or, don't wait, advance it
everybody can help! Write a release announcement!
stmuk_ I even wonder if one of either rakudo or star releases should go now we are in release phase .. monthly seems more dev than stable
moritz the star releases weren't meant to be monthly anyway 08:28
08:29 sjoshi left 08:30 sjoshi joined
stmuk_ perl5 itself has yearly stable releases 08:30
moritz I don't think we're that mature yet 08:31
stmuk_ I can start on star's ://rakudo.org/
[08:26] < stmuk_> its all very confusing
oops
08:31 FROGGS left
stmuk_ sure maybe 3 monthly releases of something at this stage 08:31
I mean to type I can start on R*'s docs/announce/2016.01.md shortly if it helps 08:32
timotimo that sounds fantastic, actually 08:33
man, the track records for "lying in bed, suddenly an optimization opportunity pops into my head" is really bad by now ... just measured how many slots in NFA's state arrays end up "wasted" after the NFA optimizer is done with it, and it's basically nothing 08:34
(because our VMArrays don't shrink their storage when they get smaller) 08:35
08:35 [Tux] left 08:36 [Tux] joined, xpen joined 08:37 vendethiel left 08:39 firstdayonthejob left, [Tux] left, [Tux] joined 08:44 xpen left 08:45 Actualeyes joined 08:50 rindolf joined 08:51 Kogurr left 08:56 wamba joined 08:57 abraxxa joined 08:59 CurtisOvidPoe_ joined 09:02 laz78 left 09:06 jonathan_ joined 09:07 jonathan_ is now known as RabidGravy, labster left 09:10 CurtisOvidPoe_ left, dakkar_ joined 09:12 labster joined
RabidGravy boom 09:12
09:15 Averna joined, abraxxa left
RabidGravy ls 09:15
moritz no such file or directory 09:17
09:22 rvchangue left 09:24 rvchangue joined
sjn no: command not found 09:27
DrForr Just a note - I got an idea on the way home from Brussels last night about how to properly do Haskell Lenses in Perl 6.
09:30 mm_ joined
[Tux] test 22.819 09:30
test-t 12.251
csv-parser 51.941
09:30 labster left 09:32 labster joined 09:33 labster left, mm_ left 09:34 abraxxa joined
RabidGravy DrForr, I may have to re-read that later, it didn't make sense to me prior to coffee 09:36
DrForr Well, it's something like action-at-a-distance which means some people won't like it, and that's fine. The writeup isn't all that great and was done just before I ran out the door, which may explain partially why it was so hard to read. 09:38
09:38 brrt joined, bioexpress joined 09:40 leont joined
bioexpress p6: my Int $c = 2 == 2; say $c.WHAT; 09:41
camelia rakudo-moar a5fe34: OUTPUT«(Bool)␤»
rindolf So I have not received a message from [email@hidden.address] since 2014 and now emailing [email@hidden.address] did not give me a reply.
bioexpress 'p6: my Int $c = 2 == 2; say $c.WHAT;'
moritz p6: my Int $c = 2 == 2; say $c.WHAT 09:42
camelia rakudo-moar a5fe34: OUTPUT«(Bool)␤»
bioexpress Why do I not get an error-message with this code?
moritz m: say Bool ~~ Int 09:43
camelia rakudo-moar a5fe34: OUTPUT«True␤»
moritz that's why
bioexpress Thx!
09:46 yeahnoob left 09:53 ely-se joined 09:55 ab6tract joined
ab6tract .seen hoelzro 09:55
yoleaux I saw hoelzro 1 Feb 2016 16:24Z in #perl6: <hoelzro> Skarsnik: it's definitely weird; I wouldn't be against a warning
09:55 virtualsue left
brrt fwiw, my 2c about the perl6-slack-channel thing 09:58
RabidGravy ab6tract, did you see github.com/jonathanstowe/Tinky/blo...code-files ? You inspired me :)
this is almost the use case I had for Tinky in the first place 09:59
brrt folks are obviously perfectly free to make a slack channel for perl6 :-)
RabidGravy (or one of them)
brrt however, i can see it become problematic when e.g. the slack channel becomes the primary channel for newcomers, and none of the core / ecosystem developers of perl6 are there
that would be irresponsible to those newcomers 10:00
RabidGravy well considering there are 344 people on the channel atm it's still quite manageable
brrt the slack channel? 10:01
RabidGravy this
brrt right :-)
DrForr I assume you're not referring to the SubGenius definition, though the app may have been inspired by "Bob"? 10:02
(I.E. no idea what the app/website is, and am curious.) 10:03
ab6tract RabidGravy: nice! 10:04
i'll give it a try later, maybe it will actually work on OS X this time :)
Slack-- -- -- -- 10:05
the job is transitioning to that crap, terrible
RabidGravy I can't make any guarantees about that but I was throwing whole directories of wav files at it at once and it was happy with that
10:05 Zero_Dogg left
moritz at $work, we're evaluating hipchat right now, because Slack isn't available on premise (iirc) 10:05
10:06 Zero_Dogg joined, Zero_Dogg left, Zero_Dogg joined
DrForr Ah, so it's another IRC wannabee. 10:06
10:06 bioexpress left 10:07 zakharyas1 left
moritz we're using Jabber right now 10:08
ab6tract moritz: ooc, what are the operational issues with jabber?
brrt DrForr: yes, slack is irc for non-nerds
timotimo brrt: i hang out on the slack channel, fwiw 10:09
moritz ab6tract: I'm not really sure. But history on the server doesn't seem to work too well
brrt that's something :-)
timotimo i am something! :)
10:09 vendethiel joined
brrt your hanging-out is something 10:09
10:09 zengargoyle joined
brrt jabber == xmpp right? 10:09
timotimo yes 10:10
ab6tract moritz: ah, i see. but with slack, there's a company that gets to keep those logs instead
seems like a terrible tradeoff.
not speaking to the decisions of your company, just venting about the situation at mine :)
10:11 cognominal joined
moritz ab6tract: we're evaluating hipchat (on premise), not slack 10:11
zengargoyle really wishes xmpp wasn't so complicated and had taken off better
ab6tract oh right, sorry for projecting :) 10:12
zengargoyle: at one point it had the largest market share of chat by far. Google Chat and Facebook were both using it
i can't say much to the complexity, but it seems more likely to me that it died because it wasn't as easy to build a silo around, which is the future that the big players all want for the web 10:13
zengargoyle i was an early fan, setup an ejabberd server for work that has been rock solid for many years now.
RabidGravy I think they both still do, massively extended of course
zengargoyle but lack of easy clients was a bit of a pain. 10:14
moritz somehow $work seems to be unable to operate a Jabber server reliably
ab6tract zengargoyle: yeah, the whole libpurple stack feels quite crufty
moritz despite having several good Linux folks 10:15
10:15 Ven left
moritz there's hardly a week without a full disconnect, and pidgin doens't seem to be able to auto-reconnect after that happens 10:15
DrForr Booking relies on Jabber, we rely on Skype here. 10:20
10:20 espadrine left
brrt that sucks 10:20
(unreliable connections)
RabidGravy I worked at a very large financial institution where all the developers used IRC 10:21
10:23 TEttinger left
timotimo libpurple is a really great Remote Procedure Call solution 10:24
though maybe that's not as true nowadays any more 10:25
flussence is mostly happy running Prosody as an XMPP server; the majority of others I've tried are awful enterprise code 10:27
nine moritz: do you know which jabber server you use? 10:29
DrForr Ah, yet another social media stream to feed.
flussence (I say “mostly”, because its SSL module has too many layers and they're all hardcoded to use openssl) 10:30
10:30 vendethiel left
leont Openssl is awful, but sadly so are the alternatives 10:30
flussence yep, though I do find gnutls's command line tools a hell of a lot more readable 10:31
leont From what I head, the API is full with fuck-ups (e.g. pointer to buffer arguments without an explicit length) 10:32
DrForr Huh. Their website only lets me create a new "team" or find one that I've already joined even though I've never used Slack before. 10:35
brrt DrForr: slack is not for folks like us 10:36
whereby 'us' is defined as 'people who can manage irc'
DrForr Seemed to be another Campfire startup thing.
brrt :-)
it is
the difference is that campfire didn't have $x billion UNICORN tagged to it, and slack does 10:37
also, 'going to kill e-mail'
nothing will kill e-mail folks
10:37 vendethiel joined
DrForr I was just trying to figure out how to find the Perl 6 community over there, and ... failed. 10:37
timotimo DrForr: you'll go via our "invitation page"
perl6.bestforever.com/ 10:38
DrForr Ah.
timotimo it's Quite Different from IRC, yeah
10:40 leont left
DrForr First time a site's told me my usual password is 'weak'. Huh. 10:40
10:40 ely-se left 10:47 araujo joined, araujo left, araujo joined 10:48 araujo left, araujo joined 10:49 kurahaupo joined, Actualeyes left, kanishka joined
DrForr Right, snoozefest over there at the moment apparently. 10:49
ab6tract RabidGravy: Tinky looks quite interesting. Thank you for the concrete example!
10:50 araujo left, araujo joined
timotimo are you keeping up with camelia? 'cause camelia is keeping up with you! 10:51
10:52 araujo left 10:53 araujo joined
DrForr Those who do not know of IRC are doomed to reinvent it poorly. 10:53
10:54 espadrine joined, araujo left 10:55 araujo joined
DrForr Aah, figured it out. Mumble. 10:55
timotimo mumble is a good thing
wait what, you're usung "your usual password" on multiple sites? :) 10:56
DrForr Usual throwaway.
10:57 araujo left, araujo joined 10:59 araujo left, araujo joined, araujo left, araujo joined 11:01 araujo left 11:02 brrt left, araujo joined 11:03 araujo left 11:04 araujo joined, araujo left, araujo joined 11:05 domidumont left, araujo left
dalek ar: b59cefe | (Steve Mynott)++ | docs/announce/2016.01.md:
import of Rakudo Star 2016.01 announce
11:07
ar: e4666ae | (Steve Mynott)++ | README:
doc that JVM doesn't actually work in main README
stmuk_ moritz: that ^^^
11:08 araujo joined 11:09 araujo left, araujo joined 11:10 domidumont joined 11:11 kurahaupo left, araujo left 11:12 araujo joined 11:13 kurahaupo joined, Ven joined 11:14 Actualeyes joined
stmuk_ oops a broken link 11:14
timotimo do all "Perl 6" have the proper evil unicode " " in between? :) 11:15
stmuk_ github.com/rakudo/rakudo/blob/nom/...2016.01.md as listed doesn't actually exist yet .. it may exist in the future depending on merges to nom 11:16
timotimo: patches welcome :)
timotimo i don't have anything nice to make the distinction visible :|
for the purpose of editing
11:18 ab6tract left 11:19 kurahaupo left
dalek ar: c7323cd | (Steve Mynott)++ | docs/announce/2016.01.md:
fix link to exist now rather than maybe in the future
11:20
11:20 vendethiel left
dalek ar: 78b4670 | (Steve Mynott)++ | docs/announce/2016.01.md:
more than one perl tutorial included now
11:23
11:24 ely-se joined 11:26 vendethiel joined
moritz stmuk_: thanks 11:32
11:47 vendethiel left 11:54 FROGGS joined 11:56 maybekoo joined 11:57 sortiz left 12:00 brrt joined 12:25 Ven left, vendethiel joined 12:30 fireartist left 12:31 SunilJoshi joined
dalek ar: 153abec | (Steve Mynott)++ | README:
bump some more years
12:32
12:33 digidog_ joined 12:34 sjoshi left, digidog left, mkz left, mkz_ joined
dalek ar: 3db4a0d | (Steve Mynott)++ | README:
Change 'is primary author' to 'was originally primary author'
12:34
12:38 Ven joined 12:39 kid51 joined 12:42 Averna left
cognominal m: my %a = 'a'..'f'; %a = Nil; # "Assigning to any entire composite container empties the container" says S02 12:42
camelia rakudo-moar a5fe34: OUTPUT«Odd number of elements found where hash initializer expected␤ in block <unit> at /tmp/Er2OxDzvSb line 1␤␤»
12:42 BenGoldberg joined
cognominal is this a fossil or is this an unimplemented feature? I suppose it would have the merit to avoid to create an empty hash just to assign it. 12:43
12:45 ab6tract joined
ab6tract cognominal: the error is in your first hash assignment 12:45
gfldex cognominal: i doubt that would be a merrit. Depending on how the memory layout of a hash is structured it can be way faster to free the whole thing and create a new empty hash.
ab6tract m: my %h = 'a'..'f' Z=> True; say %h; %h = Nil; say %h
camelia rakudo-moar a5fe34: OUTPUT«a => True␤Odd number of elements found where hash initializer expected␤ in block <unit> at /tmp/LsKKFVbRR1 line 1␤␤»
ab6tract hmm, or not :) 12:46
12:46 vendethiel left
jnthn m: my %h = 'a'..'f' Z=> True; say %h; %h = Empty; say %h 12:46
camelia rakudo-moar a5fe34: OUTPUT«a => True␤␤»
jnthn Nil stopped meaning empty list a good while ago :) 12:47
cognominal so, that's a fossil
jnthn Yeah; should say Empty, though some may prefer %h = ()
cognominal may be someone competent should go over that section : design.perl6.org/S02.html#Nil 12:49
12:50 vendethiel joined 12:52 kid51 left 12:53 maybekoo left
cognominal or should that be a s:g/Nil/Empty/ ? 12:54
12:54 wamba left
dalek osystem: da072fa | RabidGravy++ | META.list:
Rename META.info for IO::Path::Mode
12:55
RabidGravy there, IO::Path::Mode now stringifies the mode to something like what ls -l does 13:01
moritz RabidGravy++ 13:02
RabidGravy still not entirely sure about the "sticky bit" but hey how often do you see that in the wild 13:04
dalek ar: cfe8e3b | (Steve Mynott)++ | README:
more pointers to docs in README
13:06
13:06 lokien_ joined
dalek ar: c4259ad | (Steve Mynott)++ | README:
fit better in 80 cols
13:08
RabidGravy !fmt
dalek c: e9483ab | (Wenzel P. P. Peppmeyer)++ | doc/Language/list.pod:
doc Range as List/Array slice
13:09 wamba joined
hoelzro o/ #perl6 13:11
13:12 Snowdog left
gfldex m: Empty.WHAT.say; 13:14
camelia rakudo-moar a5fe34: OUTPUT«(Slip)␤»
gfldex m: ().WHAT.say;
camelia rakudo-moar a5fe34: OUTPUT«(List)␤»
gfldex jnthn: ^^^ 13:15
13:15 vendethiel left
dalek ar: 85c1046 | (Steve Mynott)++ | tools/build/Makefile.in:
bump STAR_VERSION
13:16
ar: 8d3596c | (Steve Mynott)++ | tools/build/Makefile.in:
bump a (c) year
13:19 kaare__ joined, digidog_ left
dalek ar: edd8f41 | (Steve Mynott)++ | / (2 files):
update versions in Makefile
13:26
ar: ba92e0b | (Steve Mynott)++ | LICENSE:
yet another year bump
13:29
13:33 AlexDaniel joined
jnthn gfldex: Yes? :) 13:35
13:35 skingsbu joined
jnthn I didn't say Empty and () are the same thing, just that %h = Empty and %h = () have the same effect. :) 13:35
BenGoldberg m: (|()).WHAT.say;
camelia rakudo-moar a5fe34: OUTPUT«(Slip)␤»
gfldex Empty is not the empty list and I don't really understand what that means.
BenGoldberg m: ((|()) == Empty).say; 13:36
camelia rakudo-moar a5fe34: OUTPUT«True␤»
llfourn_ m: say Slip ~~ List
camelia rakudo-moar a5fe34: OUTPUT«True␤»
gfldex since the docs dont talk about Empty I feel the urge to improve my understanding.
jnthn Well, its main use is to let list comprehensions work out
moritz gfldex: Empty is just an empty Slip
gfldex: and it's a constant, not a type
jnthn m: say ($_ * 2 if $_ > 3 for 1..10) 13:37
camelia rakudo-moar a5fe34: OUTPUT«(8 10 12 14 16 18 20)␤»
jnthn If the `if` there didn't result in Empty, then we'd end up with lots of () in the result
Uh, that wasn't quite so well put
If the `if` there didn't resulted in (), then we'd end up with lots of () in the result.
crap!
What was in my salad... :) 13:38
Take 3:
BenGoldberg Too many negatives.
zengargoyle lol
DrForr Those weren't bac-o-bits :)
jnthn If the `if` evaluated to () when the condition was False, then the result would be (() () () 8 10 ...)
Because we don't magically make an empty List disappear 13:39
An empty Slip, on the other hand, will vanish
Empty is just a nice name for the empty slip
DrForr Cr*p, now I have 'Little Black Dress' by Tim Curry in my head. 13:40
jnthn :D
gfldex i can't wait for a german Perl 6 book. see: www.amazon.de/s/ref=nb_sb_noss?__mk...words=slip 13:43
DrForr www.youtube.com/watch?v=Kq3EU-jEwrg # Woops, Richard O'Brien. 13:44
moritz gfldex' link might be NSFW for some values of W
DrForr ^^ might be slightly beyond the pale as well, but I don't think *too* far. 13:45
13:51 FROGGS left
hoelzro ab6tract: hey, I saw you were looking for me before; what's up? 13:51
dalek kudo-star-daily: 114da64 | coke++ | log/ (2 files):
today (automated commit)
13:54
kudo-star-daily: ac9e07d | coke++ | log/ (8 files):
today (automated commit)
kudo-star-daily: 09082ec | coke++ | log/ (9 files):
today (automated commit)
RabidGravy can anyone think of a better way of doing this:
dalek rl6-roast-data: 968222c | coke++ | / (6 files):
today (automated commit)
rl6-roast-data: 1d2b867 | coke++ | / (6 files):
today (automated commit)
RabidGravy m: class NoType {}; sub foo(Mu:U :$foo = NoType) { if $foo ~~ NoType { say "no foo" } else { say $foo }}; foo(); foo(foo => Int)
camelia rakudo-moar a5fe34: OUTPUT«no foo␤(Int)␤»
13:54 vendethiel joined
perlpilot RabidGravy: nope. 14:02
ab6tract hoelzro: oh, just checking to see if you had a chance to test my proc async bug on freebsd 14:04
hoelzro ab6tract: I did, but I wasn't able to reproduce the issue =/
14:05 muraiki joined
ab6tract hoelzro: :( :( :( 14:07
i usually love the experience of running OS X. but it's the odd duck in perl6 development 14:08
jnthn RabidGravy: Not other than taking the capture (|c) and looking at that
RabidGravy yeah, that's what I thought and ^ seems reasonably succinct 14:09
14:09 kanishka left
jnthn Yeah. 14:10
We do a similar trick in the QAST::Nodes fwiw :)
RabidGravy (somewhat of a hack to work round no private multis)
[Coke] .
yoleaux 03:57Z <ZoffixWin> [Coke]: The announcement is great news, but it really should include a download link :/ RE: blogs.perl.org/users/coke/2016/02/r...eased.html
[Coke] ZoffixWin: added a link to the page to the tarballs. 14:11
ab6tract what's the blocker on private multis, ooc?
[Coke] Hotkeys: don't ever use *release* from github. ew.
14:11 cdg joined
[Coke] ZoffixWin: you have many people who volunteer for the TPF in this group. what's up? 14:11
14:12 mr-foobar joined
[Coke] ZoffixWin: (documented release process) ... see docs/release* 14:12
jnthn ab6tract: Mostly just implementing them, though it'd need some MOP refactoring to do it cleanly (without code dupe) 14:13
ab6tract: I don't think there's any deep reason we can't have them.
ab6tract jnthn: ah, that's reassuring! 14:16
it felt like one of those things that should be possible
stmuk_ moritz: I've just diff'd the result of the current star build and your RC0 of yesterday and can see no interesting differences 14:17
14:17 vendethiel left
moritz stmuk_: should only be changes in README and then announcement etc. 14:19
stmuk_ moritz: yes and your perl6intro.pdf was out of date and the year changes I did this morning
nothing code related 14:21
14:23 vendethiel joined
hoelzro ab6tract: it would be nice if a macless person like me could just spin up an OS X VM to test things on it =/ 14:23
stmuk_ hoelzro: I think that maybe possible .. not sure how legal or easy it is though 14:24
llfourn_ can travis do osx? 14:25
hoelzro stmuk_: yeah, I don't think it's legal; it also would rely on me having an OS X disk
geekosaur sometimes
stmuk_ docs.travis-ci.com/user/osx-ci-environment/ 14:26
llfourn_ if you google 'OSX vps' you will find a few things
stmuk_ I'm a little sceptical it works without problems .. linux even seems a tad flaky on travis 14:27
llfourn_ stmuk_: cool.
geekosaur never mind that. OS X is always flaky in a VM
ab6tract hoelzro: agreed. i wish they would just sell licenses to the damn thing 14:28
14:28 pmurias joined
ab6tract but then again, the apple hardware ecosystem is nice and tidy, which would go straight out the window 14:29
geekosaur apple is a hardware company
stmuk_ I could probably issue a OS X shell but not at the moment since the desktop is in another city and I would have to play SSH port forwarding tricks 14:30
14:30 ZoffixW joined
ZoffixW Here's a trick: "{" -> TAB -> '[Coke]' guess how it works :) 14:32
pmurias are there Perl 6/NQP versions for the Programming Language Shootout/Computer Language Benchmark?
stmuk_ pmurias: I hope not :)
ZoffixW [Coke], well, the part of the release process I was grumbling about last night doesn't seem to be included in docs/release and I don't know if it should be the job of the release manager or someone who picks up the metaphorical torch.
14:33 ab6tract left
ugexe travis works on osx 14:33
github.com/ugexe/zef/blob/master/.travis.yml#L6 14:34
also if you want an osx vm there is macincloud.com 14:35
i use them as well. its just ok
pmurias stmuk_: why? 14:36
stmuk_: I found some in the perl6/perl6-examples repo 14:37
stmuk_ pmurias: I was partly joking
ZoffixW [Coke], step 16 (in github.com/rakudo/rakudo/blob/nom/...guide.pod) says Compiler releases are not to be announced to avoid confusion. Step 17 says to update wikipedia. And what I see—and I'm not criticizing anyone, but just making an observation—the compiler release was actually announced on blogs.perl.org, but without a download link. When asked in-channel, the link to GitHub release was given, but then I see commentary in the 14:38
backlog that it's not the right link. Wikipedia has been updated, but with the wrong year, so now it says release was "11 months ago". The guide also doesn't include any channels where developers would hear about the release (HN, Reddit, Facebook, Twitter, other social networks and blogs). And speaking of blogs, there should be some content for 3rd party bloggers to blog about. Sure, it was just a bug fix release, but what bugs? I'm having trouble f
inding the Changelog. Were there any critical/security bugs fixed or were they all minor and I don't need to upgrade if I don't feel like it? Many people were complaining about NativeCall lib version warning and now it's gone, that'd be a nice mention.
Oops.. Sorry, didn't realize I typed that much.
14:39 skingsbu left
gfldex ZoffixW: don't worry, we love your typing 14:39
ZoffixW I guess my main point is the release process should include steps to propagate the announcements to people outside of Perl community, in hopes of getting them interested in Perl 6.
stmuk_ ZoffixW: I think focusing on a few R* releases a year is the way to go 14:40
[Coke] ZoffixW: I personally wrote a blog post, mainly to mention the NY.pm event. 14:41
in channel. You got the best answer the person awake had.
the guide was written pre-christmas. some updates have to be made. suggest them. 14:42
colomon is not sure how to comment on github.com/tadzik/rakudobrew/issues/81 without sounding like he’s totally dismissing the idea.
stmuk_ ugexe: are the os x results available on travis-ci.org/ugexe/zef ? I am failing to find them
[Coke] the release announcement said what was in it.
it's in docs/announce/2016.01.md
ZoffixW wasn't aware of docs/announce stuff
[Coke] and this is all for the compiler, not R*.
ugexe stmuk_: its the 2nd build job
[Coke] changelog is also in docs. 14:43
ZoffixW: also, join the perl6-users mailing list.
stmuk_ colomon: outsiders should be using R* not rakudobrew
MadcapJake colomon: seemed an alright point until the last line...
stmuk_ ugexe: ty
[Coke] I'm sure we can start to add more notifications about the compiler release to other media; the original plan was to avoid confusing people, because we used to get, every. time. "but where's the bundled version?" 14:44
colomon stmuk_: outsiders use the p5 equivalent perlbrew all the time.
ZoffixW I don't see 2016.* in github.com/rakudo/rakudo/tree/nom/docs/announce
colomon MadcapJake: keeping one set of libraries for all your rakudobrew builds defeats one of the major purposes of rakudobrew.
ZoffixW Ah, ok, it's in the release branch/commit/tag: github.com/rakudo/rakudo/tree/f033...s/announce
stmuk_ colomon: I think rakudobrew was intended as a quick hack for development which it does very well its not a release maintaining tool 14:45
MadcapJake colomon: but at least his point makes sense, it would be *nice* to have libraries separate from versioned releases,
14:45 vendethiel left 14:46 skids joined
colomon MadcapJake: only if you envision all releases as being able to use the exact same version of a library. 14:46
ugexe but you'd only be able to keep the source
ZoffixW [Coke], fair enough, since it's compiler release and not R* a lot of my critique is unwarranted.
MadcapJake colomon: what if each module was stored with subdirs pertaining to each release? 14:47
pmurias colomon: re issue 81, points 1,2,3 are things that can be fixed while still recompiling all the libraries
ugexe sounds like what an external CompUnitRepository would do 14:48
colomon MadcapJake: how is that better than storing the module with each release?
[Coke] ZoffixW: we're waiting a merge back.
stmuk_ anyway I'm hoping R* 2016.01 is now quite close to being released
[Coke] sorry, "merge"
MadcapJake no need to reinstall, just a "rakudobrew update modules" command and it would copy prev release version over to new one and run tests again 14:49
[Coke] (since it's probably going to be a bunch of evil cherry picks.)
ugexe but then rakudobrew is package manager
MadcapJake ugexe: right, i agree, i'm just playing devil's advocate, the point is, it could be handled differently wherein modules are external 14:50
ugexe ::FileSystem are external
otherwise you need an external compunitrepository so that it can do things like install the tests 14:51
stmuk_ I think rakudobrew needs a maintainer anyway
14:51 Ben_Goldberg joined
ugexe CompUnitRepository::RakudobrewLib or something 14:51
14:52 Ben_Goldberg left
skids Other than the panda building, I could see all the rakudobrew functionality being moved into the rakudo build system, personally. (Not that I'm volunteering :-) 14:53
tadzik colomon: how about "you're right, rakudobrew is not meant for production environment"? :)
14:54 BenGoldberg left, Ben_Goldberg joined
[Coke] note that a lot of the discussion about rakudobrew might be good in -toolchain 14:54
14:55 SunilJoshi left
perlpilot tadzik: "But I thought Rakudo was production ready?!?" :) 14:55
14:55 AlexDaniel left
[Coke] Point people at the part of the christmas announcement where we said bear with us. :P 14:56
"There is still plenty of work ahead for us to improve speed, portability, and stability. Do not expect the level of perfection that you see in established products. This is essentially a .0 release of a compiler. We do not claim an absence of bugs or instabilities. We do not claim the documentation is complete. We do not claim portability to many architectures. We do not claim that all downstream software w 14:59
ill work correctly. Think of it as a first kernel release, and now we get to build and port various distributions based around that kernel.
"
TimToady++
15:00 Hor|zon joined
nine What is supposedly the issue with sharing a module installation between multiple rakudo versions? 15:05
colomon: ^^ 15:06
ugexe installing/running the tests so you know it actually works on a rakudo version you didn't install it with is one thing 15:07
15:07 _Gustaf_ left
nine If a user wants to do that she is free to do so. But that doesn't necessitate forcing all users to do that. 15:08
15:08 sammers left
ugexe right, an external compunitrepository 15:08
15:12 vendethiel joined
colomon I know I have had times where I kept an old rakudobrew build around so I could use it with modules that didn’t work with current rakudo. 15:12
I guess I’m not arguing you shouldn’t be able to do that if you’d like. But it seems like it would be a weird default for rakudobrew 15:13
(I mean, all builds sharing one install of libraries is a weird default.) 15:14
nine 1. we now promise backwards compatibility. 2. if you need an older version, you'll be able to install just that
3. installing a new version of a module does _not_ mean uninstalling the old version. Both will still be available! 15:16
15:17 ely-se left
ugexe theoretically it could get messy during build phases if any part of it is tied to the perl6 version 15:17
jnthn Also maybe worth mentioning that precomps of a module do not conflict between different compiler versions.
RabidGravy which latter does of course mean that you have to specify an version in the use if you require a specific version or later
jnthn (That is, you can move between compiler versions sharing the same installed set of modules without problems.) 15:18
RabidGravy which I don't have a problem with but it does catch me out
jnthn (Which wasn't true a copule of months back.)
15:19 Ven left
nine The only reason rakudobrew installed modules per rakudo version is that perlbrew does it that way. The only reason perlbrew does it is because in Perl 5 modules may contain a C part (called XS) that's linked against the installed perl version and does only work with that. 15:19
ugexe i thought it was because thats just default behavior for installing a module, and local::lib is available if you want it in some other path 15:21
15:23 Ven joined
mspo use lib "foo" seems so much easier than local::lib 15:24
15:25 brrt left, Ven left
RabidGravy views or attachments next, that is the question 15:26
15:26 kurahaupo joined, kurahaupo left
RabidGravy as a matter of interest does anyone use CouchDB in their work? I'm making this thing for my own purposes but I'd be interested to hear what other people use it for 15:27
15:28 ZoffixW left 15:29 Ven joined
mspo RabidGravy: I do 15:29
RabidGravy: we use it for a better memcache, and for straight up document caching mostly 15:30
RabidGravy: very little usage of views
RabidGravy cool, do you use attachments?
mspo never heard of them :)
RabidGravy: it's just an optional caching layer (for almost everything) 15:31
oh except one place where we use it as a regualr k:v database
but I think that is still just storing jobvite data
RabidGravy right, well you could use Sofa now then :)
mspo probably 15:32
RabidGravy github.com/jonathanstowe/Sofa
It's not so much "simple" now 15:33
15:34 vendethiel left 15:39 AlexDaniel joined 15:43 ely-se joined 15:44 perl6newbee joined
RabidGravy but the "better memcache" thing never occurred to me 15:44
mspo well you can open memcache buckets on couchbase and just use memcache 15:47
it was originally born from memcachedb
15:50 Relsak joined
mspo RabidGravy: I also have a group using cassandra as a cache layer; it is a terrible choice 15:52
it collects tombstones like crazy and uses tons of disk to keep itself clean
why they didn't use redis is beyond me
15:52 vendethiel joined 15:56 Ven left, AlexDaniel left 16:02 nakiro left 16:06 Actualeyes left 16:09 ely-se left
pmurias looking at some programming language shooutout benchmarks results from 2009 Perl 6 performance has improved a *lot* 16:11
16:12 khw joined
moritz pmurias: there's a mandelbrot fractal generator that I blogged about in 2009(?), whose runtime then was several minutes, and is now at the low seconds 16:13
and even lower if one uses one or two "int" annotations 16:14
16:14 AndChat380436 joined
moritz (and the code runs unmodified since then :-) 16:14
AndChat380436 p6:say 1;
16:14 vendethiel left 16:16 vendethiel joined
RabidGravy mspo++ # reminding me that Cache::Memcached hadn't been tested since Christmas 16:20
fixed again!
mspo nice
RabidGravy: woring on BSON and mongodb next?
16:22 AndChat380436 left
RabidGravy I think the MongoDB module works 16:22
mspo cassandra, then?
16:23 AndChat380436 joined 16:24 Actualeyes joined 16:29 Skarsnik joined
dalek kudo/nom: 3053598 | lizmat++ | src/core/Hash.pm:
Make basic hash access 25% faster
16:30
DrForr Well cr*p, Perl 6 qq{} interpolation is not polite to C code. Which comes as no surprise, but still... 16:31
Specifically 'c_cell->type = ZERO;' # The -> I can't seem to escape without resorting to heroic methods.
16:32 ZoffixW joined
Juerd m: say "foo->bar" 16:33
camelia rakudo-moar a5fe34: OUTPUT«foo->bar␤»
jnthn lizmat: You might be able to get it even faster
ZoffixW m: say qq{c_cell->type = ZERO;}
camelia rakudo-moar a5fe34: OUTPUT«c_cell->type = ZERO;␤»
Juerd DrForr: I don't understand.
ZoffixW DrForr, what do you mean?
lizmat jnthn: i have no doubt, but don't see right now
DrForr This is in a qq:to/FOO/ heredoc.
Juerd Still
skids So, should I leave PR#685 targettied at nom, or is there a new set of branches now? (It has a minor, untested-in-roast behavioral change that is most likely not being relied on by anyone, in addition to plain old bugfixes)
jnthn lizmat: The trick is that nqp::atpos returns null if the key does not exist
Juerd DrForr: Is it a hash key that starts with a < somewhere earlier?
ZoffixW DrForr, but -> is not special :/
lizmat jnthn: atpos? you mean atkey? and: also on JVM? 16:34
DrForr No, as I said it's C code, but embedded in a here-doc.
jnthn lizmat: well, atpos does too, but I meant atkey
lizmat: Yes, on JVM...this is relied on quite a bit by NQP code-gen.
Juerd DrForr: Do you have a full example that you can share on pastebin or a gist?
ZoffixW m: gist.github.com/zoffixznet/dfe07b2e4531c7012ba6
camelia rakudo-moar a5fe34: OUTPUT«c_cell->type = ZERO;␤␤»
lizmat ok, I'll try
DrForr Which is inside {}, which means the code is interpreted as a Perl 6 block.
<facepalm/>
ZoffixW Ah :D
jnthn lizmat: Anyway, there's also an nqp::ifnull op that returns the first thing if it's not null, otherwise evaluates the second thing.
lizmat jnthn: yes, I'm aware :-) 16:35
jnthn lizmat: so nqp::ifnull(nqp::atkey($storage, $key), nqp::p6bindattrinvres(...)) :)
Cool :)
DrForr Escaping {} does *not* make me a happy camper...
ZoffixW DrForr, there's a qq flag that removes the {} interpolation 16:36
ZoffixW digs through the design specs
DrForr looks as well.
jnthn qq:!c
DrForr Thanks.
ZoffixW m: my $x = 'meow'; say qq:!c/foo { ey } bar $x/ 16:37
camelia rakudo-moar a5fe34: OUTPUT«foo { ey } bar meow␤»
jnthn Also, lizmat++ for the 25% already :)
ZoffixW neat
jnthn Hash access is, like, common :)
16:37 vendethiel left 16:39 vendethiel joined, lokien_ left
RabidGravy the quoting doc doesn't mention any adverbs whatsoever BTW 16:40
ZoffixW Well volunteered! :}
Juerd The quoting doc is a mess imho
DrForr With that all I have to do is escape '&', something of a win.
jnthn DrForr: :!f iirc :) 16:41
16:41 skids left
Juerd I don't have any specific idea how to restructure it, but this is one of those pages where a good structure and good layout can make things much clearer 16:41
lizmat didn't I write an advent post in 2014 about the various flags in quoting ?
Juerd DrForr: Instead of disabling interpolation things, you can also start witha blank slate and enable only what you need.
lizmat: Sure, but that doesn't make great reference documentation either 16:42
lizmat true
Juerd That's more tutorial-like
DrForr What I actually need is a proper template library, but since I'm loathe to try to port TT I'm making do ATM.
Well, I'm not loathe to port, I'm more of the opinion that we can build better. 16:43
ZoffixW is porting Mojo::Template
RabidGravy Template6 could do with some love
[Tux] gist.github.com/Tux/22218a1481bf80dd4240 16:44
ZoffixW Might be done by the end of the weekend, though I might end up spending the entirety of it looking for new apartments.
RabidGravy I abandoned trying to use it for JSON::Infer
lizmat [Tux]: so about 8% of total time for compiling 16:45
[Tux] yes 16:46
16:48 cfloare joined
dalek kudo/nom: 556b829 | lizmat++ | src/core/Hash.pm:
Make simple hash access another 8% faster, jnthn++
16:49
16:49 cfloare_ left
[Tux] goes for a new make … 16:50
ZoffixW w00t lizmat++
16:54 skids joined 16:55 kaare__ is now known as kaare_
[Tux] no noticable change for my code 16:58
lizmat [Tux]: didn't expect that, really 16:59
[Tux] no harm in trying :)
lizmat I mean, it's a total of about 30% of something very little already
tony-o_ RabidGravy: interesting idea, why did you abandon it?
RabidGravy which? 17:00
tony-o_ JSON::Infer
lizmat jnthn: does it make currently sense to add --> Nil to methods that are not supposed to return a value ?
e.g. Block.fire_phasers ? 17:01
RabidGravy Oh no, that works fine, I abandoned trying to use Template6 to generate the code
jnthn lizmat: It's a good idea for avoiding accidental value leakage, yes
RabidGravy I'll probably revisit it at some point if some nicer templating thingy comes along
[Tux] but CSV::Parser is (just) under 50 again 17:02
49.7
tony-o_ RabidGravy: anything in particular you look for in a templating system? i wrote protone and walked myself into a corner with it. it's one of those projects i'm going to revisit during one of these plane rides
17:02 vendethiel left
RabidGravy I think it was the recursively calling object method s in the template that broke Template6 for me 17:04
i.e. passing one object to the template and then getting lists of the same type of object in the template (nested classes in the JSON::Infer case) 17:05
17:06 domidumont left, vendethiel joined
tony-o_ ah, protone can do recursion in the template 17:09
17:11 Kogurr joined 17:12 wamba left
RabidGravy I'll give the tires a good kick at some point :) 17:15
17:15 lokien_ joined
dalek kudo/nom: 172a92b | lizmat++ | src/core/Block.pm:
Mark methods as not returning anything
17:15
tony-o_ RabidGravy: good, i've fixed a few issues with it and had some good feedback/enhancement requests that i just saw yesterday (sorry raiph) 17:16
17:18 maybekoo5 joined
ZoffixW What's a "document nation" that we're working on? :S twitter.com/notcalledjack/status/6...5917008896 17:21
oh.. "documentation" + auto correct? 17:22
17:22 musiKk joined
ZoffixW never understood why people use that thing 17:22
DrForr, BTW, I'm a bit confused by your making a bunch of people admins this morning. What do you mean by "admin authorization of posts"? Can't any member of the group post? 17:25
(that was in reference to our Facebook group)
17:27 vendethiel left
RabidGravy I think it means that "report to admin" goes to more than one person 17:37
or maybe it's moderated now
someone post something so we can see what happens 17:38
ZoffixW :D
17:38 SCHAAP137 joined 17:39 vendethiel joined
timotimo oh god, what have i done 17:39
(on the perl6-user-experience wiki)
ZoffixW gives timotimo an invitation to the club 17:41
Had the same experience with #5 :)
17:42 musiKk left 17:48 dakkar_ left
MadcapJake I'm gonna try my hand at writing that bug reporter tool. 17:48
ZoffixW MadcapJake++ 17:50
MadcapJake it'll be a form ala StackOverflow, enter the title and it'll show related tickets
perl6 server will scrape rt.perl.org and send json data back to client, client will send completed form back to server and email it to [email@hidden.address] 17:51
17:52 sevvie joined 17:53 ZoffixW left
RabidGravy cool 17:55
MadcapJake I could even integrate other fields, like compiler version, language version, etc. 17:56
hoelzro MadcapJake: have you seen the RT command line client?
you might want to use that instead of scraping
MadcapJake no i haven't that would certainly make this easier
timotimo hey hoelzro :)
hoelzro howdy timo!
timotimo there was a release; you're the "owner" of the rakudo docker, aren't you? or is that only for rakudo star? 17:57
MadcapJake hoelzro: how do i install the RT cli? (no perl5 experience)
hoelzro timotimo: yeah, just for star 17:58
timotimo OK
so that'll happen soon enough, too. but not right now :)
hoelzro MadcapJake: which OS are you on?
MadcapJake ubuntu
hoelzro timotimo: don't worry, I'll take care of it when it happens =)
MadcapJake: ok, so I recommend you install cpanminus and local::lib; they're both probably in APT 17:59
timotimo yay
Skarsnik MadcapJake, you can inspire of the debianbug repport tools maybe
hoelzro after you have cpanm, you can install the RT client locally via cpanm -l rt-lib RT::Client::CLI
MadcapJake Skarsnik: got a link?
18:00 vendethiel left
Skarsnik wiki.debian.org/reportbug 18:00
RabidGravy Ooh 🎹
Skarsnik It's really nice by offering to search for similar bug and stuff 18:01
timotimo *sigh* our in-lining fu is rather weak sometimes
MadcapJake Skarsnik: I'm not seeing it, just see details of a cli
sevvie good localtime all.
Skarsnik it's a cli 18:02
hoelzro being able to search for similar bugs would be a godsend
MadcapJake Skarsnik: ahh xD I'm making a web reporting tool for the "pointy clicky" fans ;)
hoelzro I mean, you can in RT already, but I think that you could easily miss some duplicates that exist already
MadcapJake but perhaps I could extrapolate some of this out into a perl6-only cli
Skarsnik I used debian reportbug tool 1-2 time it was really nice
18:03 leont joined, vendethiel joined 18:04 CIAvash left, ely-se joined 18:06 yurivish left 18:08 domidumont joined 18:09 [Sno] joined 18:10 pmurias left, sno left
DrForr ZoffixWin: Around? 18:15
18:16 sjoshi joined 18:17 sjoshi left
DrForr ZoffixWin: My logic is as follows: We're large enough that spammers are doing MitM attacks on us now. All the group regulars are (or should be) admins, and able to post by themselves without causing inconvenience. Non-regulars that actually have Perl 6 content (unlikely, but possible) have to wait a bit. Spammers can be blocked, deleted from the group and anyone they "invite" can burn in hell. 18:22
MadcapJake BTW all: my perl6fe grammar is coming to github soon! github.com/github/linguist/commit/...28ef23a6f7 18:23
18:23 Zero_Dogg left
MadcapJake sevvie++ 18:23
18:24 Zero_Dogg joined 18:26 patrickz joined
DrForr .tell ZoffixWin My logic is as follows: We're large enough that spammers are doing MitM attacks on us now. All the group regulars are (or should be) admins, and able to post by themselves without causing inconvenience. Non-regulars that actually have Perl 6 content (unlikely, but possible) have to wait a bit. Spammers can be blocked, deleted from the group and anyone they "invite" can burn in hell. 18:30
yoleaux DrForr: I'll pass your message to ZoffixWin.
18:33 Actualeyes left 18:34 FROGGS joined
FROGGS o/ 18:36
AndChat380436 >say 1; 18:39
>p6:say 1; 18:40
DrForr o/ 18:41
18:42 perl6newbee left
DrForr m: say 1; 18:42
camelia rakudo-moar 172a92: OUTPUT«1␤»
AndChat380436 Thx! 18:43
m: say 2;
camelia rakudo-moar 172a92: OUTPUT«2␤»
DrForr np.
AndChat380436 cool
18:45 ely-se left 18:48 vendethiel left
jdv79 has anyone tried to install the entire eco's worth? 18:51
i wasn't that far into it and i hit a lock. i'm guessing it might be the CUR lock. 18:52
is that still an open issue?
18:53 vendethiel joined 18:56 Ben_Goldberg left, firstdayonthejob joined 19:05 awwaiid left, awwaiid joined 19:06 ely-se joined 19:10 leont left
timotimo why would the cur lock prevent you from installing stuff? isn't it only there to prevent concurrent access to the database? 19:13
19:15 vendethiel left
jdv79 not sure. maybe related to this: irclog.perlgeek.de/perl6-toolchain...i_11964808 19:17
i need to repro it again to check
19:19 lokien_ left
timotimo well, you can at least set that environment variable that makes module loading and precomp extra verbose 19:21
FROGGS stmuk_: ping
stmuk_ pong 19:22
19:22 vendethiel joined
FROGGS stmuk_: you did a lot of preparations for star, right? 19:22
stmuk_ yes 19:23
FROGGS stmuk_: thank you :o)
stmuk_: what's left to do?
stmuk_ AFAIK nothing
FROGGS stmuk_: awesome
then I'll cut the release now 19:24
stmuk_ :D
RabidGravy then at some point I may get a fedora upgrade :) 19:25
stmuk_ actually I haven't tagged 19:26
19:26 laz78 joined
FROGGS stmuk_: sure, that's fine 19:27
19:27 cdg left
stmuk_ moritz did the submodules stages and I should have done up to step 11 19:27
19:27 addison_ joined 19:28 molaf joined
jdv79 /home/jdv/cpan-api/.panda-work/1450665684_1/lib/.precomp/.lock seem to be what caused the blockage 19:28
hmm
FROGGS stmuk_: that's nice, because that gives me extra time to fix something with the msi on windows
19:28 musiKk joined
FROGGS jdv79: and it is not just a module or test that infiniloops? 19:29
19:30 dha joined
awwaiid Is there a way for concurrent threads to increment a shared counter safely? In clojure I'd do (swap! counter inc), which does software-transactional-memory stuff to try to increment the counter, and if another thread messed with the counter at the same time to to retry so no increments are lost 19:30
jdv79 working on that
stmuk_ RabidGravy: surely a fedora upgrade would be debian :P 19:31
jdv79 but. now an infiloop causes a repo lock - that's different if that's the case
this whole repo lock shenanigans thing is concerning 19:32
jnthn awwaiid: Do you want a counter, or do you just want a sequence of values that are thread-safe to obtain? 19:33
awwaiid: See github.com/jnthn/p6-concurrent-iterator if so 19:34
awwaiid thanks jnthn, will check it out. I was also just perusing 6guts.wordpress.com/2014/04/17/rac...ess-leads/ . 19:37
19:39 musiKk left 19:41 leont joined, spider-mario joined
dalek c: fcfa134 | (Daniel Perrett)++ | doc/Language/operators.pod:
Provide a simpler example for reversed operators

Reversed operators are taught before operator reduction, so give an example that does not depend on operator reduction
19:41
c: 7f4a641 | (Zoffix Znet)++ | doc/Language/operators.pod:
Merge pull request #377 from pdl/pr-r-operator-example

Provide a simpler example for reversed operators
vendethiel jnthn: we don't have atomics in p6, do we? 19:43
dalek c: 0f8387b | (Daniel Perrett)++ | doc/Language/setbagmix.pod:
Added index terms for doc/Langualge/setbagmix.pod
c: 4caa0dc | (Daniel Perrett)++ | doc/Language/operators.pod:
Trim trailing spaces in operators.pod
c: 5952ee5 | (Daniel Perrett)++ | doc/Language/operators.pod:
Add index entries for postfix and prefix call
c: bdf35b1 | (Zoffix Znet)++ | doc/Language/ (2 files):
Merge pull request #378 from pdl/pr-index-more-operators

Index more operators
19:46 FROGGS[mobile]2 left 19:51 Relsak left 19:54 telex left 19:56 telex joined 19:57 cpage_ joined 20:00 ely-se left 20:01 donaldh joined
hoelzro I'm wondering, with the advent of #perl6-{toolchain,release}, if this channel should become #perl6-dev and #perl6 should be a "welcome to the language" channel 20:02
dalek p: bc5b1df | donaldh++ | / (3 files):
Add lstat, stat_time and lstat_time on JVM
20:03
dha This assumes we wish to welcome people. :-) 20:04
20:05 Amendil left, ribasushi left 20:07 ribasushi joined 20:08 dolmen joined 20:14 donaldh left
El_Che hoelzro: an other level of indirection kind of thing? :) 20:14
hoelzro dha: =) 20:15
skids Personally I think isolating developers from users is a foot on a slippery slope.
hoelzro El_Che: I'm just thinking about how I've seen people talk about how #perl6 itself has become difficult to backlog, but the newer, topic-focused channels are easier to backlog
leont Backlog is the main reason for those channels IME 20:16
hoelzro skids: I'm not saying that everyone working on Rakudo should just stay in #perl6-dev; they would (hopefully!) still stick around to help people
leont Things tend to get lost otherwise
skids What's missing is backog summaries.
hoelzro: nice in theory but I've seen that end badly.
El_Che a mix channel will help adoption, imho
mixed
it's also easier to help people to get involved 20:17
hoelzro skids: alright, just a thought
MadcapJake any input on what should be required fields for submitting bugs? 20:19
20:20 domidumont left
perigrin subject, description, cc#, expiration date, CVV2 ... 20:20
timotimo skids: are they really missing, though? at least last time i looked someone was still using the summary feature on the irclog
dha perigrin++
geekosaur *eyeroll* 20:21
MadcapJake expiration date?
geekosaur credit card info
timotimo shipping address
perigrin MadcapJake: so you can tell if the card number is still valid.
MadcapJake lol
20:21 darutoko left
perigrin timotimo: billing address ... it's software ... adn honestly stripe doesn't require it 20:21
timotimo hm. fair enough.
hoelzro timotimo: irclog.perlgeek.de? it has a summarization feature?
timotimo yes, it does
hoelzro huh, TIL
timotimo every line has a "to be included in the summary view" field in the database, and you can edit that in summary-edit mode 20:22
MadcapJake but would something like language version or compiler version make sense on a bug report? could just be optional but having a field there would means some might use it
timotimo video.fosdem.org/2016/h2214/ - IIUC, this room was the perl devroom, and no videos have been uploaded there yet
El_Che timotimo: no, but they will be 20:23
timotimo hopefully they'll arrive this side of christmas this time
er
El_Che timotimo: there is raw video in a private url, and tom's backup video
we're golden
timotimo that sounds good!
El_Che yeah, last time is was pretty terrible 20:24
MadcapJake ok well i'll add it (lang/comp fields) and if anyone decides they don't like it, we can strip it out later
dha So... documentation question. Is there a document that covers the syntax of function signatures in detail? Last I looked, there wasn't, but that may have changed. 20:25
timotimo doc.perl6.org/language/functions#Signatures - this really doesn't cover much 20:26
20:26 yurivish joined
skids docs.perl6.org/type/Signature 20:27
dha Right. When I was working on the - now abandoned - usage statements, I often hit things in function signatures that were not at all obvious. I've been thinking for some time that there should be a document explaining the ins and outs of signatures for people who are not already Perl 6 aficionados. 20:28
20:28 laz78 left 20:29 wamba joined
dha Because, silly me, I think we might want *new* people to use Perl 6. :-) 20:30
Arguably, a document on how to use the documentation would probably not be amiss. 20:31
perlpilot A good book or three would be nice too. :)
20:32 bjz joined, bjz left, bjz joined
skids well, if type/Signature is not specific enough, then there is always the synopsis. 20:32
dha I should probably look at type/Signature more closely. 20:33
perlpilot - Yeah. I think now that Perl 6 actually exists, that will start happening.
timotimo that page is surprisingly good
20:33 franciskittu joined 20:34 franciskittu left
skids But I think when you get down to it most Signature questions are a "What is This Character doing Here?" question so if anyone wants to enhance the glossary with the old material from WITCH they are very welcome to. 20:34
MadcapJake yeah the type/Signature page info is great (I often feel a lot of it should be in the main function page)
20:35 franciskittu joined
RabidGravy yeah, the type/Signature is pretty damn thorough and linked from the functions 20:35
dha Actually, what I'm probably thinking is missing is a "this character means *this*" sort of thing. I. e. something more geared to helping people read signatures, rather than helping people write signatures. 20:36
dalek ar: de789fb | FROGGS++ | modules/ (4 files):
update modules
skids
.oO(automated svg "sentence diagram" based on parser rule names... hmm)
20:38
20:39 nowan left, mr-foobar left
dha For instance, it isn't immediately obvious what ":D" means in "returns Date:D" 20:40
It's findable in Type/Signatures, but it doesn't jump out at you. 20:41
perlpilot clearly it means we're returning a really happy Date instead of a sad one.
skids Right but if you were able to look up ":" and get something like this: www.perlfoundation.org/perl6/index....itch_colon (except up to date enough to know about :D) it would be nice 20:42
dha perlpilot++
20:43 cpage_ left
dha yeah, but I bet trying to look up something as common as ":" would be... difficult. 20:43
skids dha: see the URL?
dalek ar: f4948b4 | FROGGS++ | docs/announce/2016.01.md:
refrase support of S09 and S11
dha yes.
perlpilot one that occasionally gets me is that I'll sometimes accidentally omit the space in sub foo (@a ($alpha, $beta, *@)) { } 20:44
20:44 cpage_ joined
perlpilot though lately I've taken to always writing it with a colon @a:($alpha,$beta, *@) 20:44
skids It could probably use a "When used in a SIgnature" section, granted.
20:44 espadrine left
dha I also see that that url is at perlfoundation.org, and does not result from a search for ":" at doc.perl6.org. So... not an obvious place to look. 20:45
And, as noted, use in a signature would be needed... 20:46
20:46 laz78 joined
dha Anyway, I'm going home before the weather gets any worse. later, all. 20:47
20:47 dha left
skids Right and it is woefully out of date. What I'm saying is that material neds to be moved to docs.perl6.org/language/glossary or somesuch so the searchbox can be told "single non-alhanum character? look it up in the glossary" 20:48
20:48 cdg joined 20:49 dolmen left 20:51 neilb joined
DrForr How would one catch die()? 20:51
skids m: { die(foo); CATCH { default { } } } 20:52
camelia rakudo-moar 172a92: OUTPUT«5===SORRY!5=== Error while compiling /tmp/8wTaFPLxtr␤Undeclared routine:␤ foo used at line 1␤␤»
DrForr %*SIG<die> or has that moed?
flussence m: die "foo"; CATCH { default { say "bar" } }; say "baz";
camelia rakudo-moar 172a92: OUTPUT«bar␤»
DrForr Ah, thanks.
geekosaur I would hope the SIG hack is dead
flussence not quite what I expected...
Juerd flussence: Did you expect die not to die because of the CATCH? 20:53
flussence m: { die "foo"; CATCH { default { say "bar" } } }; say "baz";
camelia rakudo-moar 172a92: OUTPUT«bar␤baz␤»
flussence ah, scope.
Juerd Yes. die still dies :)
geekosaur die aborts the block it's in, which block had the say baz 20:54
Juerd m: say 11; { CATCH { say 42 }; die 23; } say 22;
camelia rakudo-moar 172a92: OUTPUT«5===SORRY!5=== Error while compiling /tmp/FQqzv1S_dk␤Strange text after block (missing semicolon or comma?)␤at /tmp/FQqzv1S_dk:1␤------> 3say 11; { CATCH { say 42 }; die 23; }7⏏5 say 22;␤ expecting any of:␤ infix␤ inf…»
Juerd m: say 11; { CATCH { say 42 }; die 23; }; say 22;
camelia rakudo-moar 172a92: OUTPUT«11␤42␤23␤ in block <unit> at /tmp/lYQXcTXTQZ line 1␤␤»
20:54 [Sno] left
jnthn vendethiel: No atomic ops yet, no. Perhaps for 6.d. :) 20:55
20:55 nowan joined
vendethiel okay :) 20:56
DrForr Yep, that works, thanks. Dying in the middle of a NativeCall callback corrupts the stack. 20:57
Not surprising, but I'd prefer to catch it in code rather than get surprised.
20:59 dolmen joined
DrForr Well, unwinds not corrupts, but it's still annoying. 21:01
geekosaur arguably any such unwind needs to stop at the border 21:02
21:02 donaldh joined
geekosaur could cause nasty bugs on the foreign side otherwise 21:02
DrForr I'd think so too.
RabidGravy flussence, if you want what you expected then: 21:06
m: die "foo"; CATCH { default { say "bar"; .resume; } }; say "baz"
camelia rakudo-moar 172a92: OUTPUT«bar␤baz␤»
flussence aha, thanks 21:07
(knew there was something, just couldn't remember what)
21:08 rindolf left
donaldh Ugh, Configure.pl says nqp-j is too old. It really isn't 21:09
Why is (0, 1) the result of parsing 2016.01-1-gbc5b1df 21:12
DrForr The 'c' maybe? (complex number) 21:13
donaldh Ah, parse_revision in tools/lib/NQP/Configure.pm was rewritten in the last commit. 21:14
21:15 musiKk joined, sortiz joined
donaldh that's very broken 21:15
21:16 Ben_Goldberg joined
sortiz \o #perl6 21:18
DrForr CATCH{} doesn't appear in the Search box. (I'm aware it's a phaser, just pointing out the omission.) 21:20
21:20 buharin joined
buharin hello 21:20
I have no idea how to implement this wiki.libsdl.org/SDL_RWops
that should be simple methods in class 21:21
method size(SDL_RWops)
21:22 espadrine joined
timotimo nah, you have to have a CStruct and assign function pointers 21:23
21:23 Psyche^_ joined
DrForr buharin: size() is a callback. 21:23
timotimo i don't think we have something to directly turn a piece of code into a function pointer, though? but perhaps it's enough to put a &foo in there 21:24
tony-o_ m: gist.github.com/tony-o/ef83abc5027251126804
camelia rakudo-moar 172a92: OUTPUT«5===SORRY!5=== Error while compiling /tmp/DfUtL532th␤Unable to parse expression in argument list; couldn't find final ')' ␤at /tmp/DfUtL532th:1␤------> 3ub r(Mu7⏏5 $r?) { }; ␤ expecting any of:␤ infix␤ infix stopper␤»
tony-o_ m: gist.github.com/tony-o/ef83abc5027251126804
buharin timotimo, yeah I used Cstruct class
timotimo if you just put the code in as methods, it won't end up as fields
camelia rakudo-moar 172a92: OUTPUT«(timeout)» 21:25
buharin but what I should put in class
mmm
true
timotino how to create callback then? 21:26
RabidGravy I think you are totally in uncharted territory here 21:27
timotimo m: use NativeCall; class Test is repr('CStruct') { has &.callback }
21:27 Psyche^ left
camelia rakudo-moar 172a92: OUTPUT«5===SORRY!5=== Error while compiling /tmp/6KOC0DbiFP␤CStruct representation only handles int, num, CArray, CPointer, CStruct, CPPStruct and CUnion␤at /tmp/6KOC0DbiFP:1␤» 21:27
buharin oh lol
xD 21:28
it looks like perl callback
timotimo it has to
buharin but what about argument informations
timotimo hm, how does nativecast look
probably "has &.callback:(int32 $foo, int32 $bar)" or something 21:29
donaldh PR to unbust nqp version check during Configure : github.com/rakudo/rakudo/pull/703 21:30
timotimo m: use NativeCall; nativecast(Pointer, -> int32 $a { say "test" }).perl.say 21:31
camelia rakudo-moar 172a92: OUTPUT«Native call expected return type with CPointer, CStruct, CArray, or VMArray representation, but got a P6opaque␤ in sub nativecast at /home/camelia/rakudo-m-inst-1/share/perl6/sources/075EFE4B4CDAAF73190194EA876F81A1F128D1A2 line 367␤ in block <unit> …»
timotimo nope.
m: use NativeCall; nativecast(Pointer, -> int32 $a, --> int32 { say "test" }).perl.say
camelia rakudo-moar 172a92: OUTPUT«Native call expected return type with CPointer, CStruct, CArray, or VMArray representation, but got a P6opaque␤ in sub nativecast at /home/camelia/rakudo-m-inst-1/share/perl6/sources/075EFE4B4CDAAF73190194EA876F81A1F128D1A2 line 367␤ in block <unit> …»
timotimo m: use NativeCall; nativecast(Pointer, -> int32 $a --> int32 { say "test" }).perl.say
camelia rakudo-moar 172a92: OUTPUT«Native call expected return type with CPointer, CStruct, CArray, or VMArray representation, but got a P6opaque␤ in sub nativecast at /home/camelia/rakudo-m-inst-1/share/perl6/sources/075EFE4B4CDAAF73190194EA876F81A1F128D1A2 line 367␤ in block <unit> …»
timotimo ...
perlpilot perl6advent.wordpress.com/2015/12/...-beyond-c/ 21:33
might help
21:33 TEttinger joined
timotimo perlpilot: the thing is, that only works when passed to a function as an argument 21:34
RabidGravy howzabout declaring it as a Pointer, dereference to a &thingy, apply the Native role to the result etc 21:36
buharin timotimo it looks like callbacks cant be part of CStruct 21:37
21:38 AlexDaniel joined
buharin CStruct representation only handles int, num, CArray, CPointer, CStruct, CPPStruct and CUnion 21:38
FROGGS buharin: that's true
timotimo yeah, but you need a way to turn the callback into a Pointer
FROGGS buharin: we do not (yet) support function pointers there
buharin FROGGS, wtf?
21:39 wamba left
RabidGravy you could hack round it with a tiny C function 21:39
buharin RabidGravy, what do you mean by tiny C function? 21:40
21:41 Kogurr left
Skarsnik code a small C wrapper 21:41
timotimo void *transform(void *input) { return input }
Skarsnik You can always hold the functor on a Pointer 21:42
RabidGravy yeah, basically just tat
buharin oh you mean write C functions which will invoke callbacks
RabidGravy yeah, the struct member is just a pointer to function 21:43
sortiz Nop, functions to turns callbacks into CPointer, for storage in p6 land.
buharin I will rather wait till perl6 will be complete than change library implementation :P 21:44
RabidGravy "complete" ?
DrForr At that you'll be waiting a long time... 21:45
buharin xD
DrForr You might want to look at github.com/drforr/perl6-Inline-Scheme-Guile 21:46
21:47 ELBeavers joined, kaare_ left
RabidGravy It's entirely possible that most of the code to do this is actually there, it's just not exposed in NativeCall, as, I'm guessing here, the nqp::nativecall somewhere under the hood gets a pointer to a function which it gets from the dynamic loader 21:47
and then executes it with the (marshalled) arguments 21:48
AlexDaniel skids: I'm not sure what precedence problem you are talking about here: rt.perl.org/Public/Bug/Display.html?id=127424
skids: did you miss the whitespace?
m: say ^11 .rotor: 1.5
camelia rakudo-moar 172a92: OUTPUT«This type cannot unbox to a native integer␤ in block <unit> at /tmp/ZlISoGB0MY line 1␤␤»
AlexDaniel m: say ^11.rotor: 1.5
camelia rakudo-moar 172a92: OUTPUT«Potential difficulties:␤ Precedence of ^ is looser than method call; please parenthesize␤ at /tmp/eV2ZW0dqqR:1␤ ------> 3say ^117⏏5.rotor: 1.5␤^0␤»
Skarsnik RabidGravy, probably, I should have a look one day. Not sure where is that, NQP or Moar x)
AlexDaniel skids: ↑
buharin mmm 21:49
skids AlexDaniel: indeed, missed the whitespace. 21:50
sortiz RabidGravy, I suppose that, see github.com/rakudo/rakudo/blob/nom/...l.pm6#L45, NCI uses the 'callaback' type for the marshalling. 21:51
skids reminds himself never to use that infix dot for readability reasons.
21:53 musiKk left
AlexDaniel skids: you can use more spaces to make it more readable 21:54
buharin guile next open source language ;d
skids AlexDaniel: I'll stick with parens.
21:55 gcole joined 21:58 geraud joined 22:02 skids left
DrForr buharin: Heh. 22:03
22:03 vendethiel left
ZoffixWin So how is the perl6.slack.com/ experiment going? 22:04
yoleaux 18:30Z <DrForr> ZoffixWin: My logic is as follows: We're large enough that spammers are doing MitM attacks on us now. All the group regulars are (or should be) admins, and able to post by themselves without causing inconvenience. Non-regulars that actually have Perl 6 content (unlikely, but possible) have to wait a bit. Spammers can be blocked, deleted from the group and anyone they "invite" can burn in hell.
ZoffixWin noticed an email from slack, saying I was mentioned 22:05
MadcapJake ZoffixWin: yeah I mentioned you
22:05 vendethiel joined
ZoffixWin It was my lead in to this item I saw the other day: news.ycombinator.com/item?id=10486541 22:06
Apparently Slack itself said they don't have FOSS communities in mind and their architecture is not suited for such uses :/
DrForr That was pretty obvious when trying to sign up :/
ZoffixWin :D 22:07
22:07 muraiki left
stmuk_ slack-- # hipster IRC 22:07
MadcapJake ZoffixWin: as I've said before, I don't disagree. The point *isn't* to make the slack the only community of Perl 6. It's *only* meant to promote Perl 6 in the Slack-user community, which is much larger than the IRC-stalwarts would like to admit. 22:09
much much larger
ZoffixWin MadcapJake, cool cool
yeah, the more communication channels we have the better 22:10
22:10 maybekoo5 left
AlexDaniel MadcapJake: “any input on what should be required fields for submitting bugs?” – no fields should be required :) 22:10
MadcapJake: if the user wants to submit a bug report like this, then fine… But it should be possible to edit the title 22:11
MadcapJake ZoffixWin: that's what I'm thinking, any time you want to do outreach, you have to go where the people are. Would you try to start a Esperanto speakers group in a city but only if members communicated by letter or in person? No, you're group won't survive or grow because people are interested in newer (less-free,less-principled) communication forms 22:12
RabidGravy I'd never heard of slack until it was mentioned here
AlexDaniel MadcapJake: in other words: accept any data, we'll figure it out later
MadcapJake AlexDaniel: totally agree, I'm just thinking that maybe I could "mold" the resultant email in a more usable way
jast the first rule of marketing: connect to your target audience rather than the biggest group of people you can find
ZoffixWin jast++
stmuk_ its like a less good IRC in a webpage where you have to register for each channel everytime 22:13
MadcapJake jast: i don't see how that relates, what's the Perl 6 target audience if not other programmers?
jast well, not *all* programmers are the target audience
MadcapJake jast: that i fully disagree with
jast I know a lot of programmers who find the ideas of Perl 6 completely off-putting
and it's their good right to prefer a different style of language 22:14
MadcapJake jast: doesn't mean that we shouldn't target all programmers just because a few find it "off-putting"
ZoffixWin MadcapJake, well, to give an example. Would you expend effort trying to convince a group of die-hard Pythonistas to use Perl 6 instead?
MadcapJake yes
jast if you can somehow manage to "segment the market" to reach only those people who are likely to enjoy Perl 6, that would be lots more efficient
ZoffixWin MadcapJake, but that means you have fewer resources to target other segments.
MadcapJake jast: haha right, that's exactly it, you can't do that
ZoffixWin: not if you just target *all* programmers 22:15
jast well I'm not so quick with blanket statements of impossibility
ZoffixWin MadcapJake, that's akin to trying to sell an energy drink to "*all* people"
It's too utopian a goal to achieve realistically.
MadcapJake but they do target all people! they just use their messaging style and presentation to filter
jast not really 22:16
ads for energy drinks are way more likely to run during sports events than, say, soap operas
ZoffixWin Or be posted in retirement homes
MadcapJake lol, ok fair enough
ZoffixWin :)
MadcapJake imagines an energy drink ad at a retirement home
stmuk_ is this someone here? :) 22:17
MadcapJake "do you want to *blast* through another round of jazzercise! do you want to slaughter your opponents in bingo? then drink Monster!"
jast that might actually be doable
perlpilot notices mention of perl6.slack.com ...
when did that happen?
stmuk_ perl.postbit.com/upload/109/perl-6-logo.png
ZoffixWin "Kids never visit? Nurses are mean? Sick of this miserable life? Chug MONSTER DRINK and have a heart attack!"
stmuk_ :D
MadcapJake ZoffixWin: LOL
perlpilot stmuk_: that is an angry butterfly 22:18
DrForr o/ Rockin in the rest home in 2030 o/
stmuk_ perlpilot: it's scary
ZoffixWin stmuk_++ Not a fan of that font, but the graphic is awesome :)
jast my first thought was "angry", too
22:19 [Sno] joined
ZoffixWin my first thought was "is that an owl" :) 22:19
stmuk_ its not mine! I just found it
perl.postbit.com/ is good
perlpilot so, who is using perl6.slack.com? who set it up? 22:20
22:20 perlawhirl joined
perlpilot (or is it not setup and merely a dream for now?) 22:20
ZoffixWin perlpilot, It is setup, I believe MadcapJake set it up. And based on the log in it, several people are using it :) 22:21
22:21 saaki_ is now known as saaki
perlpilot okay, how do I use it? :-) 22:21
jast I prefer my chat without a DOM and JS engine
22:22 leont left
jast I think you just open it in your browser and click 'join' or something 22:22
22:22 franciskittu left
MadcapJake perlpilot: perl6.bestforever.com to signup 22:22
jast oh. cheaty.
MadcapJake jast: you can use scudcloud, it's a really nice slack client (still has dom and js but at least it's not in a browser)
jast or I can just use IRC ;P 22:23
jdv79 what is slack for? just as an irc alternative?
jast yeah
plus snippet/file sharing, I believe
perlpilot jdv79: it's a browser-based IRC-like thingy
jast and it keeps (limited, unless you pay $$$) logfiles
perlpilot yep
MadcapJake jdv79: it allows teams, code comments, nice interface, and more 22:24
jdv79 ok
DrForr Basecamp-a-like if you've used that.
RabidGravy or even google wave
perlpilot heh
jast google what? ;)
perlpilot What's google wave? ;)
MadcapJake s/What's/What was/ 22:25
perlpilot don't speak of imaginary things
RabidGravy was a thing that nobody knew what it was for
MadcapJake lol
i think people wanted it after they'd already discontinued it
jast is wave some kind of buzz?
RabidGravy they open sourced most of the code
mspo some kind of messaging platform
jast (never even noticed buzz getting killed...) 22:26
MadcapJake one of those cultural things where people like it more after it's gone
22:26 franciskittu joined, nebuchadnezzar left, nebuchadnezzar joined
jast d'aww, orkut was killed, too 22:26
mspo wave is a lesson in platform sharecropping/relying on google 22:27
perlpilot I dunno we used wave for collaborative event planning ("Hey, let's go to the movies on Saturday! You in?" 22:31
)
22:32 vendethiel left
MadcapJake For anyone interested in the wider "Slacksteria": medium.com/@angiecois/an-incomplet....ndz11blxt 22:32
perlpilot wave was neat, but it wasn't neat enough to build something on top of
MadcapJake: that looks like a slack-ad to me :) 22:33
jdv79 i didn't know we weren't happy with irc. i am.
MadcapJake perlpilot: definitely not, they really aren't sure on how much they want slack used for communities rather than teams 22:34
jdv79: again, this is not about replacement.
buharin hey what does it mean
perlpilot indeed.
buharin adptme
adopt
adoptme
jdv79 but if something happens over there it doesn't here so its about fragmentation
perlpilot buharin: where did you see that in a Perl 6 context?
buharin blogs.perl.org/users/zoffix_znet/20...rsday.html 22:35
MadcapJake jdv79: so communities can't have several communication channels? Everything must happen on IRC?
buharin I want to help with some easy module for adopt :D
but completely don't know how to move on
perlpilot buharin: for Perl 5 modules, that's usually the signal that the maintainer has given up on maintaining their module and they want someone else to take it over. I'm guessing it's being used similarly for P6
geekosaur jdv79, we've discussed gatewaying slack<->irc 22:36
perlpilot ah, Zoffix++
geekosaur probably in a separate channel on irc that people can join or not as they wish
DrForr buharin: I think the community is a bit young for adopting modules, maybe it'd be better to look at the 'most wanted'?
buharin ohh 22:37
jdv79 having multiple places for the same purpose, or similar, just makes it harder to keep up with it all.
buharin so some guy left their code
perlpilot jdv79: just accept the fact that you won't keep up "with it all" from the outset :)
buharin and will not rewrite it to perl6
geekosaur (slack supports several kinds of gateways. what it doesn't support is "I don't want to deal with 15 different comm programs")
ZoffixWin buharin, to clarify, there are two different languages Perl 5 and Perl 6. Perl 5 has ADOPTME PAUSE user that is usually used for when a PAUSE author has passed away.
dalek c: 2e6962b | (Wenzel P. P. Peppmeyer)++ | doc/ (2 files):
doc Empty
22:38
ZoffixWin buharin, why "some guy"? Maybe it's a girl :)
geekosaur and if you want to "keep up with it all", how do you deal with stackoverflow+reddit+...?
ZoffixWin Or a hyper intelligent AI
jdv79 yeah its fine. just my 2 cents.
perlpilot &
22:38 donaldh left
geekosaur the world's already too big to keep up with everything related to some topic, unless that topic is already effectively dead (and sometimes not even then) 22:39
MadcapJake geekosaur++
buharin typical question how can I help perl?:D
DrForr buharin: Writing docs would be a *real* help.
jdv79 also, irc is public and logged and googleable. ok, i'm done now. 22:40
ZoffixWin buharin, easiest way? Blog about it and talk about it on Social Media, blogs, Web, and in person
buharin ZoffixWin, you mean learn by myself and learn the others
gfldex buharin: don't be shy with hugs and kisses
MadcapJake buharin: github.com/perl6/doc/blob/master/WANTED
not sure if that's udpated though :P 22:41
ZoffixWin buharin, a good way to learn is to teach :) You learn a new cool thing, write a blog post explaining it
MadcapJake where is that most-wanted module list?
ZoffixWin buharin, we also have the Most Wanted modules list: github.com/perl6/perl6-most-wanted...modules.md
MadcapJake haha, that was fast!
ZoffixWin I'm a mind reader :P 22:42
buharin, as well as an Issue with a consideration for cleaning up the Most Wanted modules list :P github.com/perl6/ecosystem/issues/136
buharin, this is of course for the Perl 6 programming language. You may also be interested in Perl 5 language whose modules "some guy" offered for adoption in the blog post you mentioned. The channel for that programming language is #perl 22:43
RabidGravy Are we considering that anything that is actually on the modules list need not be in the most wanted list anymore?
if so I'll cut out the ones that I have actually released at some point 22:44
buharin ZoffixWin, nooo
Perl5 is obsolete
ZoffixWin buharin, :S
jdv79 the r* list is also kinda outdated
ZoffixWin buharin, what made you think so? The latest release was 50 *days* ago
RabidGravy, that was my proposal, yes. Is there a point in keeping that sort of thing on Most Wanted, considering it already exists and thus no longer "wanted"? :) 22:45
RabidGravy I'll have a look but leave the ones that I haven't released yet
ZoffixWin buharin, that's comparable to saying C is obsolete in ##c++ channel.
22:46 buharin left
jast what just happened there 22:46
ZoffixWin I guess I convinced them :P
DrForr Must not have baited his hook well enough. 22:47
ZoffixWin I doubt there was a hook, considering the conversation started with them linking to my blog post offering module adoption :P
dalek rl6-most-wanted: b8cd9f5 | RabidGravy++ | most-wanted/modules.md:
Remove released modules
22:48
RabidGravy two really are WIP still
DrForr I'm willing to accept 'confused' as well, but I'm counting it as "weak troll"/I can't believe I wasted my time typing this sentence. 22:49
dalek kudo-star-daily: 2690fda | coke++ | log/ (8 files):
today (automated commit)
22:49 AndChat380436 left
ZoffixWin I'm betting on "confused" :) 22:49
Was R* released today? 22:50
ZoffixWin just stumbled upon the release date on en.wikipedia.org/wiki/Perl_6
geekosaur got the email about half an hour ago 22:51
22:51 buharin joined
[Coke] (documentation on the signature) that was my point originally - perl6 already embeds a LOT of information in the signature. if we can make those more obvious, we don't have to maintain separate usage. 22:52
(synopses) Don't trust those as canon at this point. they are speculative in many cases compared to the spec itself.
dalek ar: ec0dea1 | FROGGS++ | tools/build/star-product.wxs:
choose to set PATH in wizard on windows
22:53 franciskittu left
[Coke] dha: do not look things up on perlfoundation.org. 22:53
buharin hey is it like you write a module which is most wanted
then you write on irc that you have done it
and some guy annonce
it
jdv79 something like that 22:54
add it to the eco
DrForr buharin: When you're done there are instructions on the front page of modules.perl6.org on what to do.
buharin oh ok I see 22:55
;D
DrForr Talking about it on IRC doesn't hurt, but also doesn't get it added to the list :)
ZoffixWin collects the bet
[Coke] jdv79: there is already no way to keep up with it all
jdv79 i know. that was only a partial argument. i don't care enough to get into it full on. 22:56
RabidGravy right, toodles people
jdv79 nite
[Coke] I for one will very likely never sign on to slack, but go for it if someone wants to be there. I probably don't want a bridge if it's going to be noisy.
I'll keep up with the FB Perl 6 group. 22:57
sevvie [Coke]: curiously, why are you so resistant to signing onto slack? 23:00
23:00 RabidGravy left 23:01 yqt joined, buharin left
hoelzro ab5tract: booking is starting to use slack now? 23:05
AlexDaniel lol, why not make a Skype groupchat? :D
perlawhirl i'm hanging out for some perl fosdem'16 videos. still nothing uploaded from h2214 :( 23:06
23:07 cdg left
hoelzro The "Download Slack" page has a "Linux (Beta)", which has a dropdown with two distros ='( 23:09
sevvie I only ask because Slack benefits as a "install it on your mobile and get messages and notifications directly from the community" kinda thing. 23:10
</elevatorpitch>
23:10 ZoffixWin left
hoelzro sevvie: I think most people here are probably of the "slack doesn't offer anything IRC doesn't mindset", but that doesn't mean we shouldn't have a Slack team for Perl 6 23:10
AlexDaniel wow, 38 MB download 23:11
hoelzro then again, I don't mean to speak for everyone
it would be wise to have a channel topic in Slack that says "if you don't get an answer here, try IRC"
23:11 khw left
AlexDaniel where are the sources for GNU/Linux slack client? 23:13
hoelzro haha
at work, I've been using irssi + slack.pl; works pretty well
sevvie: I just saw you filed an issue against linguist for perl6; I'm sorry =( 23:14
oh wow, they got back to you within a day
that's atypical
23:15 ramon joined, ramon is now known as Guest65773
MadcapJake hoelzro: yeah it's already merged, and the dev said he'd push to github later this week! 23:15
hoelzro nice!
sevvie++ MadcapJake++
it took *forever* for them to move on requests I made
maybe it was how I asked =/ 23:16
MadcapJake hoelzro: yeah timotimo showed me a few of the PRs that took around a year to get merged xO
23:17 vendethiel joined
MadcapJake perhaps there was a change of the guard or maybe the dev has more time now to deal with linguist. I think the most likely answer is that this was an easy fix as it didn't require any change to the heuristics 23:17
23:18 laz78 left
Skarsnik AlexDaniel, I only use the webclient :( 23:18
23:19 azawawi joined, FROGGS left
azawawi hi 23:19
github.com/azawawi/atom-perl6-editor-tools # POD Preview in Atom :)
Skarsnik nice 23:20
azawawi and a couple of snippets for fun :)
script<tab> => perl6 script 23:21
hoelzro well, I'm just glad that they merged it quickly =)
Skarsnik I give an internet cookie if someone make Pg Array in DBIish work correctly (basicly returning correctly Array in a hash) 23:24
AlexDaniel Skarsnik: but the linux client is free software, right? I just cannot find the sources 23:26
Skarsnik No idea
23:27 skids joined
hoelzro AlexDaniel: for slack? 23:27
MadcapJake AlexDaniel: github.com/raelgc/scudcloud
AlexDaniel MadcapJake: that's not the official one
hoelzro the official one is not OSS, afaik 23:28
Skarsnik just write a slack module already
MadcapJake AlexDaniel: oh, didn't realize there was an official one :P, scudcloud is great though btw
There's also github.com/plaidchat/plaidchat
Hotkeys Is there a way to add a method to an imported class from within another program 23:29
without just editing the class source
AlexDaniel hoelzro: Interesting. Why did they bother making a closed-source one?
what's the point?
hoelzro AlexDaniel: I don't know; I don't really *get* slack 23:30
both the company and the software
but a lot of people seem to like it
so I guess it's just not *for* me
Skarsnik It's really nice
hoelzro it's nice, but I don't see the advantage (for me personally) 23:31
Skarsnik You understimed how annoying it can be to have something like IRC that don't render stuff the same accross client and does not offer link preview
hoelzro I understand that a lot of people like gifs to be rendered inline in their chat client; I personally don't
Skarsnik and the webclient part is good because most company block IRC 23:32
hoelzro I understand that IRC is a technical challenge for some people
AlexDaniel so, you can't just join a group and someone has to invite you? You have to register by using your email in order to get in? They provide a proprietary linux client while there are some free ones? The whole thing is stored on their servers (and we also don't have the sources of that)? Hmmmm… 23:33
okay, nvm, I'll just move on…
hoelzro AlexDaniel: yes; if anything, I favor IRC over slack because IRC gives the community a higher degree of control 23:35
but my mentality is "use what you like"
AlexDaniel hoelzro: interestingly there are ‘erc-image-mode’ and ‘erc-youtube-mode’ things that will render images inline in emacs (well if you're using ERC). These are turned out by default in spacemacs but it drives me crazy :) 23:36
and I guess that some other IRC clients may do that as well
turned on*
hoelzro that's the thing; IRC the protocol doesn't provide a lot of the features slack does; it's up to the clients to do it
AlexDaniel hoelzro: which is great 23:37
hoelzro agreed
Skarsnik nop
it's not great 23:38
just take a simple example: colors
it does not work exactly the same on every client
23:39 vendethiel left
AlexDaniel great! I don't think that everyone likes colors. And that's why it was possible (still is?) to disable colors on some channels 23:39
hoelzro Skarsnik: I'm not arguing that slack isn't good; I'm just stating that I prefer IRC, and agreeing with AlexDaniel that having protocol-client separation for features is a good thing™ 23:40
Skarsnik Yes, but IRC is a bad example
hoelzro I don't think enforcing colors is great either; I have a colorblind friend who gets driven *nuts* by popular colorschemes
(granted, I have a feeling slack provides themes that are colorblind-friendly)
Skarsnik I use IRC, Discord, Slack 23:41
When I talk with no developper people I prefer Discord/Slack over IRC
because image/link preview is fun :)
hoelzro anyway, I don't want to get into an IRC vs Slack discussion; I'm in favor of having both. If you want to keep talking about it, I'm happy to move this conversation elsewhere =) 23:42
azawawi MadcapJake: do not use farabi6
MadcapJake: do not use farabi6's codemirror mode... github.com/azawawi/farabi6/issues/46 . It is not finished nor started :( 23:43
hoelzro s/don't want to(.*);/don't want to$1 here;/
23:45 addison_ left
azawawi MadcapJake: but you're welcome to use github.com/azawawi/farabi6/blob/ma...de.js#L219 :) 23:45
23:45 virtualsue joined
Skarsnik what is really sad, it's that slack should have be something xmpp (the jabber underlaying stuff) should have beeing a lonkg time ago 23:46
because I am pretty sure it use web socket :)
23:47 vendethiel joined
azawawi Skarsnik: alright 2 Downloads already atom.io/packages/atom-perl6-editor-tools :) 23:47
Skarsnik I should try atom :)
I am using kate with an old p6 hightlighting x) 23:48
good night #perl6
23:49 Skarsnik left
azawawi MadcapJake: if you cant beat them (atom), join them :) 23:49
23:50 spider-mario left 23:55 espadrine left 23:56 wamba joined 23:58 SCHAAP137 left