Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2017-04-03

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:04 kirkins joined #perl6
00:06 ParsonsNose Zoffix: Yeah, substr was the way to go, I just sorta forgot about it. I had that same thought about .= though
00:06 Zoffix Inheritance from core classes isn't yet polished. So above you'd need to make your own .new method to make it work
00:08 ParsonsNose tnx
00:09 Exodist joined #perl6
00:09 Cabanossi joined #perl6
00:13 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 rakudo-moar dcebce: OUTPUT: «13␤»
00:13 Zoffix No idea if there's a non-nqpish way to do it...
00:15 Exodist joined #perl6
00:16 Zoffix ❗❗❗????????????❇️❇️❇️????????????????????????⚠️⚠️⚠️????????????????????????
00:16 Zoffix
00:16 Zoffix PART 2: Upgrade Information for Changes Due to IO Grant Work: http://rakudo.org/2017/04/03/part-2-upgrade-information-for-changes-due-to-io-grant-work/
00:16 Zoffix
00:16 Zoffix ❗❗❗????????????❇️❇️❇️????????????????????????⚠️⚠️⚠️????????????????????????
00:18 Guest20993 joined #perl6
00:20 _28_ria joined #perl6
00:20 gdonald joined #perl6
00:21 ParsonsNose m: role Inty { method Str { (self + 7).Str } } ; my $i = 4 does Inty ; say $i ; say $i + 1
00:21 camelia rakudo-moar dcebce: OUTPUT: «11␤5␤»
00:22 ParsonsNose That seems like the nicest way to do it. Bonus evil point, I feel.
00:26 Exodist joined #perl6
00:28 kirkins joined #perl6
00:30 BenGoldberg u: ????
00:30 unicodable6 BenGoldberg, U+1F308 RAINBOW [So] (????)
00:32 matiaslina joined #perl6
00:33 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 ????.
00:35 Zoffix left #perl6
00:36 Exodist joined #perl6
00:36 sufrostico joined #perl6
00:37 AlexDaniel mmm, ????
00:37 matiasli1a joined #perl6
00:44 labster joined #perl6
00:52 matiaslina joined #perl6
00:56 ok joined #perl6
01:01 Exodist joined #perl6
01:12 Exodist joined #perl6
01:24 Cabanossi joined #perl6
01:31 dnacat joined #perl6
01:46 wamba joined #perl6
01:46 sufrostico joined #perl6
01:52 geekosaur joined #perl6
01:56 gdonald joined #perl6
02:06 ryan_ joined #perl6
02:24 Cabanossi joined #perl6
02:28 mooch2 joined #perl6
02:29 mooch2 left #perl6
02:33 noganex_ joined #perl6
02:59 chatter29 joined #perl6
03:20 mattp_ joined #perl6
03:24 Cabanossi joined #perl6
03:52 Cabanossi joined #perl6
04:15 Geth ¦ doc: d050d4bb0a | (Zoffix Znet)++ | doc/Type/IO/Path.pod6
04:15 Geth ¦ doc: [io grant] Remove IO::Path.chdir prose
04:15 Geth ¦ doc:
04:15 Geth ¦ doc: Will write new info after fixing up the method in Rakudo.
04:15 Geth ¦ doc:
04:15 Geth ¦ doc: - &chdir is vastly different and shouldn't be listed together
04:15 Geth ¦ doc: - the prose is wrong, IO::Path.chdir doesn't use $*CWD
04:15 Geth ¦ doc: - the prose mentions now-deprecated :test param
04:15 Geth ¦ doc: review: https://github.com/perl6/doc/commit/d050d4bb0a
04:32 Geth ¦ doc: 1d0e433c2a | (Zoffix Znet)++ | doc/Type/IO.pod6
04:32 Geth ¦ doc: [io grant] Document &chdir
04:32 Geth ¦ doc: review: https://github.com/perl6/doc/commit/1d0e433c2a
04:37 Geth ¦ doc: 3fdc6dc413 | (Zoffix Znet)++ | doc/Type/IO.pod6
04:37 Geth ¦ doc: [io grant] Document &*chdir
04:37 Geth ¦ doc:
04:37 Geth ¦ doc: Hoping there won't any be 'splosions due to the `&*` in
04:37 Geth ¦ doc: the routine's name.
04:37 Geth ¦ doc: review: https://github.com/perl6/doc/commit/3fdc6dc413
04:54 curan joined #perl6
05:10 parv joined #perl6
05:22 wamba joined #perl6
05:22 travis-ci joined #perl6
05:22 travis-ci Doc build errored. Zoffix Znet '[io grant] Document &chdir'
05:22 travis-ci https://travis-ci.org/perl6/doc/builds/217941992 https://github.com/perl6/doc/compare/d050d4bb0aa2...1d0e433c2af2
05:22 travis-ci left #perl6
05:22 domidumont joined #perl6
05:26 labster joined #perl6
05:29 Geth ¦ doc: e1a299c1fd | (Zoffix Znet)++ | doc/Type/IO.pod6
05:29 Geth ¦ doc: [io grant] Reword "defined as" for &*chdir
05:29 Geth ¦ doc:
05:29 Geth ¦ doc: To avoid suggestion that `sub &*foo` is valid syntax.
05:29 Geth ¦ doc: review: https://github.com/perl6/doc/commit/e1a299c1fd
05:30 Geth ¦ doc: e5225bed8a | (Zoffix Znet)++ | doc/Type/IO.pod6
05:30 Geth ¦ doc: [io grant] Fix URL to &*chdir
05:30 Geth ¦ doc: review: https://github.com/perl6/doc/commit/e5225bed8a
05:30 TEttinger joined #perl6
05:30 Geth ¦ doc: 8d4704b4c6 | (Zoffix Znet)++ | doc/Type/IO.pod6
05:30 Geth ¦ doc: Fix typo
05:30 Geth ¦ doc: review: https://github.com/perl6/doc/commit/8d4704b4c6
05:31 xtreak joined #perl6
05:31 Geth ¦ doc: 4fc5d43dbb | (Zoffix Znet)++ | doc/Type/IO.pod6
05:31 Geth ¦ doc: Fix typo
05:31 Geth ¦ doc:
05:31 Geth ¦ doc: Copy-pasta is the root of all evil
05:31 Geth ¦ doc: review: https://github.com/perl6/doc/commit/4fc5d43dbb
05:33 spebern joined #perl6
05:35 domidumont joined #perl6
05:37 xtreak joined #perl6
05:38 tokomer joined #perl6
05:41 TimToady joined #perl6
05:42 llfourn joined #perl6
05:43 lowbro joined #perl6
05:43 lowbro joined #perl6
05:44 skids joined #perl6
05:45 IRCFrEAK joined #perl6
05:47 IRCFrEAK left #perl6
05:50 araraloren joined #perl6
05:51 RabidGravy joined #perl6
06:01 xtreak joined #perl6
06:03 araraloren joined #perl6
06:15 _28_ria joined #perl6
06:15 domidumont joined #perl6
06:18 kaare_ joined #perl6
06:20 xtreak joined #perl6
06:22 rindolf joined #perl6
06:27 xtreak joined #perl6
06:29 spebern joined #perl6
06:40 salva joined #perl6
06:43 domidumont joined #perl6
06:46 ryan_ joined #perl6
06:52 Cabanossi joined #perl6
06:57 tokomer_ joined #perl6
07:23 abraxxa joined #perl6
07:41 wamba joined #perl6
07:44 xtreak joined #perl6
07:45 zakharyas joined #perl6
07:53 Cabanossi joined #perl6
07:57 g4 joined #perl6
07:57 g4 joined #perl6
07:59 dakkar joined #perl6
08:07 wamba joined #perl6
08:10 xtreak joined #perl6
08:17 AndyDee joined #perl6
08:39 mr-foobar joined #perl6
08:46 robertle joined #perl6
08:53 perlpilot_ joined #perl6
08:53 Cabanossi joined #perl6
08:53 pictho joined #perl6
08:54 pictho left #perl6
08:56 wamba joined #perl6
08:56 mls joined #perl6
08:57 domidumont joined #perl6
09:03 xtreak joined #perl6
09:13 xtreak joined #perl6
09:15 kaare__ joined #perl6
09:20 [ptc] joined #perl6
09:42 domidumont1 joined #perl6
09:55 xtreak joined #perl6
09:55 wamba joined #perl6
10:05 astj joined #perl6
10:05 astj joined #perl6
10:07 Cabanossi joined #perl6
10:19 llfourn is splice the only way to remove an array element?
10:21 jnthn There's pop and shift also, or .= grep(...)
10:21 jnthn But yes, if it's "index in the middle"
10:22 llfourn jnthn: cool thanks.
10:22 llfourn using python recently and have gotten used to list.remove
10:23 llfourn (though, in python if you use it while iterating over it it will totally screw you)
10:24 jnthn Using indexing, or an iteration over the values themselves?
10:25 * jnthn guesses the latter
10:26 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 #perl6
10:26 lizmat for @a.keys.eager should be save I guess ?
10:27 jnthn Sure, that will be, but that's not the values :)
10:27 lizmat for @a.values.eager ?
10:27 jnthn I just don't like writing code that mutates stuff it's iterating over, though.
10:27 lizmat for @a.eager for that matter I guess
10:27 jnthn I guess .eager is safe in that it snapshots
10:27 llfourn m: my @a = <one two three four>; for @a { @a.splice(2,1) when "two"; .say } # seems ok
10:27 camelia rakudo-moar 4c3190: OUTPUT: «one␤two␤four␤»
10:28 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 jnthn Not at the point I write it...but a few months later when I have to change it :)
10:28 lizmat true: when written the smartest way possible  :-)
10:29 lizmat it will bite you in maintenance as you need to be smarter still then  :-)
10:37 pmurias joined #perl6
10:37 pmurias is it ok if I add fudges for the js backend to roast?
10:37 Cabanossi joined #perl6
10:37 tbrowder hi, #perl6!
10:37 pmurias tbrowder: hi
10:38 lizmat pmurias: yes, that's what backend specific fudges are for   :-)
10:38 lizmat #?rakudo.js I presume?
10:38 lizmat or does that need infrastructure work before that does what we hope it does ?
10:39 tbrowder i just saw bdfoy's update and read about appveyer, but i don't see it mentioned on p6 sites
10:39 tbrowder hi, pmurias
10:39 jnthn I think the compiler.backend thing was made general enough to handle new ones "naturally"
10:40 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 #perl6
10:41 tbrowder ah1 (appveyor), excellent news for the win people!
10:42 tbrowder must be some big bucks for all those MS licenses
10:42 jnthn Could do with setting appveyor up for MoarVM...
10:43 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?
10:47 jnthn 'fraid I haven't got any usage experience with it myself, but I think it's been around for a while
10:49 xtreak joined #perl6
10:54 timotimo i was asked to do that, and maybe i can do it
10:56 xtreak joined #perl6
10:57 tbrowder lizmat: i look forward to your excellent weekly p6 roundup, many kudos from the west side of the atlantic!
10:58 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)
11:02 Actualeyes joined #perl6
11:02 tbrowder question: can anyone explain diff between range operator (..) and sequence operator (...) other than seq can used for "backwards" iteration in for loops?
11:03 moritz tbrowder: the range operator just returns a Range object
11:03 moritz tbrowder: and you can't just use it for iteration, but also for range checking, for example
11:03 moritz m: say 1.5 ~~ 1..3
11:03 camelia rakudo-moar 5ae454: OUTPUT: «True␤»
11:04 tbrowder and seq doesn't return an object?
11:05 moritz tbrowder: well, it returns a Seq, which is only for iteration
11:05 moritz tbrowder: and it's much more flexible; you can give it a callable, or it can deduce other sequences
11:05 tbrowder ah, so 1..3 really is the set of real numbers between 1 and 3
11:05 moritz m: say (1, 2, 4, 8 ... *).head(10)
11:05 camelia rakudo-moar 5ae454: OUTPUT: «(1 2 4 8 16 32 64 128 256 512)␤»
11:05 moritz m: say (1, 1, &[+] ... *).head(10)
11:05 camelia rakudo-moar 5ae454: OUTPUT: «(1 1 2 3 5 8 13 21 34 55)␤»
11:06 tbrowder so would you say a "best practice is to alwas use a seq oper in for loops?
11:07 jnthn I think better advice is "use .. when it's powerful enough, ... when it isn't"
11:07 jnthn .. is much simpler
11:07 tbrowder thanks, moritz and jnthn
11:08 mattp_ joined #perl6
11:10 lizmat hmmm... perhaps ... could revert to a simple Range when it can ?
11:11 jnthn Not really, because of the return type being different :)
11:12 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
11:12 lizmat exactly  :-)
11:12 * lizmat puts it on her list
11:19 Voldenet I didn't know perl can deduce sequences :o
11:19 Voldenet m: say (1, 3, 9, 81 ... *).head(10)
11:19 camelia rakudo-moar 5ae454: OUTPUT: «Unable to deduce arithmetic or geometric sequence from 3,9,81 (or did you really mean '..'?)␤  in block <unit> at <tmp> line 1␤␤»
11:19 Voldenet m: say (1, 3, 9, 27 ... *).head(10)
11:19 camelia rakudo-moar 5ae454: OUTPUT: «(1 3 9 27 81 243 729 2187 6561 19683)␤»
11:20 xtreak joined #perl6
11:21 lizmat m: say (1, 3, { $_ * $_ } ... *).head(10)   # Voldenet
11:21 camelia rakudo-moar 5ae454: OUTPUT: «(1 3 9 81 6561 43046721 1853020188851841 3433683820292512484657849089281 11790184577738583171520872861412518665678211592275841109096961 1390084523771447327649397867896613031142188508085291379916048244300360726297664359410017691541096095218116655405488994…»
11:25 kurahaupo__ joined #perl6
11:26 kurahaupo__ joined #perl6
11:26 Voldenet m: say (0..10).map(3 ** *).head(10)
11:26 camelia rakudo-moar 5ae454: OUTPUT: «(1 3 9 27 81 243 729 2187 6561 19683)␤»
11:26 xtreak joined #perl6
11:26 kurahaupo__ joined #perl6
11:28 Voldenet m: say (1, 3, * * * ... *).head(10)
11:28 camelia rakudo-moar 5ae454: OUTPUT: «(1 3 3 9 27 243 6561 1594323 10460353203 16677181699666569)␤»
11:28 Voldenet oh, * * * in this context takes two previous elements
11:28 Voldenet m: say (1, 1, * + * ... *).head(10)
11:28 camelia rakudo-moar 5ae454: OUTPUT: «(1 1 2 3 5 8 13 21 34 55)␤»
11:30 lowbro joined #perl6
11:31 lowbro joined #perl6
11:32 pmurias joined #perl6
11:33 Actualeyes joined #perl6
11:36 lizmat Voldenet: yeah, that's why 1, 3 * * * ... *   wouldn't produce the result you wanted
11:37 lizmat m: say (1, 3, * * * ... *).head(10)
11:37 camelia rakudo-moar 5ae454: OUTPUT: «(1 3 3 9 27 243 6561 1594323 10460353203 16677181699666569)␤»
11:38 Voldenet btw, while I'm reinstalling zef packages again, should I use tinynoise or readline?
11:38 Voldenet erm
11:38 Voldenet linenoise*
11:39 xtreak joined #perl6
11:41 pmurias joined #perl6
11:41 Exodist joined #perl6
11:43 mxco86 joined #perl6
11:43 lizmat I've had good experience with linenoise on my MBP
11:47 Voldenet I wish home/end worked on it though
11:48 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
11:51 lizmat perhaps you have some keymapping active that linenoise doesn't see ?
11:52 Voldenet oh, on regular xterm it does work
11:55 bjz joined #perl6
11:55 Voldenet the pesky case of ^[[H ^[[F vs ^[[1~ ^[[4~, eh
12:00 raschipi joined #perl6
12:09 wamba joined #perl6
12:14 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
12:14 * moritz uses sqlite for file locking :-)
12:15 timotimo huh, hold on
12:15 timotimo it's at least not documented
12:15 robertle couldn't find it eaither :)
12:16 timotimo but class IO::Handle totally has a lock and unlock method
12:16 timotimo not sure if it's part of the IO rework our grantee is currently doing
12:17 robertle ok, will look at that! who is that grantee btw?
12:17 xtreak joined #perl6
12:17 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
12:18 lizmat IO::Handle.lock *is* part of the IO grant work
12:19 timotimo how's it going to change?
12:19 lizmat http://rakudo.org/2017/04/02/upgrade-information-for-changes-due-to-io-grant-work/   look for IO::Handle.lock
12:19 lizmat "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."
12:19 timotimo ah
12:19 timotimo yeah, that's a lot better
12:19 timotimo than an integer flag
12:20 robertle thanksright, but it still does not say what it actually does. e.g. around crashing without unlocking, NFS etc
12:20 robertle is it just the fcntl? there is bound to be some platform-specific madness going on as well...
12:20 timotimo one sec.
12:21 timotimo on windows we lockFileEx
12:21 timotimo and on linux we fnctl
12:21 timotimo fcntl*
12:21 ufobat joined #perl6
12:21 timotimo well, on anything-except-windows, of course
12:22 robertle this is in syncfile, right?
12:22 timotimo yup
12:22 robertle I am a noop and can never track down where stuff actually happens...
12:23 timotimo a tool could potentially be written that tracks down what exactly an nqp::foobar does
12:23 timotimo though it wouldn't necessarily be able to do the final step from "call lock on the REPR" to "these REPRs actually implement lock"
12:23 timotimo though perhaps it could
12:23 robertle I'm happy to do a lot of grepping, but frequently fail...
12:24 robertle either way, I'll cehck this out. thanks a lot!
12:24 pmurias joined #perl6
12:32 Exodist joined #perl6
12:32 zakharyas joined #perl6
12:39 tbrowder ref ranges: is it true that a range must be defined on numbers increasing monotonically?
12:40 lizmat m: dd "a" .. "z"
12:40 camelia rakudo-moar b64f21: OUTPUT: «"a".."z"␤»
12:40 lizmat m: dd |("a" .. "z")
12:40 camelia rakudo-moar b64f21: OUTPUT: «"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"␤»
12:40 lizmat increasing monotonically if iterated over, but not necessarily numbers
12:40 tbrowder ok, then increasing on whatever, gotcha
12:41 lizmat well, and then there's ranges you can't iterator over:
12:41 lizmat m: dd |(-Inf .. 0)
12:41 camelia rakudo-moar b64f21: OUTPUT: «(timeout)»
12:41 lizmat m: dd |(-Inf .. 0).head(10)
12:41 camelia rakudo-moar b64f21: OUTPUT: «-Inf␤-Inf␤-Inf␤-Inf␤-Inf␤-Inf␤-Inf␤-Inf␤-Inf␤-Inf␤»
12:41 Exodist joined #perl6
12:42 tbrowder okay, i'll stay in the simle p6 world for now, thanks, lizmat
12:51 Voldenet now I wonder
12:51 Voldenet m: dd |(-Inf .. 0).reverse
12:52 camelia rakudo-moar b64f21: OUTPUT: «(timeout)»
12:52 Voldenet uh... I broke it
12:53 Voldenet m: dd ((-Inf .. 0).reverse).take(10)
12:53 camelia rakudo-moar b64f21: OUTPUT: «Too many positionals passed; expected 1 argument but got 2␤  in block <unit> at <tmp> line 1␤␤»
12:53 Voldenet m: dd (-Inf .. 0).reverse().take(10)
12:53 camelia rakudo-moar b64f21: OUTPUT: «Too many positionals passed; expected 1 argument but got 2␤  in block <unit> at <tmp> line 1␤␤»
12:53 Voldenet m: dd (-Inf .. 0).reverse().head(10)
12:53 camelia rakudo-moar b64f21: OUTPUT: «(0, -1, -2, -3, -4, -5, -6, -7, -8, -9).Seq␤»
12:56 lizmat smart, innit :-)
12:57 Voldenet yeah, it's funny how most languages would fail on this though
13:05 mcmillhj joined #perl6
13:10 sufrostico joined #perl6
13:10 Voldenet what's the perl6 version of "std::variant<int, string>", should I always use multi instead?
13:11 Voldenet m: subset V of Any where { $_ ~~ Int or $_ ~~ Str }; my V $v = "ok";
13:11 camelia rakudo-moar 224e40: ( no output )
13:13 Voldenet hm, moderately acceptable
13:13 titsuki joined #perl6
13:16 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 rakudo-moar 224e40: OUTPUT: «42␤42␤Type check failed in assignment to $v; expected V but got Rat (4.2)␤  in block <unit> at <tmp> line 1␤␤»
13:16 jnthn That's a shorter way to write it :)
13:17 Voldenet makes sense
13:18 Voldenet but hm, now I look at it, there's no real use case where multi isn't superior
13:23 bjz_ joined #perl6
13:32 skids joined #perl6
13:37 [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)
13:38 * [Coke] waves from new $DAYJOB
13:38 timotimo greetings [Coke]
13:52 lizmat [Coke] o/
13:52 Cabanossi joined #perl6
14:02 bjz joined #perl6
14:02 geekosaur joined #perl6
14:05 jonas1 joined #perl6
14:06 geekosaur joined #perl6
14:08 mcmillhj joined #perl6
14:13 moritz [Coke]: hey, what/where are you working now?
14:19 mcmillhj joined #perl6
14:23 llfourn joined #perl6
14:26 Geth ¦ doc: bf377c7702 | (Zoffix Znet)++ | doc/Type/IO.pod6
14:26 Geth ¦ doc: [io grant] Document &indir
14:26 Geth ¦ doc: review: https://github.com/perl6/doc/commit/bf377c7702
14:29 cdg joined #perl6
14:34 bjz_ joined #perl6
14:38 mcmillhj joined #perl6
14:43 alimon joined #perl6
14:50 Exodist joined #perl6
14:51 Geth ¦ doc: 5aa614f364 | (Zoffix Znet)++ | doc/Language/5to6-perlfunc.pod6
14:51 Geth ¦ doc: [io grant] Improve suggestion for Perl 5's opendir
14:51 Geth ¦ doc: review: https://github.com/perl6/doc/commit/5aa614f364
14:54 mcmillhj joined #perl6
14:55 lowbro joined #perl6
14:58 sufrostico joined #perl6
14:59 [ptc] joined #perl6
15:09 Guest14087 joined #perl6
15:14 bjz joined #perl6
15:14 mcmillhj joined #perl6
15:17 wamba joined #perl6
15:20 bdmatatu joined #perl6
15:23 [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 zakharyas joined #perl6
15:26 [Coke] (actual job is basically tax IT. webapps, devops around various in house (er, that changed as of today) tax apps)
15:27 Voldenet >tax apps
15:27 * Voldenet runs away scared
15:27 Voldenet Actually, I'm not that scared
15:27 [Coke] Voldenet: right there with ya. :|
15:27 [Coke] "AIGH". "oh, wait, this isn't so bad."
15:28 Voldenet Got a question a propos supplies: Can I somehow make Supplier.emit stuff in a consumer's thread instead?
15:28 Voldenet m: given Supplier.new -> $s { for ^3 -> $consumer { $s.Supply().tap( -> $val { sleep(.1); ($consumer, $val).say }) }; $s.emit($_) for ^3; }
15:28 camelia rakudo-moar 1ebeeb: OUTPUT: «(0 0)␤(1 0)␤(2 0)␤(0 1)␤(1 1)␤(2 1)␤(0 2)␤(1 2)␤(2 2)␤»
15:31 Geth ¦ doc: a53015ae49 | (Zoffix Znet)++ | doc/Type/IO/Path.pod6
15:31 Geth ¦ doc: [io grant] Clarify value of IO::Path.path
15:31 Geth ¦ doc: review: https://github.com/perl6/doc/commit/a53015ae49
15:33 AlexDaniel joined #perl6
15:33 Geth ¦ doc: bdd18f14e8 | (Zoffix Znet)++ | doc/Type/IO/Path.pod6
15:33 Geth ¦ doc: [io grant] Fix desc of IO::Path.Str
15:33 Geth ¦ doc:
15:33 Geth ¦ doc: The method is just an alias for $.path attribute, but the prose
15:33 Geth ¦ doc: makes it sound you like get the complete path, which isn't true.
15:33 Geth ¦ doc: review: https://github.com/perl6/doc/commit/bdd18f14e8
15:33 mcmillhj joined #perl6
15:34 sufrostico joined #perl6
15:35 Geth ¦ doc: b78d4fd907 | (Zoffix Znet)++ | doc/Type/IO/Path.pod6
15:35 Geth ¦ doc: [io grant] Include type names in links to methods
15:35 Geth ¦ doc:
15:35 Geth ¦ doc: Since links will list all the methods with such name, make it
15:35 Geth ¦ doc: easier for reader to know which type was meant.
15:35 Geth ¦ doc: review: https://github.com/perl6/doc/commit/b78d4fd907
15:40 khw joined #perl6
15:48 mcmillhj joined #perl6
15:51 Voldenet The only solution I can think of would be starting a consuming values from a Supply and then dispatching them with a thread
15:52 Voldenet 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 Voldenet but when you need a mechanizm for multiple readers Supply is the one to use, right?
15:54 jnthn Competing?
15:54 jnthn Or publish/subscribe?
15:54 Voldenet Not competing, broadcasting
15:54 Voldenet multiple reader with all of them receiving the same set
15:54 Voldenet s/reader/readers/
15:54 jnthn OK, then supplies, and see the .schedule-on and .migrate methods
15:54 perlpilot one of these days, someone should write a cookbook for this sort of stuff.
15:55 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 :)
15:55 jnthn perlpilot: Yeah, been thinking about that ;-)
15:56 * jnthn wanders off to rest for a bit
16:01 mcmillhj joined #perl6
16:17 mcmillhj joined #perl6
16:24 travis-ci joined #perl6
16:24 travis-ci Doc build errored. Zoffix Znet '[io grant] Fix desc of IO::Path.Str
16:24 travis-ci https://travis-ci.org/perl6/doc/builds/218119192 https://github.com/perl6/doc/compare/a53015ae49f9...bdd18f14e820
16:24 travis-ci left #perl6
16:30 petercommand joined #perl6
16:30 petercommand joined #perl6
16:32 dha joined #perl6
16:34 eyck joined #perl6
16:37 mcmillhj joined #perl6
16:38 Cabanossi joined #perl6
16:50 domidumont joined #perl6
16:51 mcmillhj joined #perl6
16:59 robertle joined #perl6
17:01 mcmillhj joined #perl6
17:04 TeamBlast joined #perl6
17:07 Cabanossi joined #perl6
17:11 ChoHag joined #perl6
17:18 mcmillhj joined #perl6
17:31 mcmillhj joined #perl6
17:32 TeamBlast joined #perl6
17:34 zakharyas joined #perl6
17:51 mcmillhj joined #perl6
18:10 Exodist joined #perl6
18:11 mcmillhj joined #perl6
18:12 espadrine joined #perl6
18:27 mcmillhj joined #perl6
18:31 labster joined #perl6
18:39 RabidGravy boo!
18:41 timotimo yo gravy how's you?
18:41 dalek joined #perl6
18:41 john51_ joined #perl6
18:42 * lizmat starts with this weeks Perl 6 Weekly
18:44 AlexDaniel \o/
18:45 mcmillhj joined #perl6
18:45 vendethiel lizmat++
18:48 jeek joined #perl6
18:49 AndyDee joined #perl6
18:52 Cabanossi joined #perl6
18:56 [Coke] lizmat++
18:56 Geth ¦ specs: 5a7a5aab85 | (Zoffix Znet)++ | v6d.pod
18:56 Geth ¦ specs: Add reminder to deprecate IO::Path.chdir in 6.d
18:56 Geth ¦ specs: review: https://github.com/perl6/specs/commit/5a7a5aab85
18:56 dylanwh ^ because it can't work under the jvm?
18:57 dylanwh (all sorts of wonderfully fun things can happen if you manage to call chdir() from the JVM)
18:58 AlexDaniel dylanwh: https://github.com/zoffixznet/IOwesomeness/issues/26
18:59 sufrostico joined #perl6
19:02 AlexDaniel “Rename IO::Handle.slurp-rest to just .slurp” hmmmm
19:02 dylanwh ah
19:03 AlexDaniel it feels there was some reason for that, hm
19:05 mcmillhj joined #perl6
19:07 dha joined #perl6
19:08 dylanwh I just know large parts of the JVM will not work right if you do a real chdir()
19:08 dylanwh if that chdir is emulated, that should be safe
19:08 dylanwh but actually changing the OS's idea of the current dir will break the JVM.
19:09 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
19:09 dylanwh at least circa java 6ish
19:10 sufrostico joined #perl6
19:11 AlexDaniel timotimo: ????
19:11 AlexDaniel timotimo: https://github.com/zoffixznet/IOwesomeness/issues/9
19:14 RabidGravy timotimo, I'm fabulous
19:26 kaare__ joined #perl6
19:31 sufrostico joined #perl6
19:32 brrt joined #perl6
19:36 rindolf joined #perl6
19:37 dha Crazy question: Has anyone tried building rakudo in Ubuntu running under Crouton on a Chromebook? :-)
19:38 bjz joined #perl6
19:41 timotimo i haven't even heard of crouton yet
19:42 dha Allows you to install an actual linux dist on a chromebook.
19:42 timotimo ah, ok
19:42 timotimo like ubuntu-on-windows10 :P :P :P
19:42 RabidGravy I've got a computer called fromage if that counts
19:42 dha effectively allows you a dual boot with chromeos and (in most cases, I suspect) ubuntu.
19:42 timotimo appreciation of such computers comes from age.
19:42 dha RabidGravy - :-)
19:44 dha 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 dha But didn't want to spend a whole bunch of money on it. This was suggested as a solution, and has mostly worked well.
19:44 dha However, I'm now getting weird permission errors when trying to build rakudo
19:46 dha *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.
19:47 agentzh joined #perl6
19:47 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?
19:48 dha damn good question. I would think I would have formatted it as native, but I no longer remember.
19:49 * 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
19:50 RabidGravy mount should tell you
19:53 dha I'd have to unmount it first for that, though, right?
19:54 geekosaur 'mount' without arguments shows you what is currently mounted including fstype
19:54 dha ah.
19:55 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. :-)
19:56 dha Hm. it's saying fuseblk, which I think means NTFS. :-/
19:57 * dha wonders if this means he is screwed.
20:00 RabidGravy I'm not sure I've got anything I can format to find out
20:03 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. :-/
20:07 dha 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.
20:07 dha yep.
20:21 cdg joined #perl6
20:36 perlawhirl joined #perl6
20:37 perlawhirl m: say so Bag(<A P>) ≼ Bag(<A T>)
20:37 camelia rakudo-moar cb27bc: OUTPUT: «True␤»
20:37 perlawhirl bisectable6: say so Bag(<A P>) ≼ Bag(<A T>)
20:37 bisectable6 perlawhirl, Bisecting by output (old=2015.12 new=cb27bce) because on both starting points the exit code is 0
20:37 bisectable6 perlawhirl, bisect log: https://gist.github.com/bdb03e47065afd5746c38d675d96d1cf
20:37 bisectable6 perlawhirl, (2017-04-03) https://github.com/rakudo/rakudo/commit/928a406a8ece70a68136f378960bc7382e122777
20:41 perlawhirl .tell lizmat Baggy (<+)/(>+) Baggy commit is not working properly: https://irclog.perlgeek.de/perl6/2017-04-03#i_14370322
20:41 yoleaux perlawhirl: I'll pass your message to lizmat.
20:42 lizmat perlawhirl: will look at it after the P6W
20:42 yoleaux 20:41Z <perlawhirl> lizmat: Baggy (<+)/(>+) Baggy commit is not working properly: https://irclog.perlgeek.de/perl6/2017-04-03#i_14370322
20:42 lizmat meanwhile, adding some todoed tests to roast would be appreciated  :-)
20:42 perlawhirl np
20:43 perlawhirl 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 perlawhirl with regards to (elem) and (cont)
20:44 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?
20:45 lizmat that's actually on my list of things to look at
20:45 lizmat if we need Set semantics, we don't actually need to coerce Bags to Sets if we can look on the inside :-)
20:46 perlawhirl cool... thanks lizat++
20:51 [particle] joined #perl6
20:59 sufrostico joined #perl6
21:12 lizmat wow, a lot happened the past week
21:13 dha That's good in this context, right?
21:15 timotimo yup
21:15 dha whew.
21:15 timotimo liz is writing the weekly, so that means we got a good set of things to report
21:22 pmurias joined #perl6
21:24 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 #perl6
21:25 lizmat dha: true, and a lot won't be mentioned on the P6W  :-)
21:30 dha lizmat - And thank you for that. :-)
21:32 mcmillhj joined #perl6
21:34 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 AlexDaniel moritz: something is not right…
21:35 agentzh joined #perl6
21:37 AlexDaniel moritz: ah sorry, that's an error on my side
21:37 AlexDaniel moritz: the responsy is still weird though…
21:37 agentzh_ joined #perl6
21:38 Cabanossi joined #perl6
21:43 bjz joined #perl6
21:58 bjz joined #perl6
22:00 mcmillhj joined #perl6
22:11 mcmillhj joined #perl6
22:17 agentzh joined #perl6
22:19 lizmat and another Perl 6 Weekly hits the Net: https://p6weekly.wordpress.com/2017/04/03/2017-14-the-io-front-advances/
22:23 dha joined #perl6
22:25 mcmillhj joined #perl6
22:38 mtj_ joined #perl6
22:38 Cabanossi joined #perl6
22:38 cpage_ joined #perl6
22:39 mcmillhj joined #perl6
22:48 sufrostico joined #perl6
22:55 mcmillhj joined #perl6
23:04 kyan joined #perl6
23:17 cpage_ joined #perl6
23:22 mcmillhj joined #perl6
23:28 dct joined #perl6
23:32 Labb29 joined #perl6
23:33 Labb29 joined #perl6
23:39 nicq joined #perl6
23:39 nicq Hello o/
23:40 sammers hello
23:43 AlexDaniel moritz: and still, stuff is not parseable. Here is a gist from #perl6-dev 2017-03-17: https://gist.github.com/AlexDaniel/aff83ce7e7d45227e7b179a330d0deb9
23:43 AlexDaniel moritz: both JSON::Tiny and JSON::Fast are confused by this: "..rakudo-jvm fb4f16: OUTPUT: «[[\"I\", \" \", \"❤\", \"️\", \" \", \"P\", \"e\", \"r\", \"l\", \" \", \"6\"], 10, 6]␤»"
23:44 AlexDaniel 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”
23:45 AlexDaniel not sure if both parsers are wrong or if clog should not output stuff like this
23:46 AlexDaniel .tell moritz https://irclog.perlgeek.de/perl6/2017-04-03#i_14371056
23:46 yoleaux AlexDaniel: I'll pass your message to moritz.
23:50 mcmillhj joined #perl6
23:51 skids joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo