»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.
00:00 eyck left 00:01 eyck joined 00:03 mvorg left
Hotkeys .trans sure has some weird behaviour 00:03
m: say "a123b12332c".trans(<23> => <t>); say "a123b12332c".trans(<23 a> => <t a>); 00:04
camelia rakudo-moar c322e1: OUTPUT«a1ttb1ttttc␤a1tb1t32c␤»
Hotkeys it can replace matched strings but only if the match array has > 1 elems?
my matched strings I mean more than a single character 00:05
s/my/by/
00:05 mcmillhj joined 00:09 ovibos joined
unmatched} probably some hysterical raisings to do with perl 5's tr/// that only does single-char replacements. 00:10
rasins too
00:10 mcmillhj left 00:13 raiph left 00:14 Khisanth joined 00:17 mcmillhj joined 00:18 pierre_ joined, aries_liuxueyang joined 00:19 Nex6 left
konobi lo 00:19
yoleaux 5 Aug 2016 20:04Z <pmurias> konobi: do you have any ideas what (program):1755 might refer to in a profile I'm getting from v8?
unmatched} hel 00:20
00:21 mcmillhj left 00:32 canopus left 00:33 mcmillhj joined 00:38 mcmillhj left 00:39 canopus joined
TEttinger unicode request: curse tablet emoji www.qwantz.com/index.php?comic=2866 00:41
like a whole emoji that describes a request of a higher (or lower) power to have your enemy suffer in incredibly detailed ways. probably a whole supplementary plane, you could go nuts 00:42
00:43 mcmillhj joined 00:44 harovali joined 00:48 kshannon_ left, mcmillhj left 00:55 grondilu joined 00:58 mcmillhj joined 01:01 jack_rabbit left 01:03 mcmillhj left 01:10 mcmillhj joined 01:12 lambd0x left 01:15 mcmillhj left, cog_ left
dalek c: 97a01da | (Wenzel P. P. Peppmeyer)++ | doc/Language/io.pod6:
make examples compile
01:16
01:20 Sgeo__ joined 01:24 Sgeo_ left 01:26 mcmillhj joined 01:30 mcmillhj left 01:32 ponpon is now known as ponbiki, pduh joined 01:33 pduh left 01:34 kurahaupo_ joined 01:37 kurahaupo left 01:38 sena_kun left 01:40 mcmillhj joined 01:45 mcmillhj left, ilbot3 left 01:47 ilbot3 joined 01:50 kurahaupo_ left, kurahaupo joined 01:51 kurahaupo left, mcmillhj joined 01:55 mcmillhj left 01:59 kurahaupo joined, kurahaupo left 02:00 mohae joined 02:03 mohae_ left 02:05 Actualeyes joined 02:07 adu joined
dalek c: 1606ce2 | (Wenzel P. P. Peppmeyer)++ | doc/Language/functions.pod6:
make examples compile, fix broken examples
02:10
02:11 cpage_ joined 02:23 noganex joined 02:26 noganex_ left 02:33 cbk left, john51 joined 02:40 mcmillhj joined 02:46 mcmillhj left 02:53 grondilu left 02:54 cdg joined 02:59 cdg left 03:02 pierre_ left, pduh joined 03:03 pduh left 03:10 pierre_ joined 03:14 mcmillhj joined 03:18 mcmillhj left, Ben_Goldberg left 03:27 mcmillhj joined 03:32 mcmillhj left 03:35 rpburkholder joined, wamba joined 03:40 j75 joined 03:45 bob778 joined, nowan left, bob777 left, bob778 is now known as bob777
j75 Why does this work as a one liner ls | perl6 -e 'for (lines) { say $_ }' but ls | perl6 -e 'for lines { say $_ }' doesn't work? Why is it necessary to put lines in parentheses? 03:46
03:46 labster left, labster joined 03:47 autarch left, gregf_ left, nowan joined, gypsydav15 left, holyghost left 03:48 holyghost joined, autarch joined 03:49 gregf_ joined, hoelzro left, woodruffw left 03:50 hoelzro joined
gfldex m: sub f(&c){}; f {}; for f {} {}; 03:50
j75: ^^^ do you see the problem? 03:51
03:51 raiph joined, japhb left
gfldex m: .say for lines; 03:51
03:51 japhb joined 03:52 Cabanossi left
gfldex and this is how you solve this problem 03:52
j75 .say for lines works but why does that not require (lines)
@gfldex not sure I fully follow
gfldex because that form wont have a block as a 2nd "parameter" to for
camelia rakudo-moar 9bfbab: OUTPUT«Type check failed in binding to &c; expected Callable but got Hash (${})␤ in sub f at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
rakudo-moar 9bfbab: 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…»
03:53 Cabanossi joined
j75 but it's ok to call for @array ? 03:53
03:53 pierre_ left
gfldex subscripts are not separated by a space, so that case is fine 03:53
03:54 gypsydave5 joined, woodruffw joined, woodruffw left
gfldex m: my @a; for @a{1} {}; 03:54
camelia ( no output )
03:54 woodruffw joined, woodruffw left 03:55 woodruffw joined
j75 gfldex so what exactly does the () around lines do ? when called for (lines) { } 03:56
gfldex m: for lines() { .say }
camelia rakudo-moar 9bfbab: 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…»
gfldex it makes your intend clear not to call lines with a block as it's first parameter
j75 for lines() { .say } makes it clearer what's going on. is lines a sequence? 03:59
gfldex appreciate your help
03:59 bob777 left
gfldex m: say lines().^name; 03:59
camelia rakudo-moar 9bfbab: OUTPUT«Seq␤»
03:59 pierre_ joined 04:00 mcmillhj joined, bob777 joined
MasterDuke is there a way to create a single item list literal with angle quotes? 04:03
m: <abc>.WHAT.say; <a b c>.WHAT.say; < abc >.WHAT.say
camelia rakudo-moar 9bfbab: OUTPUT«(Str)␤(List)␤(Str)␤»
04:04 pierre_ left
MasterDuke i know the first form will always be a Str, but is there some other way to do it? 04:05
04:05 mcmillhj left
MasterDuke m: ['abc'].WHAT.say 04:05
camelia rakudo-moar 9bfbab: OUTPUT«(Array)␤»
MasterDuke that works, but is there a way without having to use both quotes and brackets 04:07
04:07 harovali left, Nex6 joined 04:08 harovali joined, skids left 04:13 Nex6 left 04:16 mcmillhj joined, Nex6 joined 04:18 pierre_ joined 04:21 mcmillhj left 04:22 khw left, pierre_ left 04:27 Cabanossi left 04:28 pierre_ joined
gfldex m: <abc>.Array.^name.say 04:28
camelia rakudo-moar 9bfbab: OUTPUT«Array␤»
gfldex m: my @a = <abc>; @a.^name.say;
camelia rakudo-moar 9bfbab: OUTPUT«Array␤»
gfldex MasterDuke: depends on where you want to use it 04:29
04:30 Cabanossi joined 04:31 mcmillhj joined 04:32 pierre_ left 04:33 itaipu joined 04:34 jack_rabbit joined 04:36 mcmillhj left 04:37 wamba left 04:41 Nex6 left 04:44 sufrostico joined 04:45 mcmillhj joined, pierre_ joined 04:50 mcmillhj left, solarbunny left, pierre_ left 04:55 itaipu left 05:01 solarbunny joined 05:04 pierre_ joined 05:08 pierre_ left 05:17 pierrot left, pierrot joined 05:18 Wiertek joined 05:22 pierre_ joined, mcmillhj joined 05:26 mcmillhj left, pierre_ left 05:31 pierre_ joined 05:32 Xliff_ joined, pierre_ left, mcmillhj joined 05:33 sufrostico left 05:35 Xliff left 05:36 pduh joined 05:37 mcmillhj left 05:38 pduh left 05:39 pduh joined 05:40 pierre_ joined 05:41 dsp- joined 05:42 zhmylove_ joined 05:43 Spot___ joined
gfldex m: say ¬10; 05:43
camelia rakudo-moar 9bfbab: OUTPUT«5===SORRY!5===␤Argument to "say" seems to be malformed␤at <tmp>:1␤------> 3say7⏏5 ¬10;␤Bogus postfix␤at <tmp>:1␤------> 3say 7⏏5¬10;␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ s…»
05:43 zostay_ joined, w4and0er96_ joined, mindos_ joined
gfldex m: say ∨10; 05:43
camelia rakudo-moar 9bfbab: OUTPUT«5===SORRY!5===␤Argument to "say" seems to be malformed␤at <tmp>:1␤------> 3say7⏏5 ∨10;␤Bogus postfix␤at <tmp>:1␤------> 3say 7⏏5∨10;␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ …»
gfldex m: say 8∨10;
camelia rakudo-moar 9bfbab: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Bogus postfix␤at <tmp>:1␤------> 3say 87⏏5∨10;␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ statement end␤ statement modifier␤ sta…»
05:44 pierrot_ joined, pierre_ left 05:45 japanoise_ joined 05:46 telex left, telex joined, Juerd_ joined 05:48 Wiertek left, bisectable6 left, zostay left, Spot__ left, dsp-_ left, mindos left, JimmyZ left, w4and0er96 left, pierrot left, zhmylove left, japanoise left, SmokeMachine____ left, avenj left, stigo left, Juerd left, avenj joined, avenj left, avenj joined, Juerd_ is now known as Juerd, SmokeMachine____ joined, Spot___ is now known as Spot__, zostay_ is now known as zostay 05:49 mindos_ is now known as mindos, JimmyZ joined, w4and0er96_ is now known as w4and0er96, stigo joined, pduh left 05:53 mcmillhj joined 05:58 mcmillhj left, harovali left, pierre_ joined 06:03 pierre_ left 06:05 rpburkholder left, movl left 06:09 movl joined 06:10 mcmillhj joined 06:14 pduh joined, mcmillhj left 06:16 pierre_ joined 06:17 adu left 06:21 pierre_ left 06:24 pduh left 06:26 mcmillhj joined
mr-foobar is there a thread pool abstraction library ? 06:28
06:30 firstdayonthejob joined, Wiertek joined
timotimo what do you mean "abstraction library"? 06:30
there is a thread pool
what do you need abstracted about it?
06:31 mcmillhj left 06:35 pierre_ joined
timotimo maybe it'd be a good start to read up on concurrency in perl6 and then see if the question about a thread pool abstraction library still stands 06:35
docs.perl6.org/language/concurrency
06:36 firstdayonthejob left 06:38 darutoko joined, mcmillhj joined 06:39 pierre_ left 06:41 Wiertek left 06:43 mcmillhj left 06:52 pduh joined 06:53 chienjo joined, pierre_ joined 06:54 chienjo left 06:57 pierre_ left 06:59 pierre_ joined 07:06 pduh left 07:07 pduh joined 07:08 pduh left, pduh joined 07:11 pduh left 07:15 sno left 07:30 brrt joined 07:34 kid51 left
j75 is the an array method similar to select in ruby? 07:34
timotimo i didn't know arrays had a method named "an"? 07:35
brrt j75: i think (not a ruby guy) we call that rep 07:36
grep
j75 ruby has a select method for arrays does perl have something similar?
brrt grep i think it what I am looking for thanks!
brrt yw :-)
07:43 movl left 07:44 zakharyas joined 07:46 dakkar joined 07:47 brrt left 07:54 rindolf joined 07:57 RabidGravy joined, CIAvash joined
j75 is there a method that combines grep and map? or do i have to use grep them map ? 07:58
nine use grep then map 08:00
It makes your code easier to follow anyway
j75 nine, thanks thought it'd be a little more efficient if i could do them at the same time rather than iterate a second time 08:01
08:01 movl joined
nine j75: never ever try to guess at the efficiency of your code. Always benchmark. In this case your error is in assuming that the chained .grep(...).map(...) will iterate twice, which it won't. 08:03
08:03 Administrateur joined
j75 nine, great to know! Thanks for the clear explanation 08:04
nine j75: both grep and map will actually just return an iterator. The real work happens once you ask for values (e.g. by assigning the result to an array). The assignment will ask map for the next value which in turn will ask grep for the next value. One by one. 08:06
That said, if you really, really insist on doing everything in the map, you can use loop control like next and last in the map, too. 08:07
08:07 Administrateur left
j75 Understood, thanks again! 08:08
08:11 zengargoyle joined
El_Che morning 08:17
08:27 pierre_ left 08:31 sno joined 08:32 j75 left
gregf_ j75: [*1..10].select(&:even?) == [1..10].grep(* %% 2 ) 08:32
08:34 pierre_ joined 08:36 pierre_ left, pierre_ joined 08:46 mvorg joined 08:55 torbjorn joined 08:59 eyck left 09:06 eyck joined 09:18 jkramer left
dalek c: 3bcbc19 | (Wenzel P. P. Peppmeyer)++ | doc/Language/unicode_entry.pod6:
show how to use digraphs (and extend them) in GNU Screen
09:37
gfldex how to do font linking on windows. This works quite well. I linked GNU unifont to FreeMono to get pretty much all possible glyphs. msdn.microsoft.com/en-us/goglobal/bb688134.aspx 09:42
TEttinger gfldex, interesting 09:52
there's a unifont upper too
gfldex can you read this: Ⓟⓔⓡⓛ ⑥ ? 09:59
timotimo yeah
10:03 pmurias joined
TEttinger yeah, I default to unifont in IRC 10:04
10:05 rand12418093821 left 10:17 lambd0x joined
lambd0x Hi fellas! 10:17
unmatched} Just fellas? :)
lambd0x unmatched}: :P
How you're doing? 10:18
unmatched} Just woke up and hung over.
lambd0x unmatched}: too.
unmatched} m: say 'test' 10:20
camelia rakudo-moar 9bfbab: OUTPUT«test␤»
unmatched} s: &say, \('test')
SourceBaby unmatched}, Sauce is at github.com/rakudo/rakudo/blob/9bfb...ors.pm#L22
unmatched} \o/ it's alive 10:21
timotimo SourceBaby: source 10:22
SourceBaby timotimo, See: github.com/zoffixznet/perl6-sourceable
10:25 pierre_ left 10:33 pierre_ joined
lambd0x finally got my little code running for multiplying matrixes after reading both from a file. \o/ 10:34
10:34 AlexDaniel joined
lambd0x guys is there a better way to the same: bpaste.net/show/6194ad04de36 10:34
?
timotimo is the code for first matrix the same code as for second matrix? itlooks that way 10:35
just using @B instead of @A
nine lambd0x: it could probably be a lot faster if you used native types. Assuming you can get away with it? 10:36
gregf_ implement a function that takes and presents a result of a x b where ,b are square matrices of order n
lambd0x nine: haven't seen them, where could I learn about it?
AlexDaniel committable6: HEAD bpaste.net/raw/6194ad04de36
committable6 AlexDaniel, Successfully fetched the code from the provided URL. 10:37
AlexDaniel, ¦«HEAD»: Cannot pop from an empty Array␤ in block <unit> at /tmp/z2wK0MsS7J line 4␤␤Actually thrown at:␤ in block <unit> at /tmp/z2wK0MsS7J line 4␤ «exit code = 1»
unmatched} lambd0x: the most obvious thing is to save matrix reading code in a function and shorten your program by like half :)
timotimo so why not either make it a function, or just put a loop around the thing
10:37 pierre_ left
lambd0x unmatched}: will try that. 10:38
10:40 labster left
unmatched} lambd0x: what's the input file? 10:41
gregf_ wow, thats handy @arr[ind1; ind2] = <whatever> 10:42
lambd0x unmatched}: something like this: bpaste.net/show/63c391aa39e7 .It's created by another code of mine that generates random order matrixes (both will always have the same order) 10:44
unmatched} And always two matrices? Looks like a fun Grammar excersise there :) 10:48
lambd0x unmatched}: In this case I limited it to be yes. Because I wasn't sure how to do it at first. After I get it nice maybe I'll try augmenting the amount of matrixes :P 10:50
10:51 japanoise_ is now known as japanoise
AlexDaniel MasterDuke: alright, it's time to make it possible to set STDIN of all our bots 10:51
unmatched} lambd0x: and if your other script can change the format of the files, I think it'd be easier to avoid specifying the length and instead specify a separator between matrices. 10:55
lambd0x unmatched}: one thing a time. But true :) 10:57
unmatched} :)
10:58 brrt joined 11:01 kaare_ joined 11:07 neuraload joined
lambd0x wow, better indeed not to work specifying sizes haahah things get more clean and sweet :P 11:11
11:11 harrison joined
lambd0x here's a somewhat improved version of my earlier code: bpaste.net/show/a3b870198815 11:12
11:12 harrison is now known as Guest54701
stmuk_ does YAPC slides 11:16
lambd0x something else inprovable in my script guys? 11:19
11:20 Guest54701 left, brrt left
timotimo no need for $file.get.words to have parenthesis 11:21
stmuk_ Perl v6.0.0 required--this is only v5.22.1 11:23
11:23 trnh joined
stmuk_ I assume perl 5 needs fixing if we aren't using 6.0.0? 11:23
11:23 trnh left, trnh joined 11:24 brrt joined
lambd0x timotimo: ok 11:24
timotimo you can use the value of the outer for loop instead of manually incrementing $i 11:25
11:26 neuraload left, sena_kun joined
dalek c: 8f8e9b5 | Altai-man++ | doc/Language/objects.pod6:
Index BUILDALL
11:27
lambd0x timotimo: I don't know how to do this. Was trying it earlier eheh 11:30
lambd0x: have a lot to learn in perl6. It's fun. 11:31
timotimo for ^$o -> $i { ... }
11:32 heatsink joined, mohae left
lambd0x timotimo: thanks. 11:43
12:06 brrt left 12:09 sufrostico joined 12:12 sufrostico left 12:14 buggable left 12:15 buggable joined 12:16 [Coke] left 12:17 [Coke] joined, kid51 joined 12:19 bazzaar joined
bazzaar \o perl6 12:20
unmatched} \o
bazzaar .oO I am wondering why any Perl 6 introduction course would not use the excellent Perl6 Introduction by Naoum Hankache as it's backbone ? 12:22
unmatched} :/ 12:23
12:23 movl left, movl joined
unmatched} benchable: sleep 60*10 12:25
benchable unmatched}: ¦«sleep»:Cannot find this revision
[Coke] bazzaar: is there a course in particular you think should have done that?
unmatched} committable6: HEAD sleep 60*10
committable6: HEAD say 'meow'
committable6 unmatched}, ¦«HEAD»: «timed out after 10 seconds, output»: «exit signal = SIGHUP (1)» 12:26
unmatched}, ¦«HEAD»: meow
bazzaar Having a themed set of code problems for users to tackle, giving the users 5-10mins breaks to research the appropriate section of Nauom's tutorial, has got to be the right approach, hasn't it?'s tutorial 12:27
unmatched} wouldn't want to attend such a course.
12:27 trnh left 12:28 grondilu joined
mst unmatched}: hrm? 12:28
why wouldn't reverse engineering slides and materials and treating perl6intro as course notes work? 12:29
12:29 kid51 left
bazzaar [Coke] ... yes ....?,?@ 12:29
unmatched} 'cause I can just do it at home, at my leasure, without having to race for the 5-10min mark. I can equally go through codeeval.com's problems using perl6intro as course notes 12:30
I'd expect the course to teach me something and the 5-10min window is where I try to apply my new knowledge to build something
[Coke] bazzaar: ok, which one are you thinking of that didn't do that? 12:31
12:31 kid51 joined 12:32 canopus left 12:33 perlpilot joined
bazzaar unmatched}: yes I know that anyone could do that, but when I look for a video on Perl6 introduction to work through, in a structured way, with some code problems that tie together, then I don't get that, I get something else 12:34
[Coke] bazzaar: I don't think anyone is making a coherent set of videos for that sort of thing, no.
mst bazzaar: oh, you want a *video*, not a course?
you never said that
[Coke] mst: does anyone other than gabor do that regularly on the p5 side? 12:35
mst not really. I do wish we had somebody who could actually write working perl doing videos, but I've not managed to convince anybody yet 12:36
12:36 edwinsage joined
bazzaar [Coke]: I don't want to point the finger, what I want is the fabulous product that is perl6, to be marketed and promoted in the best possible way 12:36
12:36 mcmillhj joined
[Coke] bazzaar: ok. you should lead with that, then. :) 12:36
also, please find us a marketer. :) 12:37
perlpilot bazzaar: wanting that makes it partly your responsibility :P
mst bazzaar: so, start trying to make a video from perl6intro.com and let people help
[Coke] notes down one request for video learnin'.
mst [Coke]: you know TPF has like an entire marketing team, right? just only liz and wendy seem to ever talk to them
unmatched} bazzaar: well, I take it it's an offer then? We do have a marketing repo in dire need of contribution: github.com/perl6/marketing/ 12:38
unmatched} tries to imagine storing videos in a github repo :o
[Coke] mst: no, I had no idea.
mst: let me downgrade to "very little idea" 12:39
mst [Coke]: mdk gives talks every couple years reminding people they exist
but without the community saying they want stuff done
TPF's treasurer (Dan Wright) will never allocate an actual budget for it
[Coke] mst: I have no way to get to the YAPCs these days. :|
mst and no access to youtube? :) 12:40
12:40 wamba joined
bazzaar O'Reilly make the video's, the perl6 community say 'who / what' should be the official approach / communicators, the perl6 code isn't being left to chance, why should it's marketing 12:40
12:40 canopus joined
mst ... 12:40
bazzaar: that's not how free software works
unless you're planning to personally pay O'Reilly to do that
[Coke] bazzaar: I think you misunderstand perl 6 and tpf's relationship with ORA.
mst because they really don't care much about perl
they're a conference company now
perl6's code is handled by volunteers. so will its marketing. 12:41
DrForr *cough* www.safaribooksonline.com/library/...491968321/
unmatched} :)
mst wonders why people are always insistant that marketing should be magically done by other people
even though they know that code doesn't get written unless somebody writes it 12:42
I wonder if this is the dilbert principle thingy in reverse
"I hear mauve gets the most pageviews"
[Coke] has found it's much easier to find good developers willing to donate their time than it is to find good marketers.
unmatched} lol
[Coke] bazzaar: anyway, yes, "we suck at marketing" is our slogan, so we get it. Constructive suggestions welcomed on how to improve the situation. 12:43
mst [Coke]: well, mdk tries, but then TPF won't let him have any money to market *with*
unmatched} Stopping saying "we suck at marketing" would be the good first step :)
[Coke] unmatched}: I tOLD YOU we WER Ebad at THIS OK!?
bazzaar I am not criticising the open source effort, that is not my point. The best perl6 video's (sound and vision) are the ones newcomers will be drawn to. 12:44
[Coke] mst: so there's half a dozen people on an unfunded committee?
bazzaar: ok. so next step: who is going to make and publish and market those videos?
bazzaar they need to be effective products
unmatched} heh 12:45
DrForr Good quality perl 6 videos will be good? Who knew?...
[Coke] mst: frustrating! (glad i'm on the one that gets some cash occasionally!)
mst [Coke]: pretty much. he manages to get money for print materials but generally has to ask on a per-print-run basis because TPF won't give them a budget
yeah, well, welcome to the world according to Dan Wright, where nobody else is allowed to decide things ever ;)
[Coke] mst: it's kind of how we do grants. we never know how much money we have, we have to ask at the end of each round what we can afford to fund.
... as I understand it. 12:46
that's actually one level above my pay grade, which I've been fine to keep there.
mst yeah, thing is, nobody except me appears to think this is worth complaining about, and if it's -only- me, it's easy to dismiss as 'mst is being an ass again' 12:47
so we continue to not actually have any way to get to functional marketing
and there's basically nothing I can do about it if nobody else will bother speaking up :(
unmatched} bazzaar: IMO that's too expensive to be the main focus of Perl 6 marketing. And tossing around several hundreds of gigs of raw data also makes collaboration by volunteers difficult. 12:48
DrForr That's why I've said 'fuck it' and started doing stuff. It's *not hard* to get in to these conferences, hell, if *I* cn. 12:49
12:49 mcmillhj left
bazzaar I am speaking up, I am going out on a limb just by voicing my thoughts on this 12:49
12:49 itaipu joined
unmatched} bazzaar: for comparison, the last 3.5 minute video I shot at work took two hours to shoot (included decisions on how it should be shot), took me about 30 minutes to edit, with the intro theme already done, and my repo for that job is 2.55GB 12:50
mst bazzaar: you're going out on a limb telling other people what to do, and refusing to help
bazzaar: that's not how open source works
unmatched}: hey, do youhave a good canonical example of perl6's errors being awesome? I'm trying to wow a rust dev 12:51
[Coke] m: my Rag $a;
camelia rakudo-moar 76be77: OUTPUT«5===SORRY!5===␤Type 'Rag' is not declared. Did you mean any of these?␤ Rat␤ Bag␤␤at <tmp>:1␤------> 3my Rag7⏏5 $a;␤Malformed my␤at <tmp>:1␤------> 3my7⏏5 Rag $a;␤␤»
unmatched} mst: funny that you mention, I kind thought rust's new error thing they announced was kinda awesome... awesomer than maybe even Perl 6 :) 12:52
bazzaar all I'm saying is, if perl6 related talks/ tutorials like those at Oscon are being professionally recorded, then it is imperative that there is coordinatination within the opensource community about how the product should be
unmatched} bazzaar: you got cut off after "the product should be"
[Coke] mst: mailed mdk 12:53
er, mk, whatever his MI is.
mst [Coke]: note if I've oversold what's doable here that's entirely my fault
I've been spending so long just trying to get people to talk to him in the first place that I may've garbled what is/isn't doable 12:54
bazzaar mst: I havn't refused to help, I have helped in a minor way previously bringing to light some bugs in the perl6 grammar and DBIish
mst I hope, under the circumstances, if that's the case, you'll forgive me
unmatched} We want a giant billboard! :D
mst bazzaar: training courses don't generally get recorded because the trainers need to get paid, sadly 12:55
[Coke] mst: eh, just pinging him to start a conversation, no worries.
DrForr bazzaar: You have control of your own presentation and style, that's it. 12:56
mst bazzaar: ok, but "we need to do X" type things don't generally help. "I think we should do X, and here's how I propose to start making that happen" helps
12:56 Actualeyes left
mst bazzaar: start doing stuff and then try and rope other people in 12:56
pmurias speaking of rust errors, blog.rust-lang.org/2016/08/10/Shap...-come.html is kindof interesting 12:57
unmatched} bazzaar: in other words, we don't suffer from lack of ideas :) It's the peoplehours and finances that are the limiting factors
mst yes, that's what led to me asking
unmatched}: anyway, got a good example?
AlexDaniel
.oO( rope! That's what we are good at )
unmatched} m: say Any eq $x
camelia rakudo-moar 76be77: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '$x' is not declared␤at <tmp>:1␤------> 3say Any eq 7⏏5$x␤»
unmatched} m: say Any eq 'x'
camelia rakudo-moar 76be77: OUTPUT«Use of uninitialized value of type Any in string context.␤Methods .^name, .perl, .gist, or .say can be used to stringify it to something meaningful. in block <unit> at <tmp> line 1␤False␤»
12:58 markk joined
unmatched} mst: nothing amazing comes to mind. 12:58
markk [Coke]: figured I'd just pop in here rather than keep it to email
The people I have worked with in the Perl6 community have generally been Liz and Wendy 12:59
And one or two others at conferences
AlexDaniel mst: perhaps try searching for “did you mean” in rakudo?
12:59 pierre_ joined
mst pmurias: possibly also of interest - 13:00
13:59 <nox> Most of our stuff comes from Elm, especially the layout for --explain with actual English sentences and whatnot.
13:59 <nox> elm-lang.org/blog/compiler-errors-for-humans
markk As mst pointed out the biggest block at the moment is that I have to present a case for funds each time. Technically I spend and they refund me, however I think I am still owed on the last print run (perl 6 and perl 5 stickerts and Perl 6 pamphlets) and I didn't get refunded for FLOSS yet
Some of this is my fault for not aggressively chasing it, Dan is a volunteer as well
jnthn mst: The ones that do Leavensh^WLeven^WLichtenstein^Wedit distance to suggest what you maybe meant are kinda nice
m: sai 42 13:01
camelia rakudo-moar 76be77: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤ sai used at line 1. Did you mean 'say'?␤␤»
markk As for producing videos, again possible. What you need to work out is what you are trying to achieve
jnthn m: class A { has $!x; method m() { $x } }
camelia rakudo-moar 76be77: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '$x' is not declared. Did you mean '$!x'?␤at <tmp>:1␤------> 3class A { has $!x; method m() { 7⏏5$x } }␤»
[Coke] ok. I mean, I trust liz and wendy with our marketing more than any of the other people that just do dev. :)
markk Do you want to promote Perl 6 or teach people how to use it - and don't say both. The issue with marketing thus far is that everyone wants to do it all
mst I did mention that liz and wendy and mdk had been talking 13:02
[Coke] markk: Bot---- oh. :)
markk You need to have one goal for the whole campaign or you get the current mess which I have trouble solving without funds
mst shakes [Coke] :P
[Coke] also, liz++ wendy++
DrForr bazzaar: I'm one of two people that's presenting Perl 6 in conjunction with OSCON. I asked people what I should talk about, and the response I got meant I basically wrote my own material, because of being Warnocked.
13:02 ptolemarch joined
markk mst: I read the backlog that [Coke] sent me :) 13:02
mst ah good
[Coke] markk: I don't know enough about marketing to know which of those two choices is better. 13:03
markk So what are you hoping to achieve?
perlpilot [Coke]: teaching people how to use Perl 6 would be best IMGO
er, IMHO
13:03 edehont joined
[Coke] I would imagine we can teach people how to perl 6 with our current resources, though. 13:03
G... Godlike? ;)
markk And teaching people may not get you more eyes looking at it 13:04
[Coke] so, see, even perlpilot and I cannot agree. :)
perlpilot yeah, that's it ;)
13:04 Actualeyes joined
DrForr And if I felt I was required to ask a committee of open source people to dictate my course content, I would ... have done nothing. 13:04
El_Che DrForr: you rebel
13:04 prammer joined
unmatched} I'm hoping to achieve: we straighten out the whole "Name Issue". That is: people thinking Perl 6 is just some mild improvement over some old language that was popular in the 90s 13:04
mst JFDI + -Ofun don't always produce the most co-ordinated results but they sure as heck produce results
13:04 pduh joined
unmatched} And I don't mean renaming anything, but getting people to know that Perl 6 is a new language. And it's the next "cutting edge" or whatever over the 25 year old Pythons and their kin 13:05
Once they know Perl 6 is something new and shiny and not a minor upgrade to an old language, I don't think we'll have trouble teaching people Perl 6. 13:06
markk Doesn't help when Larry calls it a teenager, it isn't. It is 8 months old, just had a 15 year pregnancy was all
unmatched} It's not uncommon that someone comments with "20 years too late, Perl 6" on my Perl 6 languages and that's entirely a matter of ignorance and that's what I'd aim the marketing efforts to change. 13:07
13:07 pduh_ joined
unmatched} *my Perl 6 articles 13:07
AlexDaniel markk: O_O
markk unmatched}: I think you might be devaluing people a little, it isn't new and shiny they go for at all, it is the missing out on the next big thing - Python and Javascript are neither new or shiny
mst both postmodern *and* postmature
13:07 Sgeo__ left
DrForr bazzaar: On the other hand, if you're interested in proposing talks to OSCON, FOSDEM, SXSW &c I can give you a hand writing your roposals. 13:07
AlexDaniel
.oO( ropeosals )
13:08
nine Speaking of budgets: we're now at 4500 Euros that my company did not contribute, because neither TPF, nor the YAPC Europe foundation seem to be capable of writing invoices which is the only thing I need in return for my 500 Euros / month budget.
El_Che imho, one of DrForr's project would be a nice addition for the sexyness of perl6. The atom beautifier packages uses perltidy for reformatting perl5 and perl6, what does not really work for perl6. With madcapjake perlfe's and azawawi's linter, perl6 programming on atom looks ok
nine [Coke]: ^^^
13:09 cdg joined
markk The aim is not to educate people, or change their minds, marketing will not do that for you. You have to tell them something, or repeat something, you have to make them want the thing, not give them a wiki article 13:09
13:09 pduh left, skids joined
markk It's not about showing them, it is about moving them into a position where they want it 13:09
DrForr El_Che: I'm getting whitespace handled this weekend.
13:09 cdg left, mcmillhj joined
bazzaar DrForr: It is not for me to criticise an individuals efforts, my original question, was why when we have an excellent web tutorial, why was isn't it being used in the media products that newcomers will be drawn too 13:09
13:10 cdg joined
markk It is why you have one goal and all your targetting should be towards making that goal happen 13:10
perlpilot markk: but telling them things without also having a way for them to easily learn and use Perl 6 seems a little useless to me. It ends up with frustrated potential users.
markk Coke is over a hundred and one of its marketing points is that it is a 'fresh taste'
perlpilot "Perl 6 looks cool, but I can't figure out how to do anything without #perl6" 13:11
markk perlpilot: I do not disagree
perlpilot: but we were discussing marketing and what you are trying to achieve by it
perlpilot: if you don't have the tools to support people learning then that's a different focus
El_Che DrForr: I volunteer to at least have a go for patch to atom-beautifier when you think it's ok. Not reaaly a knowledgable at JS, but I had a look yesterday and the patch seems trivial (I wrote a small one to activiate for perl6fe as it's seen by atom as a different grammar than perl6) 13:12
markk hence my question about what you are trying to achieve
unmatched} markk: what if there's nothing to be missing out on? I mean, Perl 6 doesn't have a niche, it's also slow and can be frustratingly buggy. 13:13
perlpilot A virtuous cycle of continual converts ;)
DrForr bazzaar: Well, find someone that does good voiceovers and write a script? I don't have a great voice, otherwise I'd volunteer.
markk If we want shiny videos to educate then we should focus on that but that shouldn't be confused with getting people to look at them as those are not the same
unmatched}: are you saying it has no usp whatsoever that I might as well use Perl5 or Python or JS? 13:14
[Coke] markk: ... your line about Coke REALLY confused me for about 10 seconds. :) 13:15
13:15 pduh joined
markk Because I can do the flippant taglines 'perl6: just like every other language but slower and likely to crash' - not a great usp 13:15
[Coke] nine, markk: markk, you are closer to the finance than I am, can you ping someone about nine's comment there?
markk [Coke]: oops, yeah sorry
DrForr Rust, Go and Swift users could ask the same question of perl6, python, js/node as well. 13:16
mst nine: bet EPO could do it
13:18 pduh is now known as pduh__
nine This is so frustrating. I walked to management, asked for 500€/month for supporting free software and actually got it. Now wasting a lot of work time trying to find someone who can take the money seems excessive. 13:18
markk nine: email [email@hidden.address] with your offer and I will get you an invoice for the money by the end of the day - just need the ammount, the frequency and a business address for the invoice
13:18 pduh__ left
El_Che nine++ : wow 13:18
perlpilot markk++
markk Fuck it, give me the details in query with your email address and I will get it done now
nine++ btw
mst nine: this would be one of the reasons I arranged for EPO to exist 13:19
nine++
13:19 Kaffe_ is now known as Kaffe
unmatched} markk: I can't think of any. I like the language for many of its small things, like letting me avoid using brackets all over the place. Perl 6 doesn't have a great ecosystem. It's too slow for web apps. It's too buggy for me to recommend using it as a primary language *right now* for some large business up. All of this kinda leads to lack of volunteers and points to the question: how do we get more volunteers 13:19
to be interested in this language to make it faster and less buggy?
perlpilot There's not enough ++ for nine. (I'm overly enamored of Inline::Perl5 :)
markk Nine - EPO will use the money to fund Perl-QA, CPAN Testers (all their servers) - Strawberry Perl (servers and bandwidth) and anything else that sounds good - we got MetaCPAN their server sponsorship as well
nine markk: all sound like things that help us at work (maybe except for Strawberry), so I'd be completely ok with supporting them :) 13:20
unmatched} s/business up/business app/; # can't type today >_<
Community might be Perl 6's USP, but I suspect it'll change as we grow larger and it doesn't stop people going "ROFLMAO LOOK! They made a new regex syntax that's not compatible with anything!!!!" 13:22
DrForr FWIW Inline::Perl5 needs more publicity, at least as a transition solution from perl5. 13:23
13:23 edwinsage left
unmatched} Awesome concurency too. But all of the awesome bits are currently NIY, so it's hard to sell people on them, and points again to too few volunteers implementing things. 13:23
tadzik yep, it's a huge marketing thing
mst NIY?
oh, not implemented yet 13:24
jnthn Um, what exactly is not implemented yet?
"Implemented but has a bunch of rough edges" we have in abundance.
13:24 edwinsage joined
unmatched} jnthn: all of the autothreading that's currently doesn';t autothread 13:24
mst unmatched}: yeah, this is why I'm slowly grinding towards being able to get at grammars in a way that makes me happy, and then I'm going to trick myself and other people into using perl6 via that
DrForr mst: So I need to finish up my ANTLR translator? :) 13:25
markk DrForr: that's an internal thing though and you talking about it at conferences works well
13:25 pduh_ left
unmatched} jnthn: at least I got that feel from Damian's concurrency talk. Every time he pointed out something awesome like hyper ops or feeds, he then went and said it's NIY 13:25
jnthn So there's a bunch of stuff that's both implemented and nice and he focused on a bunch of stuff that isn't? :P 13:26
13:26 bazzaar left, harmil_wk left, MasterDuke left
ugexe some feeds are implemented though 13:26
unmatched} ¯\_(ツ)_/¯
ugexe i still look forward to the double pointy feed 13:27
perlpilot ugexe: what would it do? 13:28
mst DrForr: I don't understand ANTLR at all, my favourite thing about grammars is that they actually sort of seem to make sense to me 13:29
13:31 jonas2 joined
ugexe perlpilot: allows multiple sources to be used as a single source (maybe worded poorly) 13:31
`@a1 ==>> @a2 ==> map { say $_ }` instead of `map { say $_ }, flat (@a1, @a2)` 13:33
Woodi ugexe: (@a1, @a2) ==> map {.... } would be nice 13:34
13:35 TheLemonMan joined
Woodi maybe even: my $f1 = Feed.new(@a0) ? 13:35
ugexe Woodi: but i'd prefer to have @a1 and @a2 on their own line (as well as the map) 13:36
unmatched} @a1 ⇒ @a2 ⇒ map { } would be nice :P
ugexe design.perl6.org/S03.html#Terminator_precedence shows an example of chaining ==>> and ==> in that way
Woodi I think it's similiar to Supplies but Supplies are like "ether" in Actor model without Actors, just hannels to comunicate... no subject to address 13:37
13:37 dbrock joined
Woodi ...so my brain have problems with that 13:38
13:39 TEttinger left 13:42 dbrock left, edehont left
gregf_ rust.... *wonders* 13:43
do error stacktraces use levenshtein distances in Perl6? 13:44
ugexe ==>> just iterates over each end individually but in sequence such that they are like a single sequence, so the next subsequent sink only ever sees a single sequence when it tries to pull 1 value at a time through it 13:45
which can already be done as shown, but the ==>>/==> allow the code to be formatted in a way i prefer 13:46
[Coke] gregf_: we use it for suggestion alternatives to missing names. Not sure if you mean something more specific.
Woodi anyway, last nightt unmatched} freeked ;) about Perl6 being easy to explode and it looks like pattern... eg. I once got mad about every loop iteration traversing LD_LIBRARY_PATH :) so, IMO, marketing can wait a bit
unmatched} Woodi++ I think I can agree. 13:47
gregf_ [Coke]: yep, i mean as a spell checker 13:48
unmatched} And we can use more people to do core stuff. Hint Hint. There's a course for it: github.com/edumentab/rakudo-and-nq...als-course
AlexDaniel I think that I agree too. The problem is not so much the marketing, but the compiler that is not bulletproof yet 13:49
13:54 dbrock joined, gregf_ left 13:55 prammer left
[Coke] While it's definitely a concern of mine that if we market to people and they find bugs; I don't know enough about advertising to say that we shouldn't do ANY advertising yet. 13:55
13:56 gregf_ joined
[Coke] That's why we need people who aren't developers or users to weigh in. :) 13:56
ugexe my perl6 marketing lately has been "perl6 will be a great fit for this project in 2 years". but that also involves the state of the ecosystem
awwaiid unmatched}: nice link! is it up to date enough?
unmatched} awwaiid: I'm half-way through it and it looks to be uptodate enough, with the exception of mentioning Parrot 13:57
awwaiid I think if you want marketing you should go to non-perl conferences, meetups, contests, etc. I know TimToady is going to Abstractions.io for example, which is a fantastic example.
unmatched}: ok cool
Maybe advertising is also about setting good expectations? On the one hand there are bugs. On the other hand people can influence the implementation/language through their experience :) 13:58
13:59 pduh_ joined
AlexDaniel
.oO( “Reporting rakudo bugs is fun!”… I can definitely write a post or two on that topic, but not sure if this is going to help anything… )
14:00
14:01 bob777 left, dbrock left
perlpilot awwaiid++ (setting expectations) 14:01
TheLemonMan underpromise and overdeliver then :) 14:02
14:02 n1lp7r joined
gregf_ marketing depends on the project manager *runs* 14:02
perlpilot TheLemonMan: sure, that's just one aspect though.
14:03 pduh_ left
gregf_ or product owner *ducks again* 14:03
14:04 bob777 joined
pmurias the users experience github repo seemed like a good initiative 14:04
I don't know anything about marketing stuff, but reducing the amount of times new users get annoyed seems important 14:05
gregf_ yeah sure, the product should market iself :)
pmurias finally gets round to removing annoying stuff from modules.perl6.org 14:06
perlpilot gregf_: well volunteered! (for all those "job positions" you keep mentioning :-)
gregf_ i guess if your oracle/ms/apple, you wont need marketing .. you can always bully :)
14:07 prammer joined
gregf_ perlpilot: heh 14:07
RabidGravy boom!
Woodi btw. maybe "20 years to late?" would be nice presentation on computer [scripting] languages state :) 14:08
TheLemonMan 20? I thought it was just 16 years late 14:09
Woodi TheLemonMan: I mean, generaly, not only Perl6
RabidGravy yeah a language with these capabilities should have existed twenty years ago
though I'm not entirely sure it could have
14:10 pduh joined
perlpilot Perl 6 is not late at all. I don't know what you're talking about 14:10
Woodi perlpilot: just commenting on some inet commet ;)
14:11 pduh_ joined
unmatched} perlpilot: it was in reference to my quoting a comment on one of my P6 articles: www.reddit.com/r/programming/comme...ss/d654ckz 14:11
perlpilot tends to have an auto-ignore brain filter for such comments 14:12
mst right, perl6 as the name of the language continues to screw both languages over, just like it has been doing for over a decade
but larry gets to call what he wants perl, so shrug
pmurias unmatched}: how do I install all the deps for modules.perl6.org using cpanm?
perlpilot pmurias: took me a second to understand what you were asking ... I was thinking "What?!? We can install Perl 6 modules with cpanm now?!?" at first :-) 14:14
gregf_ cpanm++
pmurias unmatched}: I figured it out 14:15
unmatched} pmurias: gist.github.com/zoffixznet/4e578ed...64a6892956
pmurias: no idea how to make it automatically use the Build.pm stuff 14:16
*Build.PL
leedo cpanm --installdeps . works with Makefile.pl, not sure about Build.PL
14:16 cdg_ joined 14:17 bob777 left
unmatched} pmurias: what annoying bits were you going to remove? 14:17
pmurias koalatee 14:18
unmatched} There's a Bit Rot branch github.com/perl6/modules.perl6.org/tree/bit-rot that I've no finished to address github.com/perl6/modules.perl6.org/issues/60
Ah... k :)
14:20 mcmillhj left, cdg left 14:21 cdg_ left 14:23 mcmillhj joined, CIAvash left
pmurias unmatched}: installing Mew fails: paste.debian.net/788300/ 14:23
mst pmurias: did you upgrade Test::Exception like the error message tells you you need to? 14:25
looks like Test::Most needs a bumped release with a higher dep, and then Mew needs to dep the new Test::Most 14:26
grondilu m: Blob.new but role {} # can't mixin a Blob?
camelia rakudo-moar 387806: OUTPUT«This REPR cannot change type␤ in any mixin at gen/moar/m-Metamodel.nqp line 1318␤ in block <unit> at <tmp> line 1␤␤»
grondilu m: Buf.new but role {} 14:27
camelia rakudo-moar 387806: OUTPUT«This REPR cannot change type␤ in any mixin at gen/moar/m-Metamodel.nqp line 1318␤ in block <unit> at <tmp> line 1␤␤»
unmatched} Would it really matter? I'm not testing exceptions on that line.
FWIW, the tests pass on 5.22 perl :/
mst it's unlikely to be anything to do with 5.22 14:28
and everything to do with what versions of things
jnthn grondilu: No
pmurias mst: I upgraded Test::Exception and it still fails 14:29
14:29 pduh_ left
mst hrmf 14:29
lemme see what happens 14:30
sigh. works here as well 14:31
ah, interesting, the chained method isn't getting chained
unmatched} FWIW, I'm not using any chained stuff on modules.perl6.org code 14:32
mst unmatched}: upgrade MooX::ChainedAttributes from 0.04 to 0.05
I think that change broke your shit 14:33
... nope
hmmm
mst downgrades Moo
yep, blammo
pmurias: upgrade Moo 14:34
Woodi mst: re: "perl6 as the name of the language continues to screw both languages" I think it is 90% not true. in apache 1.* vs 2.* it would be 100% not true, yes. but RFCs count is proof that things was/are FUBARed. perl5 is native /usr/bin citizen just like /bin/sed and awk, it's classic you can use when you need
mst unmatched}: 2.000002 (now backpan only) doesn't appear to work with the latest Moo
Woodi: I think that I've spend ten years listening to trolls use it to attack us both and to perl5 and perl6 people get upset about it
Woodi: the evidence says that it has not been good for either of us 14:35
14:35 mcmillhj left
mst you're welcome to disbelieve in reality, but reality won't care 14:35
pmurias the Perl 6 vs Perl 5 distinction is something I have to explain to people everytime I try to explain what I'm working on 14:36
unmatched} pmurias: did upgrading Moo help? 14:37
Woodi mst: pls read again. it's not same like with apache. perl5 was/is in pls-move-me-into-ex&vi-location
pmurias unmatched}: yep
unmatched} Awesome. mst++ :)
mst Woodi: please note that's nothing to do with anything I was saying. 14:38
Woodi: some of us regard perl5 as a viable language for new work, and enjoy writing it, and the existence of perl6 makes it much harder to justify it
and meanwhile, perl6 being a different language is used as a reason to ignore perl6
Woodi mst: and what you say ? you listen to trolls :)
mst ... 14:39
ah, yes, "let's insult the person making the point" as a strategy
clearly I am terribly wrong
TheLemonMan I picked up a random LTA ticket from the bug tracker and with some help from the kind guys over at #moarvm found out what triggers it, I'd love to send in a patch but I don't quite get what the correct behaviour is in this case
Woodi yes, you sayd that also, I just skipped
unmatched} TheLemonMan: what the patch is for? MoarVM NQP Rakudo? 14:40
mst you miss the point. OTHER PEOPLE listen to the trolls, and then don't use either perl.
if you're going to say "I don't want anybody who can't 100% reliably spot a troll even from the outside where the names are completely confusing to become a user", you're ruling out lots of the internet 14:41
Woodi mst: in late 90s I was surprised how ppls hate perl5. and now it is better
mst it is somewhat better. but I get very tired of explaining that "no, perl5 isn't obsolete, perl6 is a separate language" 14:42
14:42 Wiertek joined, ChoHag_ is now known as ChoHag
TheLemonMan it is the ticket #127767, the problem lies in the fact that next-interesting-index (trough AT-POS) in Backtrace.pm ends up skipping the infix:<!> frame because $settings is undef and the infix operator is actually defined in a CORE.setting (thus making is-setting return true and making the loop in next-interesting-index discard that frame when showing the backtrace) 14:43
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=127767
grondilu jnthn: why not?
TheLemonMan unmatched}, Rakudo
unmatched} TheLemonMan: you'd submit a PR against this repo: github.com/rakudo/rakudo/ do you know how do to PRs? 14:44
ugexe he wants to know what the PR should do
unmatched} Oh, sorry.
bah :) /me blushes 14:45
grondilu FYI I was looking at en.wikipedia.org/wiki/LEB128, and was considering implementing it with something like role LEB128 { method UInt(blob8:) {...}; method blob8(UInt:) {...} }
but I stumbled upon the fact that I can't mixin blobs, apparently.
TheLemonMan yep, I do know how to submit a PR :) 14:46
14:46 Nex6 joined
unmatched} :) 14:46
TheLemonMan sorry for the delay but vim is struggling with those big p6 files :|
14:46 Nex6 left 14:47 Nex6 joined
unmatched} s: &infix<!> 14:47
SourceBaby unmatched}, Something's wrong: ␤ERR: ===SORRY!=== Error while compiling -e␤Undeclared routine:␤ infix used at line 6. Did you mean 'indir', 'index'?␤␤
unmatched} s: &infix:<!>
SourceBaby unmatched}, Something's wrong: ␤ERR: ===SORRY!=== Error while compiling -e␤Undeclared routine:␤ infix:<!> used at line 6. Did you mean 'infix:<∖>', 'infix:<o>', 'infix:<⊎>', 'infix:<∩>', 'infix:<∈>'?␤␤
unmatched} feh
jnthn prefix ;) 14:49
unmatched} s: &prefix:<!>
SourceBaby unmatched}, Sauce is at github.com/rakudo/rakudo/blob/3878...ool.pm#L74
unmatched} \o/
jnthn grondilu: Because it uses a primitive representation rather than a P6opaque object representation
pmurias TheLemonMan: I sometimes find that turning of syntax highlight in vim helps when it's having trouble with too big files 14:50
14:50 tushar joined, tushar left
timotimo i don't understand why perltricks tweeted about the weekly adding "via @loltimo" at the end :\ 14:50
RabidGravy I saw that too 14:51
TheLemonMan pmurias, after a quick look at the syntax file that's used for p6 I guess I'll go down that route heh 14:54
14:57 Nex6 left
heatsink TheLemonMan: curious, just how big is the file your working with? 14:57
14:57 Nex6 joined
TheLemonMan heatsink, ~76K 14:58
14:58 Nex6 left
unmatched} :o 14:58
What is it? 14:59
14:59 Nex6 joined
timotimo perhaps the core setting 14:59
14:59 nbg joined
TheLemonMan unmatched}, src/core/Exception.pm 14:59
unmatched} It's 2578 lines....
github.com/rakudo/rakudo/blob/nom/...ception.pm 15:00
And CORE-setting is ~40K, but... do we edit it directly?
TheLemonMan oh, I was talking about the size in bytes
unmatched} Ah
RabidGravy Exception.pm opens really quickly for me in vim 15:01
15:01 pierre_ left
unmatched} I need some coffee. Can't think straight today :) 15:01
heatsink coffee++, but we drink more tea here, tea++
dalek href="https://modules.perl6.org/bit-rot:">modules.perl6.org/bit-rot: 38a0b82 | (Zoffix Znet)++ | Build.PL:
Bump Mew version

To avoid test failures it has with older Moo
15:02
[Coke] unmatched}++ is correct, coffee is the correct beverage. :| 15:03
[Coke] puts down his Coke Zero and goes to get a coffee.
ugexe time to graduate to amphetamines 15:04
RabidGravy it's the only way
moritz or, like, sleep
totally web 1.0, but works quite well
15:04 perlpilot left 15:05 perlpilot joined
TheLemonMan RabidGravy, same here, but it's excruciatingly slow when scrolling trough it 15:05
RabidGravy that's fine here too
dalek c: f8691e9 | (Christopher Bottoms)++ | doc/Type/Mu.pod6:
More clear, plus link to additional information
15:06
TheLemonMan that's weird, it's still slow even without loading my vimrc
15:06 perlpilot left
RabidGravy small memory or slow disk drice 15:07
drice? 15:08
drive
unmatched} arnsholt: weren't it you who were saying CORE-setting loads fine in vim when I made that comment about editors grinding to half in my article? Seems like it has trouble opening even a 3K-line file...
15:08 perlpilot joined
TheLemonMan there's plenty of ram and a fast ssd, I guess it's just vim that's playing tricks on me 15:09
15:09 MilkmanDan left
heatsink 76K isn't that big, is it that syntax highlighting to parsing the file multiple times i.e once per rule,or something 15:09
15:12 khw joined
literal TheLemonMan: are you using the latest perl6.vim or one that was bundled with Vim? there's a difference in performance 15:14
TheLemonMan literal, I actually get the same behaviour with both 15:18
15:18 telex left 15:19 prammer left
pmurias TheLemonMan: with syntax highlighting disabled vim shouldn't have any problems with editing the whole setting 15:19
15:20 telex joined
TheLemonMan indeed, I was just wondering what the root cause of the slowness was 15:20
15:20 harmil_wk joined, mohae joined 15:22 Ariantes_ joined
literal is DBIish (sqlite) not usable from multiple threads? I sometimes get the following output: gist.github.com/hinrik/a301b35d758...83e849c566 15:27
other times it runs without errors
15:28 prammer joined
RabidGravy the sqlite library itself may or may not be thread safe depending on how it was compiled IIRC (and/or some runtime configuration) 15:28
www.sqlite.org/threadsafe.html 15:29
Ariantes_ Hello 15:30
unmatched} \o\ 15:31
15:32 mcmillhj joined 15:34 n1lp7r left 15:39 itaipu left
AlexDaniel .u ⪥ 15:40
yoleaux U+2AA5 GREATER-THAN BESIDE LESS-THAN [Sm] (⪥)
pmurias unmatched}: I didn't specify --no-app-start and the modules.perl6.org didn't start 15:41
literal RabidGravy: from the error it looks more like the DBIish might not be accounting for the case when multiple threads try to connect before the DB driver has been initialized 15:43
15:44 mcmillhj left
literal another fun thing I see is that it occasionally just dies with "Segmentation fault" 15:45
15:45 itaipu joined
unmatched} pmurias: where is the --no-app-start from? Looks like the docs need an update. The not-starting-an-app is the default now and you need an explicit --restart-app option 15:46
pmurias unmatched}: from the README, and the DEPLOYMENT docs
unmatched} pmurias: well, "need" is not exactly accurate. You can run bin/morbo to start the dev server with the site
k, I'll be sure to go over the docs as part of the Bit Rot work. 15:47
16:02 ARM9 joined 16:04 richi235 left
pmurias unmatched}: it runs on my machine, I'll hopefully send PRs to remove koalitee and align the icons tommorow 16:05
dalek pan style="color: #395be5">perl6-examples: f6cf069 | (Aaron Sherman)++ | categories/parsers/SimpleStrings.pm:
Added a simple quoted string parser

  (which I previously posted to reddit)
pan style="color: #395be5">perl6-examples: 5ba1010 | (Aaron Sherman)++ | / (2 files):
Moved tests out to t/...
pan style="color: #395be5">perl6-examples: d6e83d6 | (Steve Mynott)++ | / (2 files):
Merge pull request #45 from ajs/master

A quoted string parser example
16:05 pmurias left, dakkar left 16:06 richi235 joined, MilkmanDan joined
harmil_wk Are examples merges automated, or was that a human being being super-responsive? 16:10
timotimo human
AlexDaniel haha
harmil_wk is impressed
dj_goku "human" 16:11
[Coke] docs folks ok with my 'xtest' fix?
unmatched} stmuk_++ 16:12
dalek c: 0c9f8ef | coke++ | Makefile:
run xt tests first

If you're running xtest, you're more likely to find issues with the xt/ tests, run them first so you can abort the test run and fix them.
