»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.
00:04 kirkins joined
ParsonsNose Zoffix: Yeah, substr was the way to go, I just sorta forgot about it. I had that same thought about .= though 00:06
00:06 Exodist left, Cabanossi left
Zoffix Inheritance from core classes isn't yet polished. So above you'd need to make your own .new method to make it work 00:06
ParsonsNose tnx 00:08
00:09 Exodist joined, Cabanossi joined
Zoffix m: use nqp; class Inty is Int { method new ($v) { method !set-v ($v) { nqp::bindattr_i(self, Int, '$!value', $v); self }; self.bless!set-v: $v; }; method ho { self + 7 } } ; my Inty $hi .= new: 6 ; say $hi.ho 00:13
camelia 13
Zoffix No idea if there's a non-nqpish way to do it...
00:14 labster left 00:15 Exodist left, Exodist joined
Zoffix ❗❗❗🌟🌟🌟❇️❇️❇️🌠🌠🌠🎉🎉🎉⚠️⚠️⚠️🚧🚧🚧🌈🌈🌈 00:16
PART 2: Upgrade Information for Changes Due to IO Grant Work: rakudo.org/2017/04/03/part-2-upgrad...rant-work/
❗❗❗🌟🌟🌟❇️❇️❇️🌠🌠🌠🎉🎉🎉⚠️⚠️⚠️🚧🚧🚧🌈🌈🌈
00:18 Guest20993 joined 00:20 gdonald left, _28_ria joined, gdonald joined
ParsonsNose m: role Inty { method Str { (self + 7).Str } } ; my $i = 4 does Inty ; say $i ; say $i + 1 00:21
camelia 11
5
ParsonsNose That seems like the nicest way to do it. Bonus evil point, I feel. 00:22
00:25 ParsonsNose left 00:26 Exodist left, Guest20993 left, Exodist joined 00:27 kirkins left 00:28 kirkins joined
BenGoldberg u: 🌈 00:30
unicodable6 BenGoldberg, U+1F308 RAINBOW [So] (🌈)
00:32 matiaslina joined
BenGoldberg One of these days, I've got to switch to an irc client which supports color emoji, cause that looks like an elbow noodle to me. And the 🎉 makes me think of 🍕. 00:33
00:34 matiasli1a left 00:35 Exodist left, Zoffix left 00:36 Exodist joined, sufrostico joined
AlexDaniel mmm, 🍕 00:37
00:37 matiasli1a joined 00:40 matiaslina left 00:44 labster joined 00:52 matiaslina joined 00:55 matiasli1a left 00:56 ok joined 00:57 ok is now known as Guest16705 01:00 Exodist left, Guest16705 left 01:01 Exodist joined 01:02 matiaslina left 01:09 Exodist left 01:12 Exodist joined 01:18 sufrostico left 01:22 Cabanossi left 01:24 Cabanossi joined 01:31 dnacat joined 01:36 dnacat left 01:45 wamba joined 01:46 sufrostico joined 01:52 geekosaur left, geekosaur joined 01:55 gdonald left 01:56 gdonald joined 02:00 _28_ria left 02:06 ryan_ joined 02:21 nadim left 02:22 Cabanossi left 02:24 Cabanossi joined 02:28 mooch2 joined 02:29 mooch2 left 02:33 noganex_ joined 02:36 noganex left 02:50 wamba left 02:59 chatter29 joined, chatter29 left 03:20 mattp_ joined 03:22 Cabanossi left 03:24 Cabanossi joined 03:27 khw left 03:35 semigloss left 03:47 kirkins left 03:50 Cabanossi left 03:52 Cabanossi joined
Geth doc: d050d4bb0a | (Zoffix Znet)++ | doc/Type/IO/Path.pod6
[io grant] Remove IO::Path.chdir prose

Will write new info after fixing up the method in Rakudo.
  - &chdir is vastly different and shouldn't be listed together
  - the prose is wrong, IO::Path.chdir doesn't use $*CWD
  - the prose mentions now-deprecated :test param
04:15
doc: 1d0e433c2a | (Zoffix Znet)++ | doc/Type/IO.pod6
[io grant] Document &chdir
04:32
04:35 MasterDuke left
Geth doc: 3fdc6dc413 | (Zoffix Znet)++ | doc/Type/IO.pod6
[io grant] Document &*chdir

Hoping there won't any be 'splosions due to the `&*` in the routine's name.
04:37
04:48 BenGoldberg left 04:50 labster left 04:52 sufrostico left 04:54 curan joined 05:10 parv joined 05:22 wamba joined, travis-ci joined
travis-ci Doc build errored. Zoffix Znet '[io grant] Document &chdir' 05:22
travis-ci.org/perl6/doc/builds/217941992 github.com/perl6/doc/compare/d050d...0e433c2af2
05:22 travis-ci left, domidumont joined 05:26 labster joined 05:27 domidumont left, Wanderer68 left
Geth doc: e1a299c1fd | (Zoffix Znet)++ | doc/Type/IO.pod6
[io grant] Reword "defined as" for &*chdir

To avoid suggestion that `sub &*foo` is valid syntax.
05:29
doc: e5225bed8a | (Zoffix Znet)++ | doc/Type/IO.pod6
[io grant] Fix URL to &*chdir
05:30
05:30 TEttinger joined
Geth doc: 8d4704b4c6 | (Zoffix Znet)++ | doc/Type/IO.pod6
Fix typo
05:30
05:31 xtreak joined
Geth doc: 4fc5d43dbb | (Zoffix Znet)++ | doc/Type/IO.pod6
Fix typo

Copy-pasta is the root of all evil
05:31
05:33 spebern joined 05:35 domidumont joined 05:36 xtreak left 05:37 xtreak joined 05:38 tokomer joined, spebern left 05:39 TimToady left, llfourn left 05:40 skids left 05:41 TimToady joined, xtreak left 05:42 llfourn joined 05:43 lowbro joined, lowbro left, lowbro joined 05:44 skids joined 05:45 IRCFrEAK joined 05:47 IRCFrEAK left 05:50 araraloren joined 05:51 RabidGravy joined 05:52 AlexDaniel left 05:58 skids left 06:01 xtreak joined, domidumont left 06:02 araraloren left 06:03 araraloren joined 06:15 kaare_ left, _28_ria joined, domidumont joined 06:18 kaare_ joined 06:19 xtreak left 06:20 xtreak joined 06:22 rindolf joined 06:24 xtreak left 06:27 xtreak joined 06:29 spebern joined 06:39 spebern left 06:40 salva joined 06:41 domidumont left 06:43 domidumont joined, ryan_ left 06:46 ryan_ joined 06:50 Cabanossi left 06:52 Cabanossi joined 06:53 grondilu left, wamba left 06:57 tokomer_ joined, tokomer left 07:07 tokomer_ left 07:10 dogbert17 left 07:23 abraxxa joined 07:40 xtreak left, mr-foobar left 07:41 wamba joined 07:44 xtreak joined 07:45 wamba left, zakharyas joined 07:51 Cabanossi left 07:53 Cabanossi joined 07:57 g4 joined, g4 left, g4 joined 07:59 dakkar joined 08:03 xtreak left 08:07 wamba joined 08:10 xtreak joined 08:17 AndyDee left, AndyDee joined 08:39 mr-foobar joined 08:46 robertle joined 08:50 Cabanossi left 08:53 perlpilot_ joined, Cabanossi joined, perlpilot left, pictho joined 08:54 pictho left 08:55 wamba left, labster left 08:56 domidumont left, wamba joined, mls joined 08:57 domidumont joined 09:01 xtreak left 09:03 xtreak joined 09:04 geekosaur left 09:11 xtreak left 09:13 xtreak joined, xtreak left 09:15 kaare__ joined 09:16 kaare_ left 09:17 [ptc] left 09:20 [ptc] joined 09:28 lowbro left 09:30 kaare__ is now known as kaare_ 09:33 wamba left 09:42 domidumont1 joined, domidumont left 09:55 xtreak joined, wamba joined 10:04 astj left 10:05 Cabanossi left, astj joined, astj left, astj joined 10:06 ryan_ left 10:07 Cabanossi joined
llfourn is splice the only way to remove an array element? 10:19
jnthn There's pop and shift also, or .= grep(...) 10:21
But yes, if it's "index in the middle"
llfourn jnthn: cool thanks. 10:22
using python recently and have gotten used to list.remove 10:23
(though, in python if you use it while iterating over it it will totally screw you)
jnthn Using indexing, or an iteration over the values themselves? 10:24
10:25 wamba left
jnthn guesses the latter 10:25
llfourn iterating over the values themselves 10:26
jnthn Yeah, I dunno how well that'll go in Perl 6 either, tbh :)
10:26 MasterDuke joined
lizmat for @a.keys.eager should be save I guess ? 10:26
jnthn Sure, that will be, but that's not the values :) 10:27
lizmat for @a.values.eager ?
jnthn I just don't like writing code that mutates stuff it's iterating over, though.
lizmat for @a.eager for that matter I guess
jnthn I guess .eager is safe in that it snapshots
llfourn m: my @a = <one two three four>; for @a { @a.splice(2,1) when "two"; .say } # seems ok
camelia one
two
four
jnthn Yeah, for me it's not a matter of whether it's OK in the language, so much as I just find such code hard to reason about 10:28
Not at the point I write it...but a few months later when I have to change it :)
lizmat true: when written the smartest way possible :-)
it will bite you in maintenance as you need to be smarter still then :-) 10:29
10:36 Cabanossi left 10:37 pmurias joined
pmurias is it ok if I add fudges for the js backend to roast? 10:37
10:37 Cabanossi joined
tbrowder hi, #perl6! 10:37
pmurias tbrowder: hi
lizmat pmurias: yes, that's what backend specific fudges are for :-) 10:38
#?rakudo.js I presume?
or does that need infrastructure work before that does what we hope it does ?
tbrowder i just saw bdfoy's update and read about appveyer, but i don't see it mentioned on p6 sites 10:39
hi, pmurias
jnthn I think the compiler.backend thing was made general enough to handle new ones "naturally"
lizmat tbrowder: it will be in tonght's Perl 6 Weekly :-) 10:40
jnthn pmurias++ # being far enough along to start running spectests :)
10:40 wamba joined
tbrowder ah1 (appveyor), excellent news for the win people! 10:41
must be some big bucks for all those MS licenses 10:42
jnthn Could do with setting appveyor up for MoarVM...
10:42 AndyDee left
jnthn Woulda caught the Windows build bustage we had recently 10:43
tbrowder is it ready enough to add info on win testing to modules docs?
10:46 araraloren left
jnthn 'fraid I haven't got any usage experience with it myself, but I think it's been around for a while 10:47
10:48 Actualeyes left, xtreak left 10:49 xtreak joined 10:52 xtreak left
timotimo i was asked to do that, and maybe i can do it 10:54
10:56 xtreak joined
tbrowder lizmat: i look forward to your excellent weekly p6 roundup, many kudos from the west side of the atlantic! 10:57
pmurias jnthn: I'll use rakudo.js, fudge can handle it without any changes the same way it does rakudo.jvm (the fugdges will be mostly the same as it's the same unicode issues jvm has I'm fudging around) 10:58
11:02 Actualeyes joined
tbrowder question: can anyone explain diff between range operator (..) and sequence operator (...) other than seq can used for "backwards" iteration in for loops? 11:02
moritz tbrowder: the range operator just returns a Range object 11:03
tbrowder: and you can't just use it for iteration, but also for range checking, for example
m: say 1.5 ~~ 1..3
camelia True
tbrowder and seq doesn't return an object? 11:04
moritz tbrowder: well, it returns a Seq, which is only for iteration 11:05
tbrowder: and it's much more flexible; you can give it a callable, or it can deduce other sequences
tbrowder ah, so 1..3 really is the set of real numbers between 1 and 3
moritz m: say (1, 2, 4, 8 ... *).head(10)
camelia (1 2 4 8 16 32 64 128 256 512)
moritz m: say (1, 1, &[+] ... *).head(10)
camelia (1 1 2 3 5 8 13 21 34 55)
tbrowder so would you say a "best practice is to alwas use a seq oper in for loops? 11:06
jnthn I think better advice is "use .. when it's powerful enough, ... when it isn't" 11:07
.. is much simpler
tbrowder thanks, moritz and jnthn
11:08 mattp_ left, mattp_ joined
lizmat hmmm... perhaps ... could revert to a simple Range when it can ? 11:10
jnthn Not really, because of the return type being different :) 11:11
11:12 Actualeyes left
jnthn May be that internally we can hand back a Seq that has whatever underlying iterator Range uses, though, if we ain't alrady doing that 11:12
lizmat but it could be a Seq running of a Range
exactly :-)
lizmat puts it on her list
11:15 xtreak left
Voldenet I didn't know perl can deduce sequences :o 11:19
m: say (1, 3, 9, 81 ... *).head(10)
camelia Unable to deduce arithmetic or geometric sequence from 3,9,81 (or did you really mean '..'?)
in block <unit> at <tmp> line 1
Voldenet m: say (1, 3, 9, 27 ... *).head(10)
camelia (1 3 9 27 81 243 729 2187 6561 19683)
11:20 xtreak joined
lizmat m: say (1, 3, { $_ * $_ } ... *).head(10) # Voldenet 11:21
camelia (1 3 9 81 6561 43046721 1853020188851841 3433683820292512484657849089281 11790184577738583171520872861412518665678211592275841109096961 1390084523771447327649397867896613031142188508085291379916048244300360726297664359410017691541096095218116655405488994…
11:25 kurahaupo__ joined, kurahaupo__ left 11:26 kurahaupo__ joined, kurahaupo__ left, xtreak left
Voldenet m: say (0..10).map(3 ** *).head(10) 11:26
camelia (1 3 9 27 81 243 729 2187 6561 19683)
11:26 xtreak joined, kurahaupo__ joined
Voldenet m: say (1, 3, * * * ... *).head(10) 11:28
camelia (1 3 3 9 27 243 6561 1594323 10460353203 16677181699666569)
Voldenet oh, * * * in this context takes two previous elements
m: say (1, 1, * + * ... *).head(10)
camelia (1 1 2 3 5 8 13 21 34 55)
11:29 xtreak left 11:30 lowbro joined 11:31 lowbro left, lowbro joined, pmurias left 11:32 pmurias joined 11:33 Actualeyes joined
lizmat Voldenet: yeah, that's why 1, 3 * * * ... * wouldn't produce the result you wanted 11:36
m: say (1, 3, * * * ... *).head(10) 11:37
camelia (1 3 3 9 27 243 6561 1594323 10460353203 16677181699666569)
Voldenet btw, while I'm reinstalling zef packages again, should I use tinynoise or readline? 11:38
erm
linenoise*
11:39 xtreak joined 11:40 pmurias left 11:41 Exodist left, pmurias joined, Exodist joined 11:43 mxco86 joined
lizmat I've had good experience with linenoise on my MBP 11:43
Voldenet I wish home/end worked on it though 11:47
lizmat perhaps a PR is on order then :-) 11:48
Voldenet it's funny, because even M-f M-b are working, but not home/end
lizmat perhaps you have some keymapping active that linenoise doesn't see ? 11:51
Voldenet oh, on regular xterm it does work 11:52
11:55 bjz joined
Voldenet the pesky case of ^[[H ^[[F vs ^[[1~ ^[[4~, eh 11:55
12:00 raschipi joined 12:08 wamba left 12:09 wamba joined 12:14 xtreak left
robertle anyone got some experience with file locks under perl6? I want lockf or so, and am wondering whether there is something built-in or a module 12:14
timotimo it's builtin
moritz uses sqlite for file locking :-) 12:15
timotimo huh, hold on
it's at least not documented
robertle couldn't find it eaither :)
timotimo but class IO::Handle totally has a lock and unlock method 12:16
not sure if it's part of the IO rework our grantee is currently doing
robertle ok, will look at that! who is that grantee btw? 12:17
12:17 xtreak joined
robertle or who else can I talk to to find out more? 12:17
timotimo it's zoffix, you can find them over at #perl6-dev
lizmat IO::Handle.lock *is* part of the IO grant work 12:18
timotimo how's it going to change? 12:19
lizmat rakudo.org/2017/04/02/upgrade-infor...rant-work/ look for IO::Handle.lock
"IO::Handle.lock no longer takes an Int:D argument, but a pair of named arguments instead. By default, it makes an exclusive, blocking lock. Use :shared named argument to make a shared lock instead and :non-blocking to make the method fail instead of waiting for lock."
timotimo ah
yeah, that's a lot better
than an integer flag
robertle thanksright, but it still does not say what it actually does. e.g. around crashing without unlocking, NFS etc 12:20
is it just the fcntl? there is bound to be some platform-specific madness going on as well...
timotimo one sec.
on windows we lockFileEx 12:21
and on linux we fnctl
fcntl*
12:21 ufobat joined
timotimo well, on anything-except-windows, of course 12:21
robertle this is in syncfile, right? 12:22
timotimo yup
robertle I am a noop and can never track down where stuff actually happens...
timotimo a tool could potentially be written that tracks down what exactly an nqp::foobar does 12:23
though it wouldn't necessarily be able to do the final step from "call lock on the REPR" to "these REPRs actually implement lock"
though perhaps it could
robertle I'm happy to do a lot of grepping, but frequently fail...
12:24 pmurias left
robertle either way, I'll cehck this out. thanks a lot! 12:24
12:24 pmurias joined 12:25 zakharyas left, llfourn left 12:26 parv left 12:28 titsuki left 12:30 Exodist left 12:32 Exodist joined, zakharyas joined 12:37 yusrideb left
tbrowder ref ranges: is it true that a range must be defined on numbers increasing monotonically? 12:39
lizmat m: dd "a" .. "z" 12:40
camelia "a".."z"
lizmat m: dd |("a" .. "z")
camelia "a"
"b"
"c"
"d"
"e"
"f"
"g"
"h"
"i"
"j"
"k"
"l"
"m"
"n"
"o"
"p"
"q"
"r"
"s"
"t"
"u"
"v"
"w"
"x"
"y"
"z"
lizmat increasing monotonically if iterated over, but not necessarily numbers
tbrowder ok, then increasing on whatever, gotcha
lizmat well, and then there's ranges you can't iterator over: 12:41
m: dd |(-Inf .. 0)
12:41 Exodist left
camelia (timeout) 12:41
lizmat m: dd |(-Inf .. 0).head(10)
camelia -Inf
-Inf
-Inf
-Inf
-Inf
-Inf
-Inf
-Inf
-Inf
-Inf
12:41 Exodist joined
tbrowder okay, i'll stay in the simle p6 world for now, thanks, lizmat 12:42
12:46 xtreak left
Voldenet now I wonder 12:51
m: dd |(-Inf .. 0).reverse
camelia (timeout) 12:52
Voldenet uh... I broke it
m: dd ((-Inf .. 0).reverse).take(10) 12:53
camelia Too many positionals passed; expected 1 argument but got 2
in block <unit> at <tmp> line 1
Voldenet m: dd (-Inf .. 0).reverse().take(10)
camelia Too many positionals passed; expected 1 argument but got 2
in block <unit> at <tmp> line 1
Voldenet m: dd (-Inf .. 0).reverse().head(10)
camelia (0, -1, -2, -3, -4, -5, -6, -7, -8, -9).Seq
lizmat smart, innit :-) 12:56
12:57 jeek left
Voldenet yeah, it's funny how most languages would fail on this though 12:57
13:05 mcmillhj joined 13:10 sufrostico joined
Voldenet what's the perl6 version of "std::variant<int, string>", should I always use multi instead? 13:10
m: subset V of Any where { $_ ~~ Int or $_ ~~ Str }; my V $v = "ok"; 13:11
camelia ( no output )
Voldenet hm, moderately acceptable 13:13
13:13 titsuki joined, pmurias left, adu left
jnthn m: subset V where Int|Str; my V $v = 42; say $v; $v = '42'; say $v; $v = 4.2; say $v 13:16
camelia 42
42
Type check failed in assignment to $v; expected V but got Rat (4.2)
in block <unit> at <tmp> line 1
jnthn That's a shorter way to write it :)
Voldenet makes sense 13:17
but hm, now I look at it, there's no real use case where multi isn't superior 13:18
13:23 bjz_ joined 13:24 bjz left 13:32 skids joined
[Coke] .seen pmurias 13:37
yoleaux I saw pmurias 10:58Z in #perl6: <pmurias> jnthn: I'll use rakudo.js, fudge can handle it without any changes the same way it does rakudo.jvm (the fugdges will be mostly the same as it's the same unicode issues jvm has I'm fudging around)
[Coke] waves from new $DAYJOB 13:38
timotimo greetings [Coke]
13:40 perlpilot_ is now known as perlpilot 13:50 Cabanossi left
lizmat [Coke] o/ 13:52
13:52 Cabanossi joined 13:56 mcmillhj left 14:00 wamba left 14:01 curan left 14:02 bjz joined, geekosaur joined, geekosaur left 14:03 bjz_ left 14:05 jonas1 joined 14:06 geekosaur joined 14:08 mcmillhj joined 14:13 mcmillhj left
moritz [Coke]: hey, what/where are you working now? 14:13
14:19 mcmillhj joined 14:23 llfourn joined 14:24 mcmillhj left 14:25 kurahaupo__ is now known as kurahaupo
Geth doc: bf377c7702 | (Zoffix Znet)++ | doc/Type/IO.pod6
[io grant] Document &indir
14:26
14:29 cdg joined, lowbro left 14:33 sufrostico left 14:34 bjz left, bjz_ joined 14:38 mcmillhj joined 14:43 alimon joined, mcmillhj left 14:50 Exodist left, Exodist joined
Geth doc: 5aa614f364 | (Zoffix Znet)++ | doc/Language/5to6-perlfunc.pod6
[io grant] Improve suggestion for Perl 5's opendir
14:51
14:54 mcmillhj joined 14:55 lowbro joined 14:57 [ptc] left 14:58 sufrostico joined 14:59 [ptc] joined, mcmillhj left 15:04 bjz_ left 15:09 Guest14087 joined 15:12 sufrostico left 15:14 bjz joined, mcmillhj joined 15:17 wamba joined 15:19 mcmillhj left 15:20 bdmatatu joined 15:22 zakharyas left
[Coke] moritz: I'm basically doing the exact same job for the same client out of the same office, except I am now getting paid by one of the "big 4" accounting firms instead of the client. (Thankfully. could have been a ton of layoffs here) 15:23
15:23 zakharyas joined 15:24 Guest14087 left
[Coke] (actual job is basically tax IT. webapps, devops around various in house (er, that changed as of today) tax apps) 15:26
Voldenet >tax apps 15:27
Voldenet runs away scared
Actually, I'm not that scared
[Coke] Voldenet: right there with ya. :|
"AIGH". "oh, wait, this isn't so bad."
Voldenet Got a question a propos supplies: Can I somehow make Supplier.emit stuff in a consumer's thread instead? 15:28
m: given Supplier.new -> $s { for ^3 -> $consumer { $s.Supply().tap( -> $val { sleep(.1); ($consumer, $val).say }) }; $s.emit($_) for ^3; }
camelia (0 0)
(1 0)
(2 0)
(0 1)
(1 1)
(2 1)
(0 2)
(1 2)
(2 2)
15:30 bdmatatu left
Geth doc: a53015ae49 | (Zoffix Znet)++ | doc/Type/IO/Path.pod6
[io grant] Clarify value of IO::Path.path
15:31
15:33 AlexDaniel joined
Geth doc: bdd18f14e8 | (Zoffix Znet)++ | doc/Type/IO/Path.pod6
[io grant] Fix desc of IO::Path.Str

The method is just an alias for $.path attribute, but the prose makes it sound you like get the complete path, which isn't true.
15:33
15:33 mcmillhj joined 15:34 sufrostico joined
Geth doc: b78d4fd907 | (Zoffix Znet)++ | doc/Type/IO/Path.pod6
[io grant] Include type names in links to methods

Since links will list all the methods with such name, make it easier for reader to know which type was meant.
15:35
15:38 mcmillhj left 15:40 khw joined 15:46 domidumont1 left 15:47 bjz left 15:48 mcmillhj joined
Voldenet The only solution I can think of would be starting a consuming values from a Supply and then dispatching them with a thread 15:51
but I hope there's more magical solution than that :) 15:52
jnthn Voldenet: Channels, not supplies, are the primarly mechanism for producer/consumer
15:53 mcmillhj left
Voldenet but when you need a mechanizm for multiple readers Supply is the one to use, right? 15:53
jnthn Competing? 15:54
Or publish/subscribe?
Voldenet Not competing, broadcasting
multiple reader with all of them receiving the same set
s/reader/readers/
jnthn OK, then supplies, and see the .schedule-on and .migrate methods
perlpilot one of these days, someone should write a cookbook for this sort of stuff.
Voldenet Ah. That tiny little line was the method I was looking for! :) 15:55
jnthn Probably you want schedule-on from the way you described things :)
perlpilot: Yeah, been thinking about that ;-)
jnthn wanders off to rest for a bit 15:56
15:57 abraxxa left 16:01 mcmillhj joined 16:02 zakharyas left 16:03 eyck left 16:05 robertle left 16:06 mcmillhj left 16:17 mcmillhj joined 16:22 mcmillhj left 16:24 travis-ci joined
travis-ci Doc build errored. Zoffix Znet '[io grant] Fix desc of IO::Path.Str 16:24
travis-ci.org/perl6/doc/builds/218119192 github.com/perl6/doc/compare/a5301...d18f14e820
16:24 travis-ci left 16:29 lowbro left 16:30 petercommand joined, petercommand left, petercommand joined 16:32 dha joined 16:35 eyck joined 16:37 Cabanossi left, mcmillhj joined 16:38 Cabanossi joined 16:41 dakkar left, mcmillhj left 16:50 domidumont joined 16:51 mcmillhj joined 16:55 mcmillhj left 16:57 aindilis left 16:59 robertle joined 17:00 tardisx_ left 17:02 mcmillhj joined, TeamBlast left 17:04 TeamBlast joined 17:05 Cabanossi left 17:06 mcmillhj left, raschipi left 17:07 Cabanossi joined 17:11 ChoHag joined 17:18 mcmillhj joined 17:23 mcmillhj left 17:28 TeamBlast left 17:31 mcmillhj joined 17:32 TeamBlast joined 17:34 zakharyas joined 17:36 mcmillhj left 17:51 mcmillhj joined 17:56 mcmillhj left 18:02 zakharyas left 18:03 mtj_ left 18:09 Exodist left 18:10 Exodist joined, wamba left 18:11 mcmillhj joined 18:12 espadrine joined, khw left 18:17 mcmillhj left 18:27 mcmillhj joined 18:31 labster joined 18:32 mcmillhj left, [particle] left 18:38 cdg left
RabidGravy boo! 18:39
timotimo yo gravy how's you? 18:41
18:41 dalek left, dalek joined, ChanServ sets mode: +v dalek, john51_ joined, john51 left
lizmat starts with this weeks Perl 6 Weekly 18:42
AlexDaniel \o/ 18:44
18:45 mcmillhj joined
vendethiel lizmat++ 18:45
18:48 jeek joined 18:49 AndyDee joined, mcmillhj left 18:50 dha left, Cabanossi left 18:52 Cabanossi joined
[Coke] lizmat++ 18:56
Geth specs: 5a7a5aab85 | (Zoffix Znet)++ | v6d.pod
Add reminder to deprecate IO::Path.chdir in 6.d
dylanwh ^ because it can't work under the jvm?
(all sorts of wonderfully fun things can happen if you manage to call chdir() from the JVM) 18:57
18:57 sufrostico left
AlexDaniel dylanwh: github.com/zoffixznet/IOwesomeness/issues/26 18:58
18:59 sufrostico joined
AlexDaniel “Rename IO::Handle.slurp-rest to just .slurp” hmmmm 19:02
dylanwh ah
AlexDaniel it feels there was some reason for that, hm 19:03
19:04 domidumont left 19:05 mcmillhj joined 19:07 dha joined
dylanwh I just know large parts of the JVM will not work right if you do a real chdir() 19:08
if that chdir is emulated, that should be safe
but actually changing the OS's idea of the current dir will break the JVM.
19:09 sufrostico left
dylanwh (subtly break; some java APIs will not realize the cwd is different and return the old one, and in some cases involving dynamically loading bytecode will fail) 19:09
timotimo AlexDaniel: just because it didn't close the handle by default, which is something .slurp itself does ... technically speaking
dylanwh at least circa java 6ish
19:10 sufrostico joined
AlexDaniel timotimo: 🤷 19:11
timotimo: github.com/zoffixznet/IOwesomeness/issues/9
19:14 silug left
RabidGravy timotimo, I'm fabulous 19:14
19:17 jonas1 left 19:22 sufrostico left 19:24 rindolf left 19:26 kaare__ joined, kaare_ left, kaare__ is now known as kaare_ 19:31 sufrostico joined 19:32 brrt joined 19:36 rindolf joined
dha Crazy question: Has anyone tried building rakudo in Ubuntu running under Crouton on a Chromebook? :-) 19:37
19:38 bjz joined
timotimo i haven't even heard of crouton yet 19:41
dha Allows you to install an actual linux dist on a chromebook. 19:42
timotimo ah, ok
like ubuntu-on-windows10 :P :P :P
RabidGravy I've got a computer called fromage if that counts
dha effectively allows you a dual boot with chromeos and (in most cases, I suspect) ubuntu.
timotimo appreciation of such computers comes from age.
dha RabidGravy - :-)
I was leaving the country and with all the shouting about the government searches of data at the border, I figured I should take something other than my main laptop. 19:44
But didn't want to spend a whole bunch of money on it. This was suggested as a solution, and has mostly worked well.
However, I'm now getting weird permission errors when trying to build rakudo 19:45
*probably* (I'm guessing) because I'm building it on the SD card I've put in it so I have a usable amount of storage. 19:46
19:47 agentzh joined
dha I'm kind of tempted to just try it with sudo and see if that resolves the issue. 19:47
RabidGravy oh, is the SD formatted as native or fat32?
dha damn good question. I would think I would have formatted it as native, but I no longer remember. 19:48
dha now tries to remember how to determine this. 19:49
RabidGravy it's just on fat32 it might not be able to do somethings to file
mount should tell you 19:50
dha I'd have to unmount it first for that, though, right? 19:53
geekosaur 'mount' without arguments shows you what is currently mounted including fstype 19:54
dha ah.
geekosaur see also /proc/mounts on linux 19:55
dha This is the problem with using things like OS X. You rarely need to deal with this stuff on the command line. :-)
Hm. it's saying fuseblk, which I think means NTFS. :-/ 19:56
dha wonders if this means he is screwed. 19:57
RabidGravy I'm not sure I've got anything I can format to find out 20:00
dha I think, since there's nothing tremendously important on this machine, I may just try sudo and see if that makes the permission problems go away. maybe the sd card thinks I'm not supposed to do stuff with it in some contexts. :-/ 20:03
Yikes. the space in "SD Card" seems to be confusing something, as I'm getting an error that refers to "/var/host/media/removable/SD" 20:07
hobbs some software is stupid.
dha yep.
20:14 TheDir left 20:21 cdg joined 20:23 brrt left, agentzh left 20:30 _28_ria left 20:36 perlawhirl joined
perlawhirl m: say so Bag(<A P>) ≼ Bag(<A T>) 20:37
camelia True
perlawhirl bisectable6: say so Bag(<A P>) ≼ Bag(<A T>)
bisectable6 perlawhirl, Bisecting by output (old=2015.12 new=cb27bce) because on both starting points the exit code is 0
perlawhirl, bisect log: gist.github.com/bdb03e47065afd5746...675d96d1cf
perlawhirl, (2017-04-03) github.com/rakudo/rakudo/commit/92...382e122777
perlawhirl .tell lizmat Baggy (<+)/(>+) Baggy commit is not working properly: irclog.perlgeek.de/perl6/2017-04-03#i_14370322 20:41
yoleaux perlawhirl: I'll pass your message to lizmat.
lizmat perlawhirl: will look at it after the P6W 20:42
yoleaux 20:41Z <perlawhirl> lizmat: Baggy (<+)/(>+) Baggy commit is not working properly: irclog.perlgeek.de/perl6/2017-04-03#i_14370322
lizmat meanwhile, adding some todoed tests to roast would be appreciated :-)
perlawhirl np
i'm actually really happy you're working on the set_operators... i actually had a commit in mind just shortly before you started your work 20:43
with regards to (elem) and (cont)
lizmat could you gist your idea ? 20:44
perlawhirl Bag's end up being coerced to a Set, which slows things down. could that be broadened to QuantHash?
lizmat that's actually on my list of things to look at 20:45
if we need Set semantics, we don't actually need to coerce Bags to Sets if we can look on the inside :-)
20:46 rindolf left
perlawhirl cool... thanks lizat++ 20:46
20:49 kst` is now known as kst 20:51 [particle] joined, perlawhirl left 20:53 sufrostico left 20:54 bjz left 20:56 robertle left 20:59 sufrostico joined 21:11 RabidGravy left 21:12 sufrostico left
lizmat wow, a lot happened the past week 21:12
dha That's good in this context, right? 21:13
21:13 skids left
timotimo yup 21:15
dha whew.
timotimo liz is writing the weekly, so that means we got a good set of things to report
21:17 setty1 left 21:18 mcmillhj left 21:22 pmurias joined
pmurias dylanwh: IO::Path.chdir is being deprecated because it doesn't do what you would expect it to do (it doesn't actually change anything) 21:24
dha In recent times, "a lot happened in the last week" has not been universally a good thing. :-/
21:25 sufrostico joined
lizmat dha: true, and a lot won't be mentioned on the P6W :-) 21:25
dha lizmat - And thank you for that. :-) 21:30
21:32 mcmillhj joined
AlexDaniel moritz: hey. I'm trying to fetch IRC logs with :Accept<application/json>, but I am getting this as a response: {"ARRAY(0x34eb040)":"/perl6/ARRAY(0x34eb040)"} 21:34
moritz: something is not right…
21:35 agentzh joined 21:36 Cabanossi left
AlexDaniel moritz: ah sorry, that's an error on my side 21:37
moritz: the responsy is still weird though…
21:37 agentzh_ joined 21:38 Cabanossi joined, mcmillhj left 21:39 agentzh left 21:43 bjz joined 21:53 bjz left 21:55 dha left 21:58 bjz joined 22:00 mcmillhj joined 22:05 mcmillhj left 22:06 bjz left 22:11 mcmillhj joined 22:13 cpage_ left 22:15 agentzh_ left 22:16 mcmillhj left 22:17 agentzh joined
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2017/04/03/...-advances/ 22:19
22:22 ChoHag left 22:23 dha joined, espadrine left 22:25 mcmillhj joined 22:29 mcmillhj left 22:36 Cabanossi left 22:38 mtj_ joined, Cabanossi joined, cpage_ joined 22:39 mcmillhj joined 22:44 mcmillhj left 22:45 cpage_ left 22:46 sufrostico left 22:48 sufrostico joined 22:52 sufrostico left 22:53 dha left 22:55 mcmillhj joined 23:00 mcmillhj left 23:04 kyan joined 23:17 cpage_ joined 23:18 kurahaupo left 23:22 mcmillhj joined 23:26 mcmillhj left 23:28 dct joined 23:32 Labb29 joined, Labb29 left 23:33 Labb29 joined 23:38 pmurias left 23:39 nicq joined
nicq Hello o/ 23:39
sammers hello 23:40
23:40 Labb29 left
AlexDaniel moritz: and still, stuff is not parseable. Here is a gist from #perl6-dev 2017-03-17: gist.github.com/AlexDaniel/aff83ce...a330d0deb9 23:43
[[\"I\", \" \", \"❤\", \"️\", \" \", \"P\", \"e\", \"r\", \"l\", \" \", \"6\"], 10, 6]
»
JSON::Fast says “at 6525: I don't understand the escape sequence \"️” and JSON::Tiny says “Input (13231 characters) is not a valid JSON string” 23:44
not sure if both parsers are wrong or if clog should not output stuff like this 23:45
.tell moritz irclog.perlgeek.de/perl6/2017-04-03#i_14371056 23:46
yoleaux AlexDaniel: I'll pass your message to moritz.
23:50 mcmillhj joined 23:51 skids joined 23:55 mcmillhj left, nicq left