»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 m: :(42) .say 00:00
camelia rakudo-moar 103781: OUTPUT«(Int $ where { ... })␤»
sortiz bbrk, till 13 without problems, need more? 00:01
bbkr sortiz: no, thanks for testing. looks like it is something OS X specific. 00:02
00:02 Skarsnik joined, tphilipp joined
bbkr sortiz: got this bug when I was trying to fix Net::ZMQ NativeCall warnings. so I just have to fix everything at once to avoid having multiple invalid signatures that leads to this compile time lock. that's motivation :) 00:05
sortiz :P 00:06
00:06 leobut joined 00:07 espadrine left
sortiz bbkr, yep working with NC requires, sometime, lots of motivation ;-) 00:07
Skarsnik true x) 00:08
sortiz \o Skarsnik!
00:09 ilbelkyr is now known as ishanyx 00:11 tphilipp left 00:12 jack_rabbit joined
AlexDaniel m: {[~] |S/' '/_/.comb(/\D/)».&{(.lc,.uc).pick},|.comb(/\d/)}(‘pa55 w0rd’).say 00:15
camelia rakudo-moar 103781: OUTPUT«pA_WRD550␤»
AlexDaniel Hotkeys: ↑
Xliff psch, thanks! 00:16
00:17 tardisx left 00:18 leobut left, leobut joined 00:19 [Tux] left 00:21 [Tux] joined 00:22 leobut left 00:23 leobut joined
AlexDaniel m: say [~|] ‘_’, ‘w’ # what's that? 00:26
camelia rakudo-moar 103781: OUTPUT«^H␤»
diakopter hunh 00:27
00:27 _nadim left 00:28 skids joined
AlexDaniel m: say [~|] ‘x’, ‘a’; 00:28
camelia rakudo-moar 103781: OUTPUT«y␤»
sortiz m: say (0x5f +| 0x7f).base(16) 00:29
camelia rakudo-moar 103781: OUTPUT«7F␤»
00:29 _nadim joined
AlexDaniel so it is just a stringy or? 00:29
sortiz Yep 00:30
00:30 sufrosti1o left 00:31 Actualeyes joined, leobut left 00:32 leobut joined 00:33 tphilipp joined 00:35 _nadim left
AlexDaniel Hotkeys: that's kinda it, I'm out of ideas now :) 00:35
m: $_ = ‘hello world’; say S:s/o w/ow/ 00:36
camelia rakudo-moar 103781: OUTPUT«helloworld␤»
AlexDaniel m: $_ = ‘hello world’; say S:s/ /_/
camelia rakudo-moar 103781: OUTPUT«5===SORRY!5=== Error while compiling /tmp/DcUqHw5_zv␤Null regex not allowed␤at /tmp/DcUqHw5_zv:1␤------> 3$_ = ‘hello world’; say S:s/ 7⏏5/_/␤»
AlexDaniel is it a bug?
it is definitely not what I meant, but is there any reason for this behavior? 00:37
m: $_ = ‘hello world’; say S:s/o /o/
camelia rakudo-moar 103781: OUTPUT«helloworld␤»
00:38 tardisx joined
sortiz m: $_ = ‘hello world’; say S:s/' '/_/ # For the record only 00:41
camelia rakudo-moar 103781: OUTPUT«Cannot unbox a type object␤ in block <unit> at /tmp/jXKGp7N7bp line 1␤␤»
AlexDaniel sortiz: well, you don't need :s for that 00:42
00:43 khw left
sortiz AlexDaniel, I know 00:43
AlexDaniel m: $_ = ‘hello world’; say Ss/o /o/
camelia rakudo-moar 103781: OUTPUT«helloworld␤»
Hotkeys AlexDaniel: what the heck is this |S/' '/_/
what is capital S///
AlexDaniel Hotkeys: just like s but not in-place 00:44
Hotkeys: | is just to slip it, so that [~] works
Hotkeys so you can .method on S///
00:44 tphilipp left
Hotkeys for the record I did try to do that, but it was with s/// 00:44
didn't know S/// existed
AlexDaniel m: $_ = ‘hello world’; S/hello/noo/; say $_ 00:45
camelia rakudo-moar 103781: OUTPUT«hello world␤»
AlexDaniel m: $_ = ‘hello world’; s/hello/noo/; say $_
camelia rakudo-moar 103781: OUTPUT«noo world␤»
00:45 tphilipp joined
sortiz Hotkeys, S is the 'general' form, like Q for quotes. 00:45
00:45 _nadim joined
Hotkeys wtb documentation 00:45
AlexDaniel sortiz: how do you make S act like s?
00:47 cdg left
sortiz AlexDaniel, I don't remember the details now. :( 00:47
00:47 cdg joined 00:48 cdg left
Hotkeys m: {[~] |S/' '/_/.comb(/\D/)».&{(.lc,.uc).pick},|.comb(/\d/)}("hello there jim").say 00:48
camelia rakudo-moar 103781: OUTPUT«hELlo_theRe JIm␤»
Hotkeys m: {[~] |TR/' '/_/.comb(/\D/)».&{(.lc,.uc).pick},|.comb(/\d/)}("hello there jim").say 00:49
camelia rakudo-moar 103781: OUTPUT«5===SORRY!5=== Error while compiling /tmp/yjpza1Rhp_␤Undeclared names:␤ TR used at line 1␤ _ used at line 1␤␤»
Hotkeys m: {[~] |T/' '/_/.comb(/\D/)».&{(.lc,.uc).pick},|.comb(/\d/)}("hello there jim").say
camelia rakudo-moar 103781: OUTPUT«5===SORRY!5=== Error while compiling /tmp/fCeTxXI8NZ␤Undeclared names:␤ T used at line 1␤ _ used at line 1␤␤»
Hotkeys m: {[~] |S:g/' '/_/.comb(/\D/)».&{(.lc,.uc).pick},|.comb(/\d/)}("hello there jim").say
camelia rakudo-moar 103781: OUTPUT«HEllO_therE_jIm␤»
Hotkeys need an extra 2 bytes for global
00:50 _nadim left 00:54 tphilipp left 00:55 dvinciguerra__ left, tphilipp joined 00:56 aindilis joined
gtodd still wants to make a stab at a perl6 version of some of these scripts for reading large files stackoverflow.com/q/36201884/2019415 00:56
00:57 leobut left
gtodd particularly like the Stream::Reader CPAN module 00:57
00:58 Ben_Goldberg joined, Ben_Goldberg is now known as BenGoldberg
gtodd it was very easy to write a perl6 script that worked with a small file by converting the string to a list (.comb) ... but this doesn't work when the string a a billion characters long 00:58
00:59 silug joined
BenGoldberg forgets, is .comb lazy? 00:59
01:01 tphilipp left, labster left, tphilipp joined 01:02 tphilipp left, labster joined 01:04 tphilipp joined 01:05 tphilipp left 01:06 tphilipp joined
gtodd I'm not sure the string was very small ... I think on a bigger string it was waiting or whatever it does when the list is huge 01:08
01:08 _nadim joined
gtodd so .... maybe that means it is :) 01:09
01:14 tphilipp left
perlawhirl hrmph 01:17
.seen sortiz
yoleaux I saw sortiz 00:47Z in #perl6: <sortiz> AlexDaniel, I don't remember the details now. :(
perlawhirl anyone online who has admin access to the ecosystem 01:19
sortiz perlawhirl, hi
Can I help you? 01:20
AlexDaniel Hotkeys: right! I forgot about global 01:21
perlawhirl umm, i'm building rakudo on a new box and i can't install my module, wig
the update.log looks ok
and the modules before and after mine are in projects.json, but not mine :~( 01:22
i ran panda update to no avail
sortiz Let me see...
AlexDaniel m: $_ = ‘hello world’; s/o/X/; say $_ 01:23
camelia rakudo-moar 103781: OUTPUT«hellX world␤»
AlexDaniel m: $_ = ‘hello world’; sg/o/X/; say $_
camelia rakudo-moar 103781: OUTPUT«5===SORRY!5=== Error while compiling /tmp/innHOYa2gH␤Missing required term after infix␤at /tmp/innHOYa2gH:1␤------> 3$_ = ‘hello world’; sg/o/X/7⏏5; say $_␤ expecting any of:␤ prefix␤ term␤»
AlexDaniel m: $_ = ‘hello world’; ss/o/X/; say $_
camelia rakudo-moar 103781: OUTPUT«hellX world␤»
AlexDaniel m: $_ = ‘hello world’; ss/o /X/; say $_
camelia rakudo-moar 103781: OUTPUT«hellXworld␤»
AlexDaniel ss is fine but sg is not, hmm… 01:24
why?
perlawhirl sortiz: i manually deleted my projects.json file and ran update
now it's there
i dunno *shrug* 01:25
it was a fresh install so dunno what went wrong
sortiz umm, well, back to work.
perlawhirl yep :D 01:26
weird that panda update didn't resolve it? how does it check for updates? does it do a checksum or just check the last module added? 01:27
i might remove a project in the middle of the projects.json and test it
AlexDaniel Hotkeys: how can I run your .flat.sort.rotor example?
Hotkeys: you say that it takes a list of lists, but 01:28
m: {$^a.flat.sort.rotor($a[0])}( ((-36, -18, 0), (8, 99, 112), (14, 6, -12)) ).say
camelia rakudo-moar 103781: OUTPUT«Cannot have elems < 1, did you mean to specify a Pair with => -36?␤ in block <unit> at /tmp/vhaLbwIQHi line 1␤␤»
sortiz perlawhirl, panda guts are a mystery to me, sorry. 01:29
perlawhirl hah, ok well i did that test and it updated just fine, so i'm perplexed, but will leave alone... just another mystery of the universe 01:30
Hotkeys AlexDaniel: oops 01:31
seems I forgot a +
m: {$^a.flat.sort.rotor($a[0])}(((-36, -18, 0), (8, 99, 112), (14, 6, -12))).say
camelia rakudo-moar 103781: OUTPUT«Cannot have elems < 1, did you mean to specify a Pair with => -36?␤ in block <unit> at /tmp/Wx1kWp7glB line 1␤␤»
Hotkeys m: {$^a.flat.sort.rotor(+$a[0])}(((-36, -18, 0), (8, 99, 112), (14, 6, -12))).say
camelia rakudo-moar 103781: OUTPUT«((-36 -18 -12) (0 6 8) (14 99 112))␤» 01:32
Hotkeys m: {.flat.sort.rotor(+.[0])}(((-36, -18, 0), (8, 99, 112), (14, 6, -12))).say 01:34
camelia rakudo-moar 103781: OUTPUT«((-36 -18 -12) (0 6 8) (14 99 112))␤»
Hotkeys nice
AlexDaniel right
Hotkeys that's as short as I can see to get it 01:35
Any ideas for shortening?
AlexDaniel Hotkeys: are you kidding me? :) 01:36
Hotkeys is that a "yes obviously I can shorten it" or "no, that's obviously the shortest it can get" 01:37
01:37 grondilu joined
AlexDaniel Hotkeys: no, it's the shortest it can get 01:37
Hotkeys alright
AlexDaniel Hotkeys: one of the ideas is to replace flat with something like |«| or something… 01:38
Hotkeys Yeah
I actually tried that lol
AlexDaniel Hotkeys: but I can't get that to work
Hotkeys i.imgur.com/WZ96Ovu.png
I also tried hoping for deepsort (see: deepmap) but that isn't a thing 01:41
AlexDaniel heh, deepsort
Hotkeys :)
01:44 kid51 left 01:50 Herby_ joined
Herby_ Evening, everyone! 01:50
\o
o/ 01:53
01:56 Actualeyes left
skids o~ 01:56
AlexDaniel @🐛@ 01:57
Herby_ \o/
AlexDaniel ↑ found that somewhere in the log. Hilarious
skids doesn't render here.
01:58 Actualeyes joined
AlexDaniel .u 🐛 01:59
yoleaux U+1F41B BUG [So] (🐛)
AlexDaniel skids: well, it was supposed to look like camelia, but: files.progarm.org/2016-04-01-04591..._scrot.png
Hotkeys here 02:03
m: sub deepsort (@a) {my @t; @a.deepmap({@t.push: $_}); @t.=sort; @a.deepmap({@t.shift})}; say deepsort [[3, [9, [2, 6], 7], [0, 1, [[[-13, 4], 5], 10], 17, -3], -42], 8]
camelia rakudo-moar 103781: OUTPUT«[[-42 [-13 [-3 0] 1] [2 3 [[[4 5] 6] 7] 8 9] 10] 17]␤»
Hotkeys :D
AlexDaniel Hotkeys: I'm more interested in shortening the password thingy 02:04
Hotkeys I wasn't doing any shortening in deepsort :p
AlexDaniel Hotkeys: if I had a bit more brains maybe I could have shortened it by a character or two
Hotkeys just having fun
AlexDaniel Hotkeys: just saying :)
02:19 Skarsnik left 02:28 cognominal left 02:38 raoulvdberge left 02:40 tardisx left 02:44 noganex_ joined 02:48 noganex left 02:53 tardisx joined
MadcapJake what does this error mean: «Cannot look up attributes in a type object»? 02:59
03:00 japhb joined
sortiz MadcapJake, your method is not receiving an instance, i.e. self is undefined 03:00
03:02 lizmat_ joined
MadcapJake I don't understand how it's getting a type though, I use the builtin new method. 03:03
03:04 lizmat left
sortiz The construction (new or BUILD) can be failing. 03:05
MadcapJake yeah it must be inside BUILD 03:06
03:06 TreyHarris joined
sortiz Inside BUILD, self is undefined yet. 03:06
Umm, wait. nop 03:07
In new is undefined. At BUILD should be defined, unless something failed. 03:09
ugexe MadcapJake: are you trying to access any members of the class that would not have been initialized until after BUILDALL? 03:11
m: class Foo { has $.a; method new { say $.a; }; }; Foo.new 03:12
camelia rakudo-moar 103781: OUTPUT«Cannot look up attributes in a type object␤ in method new at /tmp/nf2rM8TaVF line 1␤ in block <unit> at /tmp/nf2rM8TaVF line 1␤␤»
ugexe also 03:16
m: class Foo { has $.a; method bar { 1 }; method new { say self.bar; }; }; Foo.new # self can still be used to do stuff
camelia rakudo-moar 103781: OUTPUT«1␤»
skids Man I wasn't even thinking about smoking until that stupid Truth campaign commercial came on. 03:23
03:24 japhb left
ugexe the one with the catchy rap hook where they throw money makes smoking look cool as hell 03:25
03:26 japhb joined
skids m: class A { method new { self.perl.say } }; A.new; # in .new, self is just the type object (unless you do something like A.new.new of course) 03:27
camelia rakudo-moar 103781: OUTPUT«A␤»
03:28 matiaslina joined
skids There are two options, use BUILD directly, or put your own call to bless in .new and work with the object it returned. 03:28
(and be sure to return that object)
03:31 BenGoldberg left, Herby_ left
skids That campaign is mendaciously deceptive and makes kids stupider. I have to say as a smoker, the campaign with the wife-beater-wearing "bully" cigarette is both funny and discourages me from smoking, a very tiny bit, and to top it off doesn't tell lies. 03:32
03:33 geekosaur joined
MadcapJake the call that has a probem is @!tickets.push($ticket) in a separate method. I have a Ticket:D constraint on the ticket coming in, then I just push it to a constrained «has Ticket @!tickets» array, and that's where the error happens, any ideas? 03:34
I've got all the guards up, so-to-speak, but for some reason this strange type object error 03:35
03:35 tardisx left
MadcapJake actually, before entering this add-to-array method, I tried $ticket.WHAT.say and I get the same error but from within the gist method?? o_O 03:38
huh, didn't realize WHAT calls gist 03:39
tony-o do you have a gist or something somewhere MadcapJake ?
MadcapJake yeah
03:41 vike left, matiaslina left 03:42 matiaslina joined
MadcapJake gah, I really am stumped 03:42
removing that gist method gives me (Ticket) but that doesn't really help me solve this, now back to the same error but inside the add-to-array method :S 03:43
anyone want to take a gander? github.com/MadcapJake/sixbug/blob/...#L179-L182 03:46
03:49 cdg joined 03:53 vike joined 03:54 cdg left, geekosaur left 03:57 geekosaur joined
skids m: class B { }; class A { has B @!b; method foo(B:D $b) { @!b.append($b); @!b.say; self; } }; A.new.foo(B.new).perl.say; 04:01
camelia rakudo-moar 103781: OUTPUT«[B.new]␤A.new␤»
skids "RT.add($ticket);" is calling .add on the type object of RT which does not have a @!tickets. 04:03
MadcapJake skids: doh! thanks so much! totally missed that 04:04
skids Would've been nice if the error message said which type object you tried to look up an attribute in, for sure.
04:05 labster left
MadcapJake yeah agreed that is definitely a LTA error message 04:05
04:06 buharin joined 04:07 wamba joined 04:23 labster joined
perlawhirl MadcapJake: Not sure if you saw it, but it lives: github.com/0racle/p6-wig 04:24
04:34 skids left 04:35 matiaslina left 04:43 Cabanossi left 04:44 Cabanossi joined
MadcapJake perlawhirl: sweet! any word from TimToady on it? 05:13
so i'm converting a big list of hashes into json, would it make sense to use .race and call to-json on each hash? I'm curious if that would actually get me a speed increase (currently to-json on the whole array is quite expensive) 05:15
05:15 molaf joined 05:19 Vlavv_ left 05:25 remmie joined 05:32 Vlavv_ joined
MadcapJake hmm not sure that's where the bottleneck is actually 05:34
05:35 wamba left
MadcapJake lol, sooo my request to rt for tickets takes 43 seconds, building json for it takes 58 seconds! altogether a 2 minute wait to load the page! 05:41
needless to say, I am now caching the tickets and the json :P
05:43 lustlife joined 05:46 buharin left
perlawhirl MadcapJake: I haven't asked him... don't think anyone has 05:48
i dunno what the ettiquette is :S 05:49
MadcapJake hrmm 05:53
using race shaved 20 seconds off that big to-json op :D
but now I think I'm just gonna build the json myself (it's very regular)
perlawhirl i find race inconsistent, but then again, i'm on a 32-bit machine, so things are slower in general 05:55
05:56 _nadim left
MadcapJake I'm on a six core machine! OH YEAH! ;) 05:57
05:57 _nadim joined
MadcapJake well either way, converting all this to-json stuff to simple string concat should help speed things up 05:57
I wonder if race actually degrades performance on single-core machines 05:58
but wrt "where" I think your proposal (was you, right?) was really well written and compelling, I'd just message him on IRC and ask if he'd read it and whether he'd be open for it. 05:59
oh yeah! raw to-json: 58 seconds, .race to-json: 38 seconds, string ops + race: 1.5 seconds (priceless) 06:00
06:00 _nadim left
perlawhirl something also about rakudo doesn't do some JIT stuff on 32-bit... someone said in passing (probably jnthn) that flew over my head 06:01
1.5 seconds! nice
06:01 _nadim joined
MadcapJake perlawhirl: oi that's a bummer! it's surprising to me that JIT could optimize differently based on sizes of addresses, but pretty much everything suprises me about JIT :) 06:02
perlawhirl it will be nice when rakudo does things the obvious way quickly, rather than having to refactor around the slow parts
MadcapJake agreed! TBH, I find that often I reach for to-json when really the data is regular enough that I don't need it. 06:03
perlawhirl have you seen the Scientist module. How does it differ from your Test-Lab ?
MadcapJake (the real question is, is this valid json?) 06:04
perlawhirl i just noticed today there were 2 similar modules and wondered if one was better
it's ok, you can say yours is better :D
MadcapJake hehe, mine is a bit more DSL-like whereas lancew's is more perl6-ish
06:05 lizmat_ is now known as lizmat
MadcapJake I tried to mix the rubyishness of Scientist's origin with some of the cool features that Perl 6 supplies but I didn't want to do named arguments because two things, (1) you lose some of the flexibility of manipulating the experiment as you're now constrained to the named arguments provided and (2) the code to implement an experiment as named arguments (in 06:06
terms of the original Scientist API) would be much larger (and perhaps flakier) than just letting the user do it.
moritz \o 06:07
ZoffixWin++'s "Wow, Perl 6" presentation is currently No. 5 on hackernews
MadcapJake woah!
it's quite a good presentation, he hits a lot of cool points
MadcapJake goes to contribute his ▲ 06:08
lizmat commute to NLPW&
06:08 lizmat left
moritz news.ycombinator.com/item?id=11401330 06:09
MadcapJake perlawhirl: also, I implement the entirety of github's Scientist API, including every single test, it's almost an exact translation into Perl 6 06:10
06:10 pecastro_ joined 06:11 buharin joined
MadcapJake is nearing completion of the Issue Submitter 06:11
everyone hold all issues until I get this up and running! ;) 06:12
perlawhirl wow, not too much hate on HN comments. still sad that people dismiss perl off-hand for reasons that are isolated to their own prejudices 06:17
06:18 buharin left, _nadim left
MadcapJake perlawhirl: I've noticed that Perl 6 goes over quite well on HN, really poorly on Slashdot and mixed on Reddit :) 06:18
perlawhirl fair enough... i don't really visit them... reddit a tiny bit 06:19
how can you dislike a language that makes things this easy!!: say "These are my things: @things.join(', ')." 06:20
who needs printf anymore :D
MadcapJake yeah a few people were saying the ole lisper-line "why have operators when you can have a huge wall of parenthesis and function names" ugh x_x 06:22
06:28 leont joined 06:30 sheldon joined 06:31 sheldon left, pecastro_ left 06:42 perlawhirl left 06:44 geekosaur left 06:45 geekosaur joined 06:47 leont left
remmie is there perl6 syntax highlighting for sublime anywhere? 06:50
07:07 labster left 07:12 sjoshi joined 07:18 perlawhirl joined 07:23 AlexDaniel left, perlawhirl left 07:25 perlawhirl joined
sortiz "Bytecode validation error at offset 522, instruction 78: operand type 160 does not match register type 152" when using a 'uint32' constrained named argument in a method call. Without the constraint the problem vanishes. 07:26
07:26 labster joined
moritz sortiz: please submit as a bug 07:28
07:28 ufobat joined
ufobat good morning guys 07:29
07:30 RabidGravy joined, ely-se joined 07:34 abraxxa joined, leont joined
leont How do I do conditional matching in rules? 07:34
I'm trying to use multi-rules, it parses fine but doesn't seem to work 07:35
07:36 ely-se left
ufobat i think the problem is that rules and tokens dont backtrack 07:37
leont ?
sortiz moritz, Done: RT#127813
ufobat so if one is matching (but not complete) the other will not work anymore, since you have "consumed" parts of your string
07:39 fireartist joined
leont I'm not sure I understand 07:47
I'm using multi-rules on arguments (my grammar is context-sensitive like that)
07:48 zakharyas joined
timotimo oh no, is the tpm2016.zoffix.com page down? :( 07:49
ZoffixWin: It's not just you! tpm2016.zoffix.com looks down from here. 07:50
RabidGravy it appears to be the DNS 07:51
ufobat leont, i am to much of a greenhorn to help you. sorry
07:54 perlawhirl left, polyfloyd joined 07:55 lustlife` joined 07:57 g4 joined 07:59 lustlife left 08:01 perlawhirl joined 08:05 dakkar joined 08:08 lustlife` left 08:11 TEttinger left
timotimo leont: have you tried Grammar::Tracer yet? 08:12
08:13 rindolf joined 08:15 darutoko joined 08:18 donaldh joined
RabidGravy I find it really useful as I am totally crap with regex 08:20
timotimo also, the debugger lets you run code - including regexes! - in its own REPL 08:27
08:27 lizmat joined
lizmat greets from the NLPW 08:27
RabidGravy is it possible to use "zef" and "panda" interchangeably? That is if I use one am I asking for trouble using the other on the same installation?
timotimo both use the API rakudo offers to do the actual installation 08:28
RabidGravy so "in theory" it should be cool
timotimo i think it ought to be fine. though perhaps they'll get confused about what is installed already and what isn't
dalek osystem: 319caa9 | (James Albert)++ | META.list:
added JSON-WebToken

Repo: github.com/jamesalbert/JSON-WebToken
osystem: 28cd2ce | sylvarant++ | META.list:
Merge pull request #186 from jamesalbert/master

added JSON-WebToken
RabidGravy tries
08:29 perlawhirl left
timotimo i think panda has a state file to remember what's installed and what isn't 08:29
leont I'd need to simplify my use-case first before trying a tracer or debugger 08:30
timotimo i've been dreaming about building a nice little web frontend for introspecting matching process and such 08:31
Xliff regexr.com/ 08:34
It's more a general thing though. It's not up to P6 and grammars, yet.
timotimo neat
Xliff Still. It's useful in general.
RabidGravy another week still no digest authentication module 08:35
Xliff installs Grammar::Tracer
github.com/jnthn/lolsql 08:40
Xliff spits all over his keyboard.
08:40 teatime joined
RabidGravy Oooh, I had inadvertently made a module with a ver of ".0.0.1" which works but can't be uninstalled 08:42
teatime heh
stmuk_ gcc.gnu.org/ml/gcc-patches/2016-04...00001.html
timotimo oh, the presentation is back 08:43
Xliff Wow! 08:44
github.com/jnthn/zavolaj
teatime flag_never_gonna_give_you_up
Xliff ^ - This has been more helpful to me in understanding NativeCall than the examples pointed to me, earlier. 08:45
teatime --frickroll-all-loops
Xliff: the implementation code of some NativeCall::______ modules provides examples of doing stuff where the docs for same are TBD. 08:46
Xliff teatime, thanks. I'll cross that bridge when I get to it. 08:47
donaldh ZoffixWin: Your "Wow, Perl6!" talk is trending on HN news.ycombinator.com/item?id=11401330
RabidGravy also more than half the Audio::* modules are native bindings which do amusing and exciting things 08:48
timotimo it seems like some people don't realize it's a presentation and that they're supposed to use spacebar or arrow keys (sideways!) to get through the slides
Xliff ¯_(ツ)_/¯
timotimo reveal.js has the nice feature that it has clickable indicator arrows at the bottom right by default
Xliff RabidGravy, OK cool. I will have to make a note to use those as references. 08:49
RabidGravy and while you're at it can you wrap some Ogg/Vorbis codec library? I get burnout quite quickly when I'm doing those sort of things ;-) 08:50
08:54 ufobat left
Xliff :P 08:54
08:54 ufobat joined
Xliff RabidGravy, the amount of boredom I'd have to achieve to do such a thing (at this point) would be legendary. 08:55
You could throw $$$ at me and then I'd do it.
RabidGravy hahahaha
Can we all play that game? I'd like £75,000 for the 47 modules I already wrote 08:56
timotimo i don't have that kind of money :|
Xliff Just out of curiosity, WHICH Ogg/Vorbis codec library are you thinking of.... and how perlish do you want it?
I mean.... I can play the strait 1:1 game if you want.
teatime Gentlemen, do I have one number right? 08:57
Xliff thinks of writing a .h to NativeCall processor.
RabidGravy there is one
Xliff Why does that not surprise me?
RabidGravy skarsnik's most excellent gptrixie
timotimo yeah, gptrixie
Xliff RabidGravy, why don't you use that when the BQ is high?
(for the slower ones here, BQ == Boredom Quotient) 08:58
Is gptrixie in the ecosystem, or do I have to install it via package manager?
timotimo um ... that's not an "or" that goes there :D 08:59
RabidGravy well, it just takes the donkey work out of making the stubs, for most things it doesn't make what I'd call a good interface
timotimo only if it's in the ecosystem, you can install it via package manager. otherwise you'll have to git clone it manually
Xliff ==> Successfully installed App::GPTrixie
timotimo to be fair, you can then use the package manager on that local clone
Xliff Hey. It's a starting point I don't have to monkey with myself! 09:00
RabidGravy, See! That's what I meant by "perlish".
What kind of interface would you think is "good"?
Would depend partly on each lib's API. 09:01
RabidGravy I'd have to charge you consultancy fees to tell you that
Xliff See, for XQuilla (whenever I can get off my ass to start it), I would probably only have "prepare" and "evaluate" subs to start with. 09:02
Sigh.... why do people have to be difficult with their namings?!? 09:03
It's Xqilla.
WTF did "U" do to them? 09:04
Q's are naked without it!!!
timotimo qeh.
RabidGravy anyway, using panda and zef interchangeably *has* shagged my repository, but I think this is because of the foremention dodgy version
09:04 leont left
Xliff RabidGravy, WHAT? You mercenary! You'd charge me money to ask you for suggestions of something you asked me to do?!? 09:04
RabidGravy Xliff, tell that to the Basques
09:04 Vlavv_ left
Xliff boggles! 09:04
o_0 09:05
Xliff flogs RabidGravy with a foam "U"
Aaannnd.... with that, it's probably time for zleep!
RabidGravy nukes the rakudo and rebuilds 09:06
timotimo good night xliff
Xliff RabidGravy, Once I get my XQuery parser project started I may switch gears for the halibut. 09:07
RabidGravy, you'd need to at least suggest a ogg/vorbis library to start with, though.
09:08 Xliff is now known as Xliff_ZZZzzzzz
RabidGravy It appears that I really am doomed to write all the things 09:09
teatime RabidGravy: I'm bored; pick a library and I'll consider it. 09:10
Xliff_ZZZzzzzz teatime, that's what I'm telling him! 09:12
RabidGravy I'm calling this "the first hurdle" ;-p 09:14
Xliff_ZZZzzzzz hurdles RabidGravy 09:15
09:18 Vlavv_ joined
gregf_ ]foo[ 09:21
]bar]
teatime [baz[
timotimo floof 09:22
09:30 sjoshi left 09:40 sue_ joined 09:41 espadrine joined, wamba joined 09:44 sjoshi joined 09:53 sue_ left
RabidGravy tadzik, any chance you could merge github.com/tadzik/JSON-Unmarshal/pull/18 I'd forgotten that something I am doing depends on it 09:57
Also as JSON::Unmarshal has flipped to JSON::Fast I may as well go with it for JSON::Marshall and JSON::Infer now 10:01
tadzik merged :)
RabidGravy you beauty :-*
lizmat will be doing a 20 minute ad-hoc Q&A session at the NLPW this afternoon at 16:00
timotimo i'm glad JSON::Fast gets some love :) 10:02
RabidGravy lizmat, I'd love to say I'll come along but I don't think I'll have time after doing the shopping ;-)
timotimo hm, is its test suite up to snuff? i wonder ...
lizmat RabidGravy: I'm just saying this in the hope that people will come up with some Questions I can Answer there, in case nobody has any Questions at the session itself :-) 10:03
RabidGravy timotimo, don't worry I'm sure some of the things I'm doing with JSON will find any bugs if indeed there are any ;-) 10:06
timotimo i wonder how much faster we could get by not doing stuff recursively 10:12
on the other hand we'll get faster invocation soon anyway
10:15 tphilipp joined
RabidGravy It's just use the power of Greyskull 10:21
timotimo the force doesn't work like that! 10:22
10:23 salva joined 10:24 tphilipp left 10:25 tphilipp joined 10:27 tphilipp left, tphilipp joined
RabidGravy well, it should and needs fixing ;-) 10:28
10:30 Guest38483 joined
Guest38483 Trying to clickbait lizmat++ with irclog.perlgeek.de/perl6/2016-03-31#i_12267194 10:31
10:31 perlawhirl joined 10:33 wamba left 10:35 tphilipp left
perlawhirl Guest38483: certainly looks bug-ish 10:38
:k works, :kv doesn't
lizmat is at NLPW and probably busy. i'm gonna go check out the 'first' method implementation source and see if :kv in there 10:39
RabidGravy arguably a doc bug 10:40
10:40 tphilipp joined, Guest38483 left
RabidGravy anyway, the plans for world domination have to go on hold while we go shopping 10:41
10:41 robertle joined 10:43 dogbert1 joined
dogbert1 masak and moritz thought that :kv should be added according to the discussion from yesterday 10:43
perlawhirl yeah i don't see a reason why it shouldn't be implemented 10:44
dogbert1 was Guest38483 before my computer decided to reboot on me :-(
perlawhirl fwiw, first calls first-result, which is implemented here: github.com/rakudo/rakudo/blob/6d12...ds.pm#L498
no conditional for %a<kv> 10:45
teatime anything come of the .where() bikeshed yet?
10:45 labster left
perlawhirl lolol... no... TimToady answered an unrelated question of mine this morning... 10:45
i was about to ask him to take a gander, but then $day-job got in the way 10:46
dunno whether i should just notify him via yoleaux??
10:47 wamba joined
jnthn perlawhirl: I'd agree :kv should work on .first as well as .grep, suspect just forgotten 10:49
-1 from me to .where as well as .grep; we don't need two ways to do exactly the same thing. :) 10:51
Learning a language's vocabularly is part of learning a language.
robertle gentlemen, I am trying to figure out to what degree I can use == and != on numbers in perl6. is it fair to say that you of course still need to be careful with comparisons of reals, you need to find the difference and compare to some threshold, but that many of those will actually be rationals in practice, so you get away with == and != more often?
perlawhirl Guest38483: for the meantime, this will give the same result as what i :kv should (probably) do
m: say (1, 22/7, 42, 300).first( * > 5, :end, :p).map({$_.key, $_.value}).flat 10:52
camelia rakudo-moar 103781: OUTPUT«(3 300)␤»
10:52 tphilipp left, lizmat left 10:53 tphilipp joined
jnthn robertle: You still need to be careful about comparison of Num (which is floating point), but yes, you encounter those less often because things like 2.5 are rationals 10:53
robertle: There's a =~= operator (approximately numerically equal), iirc you can also control its tolerance 10:54
m: say 0.1 + 0.2 == 0.3 # Rat, no problem 10:55
camelia rakudo-moar 103781: OUTPUT«True␤»
jnthn m: say 0.1e0 + 0.2e0 == 0.3e0 # Num, loss of precision by floating point makes for a surprise
camelia rakudo-moar 103781: OUTPUT«False␤»
jnthn m: say 0.1e0 + 0.2e0 =~= 0.3e0 # But can compare with a tolerance
camelia rakudo-moar 103781: OUTPUT«True␤»
perlawhirl Guest38483: shorter... 10:57
m: say (1, 22/7, 42, 300).first( * > 5, :end, :p).flatmap({.key, .value})
camelia rakudo-moar 103781: OUTPUT«(3 300)␤»
robertle jnthn: wow, thanks! I didn't know about the approximate operator, and this is exactly what I need 11:00
jnthn: we are doing a perl6 study group at work...
jnthn Sounds fun :) 11:01
11:02 wamba left 11:04 kaare_ joined 11:05 domidumont joined
chris2 what does the >+ operator do? i only find +> in the docs 11:08
perlawhirl docs.perl6.org/language/setbagmix#i...8%3E%2B%29
Baggy superset
chris2 thanks 11:09
i guess a "poogle" is needed in the future :)
perlawhirl the search box on the docs page is fine for searching for exact operators... but doesn't search more vague terms 11:11
11:11 tphilipp left 11:14 tphilipp joined
chris2 ah good idea :) 11:16
11:19 tphilipp left 11:20 wamba joined 11:21 tphilipp joined
hoelzro morning #perl6! 11:26
I was reading this blog post about OOP and what sounds like MMD, and I thought I'd share it with the channel: devblog.avdi.org/2016/03/31/a-late-...-dispatch/ 11:27
11:29 tphilipp left 11:30 tphilipp joined
tphilipp 5 11:30
11:31 donaldh left
moritz hoelzro: I already disagree with item 1, because it ignores the issue of namespaces 11:31
perlawhirl yeah... also... i've come to reason about things such that Classes != OOP 11:32
just cause i make a class, doesn't mean i'm writing purely OOP code
i'm not "smart" enough to write proper OOP code... but a Class is a nice way to package up related methods on a data structure 11:33
11:34 tphilipp left, tphilipp joined 11:35 tphilipp left, tphilipp joined 11:37 tphilipp left, tphilipp joined 11:38 tphilipp left 11:39 tphilipp joined 11:40 tphilipp left 11:41 tphilipp joined
dogbert1 perlawhirl++, thx for the workaround 11:41
11:42 kid51 joined
perlawhirl erm... were you the Guest? 11:42
11:42 fireartist left
timotimo yes 11:43
perlawhirl cool.. well, np
dogbert1 guess a should get a nick that sticks :-) 11:44
11:44 tphilipp left 11:45 tphilipp joined
perlawhirl isn't that a song by Huey Lewis and the News - I Need a Nick that Sticks 11:45
11:50 lizmat joined 11:51 kid51 left
lizmat perlawhirl: looks to me first( :kv) is totally implementable 11:51
perlawhirl yeap
lizmat probably should live in !first-result 11:52
11:52 leont joined
perlawhirl I feel like you're motioning me to submit a PR :D 11:53
will have a look at it later tonight (aussie time)
lizmat ok, cool...
thanks in advance!
:-)
11:54 felher left 12:03 webmind joined
webmind ohai 12:03
when executing the perl6 binary, it suggests to run panda install Readline
but that gives the following:
share.riseup.net/#HuQ1JHSPffAb6OzEO3Ha1Q
anyone know why?
timotimo can you paste the contents of that file? 12:04
/opt/rakudo-star-2016.01/share/perl6/site/panda/projects.json
actually, is this a rakudo star? i thought that comes with Linenoise and/or Readline by now
hoelzro star comes with Linenoise...but I don't know about 2016.01 12:05
timotimo OK
anything in particular keeping us from a rakudo star for .03, btw? i expect it's just no volunteer to do it
webmind timotimo: it's empty 12:07
timotimo is that the file that gets filled by "panda update"? 12:08
webmind I just followed a manual on installing perl6 :)
timotimo maybe that helps
hoelzro moritz: I didn't think of that; I was interested because of Perl6's 'multi sub' and 'multi method' distinction
timotimo i wonder why it's an empty file, though. it should be either with the correct contents or completely missing
webmind running panda update solves the problem :)
moritz timotimo: I'm not aware of a blocker 12:12
timotimo i feel like just giving it a shot for the hell of it. 12:13
m: my @a; Buf.new(@a) 12:19
camelia ( no output )
timotimo m: my @a; Buf.new(@a).say
camelia rakudo-moar ae43ba: OUTPUT«Buf:0x<>␤»
timotimo star: my @a; Buf.new(@a).say
camelia star-m 2016.01: OUTPUT«Buf:0x<>␤»
timotimo i wonder what was b0rked there in between now and the beginning of march
moritz hoelzro: well, the main difference between multi sub and multi method dispatch (in Perl 6) is where the code objects are being searched 12:20
hoelzro: in the lexical scope(s) vs. in method tables
hoelzro ah, good point 12:21
timotimo hum. do i need to use an older-ish version of panda to be compatible with our 2016.03 release? 12:22
don't think so.
no major changes since 2016.02
star will need to get File::Which 12:23
oh, #perl6-release is prime for this 12:24
12:27 Sgeo__ left
ZoffixWin Hacker news, eh 12:35
I wish my blog post and not just my slides were the link :/ 12:36
12:38 cdg joined
timotimo aye 12:41
12:44 ufobat left 12:45 ufobat joined
ZoffixWin I'll just add another slide in the front 12:46
12:46 sue joined
[Coke] hoelzro: how hard is RT #70927 with your new repl? 12:46
lizmat timotimo: what is exactly the problem with Buf.new??? 12:49
timotimo i have no idea, but there's a module that has a workaround in its source for that
probably not important
lizmat timotimo: ah, that, because it's all 6.c 12:50
and some 6.c support some Buf.new features, and some don't :-( 12:51
[Coke] (empty projects.json) I used to see that all the time when trying to panda from behind a firewall. 12:52
timotimo oh, ugh
12:52 tphilipp left, tphilipp joined
ZoffixWin Well... seems like a good chunk of those comments on HN are positive-ish. I really only spotted people crying about custom emoticon operators or that the UI on slides is unintuitive... 12:53
12:53 zakharyas left
ZoffixWin We're making progress! :D 12:53
jnthn 6.c unambiguously does not, because there weren't roast tests for those Buf.new features :P
12:54 ufobat left
perlawhirl Zoffix: yeah, people already have some unfounded fear of sigil'd variables... i think throwing more unicode frightens them 12:55
when i'm "selling" sigils to non-perlers... i always show them string interp first: say "hi $name" 12:57
it's one of those little things that make sigils worthwhile
12:58 sue left
perlawhirl once they are impressed with that, then you can hit them on the head with more unicode :D 12:58
oh! ÷ and × 12:59
those too
12:59 leont left
avalenn m: my @inf = (Inf,Inf...Inf) ; my @o = (0,0...Inf) ; my @max, @min, @sum ; for lines>>.split(" ")>>.chars -> @cur { @min = (|@min, |@inf) Zmin @cur ; @max = (|@max, |@o) Zmax @cur ; } ; say @min ; say @max 12:59
camelia rakudo-moar ae43ba: OUTPUT«5===SORRY!5=== Error while compiling /tmp/1AZ0ZQxo8d␤Variable '@min' is not declared. Did you mean '&min'?␤at /tmp/1AZ0ZQxo8d:1␤------> 3..Inf) ; my @o = (0,0...Inf) ; my @max, 7⏏5@min, @sum ; for lines>>.split(" ")>>.ch␤»
avalenn m: my @inf = (Inf,Inf...Inf) ; my @o = (0,0...Inf) ; my @max ; my @min ; for lines>>.split(" ")>>.chars -> @cur { @min = (|@min, |@inf) Zmin @cur ; @max = (|@max, |@o) Zmax @cur ; } ; say @min ; say @max 13:00
camelia rakudo-moar ae43ba: OUTPUT«[0 2 1 2 1 2 2 3 1 4 5 2 4 0]␤[6 4 8 7 5 7 4 7 3 11 6 6 8 6]␤»
gtodd custom emoticon operators ++
13:00 tphilipp left
teatime so, are they called texas variants, because everything is bigger in texas? 13:00
perlawhirl yes 13:01
timotimo yes
13:01 mr-foobar left, mr-foobar joined
dalek osystem: 95d6ea8 | kuerbis++ | META.list:
Add Term::Form to ecosystem

See github.com/kuerbis/Term-Form-p6
13:02
osystem: cea93ce | jnthn++ | META.list:
Merge pull request #187 from kuerbis/master

Add Term::Form to ecosystem
ZoffixWin perlawhirl, but that isn't impressive. I believe it's Ruby that has sigiless variables, but it lets you interpolate them by adding a sigil. Interpolation isn't an excuse to have sigils all the time.
13:03 dvinciguerra__ joined
ZoffixWin To me, they're an indication of a variable, as opposed to a sub or a keyword. Good enough for me to like 'em. 13:03
13:03 felher joined
gtodd sigils make it possible for humans to read code more quickly :) 13:04
perlawhirl yeah that's reason No1
gtodd tiwigils of course are part of the secret plan
perlpilot I think people's fear of sigils is rooted in community memories of BASIC :) 13:05
13:05 skids joined
gtodd or people who never used unix shells 13:05
powershell has sigils so ... 13:07
:-P
param( [string] $infile = $(throw "Please specify a filename.") )
moritz I'm not quite sure how much I like sigils
gtodd so clean .... so ... "power"ful
moritz but I do know that it infuriates me that in python "id" is a built-in symbol (the identity function) 13:08
gtodd hah
moritz and I work with lots of objects that have unique ids
13:08 zakharyas joined, sue joined
perlawhirl gtodd: psh is fine if you don't have perl and you need to script some stuff... but i really can't stand the dynamic scoping 13:08
13:09 ufobat joined
moritz so either I get name it 'id' and get lots of linter warnings, and have a bad conscience for doing something Not Right, and override the builtin, or I have to find a different name 13:09
timotimo "uid" or "oid" :\
doesn't feel so good
"idx" :P
perlawhirl ident?
gtodd maybe there could be a bunch of macros or a slang/grammar/dialect of perl6 that is sigil-less and uses tons of parens instead!
timotimo or do it like other places in python do it: jd or di or id_
moritz gtodd: Inline::Scheme? :-) 13:10
gtodd :-D
jnthn Sigils also make the data stand out, so aiding readability. :)
moritz timotimo: I hate all of those
gtodd $igil + syntax highlighting = super powers
timotimo :)
13:10 edjsu left
perlawhirl speaking of syntax highlighting... in vim, at least, sometimes ». causes all trailing methods to be blue 13:11
timotimo :o
perlawhirl anyone other vim users get that?
moritz it's like, I don't like sigils, but I tend to not like the absence of sigils as well :-)
perlawhirl: haven't noticed it
hoelzro [Coke]: could you paste a link? I can't seem to load that ticket
gtodd besides isn't there a rule about TimToady being right even if wrong? therefore if sigils are a bad idea they are still good so .... 13:12
perlawhirl i have no problem dropping sigils when writing python, but i can't stop typing 'my' :)
gtodd heh
perlawhirl i do it in powershell too 13:13
gtodd anyway a good editor could make sigils disappear unless you hover a mouse for 10 seconds or whatever 13:14
best of both worlds
[Coke] rt.perl.org/Ticket/Display.html?id=70297 , maybe I typod it 13:15
teatime perlawhirl: I do not...
perlawhirl gtodd: lol, keymap to toggle sigil visability. not for me, i'm in the "i like 'em" camp
teatime (get that behavior from vim)
perlawhirl teatime: i'll update vim perl and check again 13:16
gtodd perlawhirl: in vim you could use texas variant maybe ? >>
teatime perlawhirl: presumably it's seeing it as a quote/begin string?
or comment...
perlawhirl well, another workaround is add the trailing brackets, so lines()».trim instead of lines».trim
teatime perlawhirl: I'm pretty sure I'm using the crappy bundled perl6 syntax files
perlawhirl hmm, i thought vim-perl came with modified syntax files 13:17
gtodd perlawhirl: sometimes when I try to open powershell scripts they open in a postscript viewer
13:18 lizmat left
perlawhirl righto... will remove vim-perl then and see what happens 13:18
arnsholt A question for the native-speakers of English: When speaking of Biblical texts, is "the begats" a term that's actually used (or at least understandable)? 13:19
13:19 ocbtec joined
teatime um.. tthere might be people at HN who think smoe of the syntax is a Apr 1 joke 13:21
like the whatever operator
perlawhirl teatime: well removing vim-perl fixed it
some colors are slightly different too
13:21 lizmat joined, cdg left
teatime arnsholt: understandable, yes I know exactly what is being referred to by that 13:22
13:22 cdg joined
teatime arnsholt: some people might not be familiar enough w/ the bible to recognize it though 13:22
perlawhirl .d begats
yoleaux Sorry, I couldn't find a definition for 'begats'.
perlpilot teatime: really? What is "the begats"? 13:23
perlawhirl .ety begats
yoleaux Sorry, I couldn't find the etymology of that.
teatime perlawhirl: you could probably fix the syntax file for vim-perl
perlawhirl lousy yoleaux!!
teatime perlawhirl: the default one sucks too, just in other ways
perlpilot unless you mean "begets" and even then, it's not very nouny
perlawhirl .d begat
yoleaux perlawhirl: Sorry, that command (.d) crashed.
perlawhirl hah!
.ety begat
yoleaux begat (v.): "archaic past tense of beget." — etymonline.com/index.php?term=begat
perlawhirl .d begat
yoleaux perlawhirl: Sorry, that command (.d) crashed.
teatime perlawhirl: early in the old testament, there's a long, long string of X begat Y begat Z begat A begat B...
begat means parented 13:24
perlpilot so "the begats" is referring specifically to that long string of A begat B begat C ?
teatime I assume so.
perlawhirl yes i'm aware of it's use in the bible
teatime www.readbookonline.net/read/933/10945/
also www.biblegateway.com/passage/?sear...ersion=WYC 13:25
perlawhirl i think it's just to assert the lineage to... Abraham?
arnsholt Basically
There's another one in one of the Gospels, tracing the lineage of Jesus
teatime biblehub.com/commentaries/pnt/matthew/1.htm
perlawhirl if someone joined now they might think they were in the wrong channel =D 13:26
teatime lol
sjoshi lol
moritz man, they should've just drawn a freakin' graph
13:27 cdg left
perlpilot Have you accepted Perl 6 as your personal savior? 13:27
teatime And lo, Abraham begat the following DAG:
tony-o m: { sleep $_; .say }.(.2|.3|.1);
camelia rakudo-moar ae43ba: OUTPUT«any(0.2, 0.3, 0.1)␤»
tony-o does the output of that get serialized?
timotimo it's not actually spreading out to multiple threads 13:28
lizmat no, at the moment all elements of the junction are executed in turn
perlpilot Since they only mention the "important" male-children, the DAG would be very linear.
13:28 sortiz left
tony-o ahh, this explains. thanks lizmat timotimo 13:29
arnsholt teatime: Right, so I'm not completely insane when using that term. Still, since both my advisors commented "what?" on it, I should probably add an explanatory footnote =) 13:30
13:30 leont joined
teatime arnsholt: is it not a common term among people who study the bible? 13:30
somehow, it seems like it would be.
timotimo oh hey arnsholt!
perlpilot arnsholt: What's the sentence where you used it? 13:31
arnsholt perlpilot: "the latter text is also written in a highly repetitive style 13:32
with key phrases and constructions being very frequent, such as the begats and short sentences
introducing quotations."
teatime hrm.
timotimo it's basically written in a very antique variant of JSON
arnsholt The latter text being the Latin Vulgata text (contrasting with Caesar's Bellum Gallicum)
perlpilot I think had you mentioned the bible, it would have been clear
timotimo they didn't have ASCII yet so it's all words instead of symbols 13:33
[Coke] I wouldn't say that's an english question so much as a bible question.
teatime arnsholt: doesn't exactly roll off the tongue, does it.
[Coke] m: say 3 !=3; say 3 != 3;
camelia rakudo-moar ae43ba: OUTPUT«Cannot modify an immutable Int␤ in block <unit> at /tmp/CqnfXDXKt6 line 1␤␤»
[Coke] m: say 3 != 3; say 3 !=3; 13:34
camelia rakudo-moar ae43ba: OUTPUT«False␤Cannot modify an immutable Int␤ in block <unit> at /tmp/V5KKtlflkY line 1␤␤»
teatime arnsholt: 'such as' also indicates that the begats and short sentences are part of the later text, if they're not something like "like" would be better.
[Coke] if that's OK (due to whitespace requirements), someone can close rt.perl.org/Ticket/Display.html?id=121108
timotimo "Ha! Nice april fools, the best part was those operators.
^- best comment
arnsholt Yeah, they're particular to the Bible text. Caesar isn't particularly interested in genealogy =)
[Coke] Today is the day we reveal the original AF joke is now complete, and turn off the lights and go home? 13:36
.seen gabor 13:38
yoleaux I haven't seen gabor around.
avalenn m: my @max ; my @min ; for lines>>.split(" ")>>.chars -> @cur { @min = (|@min, |(Inf xx (@cur - @min))) Zmin (|@cur, |(Inf xx (@min - @cur))) ; @max = (|@max, |(-Inf xx (@cur - @max)) Zmax (|@cur, |(-Inf xx (@max - @cur)))) ; say @cur } ; say @min.perl ; say @max.perl
camelia rakudo-moar ae43ba: OUTPUT«(4 4 2 8 6 7 4 2 5)␤(4 3 5 4 2 2 8 3 3 4 2 5 4 2 5)␤(5 1 7 4 5 2 4 4 6 2 5 2 7 3 3)␤(1 2 7 3 10 6 6 8 2 4 3 5 2 6)␤(0)␤(2 8 3 5 3 6 7 7 2 2 6 2 2 4)␤(2 5 3 1 2 2 4 2 2 3 3 7 1 4 4 5)␤(2 8 4 1 3 2 7 1 5 4 2 9 5)␤(6 2 4 2 5 4 2 3 2 5 2 2 10 6…»
moritz .seen szabgab
yoleaux I saw szabgab 19 Dec 2015 06:47Z in #perl6: <szabgab> thanks for the links
lizmat .see gabzab
[Coke] danke. 13:39
lizmat .seen gabszab
yoleaux I haven't seen gabszab around.
lizmat .seen gabzsab
yoleaux I haven't seen gabzsab around.
lizmat hmmm....
[Coke] .ask szabgab - Can you please double check rt.perl.org/Ticket/Display.html?id=123497 against a 2016 R*?
yoleaux [Coke]: I'll pass your message to szabgab.
moritz [Coke]: he's not much on IRC anymore; if you want an answer, best write him an email
avalenn m: my @max ; my @min ; for lines>>.split(" ")>>.chars -> @cur { @min = (|@min, |(Inf xx (@cur - @min))) Zmin (|@cur, |(Inf xx (@min - @cur))) ; @max = (|@max, |(-Inf xx (@cur - @max)) Zmax (|@cur, |(-Inf xx (@max - @cur)))) } ; say @min ; say @max ; 13:40
camelia rakudo-moar ae43ba: OUTPUT«[0 1 1 1 1 2 1 1 1 1 2 2 1 0 3 4 3]␤[6 11 8 8 10 10 8 8 10 11 8 9 10 6 6 8 3]␤»
moritz avalenn: wow, what's that?
[Coke] moritz: Not enough to compose an email, no. :) 13:41
avalenn try to do max and min accross several lines
[Coke] .ask nine is rt.perl.org/Ticket/Display.html?id=123276 done?
yoleaux [Coke]: I'll pass your message to nine.
perlpilot avalenn: don't we have a minmax routine/method/whatever? 13:42
avalenn I don't know, I could be interested in one
perlpilot I think we do
lizmat m: say ^10 .minmax
camelia rakudo-moar ae43ba: OUTPUT«(0 9)␤»
perlpilot m: say (7,9,3,8).minmax; # for some variation 13:43
camelia rakudo-moar ae43ba: OUTPUT«3..9␤»
skids Note Range.minmax and List.minmax are different.
avalenn m: lines>>.split(" ")>>.chars.minmax
skids jinx
camelia ( no output )
avalenn m: lines>>.split(" ")>>.chars>>.minmax 13:44
camelia ( no output )
skids need to say
avalenn m: lines>>.split(" ")>>.chars.minmax.say
camelia rakudo-moar ae43ba: OUTPUT«0..11␤»
avalenn m: lines>>.split(" ")>>.chars>>.minmax.say
camelia rakudo-moar ae43ba: OUTPUT«(2..8 2..8 1..7 1..10 0..0 2..8 1..7 1..9 2..10 0..0 1..7 1..8 1..10 1..10 2..11 0..0 1..7 2..8 1..6 0..11)␤»
perlawhirl m: chars 13:45
camelia rakudo-moar ae43ba: OUTPUT«5===SORRY!5=== Error while compiling /tmp/Nzp_dcX_Lw␤Calling chars() will never work with proto signature ($)␤at /tmp/Nzp_dcX_Lw:1␤------> 3<BOL>7⏏5chars␤»
lizmat m: chars("foo").say
camelia rakudo-moar ae43ba: OUTPUT«3␤»
perlawhirl m: lines».words».chars».minmax.say
camelia rakudo-moar ae43ba: OUTPUT«(2..8 2..8 1..7 1..10 Inf..-Inf 2..8 1..7 1..9 2..10 Inf..-Inf 1..7 1..8 1..10 1..10 2..11 Inf..-Inf 1..7 2..8 1..6 2..11)␤»
perlpilot heh 13:46
gregf_ m: lines
camelia ( no output )
perlpilot I like that Inf..-Inf
perlawhirl haha i dunno whats up with that
lizmat m: .say for lines # gregf_
camelia rakudo-moar ae43ba: OUTPUT«Céad slán ag sléibhte maorga Chontae Dhún na nGall␤Agus dhá chéad slán ag an Eireagal ard ina stua os cionn caor is coll;␤Nuair a ghluais mise thart le Loch Dhún Lúich’ go ciúin sa ghleann ina luí␤I mo dhiaidh bhí gleanntáin ghlas’ G…»
gregf_ m: lines.^name
camelia ( no output )
perlawhirl i was just trying to replace avelenn's .split with .words
skids m: (1,2,3,0)».minmax.say
camelia rakudo-moar ae43ba: OUTPUT«(1..1 2..2 3..3 0..0)␤»
skids weird, what's that .chars putting out. 13:47
avalenn m: say ines>>.split(" ")>>.chars # skids
camelia rakudo-moar ae43ba: OUTPUT«5===SORRY!5=== Error while compiling /tmp/GP_1gn85DJ␤Undeclared routine:␤ ines used at line 1. Did you mean 'lines'?␤␤»
perlawhirl m: lines».words».chars
camelia ( no output )
perlawhirl m: say lines».words».chars
camelia rakudo-moar ae43ba: OUTPUT«((4 4 2 8 6 7 4 2 5) (4 3 5 4 2 2 8 3 3 4 2 5 4 2 5) (5 1 7 4 5 2 4 4 6 2 5 2 7 3 3) (1 2 7 3 10 6 6 8 2 4 3 5 2 6) () (2 8 3 5 3 6 7 7 2 2 6 2 2 4) (2 5 3 1 2 2 4 2 2 3 3 7 1 4 4 5) (2 8 4 1 3 2 7 1 5 4 2 9 5) (6 2 4 2 5 4 2 3 2 5 2 2 10 6 6 8) () (6 1 2 …»
perlawhirl m: say lines».words».chars».minmax 13:48
camelia rakudo-moar ae43ba: OUTPUT«(2..8 2..8 1..7 1..10 Inf..-Inf 2..8 1..7 1..9 2..10 Inf..-Inf 1..7 1..8 1..10 1..10 2..11 Inf..-Inf 1..7 2..8 1..6 2..11)␤»
perlawhirl figure that out?
skids Yay lols. 13:49
13:49 domidumont left
skids I'm off to go glue various small metal items to a door with a sign warning of strong magnetic fields inside. 13:50
13:50 domidumont joined
lizmat
.oO( the power of Perl 6 ! )
13:50
perlawhirl .words shoud do the same as .split(' ') asuuming single spaced words 13:51
avalenn quite, the difference is on empty lines
13:51 _4d47 joined
avalenn m: say "".words 13:51
camelia rakudo-moar ae43ba: OUTPUT«()␤»
avalenn m: say "".split(" ")
camelia rakudo-moar ae43ba: OUTPUT«()␤»
avalenn say lines[4] 13:52
m: say lines[4]
camelia rakudo-moar ae43ba: OUTPUT«␤»
avalenn m: say lines[4].words
camelia rakudo-moar ae43ba: OUTPUT«()␤»
avalenn m: say lines[4].split(" ")
camelia rakudo-moar ae43ba: OUTPUT«()␤»
13:53 domidumont left, domidumont joined
[Coke] rt.perl.org/Ticket/Display.html?id=127127 - Perl 6 advent site needs an archive widget in the theme. 13:54
perlawhirl m: say lines[2].split(' ').elems
camelia rakudo-moar ae43ba: OUTPUT«15␤»
13:54 leont left
perlawhirl m: say lines[2].words.elems 13:54
camelia rakudo-moar ae43ba: OUTPUT«15␤»
timotimo how many of you are interested in testing a rakudo-star tarball?
moritz is
avalenn m: say lines[4].words.perl
camelia rakudo-moar ae43ba: OUTPUT«().Seq␤»
avalenn m: say lines[4].split(" ").perl
camelia rakudo-moar ae43ba: OUTPUT«("",)␤»
moritz m: say lines[4].perl 13:55
camelia rakudo-moar ae43ba: OUTPUT«""␤»
lizmat m: say lines[4]
camelia rakudo-moar ae43ba: OUTPUT«␤»
moritz m: say lines[3].perl
camelia rakudo-moar ae43ba: OUTPUT«"I mo dhiaidh bhí gleanntáin ghlas’ Ghaoth Dobhair, is beag nár bhris mo chroí."␤»
timotimo oh, look at that, that string is totally corrupted!
moritz lol
[Coke] timotimo: *thbbthp*
teatime is that Klingon? 13:57
13:57 sQuEE` is now known as sQuEE
timotimo welsh? 13:57
avalenn seems gaëlic or something similar from my untrained eye 13:58
timotimo garlic?
t.h8.lv/rakudo-star-2016.03.tar.gz
avalenn en.wikipedia.org/wiki/Gaelic
[Coke] m: DOC INIT { use barf }
camelia rakudo-moar ae43ba: OUTPUT«===SORRY!===␤Could not find barf in:␤ /home/camelia/.perl6␤ /home/camelia/rakudo-m-inst-2/share/perl6/site␤ /home/camelia/rakudo-m-inst-2/share/perl6/vendor␤ /home/camelia/rakudo-m-inst-2/share/perl6␤ CompUnit::Repository::Absolu…»
teatime taH pagh taHbe' 13:59
13:59 lizmat left
timotimo so it'd be nice if you'd all get that tarball, unpack, install, test :) 14:00
14:00 sQuEE is now known as sQuEE`
timotimo maybe then i'll go write an announcement and see what else there is to do for a release 14:01
AFK for a bit
14:02 sue left 14:04 lizmat joined, sue joined, g4 left
moritz timotimo++ # star RC 14:07
timotimo++ # server reachable by IPv6 :-) 14:08
avalenn is there any "perldoc -f" equivalent for perl6 ? 14:16
moritz .tell timotimo re star RC, I get this error during modules-install-m: perlpunks.de/paste/show/56fe8298.655e.1d5
yoleaux moritz: I'll pass your message to timotimo.
14:17 frederico joined
moritz .tell timotimo this is on a boring Debian Jessie (stable) amd64 14:17
yoleaux moritz: I'll pass your message to timotimo.
14:17 cognominal joined
avalenn teatime: www.gaelicmatters.com/irish-songs-l...#gleanntan 14:18
14:18 g4 joined 14:23 g4 left
tony-o oh cool, i'm in the perl6 FAQ :-) 14:24
perlpilot avalenn: there's a p6doc command you can install, but I think you'll have better luck with doc.perl6.org
14:30 abraxxa left, abraxxa joined
timotimo moritz: inconthievable! 14:30
yoleaux 14:16Z <moritz> timotimo: re star RC, I get this error during modules-install-m: perlpunks.de/paste/show/56fe8298.655e.1d5
14:17Z <moritz> timotimo: this is on a boring Debian Jessie (stable) amd64
tony-o m: sub a (--> Rat:D) { Rat; }; a.say; 14:31
camelia rakudo-moar ae43ba: OUTPUT«Type check failed for return value; expected Rat but got Rat␤ in block <unit> at /tmp/uWs7B6Wj2g line 1␤␤»
timotimo moritz: on my end, that line reads "mkpath 't/dir2';"
tony-o i like that error message
14:34 jferrero joined
perlawhirl besides (1,2).Seq, is there any other syntax do define a Seq 14:35
14:35 g4 joined
rjbs > $x = (1, 2 ... *) gets a seq FWIF. 14:36
FWIW, even.
$x = (1 ... 2) also 14:37
gregf_ m: say (1 ... 3).^name 14:38
camelia rakudo-moar ae43ba: OUTPUT«Seq␤»
14:39 dvinciguerra__ left 14:40 pmurias joined, sue left
jnthn perlawhirl: gather/take blocks also return a Seq 14:40
In fact, *most* things that produce values you can iterate return a Seq :)
14:42 lizmat left
RabidGravy If I create a new type on the fly using the MOP does it stick around for ever or just until the final place it is used go out of scope? 14:42
or to put it another way if I create a bunch of dummy classes in a long running program am I goiing to run into trouble eventually 14:44
perlawhirl jnthn: that's my point i guess. i'm working on a PR (and tests!) to add a :kv adverb to 'first'. it's not in S32 - is in the docs - but i think there's a reasonable user expectation that first would have the same adverbs as grep 14:45
jnthn perlawhirl: yes, me too
perlawhirl so... grep :kv returns a Seq, so with first, is it enough that i just return (index, value).Seq or is there more to returning a Seq of definite values
jnthn RabidGravy: Types can be GC'd 14:46
RabidGravy: So long as they're not referenced anywhere
Meta-objects really are just objects
pmurias does publishing nqp-js-on-js on npm seems like a good idea (to mostly see how what is needed for that it's not really ready for "serious" use) 14:48
yoleaux 31 Mar 2016 21:02Z <[Coke]> pmurias: I can do it when not at work, but I can also do it at work if I use certain URLs.
RabidGravy cool, I'm just doing an experiment with a web proxy thing for Lumberjack and to make it really simple I'm serialising the class name and its log-level and when it is receoved on the other end if the web part doesn't know about the class it creates an empty one and applies the LOgger role to it
the lifetime of which is only going to be as long as the dispatchers get to handle it 14:50
moritz timotimo: that's probably from modules/panda/ext/Shell__Command/t/02-shell-command.t 14:51
timotimo oh!
moritz timotimo: and not from modules/shell-command/t/02-shell-command.t
ok which('perl6').IO.x, 'which - perl6 is found'; 14:52
MadcapJake This is pretty great: github.com/crystal-lang/crystal/issues/2394
moritz timotimo: of note: I have no perl6 in my path
timotimo oh, haha
well, that's clever!
that must be "we'll just search for something we totally know must be in the path!"
*sigh* 14:54
i'm not sure how to deal with this in a sensible way
moritz neither 14:55
well, a sensible first step would be to open an issue
confront the maintainer. TADZIK!!!! 14:56
timotimo if we bring the submodule up to latest, we'll also have to add another ext, File::Which
that one may have better tests
14:56 cuthere joined
RabidGravy put the path of $*EXECUTABLE in %*ENV<PATH> ? 14:56
moritz RabidGravy: might be an option as well
14:56 sue joined
MadcapJake is surprised that Perl 7 wasn't announced today ;) 14:56
14:57 robertle left 15:00 andreoss joined
perlawhirl so let me ask... if i'm submitting a PR for previously untested functionality... where do my tests go? do i submit a PR to roast as well and reference my rakudo PR? 15:01
jnthn perlawhirl: Yes, or just ask for a commit bit to roast and push them afterwards
timotimo onto the master branch, aye 15:02
andreoss was p6 feed operator borrowed from somewhere? It doesn't seem to be sematically close to what haskell pipe does.
jnthn andreoss: ASCII art, perhaps :P 15:03
15:05 jolts joined, jolts left 15:06 jolts joined
geekosaur feed operator is more or less inspired by unix shell, no? 15:07
15:08 abraxxa left
perlpilot Maybe it says somewhere around S06:1361 15:10
tadzik what what :o
geekosaur hm., are we talking about the syntax or the semantics? 15:11
timotimo yo tadzik 15:18
Shell::Command - only the version bundled with panda - doesn't pass its tests when there's no "perl6" in the PATH
which is usually the case when building rakudo-star 15:19
could you bring it up to date with the newest Shell::Command, which'd also mean to pull in File::Which?
tadzik I think so, yes 15:22
timotimo cool
i'm in the middle of making a rakudo star release
andreoss perlpilot: it doesn't say where it was borrowed from, but i assume shell
tadzik I really have no idea how to work with submodules ;D 15:23
andreoss though pipe in shell can have any command on either of sides 15:24
geekosaur sure. shell is untyped :)
timotimo tadzik: well, that's why you use subtree, right?
geekosaur | won't know that some command produces no output or takes no input, for example,
andreoss it would feel much more like shells pipe, if ==> was just a another syntax for method call 15:25
assuming method call won't perform a mutation on object
tadzik oh, I do indeed 15:26
should be good now
moritz++ for update_subtree.pl
15:26 ptolemarch joined
andreoss m: (1..10) ==> reduce * + * ==> elems ==> say; 15:26
camelia rakudo-moar ae43ba: OUTPUT«5===SORRY!5=== Error while compiling /tmp/2j7sv5dmM4␤Preceding context expects a term, but found infix ==> instead␤at /tmp/2j7sv5dmM4:1␤------> 3(1..10) ==> reduce * + * ==> elems ==>7⏏5 say;␤»
15:26 g4 left
andreoss m: (1..10).reduce( * + * ).say; 15:27
camelia rakudo-moar ae43ba: OUTPUT«55␤»
andreoss sadly there's no alernative syntax for ., method chaining could be more readble 15:28
psch m: sub ssay(|c) { say c }; (1..10) ==> reduce * + * ==> ssay; 15:29
camelia rakudo-moar ae43ba: OUTPUT«\(55)␤»
psch m: say; # this error break feeds there, apparently
camelia rakudo-moar ae43ba: OUTPUT«5===SORRY!5=== Error while compiling /tmp/AF7ajdzoTC␤Unsupported use of bare "say"; in Perl 6 please use .say if you meant $_, or use an explicit invocant or argument, or use &say to refer to the function as a noun␤at /tmp/AF7ajdzoTC:1␤------> 3…»
andreoss psch: because ==> does not call it as a method
psch andreoss: no, because the grammar parses every base "say" as wrong 15:30
*bare
as in, when not in a dottyop or without arguments
andreoss m: (1..10) ==> reduce * + * ==> print;
camelia rakudo-moar ae43ba: OUTPUT«5===SORRY!5=== Error while compiling /tmp/V9NeLs8IBc␤Unsupported use of bare "print"; in Perl 6 please use .print if you meant $_, or use an explicit invocant or argument, or use &print to refer to the function as a noun␤at /tmp/V9NeLs8IBc:1␤----…»
psch same error for print, yes :) 15:31
jnthn m: (1..10) ==> reduce * + * ==> say(); 15:32
camelia rakudo-moar ae43ba: OUTPUT«55␤»
psch that i didn't try because i wasn't sure of the feed semantics... :)
jnthn++
andreoss m: role Foo { method say-it { self.say } } ; ( ( 1..10 ==> reduce( * + * ) ) does Foo) ==> say-it() 15:34
camelia rakudo-moar ae43ba: OUTPUT«5===SORRY!5=== Error while compiling /tmp/yvJNRMs4dG␤Undeclared routine:␤ say-it used at line 1␤␤»
15:34 domidumont left
perlawhirl ok... my first rakudo PR submitted... which also happens to be for 'first' 15:35
it's 2:30 in the am here... i think bedtime 15:36
jnthn andreoss: It does sub calls, not method calls 15:37
andreoss jnthn: why so? many subs indented for ==> have a coresponding method, e.g map, grep 15:39
I mean ==> feels like chainging method calls with this subs, why doesn't it call methods in a first place? 15:40
jnthn andreoss: I don't know, I didn't design it. :-)
andreoss: But if you want to chain method calls, then that's already perfectly easy 15:41
It'd be a bit odd to introduce a more verbose syntax just for that. :)
15:41 dvinciguerra joined
perlawhirl method calls already chain left the right... whereas chained subs kinda read right to left 15:42
jnthn Right
:)
15:43 zakharyas left
psch way i see it, ==> is an alternative to .&sub 15:44
plus choice of direction, seeing as <== is around too and means the same afair
15:45 sue left
psch it's what this one HN commenter called "more than one way to make it look" 15:45
15:45 domidumont joined
jnthn m: multi sub cross () {} 15:45
camelia rakudo-moar ae43ba: OUTPUT«(signal SEGV)»
15:46 sjoshi left
[Coke] BOOM 15:46
is there a ticket for that?
jnthn [Coke]: yes, that's why I was trying it out
[Coke] (please add [SEGV] to titles for segfaulting tickets)
jnthn While my local Rakudo rebuilt after fixing something else :)
15:46 sue joined
_4d47 psch: what is .&sub ? i've seen it recently, still confused by it 15:47
ugexe foo(1, 2, 3) is the same as 1.&foo(2,3) 15:48
15:48 perlawhirl left
jnthn alrady has 2 SEGV fixes locally, and is thinking he might hunt this one down for a third :) 15:48
RabidGravy jnthn++ 15:49
15:51 khw joined
[Coke] jnthn: nice. Thanks! 15:52
_4d47 ok ugexe like just a shorcut, cool thx 15:53
jnthn m: say &cross.derive_dispatcher
camelia rakudo-moar ae43ba: OUTPUT«(signal SEGV)»
[Coke] Someone can go through and retry the other SEGVs after to make sure they weren't dupes. :)
jnthn Golfs to that heisensegv
If anyone wants to check something: please try rt.perl.org/Ticket/Display.html?id=126956
_4d47 makes me think of github.com/tj/luna#method-syntactic-sugar
jnthn It doesn't leak at all for me
15:56 telex left
[Coke] fires up a fresh build of rakudo to test 15:57
15:58 telex joined, dakkar left
ugexe _4d47: its handy for using with hypers `(1..10)>>.&sin` or whatevers `(1..10).map(*.&sin)` 15:59
[Coke] jnthn: it's leaking here. 16:01
100% CPU, already up to 22% of my 16G box.
OS X 16:02
Woodi same here. how it do this ?? :)
16:03 cur8or joined
[Coke] jnthn: ... it does eventually level off at about 18% of 16G 16:04
... 10.5 % 16:07
er, sorry, 13%. but it does seem to be freeing some up now, anyway. 16:08
jnthn [Coke]: Odd, here on Windows its flat at 106MB
Woodi: OSX or Linux? 16:09
timotimo it grows comfortably to 5gb on my box
Woodi Linux, 32bit
jnthn OK, good
timotimo what happens when we sink a Supply like that? 16:10
jnthn (Good it's reproducable on Linux)
timotimo: Well, it fires off the events, which go nowhere :)
andreoss jnthn: you meant something else that .?
timotimo that could be thrown owt ;)
andreoss *than 16:11
jnthn andreoss: No, I meant . is already a perfectly fine way to chain method calls 16:12
Especially since you can spread them over lines now
timotimo yeah 16:13
Woodi so, working like requested ? Supply accumulate forever ? 16:15
jnthn Woodi: No, we don't store the messages anywehre 16:16
The fact it doesn't leak on Windows makes it highly likely it's something platform specific going on and probably fairly low level 16:17
timotimo want me to pull a heap snapshot and take a look? 16:20
jnthn The third SEGV turns out to be a bit more interesting to hunt down than the previous two, and it's time to think about cooking dinner...I'll bbiab
timotimo: You can but I suspect it'll be C-level leaking
timotimo: Because once we're in managed memory space, why'd it behave any differently across platforms? :)
So I'd valgrind it but with a sleep 5; or something 16:21
timotimo well, ... fair enough
jnthn As a starting point
timotimo++
I'll probably be back to write tests for the two things I did fix, once the cooking gets to the ignore-it-a-while step :) 16:22
&
16:22 sue_ joined, sue left 16:23 sue_ is now known as sue
Woodi btw. new OpenBSD release as one of highlights have "rewrite of less". so, it can be considered exploitable :) 16:23
RabidGravy so all the while Supply was a role I never cared about its roley-ness, now it isn't anymore I want it to be a role again :-\
oh well 16:24
Woodi ...but i realy didn't consult code. just probability :)
timotimo jnthn: seems like we're just not reaching a safe point, so our list for what to free at a safe point grows into the million-digits 16:28
==10310== 38,592,512 bytes in 18,844 blocks are still reachable in loss record 1,937 of 1,938
==10310== by 0x4FB6488: MVM_fixed_size_free_at_safepoint (fixedsizealloc.c:249)
==10310== 57,885,696 bytes in 18,843 blocks are still reachable in loss record 1,938 of 1,938
==10310== by 0x4FA26E8: allocate_frame (frame.c:270)
lots of stuff omitted, obviously, but i think that's why we do them leaks
16:29 Relsak joined
timotimo gist.github.com/timo/fffc13ab92d42...fe47dc9a14 16:30
16:32 Wlattin joined, cdg joined
Wlattin Suggestions for detailed help on MAIN??? 16:33
timotimo you can add Pod to your candidates as well as the parameters
moritz man, git submodule has "init", "update" and "sync". I can never rememer which one I need
timotimo and you can generate your own usage
update --init :)
moritz timotimo: was it intential that you haven't done that yet for star? 16:34
timotimo i have all the changes on my machine
moritz timotimo: erm, I meant the git submodule foearch pull thing
timotimo just not pushed
moritz oh, just not pushed 16:35
timotimo: push, push, push
timotimo can do
i just wanted to look first if things will be fine
moritz well, if you push now, others can chime in :-)
in particular, I wanted to see if tadzik++'s panda commit fixed star for me 16:36
16:36 sue left
dalek ar: 93f7022 | timotimo++ | / (10 files):
bring modules up to date, add File::Which
16:36
ar: b8ba246 | timotimo++ | tools/star/Makefile:
bump versions to 2016.03
ar: 26acc3f | timotimo++ | / (2 files):
remove ufo, as rakudo now manages precompilation itself.
ar: f6dc6cf | timotimo++ | / (3 files):
Change in the readme and more modules.
16:37 geekosaur left, sue joined, geekosaur joined 16:42 Wlattin left
Woodi adx 16:43
stmuk_ I did wonder if balidor should be removed from star since there are competing frameworks now 16:47
gtodd m: my %hash; for ( "abcdefabcgbacbdebdbbcaebfebfebfeb".comb.rotor(5 => -4) ) -> $a { %hash{$a}++ } ; .say for %hash.sort(-*.value ).[^5]
camelia rakudo-moar ae43ba: OUTPUT«e b f e b => 3␤b f e b f => 2␤f e b f e => 2␤e b d b b => 1␤f a b c g => 1␤»
stmuk_ Bailador even 16:48
tadzik as a Bailador author and maintainer, +1 16:49
it doesn't feel "core enough" 16:50
16:50 pmurias left
tadzik I agree that it may have used to be some of the "here's something useful you can do with perl 6 even today!", but it doesn't seem particularly fitting these days 16:50
ufobat what other frameworks are actually there?
[Coke] Prancer is under development, IIRC. 16:51
kmwallio tadzik: I was actually really confused when I went to install Bailador and got an error, it took me a while to read the error and see it was already installed
tadzik hahah
that's panda being LTA though
16:51 pmurias joined
kmwallio LTA? 16:51
ufobat and Web::App::Ballet 16:52
tadzik S99:LTA 16:53
...does this work?
kmwallio: design.perl6.org/S99.html#LTA
kmwallio ... 16:54
tadzik :) 16:55
RabidGravy yeah, it probably wants to catch the exception and emit something a bit more user friendly 17:01
gregf_ m: print (1...5).reduce: -> $sum,$v { $v % 2 == 0 ?? $sum + $v !! $sum }
camelia rakudo-moar ae43ba: OUTPUT«7»
gregf_ ^^ i was expecting the result to be 6 :| 17:02
timotimo buh, is synopsebot offline again?
RabidGravy Boo! I go to actually make something using WebSocket and it's failing its tests again :( 17:03
assuming nothing changed in the module since this morning I can only assume it's a change in rakudo
17:04 pmurias left
skids m: print (0...5).reduce: -> $sum,$v { $v % 2 == 0 ?? $sum + $v !! $sum } 17:04
camelia rakudo-moar ae43ba: OUTPUT«6»
skids reduce has to start somewhere.
gregf_ aha 17:05
skids: cheers 17:06
17:07 Relsak left
dalek ar: 63da543 | (Steve Mynott)++ | modules/MODULES.txt:
remove Bailador
17:13
17:14 cur8or left, prammer joined 17:15 Xliff_ZZZzzzzz is now known as Xliff
Xliff Hello #perl6. 17:16
timotimo i'm not the only one who works on this release \o/
Xliff tpm2016.zoffix.com/#/61
^ In that second example, am I the only one that can't see the operator?
timotimo i can see it
tadzik hm. I found github.com/tadzik/perl6-Acme-Meow/...eow.pm#L65 in my code, and now I'm wondering what's the backstory... 17:17
timotimo that's fantastic :D 17:18
they told me i could never teach a carrot to fly!
jnthn tadzik: I...really can't remember :D
stmuk_ timotimo: oh we probably have to sync with github.com/tadzik/Task-Star
timotimo oh 17:19
thta's a point, yeah
Xliff timotimo, say <foo bar> ⊖ <bar meow>;
Oh! I can't see it in my bloody browser!
That's so weird. 17:20
timotimo maybe he used Sans sans for that
jnthn Urgh 17:21
17:21 Relsak joined
tadzik stmuk: oh, indeed 17:21
jack_rabbit (IRC):
argh, sorry
jnthn We still have that hang on Windows in parallel spectest around S10/S11
tadzik jnthn: I assume something with dashes in pod because of some parsing... stuff? :)
jnthn Maybe... 17:22
17:23 buharin joined
dalek ar: b2cbb4e | (Steve Mynott)++ | / (2 files):
remove Bailador directory
17:25
stmuk_ I'm also tempted to add Grammar::Debugger to star
oh its in 17:26
timotimo yeah. star also has the debugger commandline, but task::star doesn't
tadzik //o\ 17:27
one should be generated from the other
Xliff tpm2016.zoffix.com/ <- REALLY GOOD presentation?
And now I am off for glasses.
17:27 Xliff is now known as Xliff_OFF 17:31 sue left
RabidGravy fixes the tests of WebSocket for a third time 17:32
17:33 espadrine left
jnthn RabidGravy: Is it breaking because of some Rakudo regression, or? 17:33
17:33 _nadim joined
RabidGravy no, I think it may be failing now because it has sped up :) 17:34
17:35 wamba left
RabidGravy there was a potential race in the way the tests were done in that they were in callbacks in the websocket server and client, it was fine before, now the tests are passing, it's just the numbers are duplicated, out of order or just weird 17:36
sticking the results in an array and doing for @tests -> $test { ok $test<result>, $test<text> } is much more reliable 17:37
jnthn ah, ok :) 17:41
Damn us making things faster :P 17:42
dinner, rest, etc. & 17:45
ZoffixWin Xliff_OFF, those are just my slides. You can get the recording of the actual presentation and the answers to some of the questions I didn't answer during it in this blog post: blogs.perl.org/users/zoffix_znet/20...tions.html 17:47
Xliff_OFF, as for seeing operators. You likely just have a poor font in your browser that does not have that character available in it. 17:48
TimToady m: sub summer ($sum,$v) { $v %% 2 ?? $sum + $v !! $sum }; say [[&summer]] 0..5 17:52
camelia rakudo-moar a87fb4: OUTPUT«6␤»
dalek osystem: d7045f4 | tadzik++ | META.list:
Add Geo::Coder::OpenCage
17:55
tadzik so, in this module I wrote a fairly idiomatic (imho) Pod
masak .oO( iiiin &summer... )
RabidGravy POD-TASTIC! 17:56
tadzik unfortunately, it renders like like shit, both To::Text and To::HTML
masak how utterly thankless of it
tadzik and I wonder if that's renderers fault, or am I just using #| docs beyond what they were designed to be
and if the latter, could it maybe be better 17:57
RabidGravy I stopped using #| because of how bad everything looked
tadzik right. Who can we blame?
I'll look at the speculation
long story short: I think whatever is in #|'s should be respected exactly as if it was inside a =begin pod/=end pod block 17:58
but should also give you the method/sub/whatever signature for free
yeah, they're speculated to be simple text 17:59
but both mine and RabidGravy's experience shows that people expect more of it
RabidGravy tadzik, while I remember and you're here - would you be up for a patch to JSON::Unmarshal whereby if the top level object is a [ ] then it actually tries to return an array of the objects? 18:00
right now it just doesn't do anything and keeps quiet about it 18:01
tadzik RabidGravy: hmm, the later is clearly suboptimal 18:03
but I think I'd rather have it react to Object[] properly
if possible
18:03 _nadim left
tadzik as in: unmarshal to an array if the user expects a typed array 18:03
if you want an object, you should not get an array in return 18:04
timotimo "Texas Since Remarks" ?!? 18:06
tadzik wat 18:07
18:07 spanner joined
timotimo docs.perl6.org/language/unicode_texas - see the table 18:09
geekosaur thinsk that was 3 column headers at one point? 18:11
18:11 Relsak left
RabidGravy tadzik, ah yes of course, it doesn't actually need "fixing" as if you have data of "[ { .. }, { ... } ]" and tell it Array[Object] then it will DTRT 18:11
ptolemarch What is "Texas" about those? "Bigger [and better]"? 18:12
geekosaur texas form, date added, remarks. but nothing has a date and the long form and remarks are run together
timotimo ptolemarch: only bigger, otherwise equivalent 18:13
tadzik ah, right
ptolemarch nods 18:14
TimToady hmm, it doesn't say you can use Lo for alphabetics 18:15
tadzik then I'd just take care of the "silently does nothing" issue :) 18:16
timotimo should we point out Slang::Roman in that page?
TimToady m: say '山'.uniprop 18:17
camelia rakudo-moar a87fb4: OUTPUT«Lo␤»
TimToady m: my \山 = 'mountain'; say 山 18:18
camelia rakudo-moar a87fb4: OUTPUT«mountain␤»
18:19 spanner left
RabidGravy tadzik, yaw. Also I have to "fix" JSON::Infer to generate something "constant Foo = (Array[Bar] but JSON::Class); to deals with that 18:19
18:20 spanner joined 18:21 roguelazer left 18:22 roguelazer joined
RabidGravy tadzik, yeah just tested with the JSON that gave rise to the query in the first place and it *just* works, which is fabulous 18:32
tadzik :) 18:34
moritz why was Bailador removed from star? 18:38
tadzik I think both people who participated in that particular discussion agreed 18:39
one of the points being "it has competition now, why should it be the one shipped with Star?"
perhaps some more detailed discussion is required
moritz panda has competition too :-)
tadzik correct :0 18:40
:)
I also wonder why LWP::Simple is there rather than HTTP::UserAgent
moritz first to market
timotimo aye
18:43 Possum joined
tadzik so now the question is: is it also best to market :) 18:44
timotimo probably not :) 18:45
ufobat what is the "nom" branch, currently?
or why is this the default branch, if you dont specify a perl6 version for travis-ci 18:46
masak ufobat: only historical reasons and inertia. 18:47
skids It's the default one that people PR/push to. Eventually supposed to get something more like "master" IIRC.
masak ufobat: the "n" in "nom" used to stand for "new", long ago. it replaced something that everyone's forgotten now.
18:47 labster joined
RabidGravy well, if someone comes along and says "why can't I get/post binary/gzipped data" then you can point them at HTTP::UserAgent 18:48
timotimo thank god the thing it replaced isn't remembered any more :)
diakopter or the several things prior to that
18:48 photog13 joined
ufobat masak, yeah that's what i read from articles from 2011. so makes this sense for travis-ci? 18:48
i am asking because on travis my perl6 with - latest is not working :/ 18:49
skids ufobat: if your objective is to test against the bleeding edge, yes.
ufobat ah! so nom is still bleeding edge :)
timotimo yeah 18:50
18:50 photog13 left
skids Well, not so much as experimental side-branches. 18:50
But it's where those get pushed when they are good enough.
ufobat good to know, thanks :) 18:51
and maybe someone has got an idea what i was doing wrong with travis-ci pastebin.com/5Qvzm0Pj 18:52
timotimo why don't you just link to the travis log site? :) 18:54
oh, someone didn't update panda properly
skids (for those that don't like scrolling, panda build step failed due to Shell::Command step failing due to not finding File::Which) 18:55
timotimo yeah 18:56
tadzik: didn't i tell you you'll have to include File::Which?
now how do we do that ...
tadzik oh gawd
that's correct
ufobat sorry, didnt know what the travis log seite is public to all. 18:58
timotimo tries 18:59
tadzik: want me to PR?
tadzik yes please :)
19:00 azawawi joined
azawawi hi 19:00
yoleaux 31 Mar 2016 09:03Z <RabidGravy> azawawi: RARRRR!
azawawi ducks
timotimo yo azawawi 19:01
azawawi timotimo: i noticed File::Which is getting a lot of attention lately... saw the travis logs
timotimo hm?
the travis logs of File::Which¿ 19:02
tadzik: i'm not sure if i was supposed to squash the subtree commit history into a single commit. i can do that, though
tadzik hmm, I usually don't
since that's what update-subtree does :)
timotimo you usually don't what?
azawawi timotimo: github.com/tadzik/Shell-Command/bl...mand.t#L45 # the start of the problem on win32 19:03
tadzik squash
timotimo so i shouldn't squash either?
19:03 labster left
timotimo azawawi: the problem is when we're just installing rakudo-star and perl6 isn't in $PATH yet 19:03
azawawi timotimo: i know
timotimo oh, you linked to S:C 19:04
azawawi my recommendation deprecate which in Shell::Command, recommend File::Which, drop File::Which from Shell::Command
RabidGravy if an object already has the role applied does "but" not do anything or do the copying part but skip the role applying part, or indeed neither
azawawi and on win32, fail that api on purpose since it is not meant to work on win32 19:05
nor Mac OS X
timotimo S:C already depends on F:W
i can kind of see why people would think a release is strenuous 19:06
azawawi only for the which command which was added in github.com/tadzik/Shell-Command/co...924fc23802
tadzik: what do you think? 19:07
timotimo well, yeah
azawawi re "deprecate which in Shell::Command, recommend File::Which, drop File::Which from Shell::Command"
timotimo oh, drop F:W from S:C?
tadzik hm
azawawi yup
which is not portable 19:08
in S:C
want me to prepare a PR?
19:08 sue joined
tadzik I don't actually have a clear philosophy of what should be included in S::C, but I know I don't want it to end up being an aggregator of modules that implement stuff 19:09
since which() proved to be tricky enough, I think I agree
(to deprecate it)
azawawi the funny thing is metacpan.org/source/FLORA/Shell-Co...Command.pm is exactly that :) 19:11
azawawi starts a fork 19:12
tadzik haha 19:13
maybe we should follow that :P
19:17 sQuEE` is now known as sQuEE, znpy joined 19:19 labster joined, sQuEE is now known as sQuEE` 19:23 leobut joined 19:25 leobut left
[Coke] jnthn: rt.perl.org/Ticket/Display.html?id=115618 - looks like this just needs a compose call, yah? Maybe this one's an easy close. 19:25
19:25 leobut joined, ChoHag joined
jnthn [Coke]: nah, we should actually fix it proper at some point :) 19:26
I'm pondering re-working how we do method caching a bit anyway, for other reasons 19:27
stmuk_ is anyone working on "panda uninstall"?
El_Che pretty nice perl6 thread at hackernews 19:28
ZoffixWin for the win
MadcapJake andreoss: you can totally use ==> in place of method chaining you just have two syntax things to watch out for: a closing curly will end a statement (so wrap in parentheses) and no argument subs need to be given empty postcircumfix parentheses (though this is ok to leave out at the end of a chain of them)
check out the docs on them (I recently added some details) doc.perl6.org/language/operators#Se...Precedence
19:29 leobut left
tadzik stmuk: no one that I know of 19:29
it's been requested recently though
19:29 leobut joined
nine I'd say it's been requested again and again since it's inception... 19:29
yoleaux 13:41Z <[Coke]> nine: is rt.perl.org/Ticket/Display.html?id=123276 done?
19:30 yqt joined
MadcapJake awwaiid++ # for the feed op docs too! 19:30
nine [Coke]: WRT #123276 I'd guess that it's been fixed. The reporter thinks so and it looks very much like the HTTP::UserAgent issue fixed by commit b384d8f777f6169052c25e5cc37353f01b8f5f1a 19:33
tadzik it now becomes a bit trickier with CUR and all
nine tadzik: you sure? github.com/rakudo/rakudo/blob/nom/...on.pm#L255 19:34
sjn \o 19:35
azawawi hmm, you cant use 'is DEPRECATED' with 'is export'?
tadzik oo:o
I had no idea
something for me to do on the qah then :)
19:36 brosenberg joined 19:37 leobut left 19:38 leobut joined 19:39 leobut left
moritz timotimo: do we actually need modules in top-level star if they are also provided by panda's bootstrap? 19:39
timotimo moritz: potentially not
not 100% sure how it works
moritz ==> Testing Shell::Command 19:40
Could not find File::Which in:
and then a list of directories
timotimo is that while trying to install panda?
moritz this is with newest panda imported into star
timotimo: yes
timotimo does that newest panda already have File::Which at all? 19:41
nine there's an open PR
timotimo and, if so, does it know to install that before trying Shell::Command
nine: what, the one i made? :)
nine yes
timotimo huh, some other commits sneaked by when i wasn't looking 19:43
azawawi tadzik: please review github.com/tadzik/Shell-Command/pull/13
azawawi waits for AppVeyor and Travis CI results
19:46 cdg left
azawawi moritz: 'is DEPRECATED' with 'is export' is a bad mix somehow 19:46
19:46 cdg joined 19:51 cdg left 19:53 cdg joined, cdg left, oceanwhit joined
moritz tadzik, timotimo: I've opened github.com/tadzik/panda/pull/304 which is basically the same as timotimo++'s 303, but it actually works :-) 19:54
and uses a squash merge
azawawi moritz: github.com/tadzik/Shell-Command/pull/13 ? 19:56
19:56 andreoss left
azawawi moritz: fix the problem from the source? :) 19:56
timotimo that sounds good
moritz *shrug* works for me too 19:57
just why was the dependency on File::Which added in the first place? 19:58
timotimo i mean moritz' suggestion sounds good
i guess so that Shell::Command can still provide File::Which
with just the one use statement
azawawi moritz: because of failed which() non-portable implementation on win32 19:59
moritz: panda was broken since 2016.01 as far as i recall
19:59 domidumont left
azawawi moritz: these days i use zef on windows 19:59
19:59 domidumont joined
moritz azawawi: so how does your pull request solve that? 20:00
I mean, a warning doesn't make it more portable, does it?
azawawi moritz: drops file-which, deprecates it... 20:01
moritz azawawi: but does that fix panda on windows?
azawawi moritz: tried 'is DEPRECATED' and it did not work with 'is export'
moritz: yup
moritz: it skips it on win32 and macosx
dalek osystem: ebc08af | RabidGravy++ | META.list:
Rename META for JSON::Marshal
20:02
moritz azawawi: so panda doesn't actually use which, it was only the tests that failed?
timotimo potentially, yeah
azawawi moritz: yup but tadzik was worried that people already depended on non-portable which implementation
moritz: so we need to deprecate it, then remove it later 20:03
timotimo twitter.com/typed/status/715988070177370114 - in this thread: people are really angry about that we call our ascii operator variants "texas"
stmuk_ hahaha 20:04
timotimo really angry is an overstatement
[Coke] nine - do we need tests on that one?
timotimo maybe "irritated"?
"unconvinced of the merit"? 20:05
[Coke] "who gives a f***" ?
geekosaur well, it is us-centric >.> 20:06
stmuk_ timotimo: qntm ironically claims "I have no strong feelings one way or the other" on his twitter profile
[Coke]++
timotimo it doesn't seem like he has strong feelings in that particular thread 20:07
today: perl6 calls operators either "unicode" or "texas". that's a thing that happens. also, people found that out and derided it on twitter. that's also a thing that hppens. move along, it's just the internet. 20:08
geekosaur "if that's the biggest complaint they can come up with about perl 6..." 20:09
timotimo that applies all the time
20:11 domidumont left
azawawi good night everyone 20:12
20:13 azawawi left 20:14 ufobat left
skids Shame on us for implying texans are intolerant of the french :-) 20:19
20:21 spider-mario joined
ugexe how dare they call it the goatse operator and not the man spreading his anus open operator 20:21
20:21 prammer left 20:22 prammer joined
kmwallio ugexe: to call it a "man" operator would alienate a lot of people I think... 20:22
Xliff_OFF So.... 20:23
(•_•)
( •_•)>⌐■-■
(⌐■_■)
....better glasses! Boo yah!
timotimo well, the operator itself doesn't have a representation of the primary sexual characteristics
you know, like the original picture
ugexe it also doesnt have the wedding ring
timotimo so it's gender neutral, really
which is nice and inclusive
20:24 Xliff_OFF is now known as Xliff
ugexe the perl6 goatse operator needs a ring on the left hand 20:24
timotimo we don't actually have the goatse operator, do we? 20:25
ugexe its still april 1...
20:25 cjdbikeit joined 20:26 _nadim joined, cjdbikeit left
jnthn m: say (^0x10FFFF).map(*.uniname).grep(*.contains('COMBINING RING')) 20:26
camelia rakudo-moar a87fb4: OUTPUT«(COMBINING RING ABOVE COMBINING RING BELOW COMBINING RING OVERLAY)␤» 20:27
jnthn No problem to put a ring on it :)
20:27 TEttinger joined
dalek osystem: 2b33fac | RabidGravy++ | META.list:
Rename META file for the JSON::Class
20:29
20:29 pmurias joined
gtodd what is slurp-rest 20:30
perlpilot unfortunate? 20:31
20:31 dvinciguerra left
gtodd as in: $handle.slurp-rest vs. $handle.IO.slurp 20:32
timotimo gtodd: the semantics of .slurp on an opened file weren't definable without problems, so we decided to call it "slurp-rest" and have very clear semantics
gtodd errm sigh ....
perlpilot really wishes we didn't have slurp-rest.
gtodd is there a slurp-some ? 20:33
timotimo oh, really?
20:33 darutoko left
timotimo that'd be .read($n) 20:33
gtodd :-D
perlpilot I wish it were just slurp with some modifiers
cfedde .slurp-enough
timotimo also supports :bin, i expect?
read, i mean 20:34
BBIAB
gtodd perlpilot: I thought it was ...
.slurp-dwim
skids Damn, my plan to make a "french fries" operator and then call "//" "the freedom fries operator" fizzles due to no fonts having ⫽. There's ᜶, but that would be Philippine fries. 20:35
cfedde just sorten that one to .dwim
gtodd cfedde: excellent
cfedde 'sorten'.dwim ==> shorten
gtodd perl6 -e '.dwim ;' end of golf forever 20:36
RabidGravy there
mst alias d="perl6 -e '.dwim'" # I beat gtodd
gtodd crap 20:37
gtodd d
Xliff Oh, god. It is 4/1... isn't it?
You know you are old when you forget April Fool's day.
Because you are just too effin tired to care. 20:38
gtodd Xliff: so you heard that perl6 has been discontinued
RabidGravy TreyHarris, in case you don't look at GH very often I have updated JSON::Class and JSON::Marshal so they deal properly with data like you have, and left an example on the GH issue which will work better than the earlier workardound
gtodd perl5 was renamed
BIG changes today 20:39
Xliff gtodd: Pull the other one, d00d.
Or then again.... maybe don't =P
RabidGravy haven't got anywhere with what I actually set out to do this morning but hey 20:40
Xliff RabidGravy: Have you decided what Vorbis lib you want, yet?
RabidGravy and there I was thinking it was a documented, open encoding scheme 20:43
20:43 pmurias left
RabidGravy why does it matter what lib? 20:43
20:43 buharin left
RabidGravy infact afaik there is only one full codec library, but you knew that already right? 20:47
20:52 leont joined, arc__ joined 20:53 arc__ is now known as arcetera, arcetera left, arcetera joined 20:54 buharin joined 20:56 sufrostico left 20:57 skids left
Xliff RabidGravy, no. I was hoping you'd do all the research for me and leave me with just the coding. =p 21:02
www.linuxsecurity.com/content/view/167517?rdf <- The mind boggles.
21:03 brosenberg left, ptolemarch left
Xliff I think I'll just go with libvorbisenc, then =) 21:05
garu tadzik: bump! github.com/tadzik/rakudobrew/pull/84 :) 21:06
tadzik garu: right! 21:07
21:09 go|dfish joined
garu \o/ 21:10
tadzik on it now :)
21:12 rindolf left 21:13 buharin left, tphilipp joined
Xliff Hrm. libvorbisenc is just a helper for libvorbis. *sigh* 21:14
garu tadzik: thanks! as usual, just let me know if you need anything
Xliff Actually could still be useful, though. 21:15
tadzik Done, built panda 2015.11 for moar-2015.11 21:16
seems to work :)
it could perhaps qualify for bonus points if installing panda for nom didn't result in a line saying "error: pathspec 'nom' did not match any file(s) known to git." 21:17
but it's a minor git output thing and I'm gonna let that slide :)
21:18 sjakke joined
sjakke check, check, check, is this thing on? 21:18
ZoffixWin yup 21:19
21:19 sufrostico joined
sjakke i'm checking in with some noob problems... can't install panda with rakudobrew. What info do you need to help me? 21:20
jdv79 ZoffixWin: nice presentation. i saw it on HN earlier today. it was confusing how to get it going though. 21:21
mst jdv79: he added instructions after realising it was posted 21:23
originally it was just his slides, which of course -he- already knew how to use ;)
ZoffixWin :) 21:24
sjakke am I on the right channel? is there a dedicatet noob channel?
ZoffixWin sjakke, this is a channel for everything. 21:25
Xliff is dissecting Audio::Libshout
garu tadzik: we can achieve that by tagging nom on panda :)
tadzik: I'd also love it if we tagged v6.c as well
ZoffixWin sjakke, what's the full output from rakudobrew when you try to install panda?
pastebin it 21:26
Xliff Rut Roh 21:29
21:32 cognominal_ joined
Xliff pastebin.com/mkQHe5fk 21:35
21:35 cognominal left
Xliff kills Pastebin devs for putting Comic Sans in for 4/1 21:35
tphilipp lol
Xliff It's in an april1st.css file, so hopefully it will go back to normal, shortly.
So gptrixie is borking for vorbis/codec.h. 21:36
Any ideas?
21:37 oceanwhit left 21:38 kaare_ left
RabidGravy that does look a bit special, I recall that skarsnik said that at one there was a problem if it wasn't the "real" gccxml but castxml instead 21:38
can't remember what the issue was
cognominal_ is parrot still supported on nqp ? 21:39
RabidGravy ah, "error: '--castxml-gccxml' does not work with c11" 21:40
21:42 eam joined, AlexDaniel joined
sjakke ZoffixWin, here it is pastebin.com/vtx2rghV 21:43
RabidGravy I think some distributions use castxml as an "alternative" to the real gccxml, which don't work 21:44
AlexDaniel I hate April 1st but this is pretty funny: mail.python.org/pipermail/python-d...43603.html
eam I'm encountering the exact bug described here rt.perl.org/Public/Bug/Display.html?id=123497 -- #!/usr/local/bin/perl6 executes as a shell script - using env to dispatch works as expected
21:45 sue left
eam oh, and I think I see the issue which isn't mentioned in the ticket -- bin/perl6 from osx brew isn't a binary. It's a shell script 21:45
I guess I just solved my own problem
RabidGravy sjakke, ah, I think that is something someone changed today, not sure if they are still around
sjakke Tried zef, and it worked fine. But it borked on me when trying to install Task::Star 21:46
I prefer the name zef.... and it reminds me of the antwoord 21:47
RabidGravy sjakke, if you have zef then you can fix the panda problem by installing File::Which first
it is entirely possibly that Task::Star has the same problem 21:48
sjakke could I maybe just stick with zef.... or is panda recommended for newcomers?
RabidGravy it shouldn't matter tbh
panda has just been around longer, more modules have been tested with it etc, etc 21:49
ugexe zef actualy doesnt mess up on star. its that some distributions in star do not declare all their build dependencies
one of which is panda, so if you are installing with panda you wouldnt notice 21:50
21:51 perlpilot left
ugexe sjakke: its name is indeed a homage to antwoord 21:51
RabidGravy maybe at some point there should be a common role for the Build.pm
21:52 perlpilot joined, sue joined, Vitrifurr left
sjakke ugexe: then i'm def sticking to zef ;) 21:53
RabidGravy it does seem a bit strange having a dependency on a particular package manager
21:54 Vitrifur joined
Xliff RabidGravy, so if the Great and Powerful Trixie is borking on vorbis/codec.h --- I'm kinda at an impasse. 21:55
Here's hoping I can do better with Xqilla.
sjakke thanks for the help guys. I'll most likely be back for more help :)
RabidGravy Xliff, at a guess I'd say your "gccxml" is actually castxml masquerading as gccxml, which I can confirm doesn't like that file 21:56
Xliff Yup. Same error.
Where can I get proper GCCXML? 21:57
And where is that file located?
RabidGravy what OS distribution are you using? 21:58
Xliff Ubuntu 21:59
And it would help if gccxml was installed. =P
RabidGravy it's entirely possible that you can install the "real" gccxml using whatever package manager and then flip the alternatives
Xliff Which it wasn't.
And it was disheartening seeing castxml as a dependency.
RabidGravy :(
21:59 Wiertek joined
RabidGravy for some reason everyone seems to want to flip to castxml 22:00
Xliff #! /usr/bin/python
#
# Emulate the gccxml command using castxml.
Fuuuuu.....
22:01 Wiertek left
RabidGravy fwiw fedora has gone completely over to castxml 22:01
Xliff "Note: GCC-XML has been succeeded by CastXML. 22:03
"
Right there on the front page...gccxml.github.io/HTML/Index.html
22:03 lizmat joined
Xliff Hrm. "apt install x" hangs at 6% consistently. 22:05
22:06 perlawhirl joined 22:07 Wiertek joined
sjakke Just did my first 'hello world' with perl6. Happy times 22:07
perlpilot sjakke: congrats :) 22:08
ZoffixWin \o/ 22:09
Xliff Calling GCCXML : gccxml /usr/include/vorbis/codec.h -fxml=plop.xml[]
Error: Could not run /usr/local/bin/gccxml:
Argument list too long
*sigh*
22:09 _4d47 left 22:15 molaf left
Xliff gccxml has some fucked up build instructions. 22:21
Aaaand... the build fails. 22:22
22:23 sufrostico left 22:24 Wiertek left 22:31 cognominal_ left
Xliff Can anyone here build gccxml? 22:35
perlawhirl .botsnack 22:37
yoleaux :D
perlawhirl just checking i was online :)
22:40 _nadim left, Sgeo__ joined
sjakke RabidGravy: So to install Task::Star, from zef, I'm required to install another package manager. Am I correct? 22:42
Xliff Downgrading to gcc 0.6.0...just to see if it builds.
s/gcc/gccxml/
RabidGravy sjakke, that is a workaround for the situation you find yourself in at the moment, yes. It shouldn't really be the case but one of the bundled dependencies has a dependency itself which didn't get bundled by mistake 22:44
or something like that
to be honest I've never installed Task::Star as it doesn't have anything in it that I would commonly use 22:45
sjakke that just seems wrong. Anywho.... I'll stick to zef. I'll manage without it. Just wanted it because the intro material mentions it. 22:47
RabidGravy it's a mistake
also Task::Star came about when there were less than a third of the module than there are now and hasn't really been adjusted since 22:49
Xliff Would compiling with gcc-4.6 and g++-4.9 cause problems? 22:51
22:51 kid51 joined
RabidGravy dunno 22:52
Xliff Wheee! 22:53
Xliff had to drop back to gcc and g++ v4.9 before gccxml would compile
Aaaaand.... still get the same error. 22:55
*SNARL*
sjakke I just want to see if I can connect and send files to my chromecast. I'm using an application called castnow and I'm unhappy with it.
RabidGravy send files how? 22:56
sjakke It's a great excuse to reboot my programming interest. Started Comp.Sci a couple of years ago, but dropped out 'cause java hurts my brain. 22:57
Xliff sjakke, It took me a long while to become cognizant in Java. 22:58
Over a decade, in fact.
It wasn't until modern IDEs started grokking Java that I finally got into it. (Android helped) 22:59
So... new development.... gccxml seems to work in an -non-root- shell.
sjakke I'm not sure what protocol castnow uses.... I need to read some of its code
Xliff WTF?
23:00 perlawhirl left
Xliff Aaaaannnd.... all of perl6 is build into root's homedir. 23:02
*sigh*
23:03 sue left 23:04 kid511 joined
Xliff This most likely means a complete reboot of perl6 in a non-root location. 23:04
stmuk_ Task::Star has documentation and debuggers!
23:05 kid51 left
RabidGravy sjakke, castnow appears to use a special chromecast protocol, you'll be needing to make an implementation yourself if you want to do it in Perl 6 23:05
Xliff Aaand I just blew away some uncommitted code for Match.pm -- fortunately I keep my gist updated. 23:07
Xliff is having a bad evening. Time for beer and loud music.
sjakke Well if someone can point me in the right direction I'll get cracking
23:07 perlawhirl joined
RabidGravy github.com/thibauts/node-castv2#pr...escription 23:09
there you go :)
23:11 edehont joined
RabidGravy if it can be implemented in javascript it definitely can be implemented in Perl 6 ;-) 23:11
23:11 edehont left
sjakke RabidGravy: Thanks man! A thank you all the way from the Faroe Islands. Probably the only Perl 6 user here ;) 23:11
RabidGravy lawks, I didn't know they had the internet there ;-) 23:12
sjakke We have great internet
Xliff rakudobrew build panda -- Fails when installing Shell::Command 23:13
ilmari it was a bit congested around the eclipse
Xliff Should I pastebin the error screen or is this known?
RabidGravy I guess half the trans-atlantic cables go that way
ilmari (mobile internets, at least) 23:14
sjakke I missed it :( it was cloudy, where I am
Xliff ==> Testing Shell::Command
Could not find File::Which in:
RabidGravy Xliff, no it's known, missing dependency on File::Which
Xliff JINX!
ilmari I only caught the last few seconds of the totality through a fortuitous gap in the clouds
Xliff Will that be fixed?
ilmari from in torshavn
-from
RabidGravy the Shell::Command that is bundled got updated with one that depends on File::Which and File::Which didn't get bundled 23:15
sjakke You were here. Cool.
RabidGravy anyhow I really was going to crash, going to plant some trees tomorrow 23:16
Xliff Yeah. So that means that panda is kinda non-installable.
RabidGravy toodles
Xliff nn, RabidGravy
I guess that sets the tone for me, as well.
Will start fresh tomorrow. Putting the perl6 stuff to bed, for now.
sjakke A local band made a music video with the eclipse in the background. One shot take 23:17
First ever to do so I think 23:18
ilmari yeah, i saw that one, impressive
23:20 RabidGravy left
jdv79 is there some sort of html parsing module yet that's fast enough? 23:20
23:21 sjakke left
leedo there is a partially completed libxml module, i started using Inline::Perl5 to use perl5's XML::LibXML 23:22
it's fast once things get loaded, but takes about 2 seconds to get it all loaded 23:23
23:27 DarthGandalf left
jdv79 yeah. ok. 23:35
thanks
23:36 wamba joined
AlexDaniel jdv79: yes, Gumbo 23:45
jdv79: github.com/Skarsnik/perl6-gumbo 23:46
Skarsnik++ :) 23:48
23:53 perlawhirl left 23:54 DarthGandalf joined
jdv79 oh, cool. 23:54
AlexDaniel jdv79: it is really fast and it does things correctly. I've used it in the past and was surprised by how well it worked 23:55
23:59 arcetera left