16:15
c: 0b7fb97 | coke++ | doc/ (3 files):
whitespace fixes
16:16 bob777 joined 16:17 kaare__ joined 16:18 kaare_ left 16:19 kaare__ is now known as kaare_
AlexDaniel [Coke]: maybe! :) 16:20
16:23 prammer left, travis-ci joined
travis-ci Doc build errored. Will "Coke" Coleda 'whitespace fixes' 16:24
travis-ci.org/perl6/doc/builds/151557167 github.com/perl6/doc/compare/f8691...7fb9751684
16:24 travis-ci left
jdv79 I know very little about it but would it be possible to sub in a more erlang-ish async subsystem? 16:24
16:24 prammer joined
jdv79 i imagine it would involved a special scheduler and remote support and probably a few new/altered primitives 16:24
sena_kun m: fail; CATCH { default { say .WHAT } }; # How to generate Failure? 16:26
camelia rakudo-moar ef04e1: OUTPUT«(AdHoc)␤»
hoelzro sena_kun: Failures in sink context throw their underlying exception
m: my $f = fail 'fail!'; say $f.WHAT 16:27
camelia rakudo-moar ef04e1: OUTPUT«fail!␤ in any at /home/camelia/rakudo-m-inst-2/share/perl6/runtime/CORE.setting.moarvm line 1␤ in block <unit> at <tmp> line 1␤␤»
sena_kun hoelzro, yes, I already read that line, but. Ah, thanks!
hoelzro .oO( does Failure.WHAT trigger the throw too? )
unmatched} m: fail.^name.say 16:28
camelia rakudo-moar ef04e1: OUTPUT«Failed␤ in any at /home/camelia/rakudo-m-inst-2/share/perl6/runtime/CORE.setting.moarvm line 1␤ in block <unit> at <tmp> line 1␤␤»
unmatched} bisect: m: fail.^name.say
bisectable unmatched}: Exit code is 1 on both starting points (good=2015.12 bad=ef04e1e), bisecting by using the output
unmatched}: bisect log: gist.github.com/ebbbd830e1c0d115a0...c4df94c746
unmatched}: 'bisect run' failure
sena_kun m: my $f = fail 'fail!';
camelia rakudo-moar ef04e1: OUTPUT«fail!␤ in any at /home/camelia/rakudo-m-inst-2/share/perl6/runtime/CORE.setting.moarvm line 1␤ in block <unit> at <tmp> line 1␤␤»
unmatched} m: sub f { fail }; f.^name.say
camelia rakudo-moar ef04e1: OUTPUT«Failure␤»
unmatched} fail is like return innit? so it returns before you get to .WHAT so you need to return it from somewhere 16:29
hoelzro unmatched}: oh, good point, thanks!
hoelzro just realized that bisectable is like wiki.documentfoundation.org/QA/Bibisect
sena_kun m: sub f{ fail }; f.handled;
camelia ( no output )
sena_kun m: sub f{ fail }; say f.handled;
camelia rakudo-moar ef04e1: OUTPUT«False␤»
unmatched} m: sub f{ fail }; my $v = f; $v.handled.say; $v.Bool; $v.handled.say; 16:30
camelia rakudo-moar ef04e1: OUTPUT«False␤True␤»
16:31 titsuki joined
sena_kun Great. Thanks! 16:31
16:32 TimToady left 16:34 TimToady joined 16:36 bob777 left, bob777 joined 16:37 andrzejku joined 16:40 n1lp7r joined, mcmillhj joined
titsuki m: for ^3 { .map(-> $e { say $e + 3 }) }; # map called in the sink context (Sorry it's just a experiment) 16:45
camelia ( no output )
16:45 dogbert17 joined
titsuki m: for ^3 { 10; }; # check context (Sorry it's just a experiment) 16:46
camelia rakudo-moar ef04e1: OUTPUT«WARNINGS for <tmp>:␤Useless use of constant integer 10 in sink context (line 1)␤»
16:46 mcmillhj left 16:47 mcmillhj joined
sena_kun m: (^3) .map({ say $_ + 3 }); 16:47
camelia rakudo-moar ef04e1: OUTPUT«3␤4␤5␤»
dogbert17 o/ #perl6
sena_kun dogbert17, \o
dogbert17 hi sena_kun
sena_kun Map is already cycle in some sense. 16:48
b2gills Seems to me that Perl 6 is 20 years too early
ARM9 or 15 years too late 16:49
dogbert17 The paragraph about 'Versioning_and_Authorship' is duplicated in docs.perl6.org/language/typesystem. Should I clean it up?
ARM9 what's the koalatee % thing on modules.perl6.org ? 16:50
16:50 bazzaar joined
sena_kun dogbert17, yeah, it will be good. 16:50
bazzaar \o perl6
dogbert17 sena_kun, fix is on its way :)
sena_kun dogbert17++
titsuki m: for ^3 { sink .map(-> $e { say $e + 3 }) }; # map called in the sink context (Sorry it's just a experiment) 16:51
camelia rakudo-moar ef04e1: OUTPUT«3␤4␤5␤»
bazzaar Hi, re. earlier discussion on Perl6 introduction media content, I put down a few thoughts nopaste.linux-dev.org/?1119698 16:52
andrzejku hi 16:53
sena_kun titsuki, you can write messages to camelia in private, it'll answer you, if you want just osme local experiments.
titsuki sena_kun: I understand that, but I couldn't understand behavior of the map exactly.
dalek c: ff87518 | (Jan-Olof Hendig)++ | doc/Language/typesystem.pod6:
Removed duplicated paragraph about Versioning and Authorship
16:53 nicq20 joined
sena_kun titsuki, I'm not sure about the sinking now, but generally, map just applies function to every element of positional. 16:54
afk&
16:54 j75 joined
unmatched} titsuki: note that that area currently has some bugs, so you might see some weirdness. 16:57
AlexDaniel hoelzro: “A bibisect can take as little as 15 minutes” 15 minutes? Why is it so long?
titsuki sena_kun: Sorry, I'll use the cameria's private message feature from now on.
unmatched} m: for ^3 { .map: *.say };
camelia ( no output )
unmatched} m: for ^3 { ^3 .map: *.say };
camelia rakudo-moar ef04e1: OUTPUT«0␤1␤2␤0␤1␤2␤0␤1␤2␤»
hoelzro AlexDaniel: good question! 16:58
maybe longer history? or maybe they build OO for each bibisect =P
gregf_ m: say (^3).map:{ ->$x { $x + 3 }.($_)}
camelia rakudo-moar ef04e1: OUTPUT«(3 4 5)␤»
unmatched} m: say ^3 .map: *+3 16:59
camelia rakudo-moar ef04e1: OUTPUT«(3 4 5)␤»
unmatched} Death to parenths!
gregf_ m: sub add($v){ $v+ 3; }; say (^3).map(&add)
camelia rakudo-moar ef04e1: OUTPUT«(3 4 5)␤»
AlexDaniel MasterDuke: hey, what if we put all rakudo builds into one git repo? 17:00
is it going to save some space? Is it going to be slower?
unmatched} m: my &add = * + 3; say ^3 .map: &add 17:02
camelia rakudo-moar ef04e1: OUTPUT«(3 4 5)␤»
unmatched} Death to parenths! :D
AlexDaniel unmatched}: by the way, the thing you tried to bisect there seems to be another effect of return-without-lexotic 17:03
unmatched}: see also #128872
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=128872
unmatched} AlexDaniel: what thing? 17:04
dalek c: 40edfc8 | (Jan-Olof Hendig)++ | doc/Language/typesystem.pod6:
Revert "Removed duplicated paragraph about Versioning and Authorship"

This reverts commit ff87518be5fe7bb9d1fe8adfe8dc0a40e3bc7c88.
17:05
c: fd3b8c1 | (Jan-Olof Hendig)++ | doc/Language/typesystem.pod6:
Fix typos
AlexDaniel bisect: fail.^name.say
bisectable AlexDaniel: Exit code is 1 on both starting points (good=2015.12 bad=ef04e1e), bisecting by using the output
AlexDaniel: bisect log: gist.github.com/97de31fc5372219ada...29cbbf5afc
AlexDaniel: 'bisect run' failure
AlexDaniel unmatched}: this one ↑
unmatched} That works correctly. I bisected it before I realized that it works correctly :)
AlexDaniel oh, hm
17:05 cdg joined
unmatched} unmatched} │ fail is like return innit? so it returns before you get to .WHAT so you need to return it from somewhere 17:05
AlexDaniel ok :) 17:06
17:06 zakharyas left 17:07 nicq20 left 17:12 brrt joined, brrt left 17:13 firstdayonthejob joined 17:19 travis-ci joined
travis-ci Doc build passed. Jan-Olof Hendig 'Removed duplicated paragraph about Versioning and Authorship' 17:19
travis-ci.org/perl6/doc/builds/151567240 github.com/perl6/doc/compare/0b7fb...87518be5fe
17:19 travis-ci left 17:22 setty1 joined 17:24 travis-ci joined
travis-ci Doc build passed. Jan-Olof Hendig 'Fix typos' 17:24
travis-ci.org/perl6/doc/builds/151570438 github.com/perl6/doc/compare/ff875...3b8c103604
17:24 travis-ci left 17:27 prammer left 17:28 aries_liuxueyang left 17:29 prammer joined 17:33 ptolemarch left, ptolemarch joined 17:34 Actualeyes left
harmil_wk Was github:perl6/rakudo a thing before a couple seconds ago? I fear I just did something I didn't mean to... 17:34
unmatched} hahaha 17:35
geekosaur says forked from rakudo/rakudo
"whoops"
unmatched} harmil_wk: if you go to settings you can delete that fork
harmil_wk I just want to be REALLY sure that that's what I should be doing. Is there someone here who feels confident that that repo did not and should not exist? 17:36
unmatched} harmil_wk: yes, I'm confident. 17:37
harmil_wk k, thanks
Just re-stating because scary: perma deleting "perl6/rakudo"
geekosaur harmil_wk, you can tell stuff like this by looking through the metadata on the web interface 17:38
ovibos if you're *really* scared, you could make a local copy of perl6/rakudo
geekosaur which is how we can be sure it was not there to begin with (in any case you'd likely have gotten LOTS of conflicts if it had)
17:39 mcmillhj left, mcmillhj joined
ovibos is there a way to mattern-match assign from a hash? 17:40
harmil_wk Done
Sorry about that
ovibos like, my { status => $status, content => $content } = ...
17:40 mcmillhj left, mcmillhj joined
unmatched} ovibos: what would be on the right side? 17:40
ovibos well, in my case it's a function that returns a hash 17:41
something like { status => 'ok', content => 'content' } though
jnthn m: my %h = status => 'foo', content => 'bar'; my (:$status, :$content) := %h; say $status; say $content;
camelia rakudo-moar c587b9: OUTPUT«foo␤bar␤»
unmatched} m: my ($status, $content) = funct<status content>;
camelia rakudo-moar c587b9: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤ funct used at line 1␤␤»
ovibos oh, thanks
unmatched} jnthn++ 17:42
jnthn Perl 6 destructuring in general using signatures syntax
*uses
17:43 sno left 17:44 aries_liuxueyang joined
[Coke] PERL SIX IS SELF DESTRUCTING!? 17:44
jnthn Oh man...this is just asking for... 17:45
m: class A { has $.x; has $.y; method m((:$x, :$y):) { say $x; say $y } }; A.new(x => 42, y => 100).m # PERL 6 IS self DESTRUCTURING! 17:47
camelia rakudo-moar c587b9: OUTPUT«42␤100␤»
unmatched} lol
AlexDaniel 🦋🔫
geekosaur is that better or worse than self distracting? 17:50
17:52 jonas2 left
AlexDaniel m: sub term:<🦋🔫> { 42 xx (2 ** 62) }; say ‘Hello world!’; 🦋🔫 17:52
camelia rakudo-moar c587b9: OUTPUT«(signal SEGV)Hello world!␤»
17:53 bob777 left
sena_kun dogbert17, I'm a bit stupid here, but why did you revert your commit? 17:53
Ah.
unmatched} m: "🦋".uninames.say
camelia rakudo-moar c587b9: OUTPUT«(<reserved>)␤»
sena_kun It was my mistake. Sorry for the wrong advice.
unmatched} m: say 42 xx (2 ** 62) 17:54
camelia rakudo-moar c587b9: OUTPUT«(signal SEGV)»
17:54 mcmillhj left, bazzaar left 17:55 cpage_ left
AlexDaniel unmatched}: there's a ticket for this segfault but I'm not sure if there is a ticket for Unicode 9.0 support 17:55
#126014
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=126014
unmatched} m: say "🦋".ord.base: 16
camelia rakudo-moar c587b9: OUTPUT«1F98B␤»
unmatched} oh hey! that's my adopted character! 17:56
17:57 mcmillhj joined, n1lp7r left 18:00 Sgeo__ joined
AlexDaniel perl6 say (:<16>1F98C..:16<1F98F>)».chr 18:00
m: say (:<16>1F98C..:16<1F98F>)».chr
camelia rakudo-moar c587b9: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Two terms in a row␤at <tmp>:1␤------> 3say (:<16>7⏏051F98C..:16<1F98F>)».chr␤ expecting any of:␤ infix␤ infix stopper␤ statement end␤ statement modifier␤…»
AlexDaniel m: say (:16<1F98C>..:16<1F98F>)».chr
camelia rakudo-moar c587b9: OUTPUT«(🦌 🦍 🦎 🦏)␤»
TimToady m: say "\x1f98c" ... "\x1f98f" 18:01
camelia rakudo-moar c587b9: OUTPUT«(🦌 🦍 🦎 🦏)␤»
AlexDaniel right 18:02
TimToady after backlogging, I hope everyone is done being quite so lugubrious for another week or two... 18:03
18:04 spider-mario joined, mvorg left
moritz has upcoming vacations, and is quite happy 18:05
unmatched} TimToady: why a week? :) 18:06
TimToady then I'll be on the road again, with an excuse for not backlogging :)
unmatched} heh
TimToady but I do think the pessimism was a bit overdone overnight 18:07
edwinsage Is there a convenient way to stop a line from running in interactive mode without killing the interpreter itself?
^C kills perl6. 18:08
[Coke] nope
TimToady we're aiming for a natural growth curve, neither too fast nor too slow, and so far this year I think we're pretty close to a decent growth curve
ovibos shouldn't ^D kill perl6? 18:09
TimToady sure, there's lotsa stuff left to do, and we knew this year would be bugfixes and speedups mostly
moritz edwinsage: you can catch SIGINT
[Coke] ovibos: not kill. tell it you're done
geekosaur edwinsage, the interactive mode is not very good at all currently
it
[Coke] geekosaur: that's a lie.
ovibos i mean, yes, but the behavior is the same [Coke] 18:10
moritz edwinsage: docs.perl6.org/type/Supply#sub_signal
[Coke] no, ^D doesn't behave the same as ^C.
moritz not sure how well it works in the REPL
nine can't get enough bugfixes and speedups
unmatched} TimToady: fair enough :)
edwinsage Ok, I'm just running stuff like "(1..*).eager" to fiddle around.
TimToady unmatched}: it wasn't just you :)
ovibos [Coke]: what's the difference?
geekosaur [Coke], we have problems ranging from this one to scoping problems with previously defined names. this is not what people expect of a decent REPL 18:11
TimToady goes to look at the bug he promised two
perlpilot TimToady++ (for the optimism injection :)
unmatched} ovibos: ^D sends EOF. If you're not in the process of reading from a file, it won't care if you sent it and being in an infinite loop is one of those cases 18:12
ovibos oh, you're right 18:13
still, i think it would be saner if ^C behaved like it did in other REPLs 18:14
perlpilot ovibos: Which other REPLs? Should it say "KeyboardInterrupt" like Python? Should it do nothing like Ruby? Or ... should it exit the REPL like lua? 18:16
18:17 nbg left
ovibos perlpilot: it does nothing in ruby and zsh, it prints some sort of error and returns to prompt in python and racket 18:17
which were what i was basing my "other REPLs" on
i don't think one is better than the other, but, i think they're both better than exiting the REPL entirely 18:18
moritz exiting the REPL is exactly what I want
ctrl+c = exit that program, unless you have a *very* good reason not to
unmatched} ^C -> abort the running line; ^C^C -> exit 18:19
ovibos i'm mostly used to ^C's behavior because of zsh/bash's handling of it (as that's the shell i'm in the most)
perlpilot unmatched}: that would work for me.
unmatched} (and so it'd naturally end up being the current ^C -> exit if no line is running)
lizmat fwiw, I think it should be pretty trivial to catch SIGINT in a signal handler, then die("SIGINTED") in there, and then let the CATCH handler cleanup 18:21
yoleaux 10 Aug 2016 22:24Z <japhb> lizmat: Thanks for merging (and fixing) the perl6-bench fix!
lizmat especially since the REPL is now mostly in Perl 6
afk& # some celebrating
18:24 itaipu left 18:27 sjoshi joined 18:32 sjoshi left
unmatched} Rakudo bugged as LHF: rt.perl.org/Ticket/Display.html?id=128900 18:35
(the ^C in REPL)
18:37 dbrock joined 18:42 labster joined 18:43 cpage_ joined, nbg joined 18:44 prammer left 18:46 prammer joined
hoelzro I think we need support from the line editor for that 18:46
possibly relevant: github.com/hoelzro/p6-linenoise/issues/12 18:47
dalek c: c646e98 | (Jan-Olof Hendig)++ | doc/Type/Bool.pod6:
Added docs for Bool.pick
19:03
19:03 z8 joined
arnsholt unmatched}: Yeah, that was me. Works fine on my machine, but I guess it'll depend on the details of the hardware and config we're running 19:06
19:06 z8 left 19:16 dbrock left 19:30 mr-foobar left 19:31 darutoko left 19:32 zacts joined 19:36 edwinsage left 19:37 heatsink left
grondilu just red the git log. You guys. 19:41
"IterateOneNotSlippingWithoutPhasers"
^srsly?
unmatched} grondilu: what about it? 19:43
Saying "srsly" isn't overly helpful with the problem we're trying to solve :) 19:44
jdv79 thank god for autocomplete
unmatched} It's a private Rakudo-internal class.
gfldex move along! nothing to see here! 19:45
dalek c: a57a821 | (Jan-Olof Hendig)++ | doc/Type/Bool.pod6:
Added docs for Bool.roll
19:46
19:48 pmurias joined
unmatched} m: my \term:<🔂✗⛸✗🔫> = Seq; say 🔂✗⛸✗🔫.new(1, 2, 3); 19:50
camelia rakudo-moar c587b9: OUTPUT«Too many positionals passed; expected 2 arguments but got 4␤ in block <unit> at <tmp> line 1␤␤»
unmatched} m: my \term:<🔂✗⛸✗🔫> = Seq; say 🔂✗⛸✗🔫.new([1, 2, 3]);
camelia rakudo-moar c587b9: OUTPUT«Type check failed in binding to $iter; expected Iterator but got Array ($[1, 2, 3])␤ in block <unit> at <tmp> line 1␤␤»
unmatched} grondilu: how about 🔂✗⛸✗🔫 ? :D
pmurias unmatched}: is it expected that some tests fail for modules.perl6.org? 19:53
[Coke] geekosaur: please make sure that there are tickets open for any outstanding issues. 19:54
(danke)
19:56 mcmillhj left 19:59 wamba left 20:02 cpage_ left 20:03 TEttinger joined 20:04 mcmillhj joined, AlexDaniel left 20:05 kaare_ left
unmatched} pmurias, they were all passing when I was done with the rewrite last December, but I don't know if people added code without running the tests. I know the GitLab DistSource was added without tests covering. They may have also bit rotted due to major Mojolicious Donut release. Lastly, they do check for koality, so if you removed it without adjusting the tests, then they will fail 20:06
20:12 mcmillhj left, sno joined, heatsink joined 20:16 zacts left 20:18 cpage_ joined 20:20 mcmillhj joined 20:22 andrzejku left 20:24 labster left 20:25 mcmillhj left
dalek href="https://modules.perl6.org:">modules.perl6.org: c99d321 | (Pawel Murias)++ | Build.PL:
perl Build test now runs tests.
20:27
20:30 zakharyas joined
El_Che maybe a silly question. When using NativeCall, is it customary to write a Perl6 class wrapper to sanitize/convert the input/output to the NativeCall functions in the same distribution? 20:33
skids Personally I'd say yes, the only reason I'd split those up is if the bindings are very large and there may be more than one high level package using them. 20:36
20:38 harovali joined
skids Having the bindings in a separate compunit can be nice. Though, the fact that CStruct reprs can directly have methods may be a good reason not to. 20:38
I don't think we've firmly established a "customary" though. 20:39
El_Che skids: thx 20:40
pmurias unmatched}: a lot of the failures seems to be bitrot in tests fetching metadata 20:42
20:42 Wiertek left 20:44 labster joined
unmatched} pmurias: ok, I'll take a look as part of the Bit Rot issue fixing 20:47
20:50 n1lp7r joined
pmurias unmatched}: should I make a PR or will you revert if anything is too horrible? 20:51
20:53 cpage left 20:54 raiph left 20:55 mcmillhj joined 20:59 n1lp7r left, skids left, mcmillhj left
pmurias unmatched}: pushed my removal commits, when will modules.perl6.org be updated? 20:59
ahh, found it in the REAME 21:00
21:01 ptolemarch left, TheLemonMan left
gfldex m: my $s = '1a 1b 1c'; my @a = <a b c>.any; say @a ~~ /<@a>/ 21:02
camelia rakudo-moar c587b9: OUTPUT«This type (Junction) does not support positional operations␤ in block <unit> at <tmp> line 1␤␤»
gfldex m: my $s = '1a 1b 1c'; my @a = <a b c>; say @a ~~ /<@a>/
camelia rakudo-moar c587b9: OUTPUT«「a」␤»
21:03 cpage__ joined
gfldex m: my $s = '1a 1b 1c'; my @a = <a b c>; say $s ~~ m:g/'1'<@a>/ 21:03
camelia rakudo-moar c587b9: OUTPUT«(「1a」 「1b」 「1c」)␤»
21:04 mcmillhj joined
gfldex m: my $s = '1a 1b 1c'; my @a = <a b c>; say ($s ~~ m:g/'1'<@a>/).perl 21:04
camelia rakudo-moar c587b9: OUTPUT«(Match.new(ast => Any, list => (), hash => Map.new(()), orig => "1a 1b 1c", to => 2, from => 0), Match.new(ast => Any, list => (), hash => Map.new(()), orig => "1a 1b 1c", to => 5, from => 3), Match.new(ast => Any, list => (), hash => Map.new(()), orig => …»
21:04 cpage_ left, cpage__ is now known as cpage_ 21:06 mvorg joined
dalek osystem: 2372851 | (Zoffix Znet)++ | META.list:
s/META.info/META6.json/

For List-Combinations and Text::Diff::Sift4
21:06
21:06 zakharyas left 21:07 nbg left
gfldex m: my $s = '1a 1b 1c 1d'; my @a = <a b c>; say $s ~~ m:g/'1'<@a>/ 21:07
camelia rakudo-moar c587b9: OUTPUT«(「1a」 「1b」 「1c」)␤»
unmatched} pmurias: probably in an hour. Last job just finished ( modules.perl6.org/update.log ), next one will start at 5:20EST and will probably take like 40 minutes
gfldex m: my $s = '1a 1b 1c'; my @a = <a b c d>; say $s ~~ m:g/'1'<@a>/
camelia rakudo-moar c587b9: OUTPUT«(「1a」 「1b」 「1c」)␤»
gfldex is there an easy way to get the element of @a that didn't match? 21:08
unmatched} pmurias: I don't see you pushing anything tho
other than 1 small change in Build.PL
unmatched} for rest of the night 21:11
I meant & :)
21:13 nbg joined
dalek href="https://modules.perl6.org:">modules.perl6.org: 8467077 | (Pawel Murias)++ | / (15 files):
Remove Koalatee.
21:13
21:18 n1lp7r joined 21:19 LegalResale left 21:23 LegalResale joined 21:26 cpage_ left 21:29 n1lp7r left 21:43 cognominal joined
sena_kun m: my &a = {}; 21:48
camelia rakudo-moar c587b9: OUTPUT«Type check failed in assignment to &a; expected Callable but got Hash (${})␤ in block <unit> at <tmp> line 1␤␤»
sena_kun Does this syntax worked? 21:49
Nevermind. 21:50
geekosaur m: my &a = {;} 21:51
camelia ( no output )
geekosaur perl 5 has the same ambiguity of needing to figure out if {} is a hash(ref) or a block
sena_kun geekosaur, ah, neat. 21:52
21:55 rpburkholder joined 21:57 kid511 joined, kid51 left 21:58 mcmillhj left
gfldex is there a way to list methods that are :local and not mixedin by a role? 21:59
m: say Date.^methods(:local)[9].package ~~ Date 22:02
camelia rakudo-moar c587b9: OUTPUT«False␤»
gfldex :)
jnthn Not aware of a better way than filtering them by .package 22:03
22:04 prammer left
gfldex m: NQPRoutine.^name 22:04
camelia rakudo-moar c587b9: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared name:␤ NQPRoutine used at line 1␤␤»
22:05 chris2 left, n1lp7r joined 22:07 cpage_ joined 22:08 chris2 joined
gfldex m: say Cursor.^methods[11].package 22:10
camelia rakudo-moar c587b9: OUTPUT«Method 'package' not found for invocant of class 'NQPRoutine'␤ in block <unit> at <tmp> line 1␤␤»
22:11 firstdayonthejob left
gfldex jnthn: any way to do that for NQPRoutine (i am willing to descend into nqp)? 22:11
22:13 labster left
jdv79 anyone have any grammars working async? 22:17
concurrently i guess is the actual aspect.
22:19 skids joined
dalek c: e83de06 | Altai-man++ | doc/Type/ (9 files):
Make examples compile
22:24
22:28 cpage_ left
jnthn gfldex: Umm...I don't *think* so, though in the Cursor case all of them are certainly from a role :) 22:28
22:28 cpage_ joined, cpage_ left
jnthn jdv79: I'm pretty sure there's a spectest involving grammars being used from multiple threads. 22:28
Ran across it the other day while doing stability work. 22:29
22:29 cpage_ joined
jdv79 k 22:29
jnthn (Though it was actually another test in the file that was the problem child.)
I think there's an RT relating to grammars and concurrency recently filed.
And I know EVAL has trouble with reentrancy, which may or may not be grammar related. 22:30
Guess those'll be a day or so's debugging each...
jnthn is fairly tired of the concurrency bugs spoiling our fun and is trying to hunt them down 22:31
22:31 AlexDaniel joined
jdv79 i'm still clueless as to how "Use of uninitialized value $!made" is happening 22:31
jnthn o.O
jdv79 probably beyond my pay grade
jnthn That sounds pretty wrong.
Do you have actions embedded in the grammar, or in a separate actions class?
jdv79 i'm poking at it though 22:32
its seperate. its the XML dist. 22:33
jnthn OK, that's one hypothesis down then :)
jdv79 the cross thread write log didn't seem to illumniate anything startling to me 22:34
gonna try some other grammar using things
22:35 labster joined
jnthn tries to figure out the order modules are listed in on modules.perl6.org/ 22:35
timotimo it was alphabetically at one point 22:37
jdv79 oh, the grammar and conc rt is mine probably 22:38
22:39 cpage_ left 22:40 cpage_ joined 22:43 spider-mario left 22:45 pmurias left 22:53 cpage_ left 22:55 lambd0x left 22:58 rindolf left 23:08 mcmillhj joined 23:18 itaipu joined, mcmillhj left 23:19 setty1 left 23:20 pduh left 23:27 n1lp7r left 23:28 n1lp7r joined 23:29 RabidGravy left 23:30 jambun joined 23:33 n1lp7r left 23:35 mcmillhj joined
dalek c: b6f82b6 | (Wenzel P. P. Peppmeyer)++ | util/list-missing-methods.p6:
add util/list-missing-methods.p6
23:38
gfldex Hash append is missing what I found with ^^^ 23:43
dalek href="https://modules.perl6.org:">modules.perl6.org: fd90ee5 | (Zoffix Znet)++ | public/js/main.js:
Restore alphabetical module listing

Fixes fallout from koality removal
sena_kun m: my %a = one => 1, two => 2; for %a.pairs -> ($a, $b) {}; 23:47
camelia rakudo-moar c587b9: OUTPUT«Too few positionals passed; expected 2 arguments but got 0 in sub-signature␤ in block <unit> at <tmp> line 1␤␤»
timotimo you get pair objects from .pairs 23:48
they do not
destructure into a list
m: my %a = one => 1, two => 2; for %a.pairs -> (:$key, :$value) { }
camelia ( no output )
sena_kun Hmmm... 23:49
dalek href="https://modules.perl6.org/bit-rot:">modules.perl6.org/bit-rot: 469195d | (Zoffix Znet)++ | Build.PL:
List Data::UUID in prereqs

It's listed as prereq by Data::GUID, but for some reason did not get installed on a clean system, even though Data::GUID did install fine.
23:50
href="https://modules.perl6.org/bit-rot:">modules.perl6.org/bit-rot: 88742d4 | (Zoffix Znet)++ | Build.PL:
Add Time::Moment to prereqs
href="https://modules.perl6.org/bit-rot:">modules.perl6.org/bit-rot: f6f8f50 | (Zoffix Znet)++ | Build.PL:
Add Imager::File::PNG to prereqs

It may have been a prereq on ::AssetPack at some point, but is no longer
href="https://modules.perl6.org/bit-rot:">modules.perl6.org/bit-rot: 8fe1075 | (Zoffix Znet)++ | README.md:
Mention Edge in browser support
href="https://modules.perl6.org/bit-rot:">modules.perl6.org/bit-rot: cd7572b | (Zoffix Znet)++ | README.md:
s/browers/browsers/
href="https://modules.perl6.org/bit-rot:">modules.perl6.org/bit-rot: c99d321 | (Pawel Murias)++ | Build.PL:
perl Build test now runs tests.
href="https://modules.perl6.org/bit-rot:">modules.perl6.org/bit-rot: 8467077 | (Pawel Murias)++ | / (15 files):
Remove Koalatee.
href="https://modules.perl6.org/bit-rot:">modules.perl6.org/bit-rot: fd90ee5 | (Zoffix Znet)++ | public/js/main.js:
Restore alphabetical module listing

Fixes fallout from koality removal
href="https://modules.perl6.org/bit-rot:">modules.perl6.org/bit-rot: dbb8e76 | (Zoffix Znet)++ | / (18 files):
Merge branch 'master' into bit-rot
sena_kun Oh wow.
Thanks, timotimo. 23:51
23:51 pduh_ joined
unmatched} gives dalek a sock 23:52
23:52 pduh_ left 23:53 cpage_ joined
gfldex m: my %a = one => 1, two => 2; for %a.kv -> $a, $b { say $a, $b }; 23:56
camelia rakudo-moar c587b9: OUTPUT«two2␤one1␤»
gfldex sena_kun: you may want .kv instead of .pairs
timotimo m: my %a = :1one, :2two; for %a -> (:$key, :$value) { say $value } 23:57
camelia rakudo-moar c587b9: OUTPUT«2␤1␤»
timotimo ^- iterating over the hash directly behaves just like iterating over .pairs
sena_kun gfldex, I want(or, rather, author of now broken example wanted .ks). Oh, without brackets it should do the right thing.
*.kv
23:58 cpage_ left
timotimo no, without brackets you get 2 pairs per iteration 23:58
m: my %a = :1one, :2two; for %a -> $a, $b { say "a: $a"; say "b: $b" }
camelia rakudo-moar c587b9: OUTPUT«a: two 2␤b: one 1␤»
sena_kun The original line is "for %intervals.pairs.sort(* cmp *) -> (Instant $instant, $note) { ... }".
timotimo isn't cmp the default for sort? 23:59
dalek href="https://modules.perl6.org/bit-rot:">modules.perl6.org/bit-rot: 80b12cc | (Zoffix Znet)++ | / (2 files):
Fix Build.PL to make cpanm --installdeps -vn . work
sena_kun It's an old example. (: