»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_log/perl6 | UTF-8 is our friend! 🦋
Set by Zoffix on 25 July 2018.
00:03 Xliff joined 00:06 lichtkind_ joined 00:08 lichtkind left 00:11 yqt left
AlexDaniel somebody suggested “Pelr” xD 00:26
00:26 Sgeo_ left 00:27 Sgeo_ joined 00:29 reportable6 joined 00:30 irced joined
vrurg Oh my... (facepalm) 00:36
00:36 Actualeyes left
Grinnz ... ok, that's one of the more creative "I didn't read the discussion" suggestions :P 00:37
vrurg I wish to see the voting results and be over with it one way or another
AlexDaniel vrurg: it may take some time 00:43
I think we'll start this week, then it'd be at least two weeks from now 00:44
if there are requests for changes, then it'll take a bit more
vrurg AlexDaniel: I know. I'm patient. Just wanna something to happen because soon it'll start causing reputational harm. 00:46
AlexDaniel: BTW, back to the yesterday discussion. I now recall more details of what I spoke about with timotimo and it was mostly related to Configure and to the possibility of rewriting it in NQP. make.nqp can be used after moar is built. 00:49
AlexDaniel vrurg: well, there's this ticket now github.com/perl6/problem-solving/issues/100 00:50
which is kinda related
we'll see how it goes
vrurg But then another thing crossed my mind: so far, our Makefile generation convers perhaps 99% of all use cases by supporting BSD, GNU makes and NMAKE. There is just no point of using something non-standard.
discord6 <RaycatWhoDat> Grammars are doin' my head in. :< 00:51
tellable6 2019-09-09T06:10:22Z #perl6 <SmokeMachine> RaycatWhoDat if you wanna see a example of a simple toy language: github.com/FCO/bernaLang
discord6 <RaycatWhoDat> oh hey
vrurg Oh, BTW, you once took #3000 on rakudo, now it's #100 on problem-solving – do you hunt for these numbers? ;) :D 00:52
AlexDaniel hah
discord6 <RaycatWhoDat> How do you make a grammar parse recursively? 00:54
vrurg AlexDaniel: as to #100 – yep, I read it. There is nothing to add to what jnthn wrote. Besides, prventing stage0 nqp from being bundled into a pacakge is like preventing from bundling with images. Or, other way around, how do they distribute precompiled Java code? 00:56
stage0 nqp is both code and resourse, from this point. 00:57
AlexDaniel vrurg: there's so much wrong in your message :D 00:59
vrurg: you must be able to get your package from source code, what does this have to do with precompiled java code? 01:00
and how images are related to executable stuff?
I understand it's a hard problem, I understand it's pain for us 01:01
vrurg AlexDaniel: if one creates a package of an application acting as a web server too and implemented in java then the package would have to be built with some images and likely some java bytecode. Isn't it? 01:02
AlexDaniel we should just figure out if debian is OK with this, as soon as possible, and then we can skip the denial stage :) 01:03
vrurg: sure, and debian will only take source code and will build the app themselves
vrurg: users will get it in the compiled form, yes, and that's irrelevant 01:04
a package without source code won't make its way into debian
drivers with blobs might get into nonfree but that's different 01:05
vrurg: and images are fine, you're not executing images…
vrurg AlexDaniel: I'm not denying, but looking for a way to get it solved by, perhaps, finding workarounds. I think I was misunderstanding it. Nobody needs stage0 to be included into the package, but debian wouldn't compile NQP on their side because it's using a blob to be compiled. Right? 01:07
^ there is :) after 'workaround' ;) 01:09
Xliff RayCatWhoDat: What do you mean? 01:10
AlexDaniel vrurg: well, that's the question. I'm not sure, which is why I asked robertle to clarify it colabti.org/irclogger/irclogger_lo...-09-09#l83 01:11
Xliff Grammars can parse recursively, as long as you have rules that advance the position.
AlexDaniel vrurg: from my point of view it's irrelevant if stage0 makes it into the final package or not
discord6 <RaycatWhoDat> Oh. Then, I probably did it wrong, Xliff.
Xliff Got code?
discord6 <RaycatWhoDat> I'm trying to make a Lisp grammar.
Xliff Urg! I'm List-alergic. ;)
discord6 <RaycatWhoDat> Welp. 01:12
<RaycatWhoDat> I'll just see myself out
<RaycatWhoDat> 😛
Xliff Hah! 01:13
Show me da code!
discord6 <RaycatWhoDat> gist.github.com/RayMPerry/f8e1f804...5a3836fb20
Xliff RaycatWhoDat: My, those are sharp teeth you have.
discord6 <RaycatWhoDat> wat 01:14
irced throws peanuts into the gallery.
Xliff throws the gallery at irced
discord6 <RaycatWhoDat> EY. That's good food gone to waste.
irced dodges the gallery, yes, the gallery.
Xliff What? You levelled up and put points into dodge? You sly one, you!
vrurg AlexDaniel: I think robertle replied on your direct question without taking into account all aspects. So, yes, further investigation is needed.
AlexDaniel vrurg: which is why I asked for more clarifications 01:15
vrurg: but if I were them I'd tell us to fuck right off :P 01:16
vrurg AlexDaniel: meanwhile, I have removed BLOCKER from R#3100. I see no bug on my macos.
synopsebot R#3100 [open]: github.com/rakudo/rakudo/issues/3100 [macOS] IO::Notifications for changes to file contents (2019.03 vs 2019.07.1)
AlexDaniel vrurg: interesting! OK 01:17
vrurg AlexDaniel: but when we'll grow up with fuck them back! Yeah! :D (hope they don't take this seriously...)
AlexDaniel vrurg: I haven't figured out what they do with go or haskell 01:18
vrurg: the guide mentions them but I'm not sure what's the workaround
vrurg AlexDaniel: Also, with R#3132 – it's still uncertain what happened with Blin and why it reported old problems. 01:19
synopsebot R#3132 [open]: github.com/rakudo/rakudo/issues/3132 [BLOCKER] [WIP] Blin 2019-08-20
AlexDaniel vrurg: oh, that's ok then, ignore it for now
vrurg AlexDaniel: I know nothing how those two are compiled. As far as I understand, they compile themselves?
Xliff RaycatWhoDat: Looking. 01:20
repl.it/repls/VapidBountifulRobots 01:25
<atom> needed to be first to advance the cursor.
discord6 <RaycatWhoDat> Oh
<RaycatWhoDat> Interesting. And what does the change to sexp do? 01:26
Xliff It's just a rewrite of what you did. It helps when dealing with nested pairs of tokens.
discord6 <RaycatWhoDat> Ah. Gotcha. Thank you for that. 01:27
Xliff 😛
O. Wait. ☺
m: '😛'.uniname.say 01:28
evalable6 FACE WITH STUCK-OUT TONGUE
irced wow!
m: '😃'.uniname.say
evalable6 SMILING FACE WITH OPEN MOUTH
Xliff 😀 01:29
Wow. That looks freaky in this font.
irced m: '😀'.uniname.say
evalable6 GRINNING FACE
Xliff 😛 <- Thought this was a dracula smile!
irced m: '😀'.unicode.say
evalable6 (exit code 1) No such method 'unicode' for invocant of type 'Str'. Did you mean 'encode'?
in block <unit> at /tmp/cXVrQHXnPP line 1
discord6 <RaycatWhoDat> o
Xliff The tongue is not really rendered well. It looks like fangs in the mouth on this end. LOL! 01:30
irced m: printf("%x", '😃')
evalable6 (exit code 1) Directive x not applicable for value of type Str (😃)
in block <unit> at /tmp/EU47X4Vakv line 1
Xliff m: '😃'.ord.fmt('%x').say
evalable6 1f603
irced m: '😀'.ord.fmt('%x').say 01:31
01:31 MasterDuke left
evalable6 1f600 01:31
irced 😀😀😀
Xliff <😀😀😀>.map( .ord.fmt('%x') ).join(' ').say
m: <😀😀😀>.map( .ord.fmt('%x') ).join(' ').say
evalable6 (exit code 1) No such method 'ord' for invocant of type 'Any'
in block <unit> at /tmp/QlmXykqUhY line 1
timotimo needs a * 01:32
Xliff m: <😀 😀 😀>.map( .ord.fmt('%x') ).join(' ').say
evalable6 (exit code 1) No such method 'ord' for invocant of type 'Any'
in block <unit> at /tmp/AQ4UjQrm8T line 1
Xliff m: <😀 😀 😀>.map( *.ord.fmt('%x') ).join(' ').say
evalable6 1f600 1f600 1f600
Xliff D'oh
irced m: say '😀' xx 3;
evalable6 (😀 😀 😀)
Xliff timotimo: o/ -- Hey u! How are things?
irced say '😀'.ord.fmt('%x') xx 3 01:33
evalable6 (1f600 1f600 1f600)
Xliff timotimo: How difficult would this be? -- github.com/perl6/problem-solving/issues/96
timotimo starting at it should be easy, then you get to fix all the problems that we didn't foresee 01:35
Xliff /o\
discord6 <RaycatWhoDat> m: '"'.unicode.say 01:37
evalable6 (exit code 1) No such method 'unicode' for invocant of type 'Str'. Did you mean 'encode'?
in block <unit> at /tmp/gDYeGE5N_P line 1
Xliff m: '"'.ord.say
evalable6 34
Xliff m: '"'.uniname.say
evalable6 QUOTATION MARK
discord6 <RaycatWhoDat> Thanks
irced say "\x[1f600]" 01:38
evalable6 😀
irced say "\c[QUOTATION MARK]and I quote!\c[QUOTATION MARK]"
evalable6 "and I quote!"
irced print "\c[QUOTATION MARK]and I quote!\c[QUOTATION MARK]"
evalable6 "and I quote!"
Xliff timotimo: Can you point me to where a ThreadContext is created?
timotimo MVM_thread_create or whatever, should be the function called by the OP(newthread) 01:39
and also at startup a TC gets created, which is either in the main.c or whereever the MVMInstance is created
a nice trick i found if oyu want to see where a certain type gets allocated, look for sizeof(that type) in the code
Xliff OK, thanks! 01:40
ThreadContext is the type?
Xliff wishes you could grep the code using github.
timotimo MVMThreadContext 01:41
Xliff Cool
timotimo i wonder what needs to be done to make a webpage with cross-referenced moarvm source code 01:42
containing every regular go-to-definition that C knows, plus some stuff we come up with ourselves
Xliff O! Github can search in repositories! 01:43
github.com/MoarVM/MoarVM/search?q=...text%29%22
timotimo its search isn't fantastic 01:44
01:45 molaf left
Xliff OK. So I think I have MUCH of what goes into a MVMThreadContext. Now I need to figure out what else goes into it aside from a thread_id 01:48
I can see this being a lot more than I expected, but I need it for GStreamer. 01:49
timotimo you are brave 01:51
Xliff LOL
I'll start by coming up with a spec. I'm not THAT crazy. 01:52
01:57 Itaipu left, molaf joined 02:10 Itaipu joined 02:19 cpan-p6 left 02:20 cpan-p6 joined, cpan-p6 left, cpan-p6 joined 02:54 Cabanoss- joined 02:56 Cabanossi left, Cabanoss- is now known as Cabanossi 03:14 MilkmanDan left 03:16 MilkmanDan joined 03:19 Xliff left 03:20 Kaypie left 03:22 Kaiepi joined 03:55 aborazmeh joined, aborazmeh left, aborazmeh joined 03:57 sno left 04:04 aborazmeh left 04:10 molaf left
irced disembarks. 04:31
04:31 irced left 04:48 Kaiepi left 05:23 squashable6 left 05:28 squashable6 joined 05:32 sauvin joined 05:59 jmerelo joined 06:04 HoboWithAShotgun left 06:12 sena_kun joined 06:16 wamba joined 06:18 domidumont joined, ufobat joined
jmerelo m: say Cursor.^mro 06:29
evalable6 ((Match) (Capture) (Cool) (Any) (Mu))
06:32 leont joined
Geth doc: 49b5a4e902 | (JJ Merelo)++ | xt/type-graph.t
Removes unneeded path refs #3001
06:40
doc: eaf6652ce5 | (JJ Merelo)++ | type-graph.txt
Eliminates wrong mro refs #3001
doc: 5fb93693f6 | (JJ Merelo)++ | README.md
Revises documentation, closes #3000
jmerelo notable6: weekly Lots of work done for upgrading the documentation processing and generation system, mainly around this issue: github.com/perl6/doc/issues/2983 06:46
notable6 jmerelo, Noted! (weekly)
07:02 aborazmeh joined, aborazmeh left, aborazmeh joined 07:04 robertle_ joined 07:08 leont left 07:11 wamba left 07:16 jmerelo left 07:22 Itaipu left
El_Che funny how an localized mindset means you don't see a lot of possibilities. I never thought about the script/module extension difference that other languages don't have and why raku should keep it. Mind blown. 07:23
the ending a pm with "1" is an artifact from the old days. Otherwise, I wouldn't know why. 07:24
07:30 dakkar joined
SmokeMachine RaycatWhoDat: have you seen the bernaLang example? 07:30
07:32 Itaipu joined
daxim AlexDaniel, if you want to add a pull req to address <github.com/perl6/problem-solving/p...81879>, then you can use <old.reddit.com/r/perl6/comments/9t...djoj/>. maybe add a note that approximate pronunciation is fine so that it's less intimidating for those who never learnt ipa. 07:35
07:38 Itaipu left 07:39 rindolf joined 07:43 Itaipu joined 07:45 wamba joined 07:46 lnx joined
sena_kun re "how to pronounce": while not being the one who suggested the name or the one who holds the Great Truth about pronunciation, e.g. www.youtube.com/watch?v=uk1CZSogZ6I shows what sounds close enough to "楽" (from English perspective, though). of course, there are always accents to strike later (I'd pronounce it in a more clean/rough way), but that shouldn't be any more of an issue as with any other name. 07:46
tellable6 2019-09-09T10:23:16Z #perl6 <SmokeMachine> sena_kun: what about adding a submodule on `docs/` pointing to the wiki repo? make sense?
2019-09-09T10:25:51Z #perl6 <SmokeMachine> sena_kun: or maybe make the entire `docs/` be the submodule. That way, when generating the docs (if you push it) it will be on the wiki as well
SmokeMachine sena_kun: I tried the submodule and didn't work... 07:47
07:48 Itaipu left
SmokeMachine sena_kun: but: fco.github.io/Red/ 07:48
sena_kun SmokeMachine, it depends on if you want to keep the wiki (in the long term perspective) or not. If you do, then any path that keeps the maintainence low sounds great to me. OTOH, I thought that maybe a more clean / short readme is needed, like "We are cool Perl 6 ORM, you can find docs at $URL (instead of the long cheatcheat, but we can only remove it after the docs will be written in its entirely), you can contribute with this and that, 07:49
license is foo, thanks".
07:50 zakharyas joined
sena_kun and if we have an inline docs (for tools) and the dedicated pages for rendering, and a way to contribute, I don't see where wiki is better than that, but maybe I am just ignorant. 07:50
SmokeMachine sena_kun: I agree...
sena_kun: do you know any way to make gh-pages highlight perl6 code? 07:52
sena_kun SmokeMachine, are generated API pages use "```perl6```" in code chunks, for just ```?
s/use/using/ 07:53
SmokeMachine sena_kun: it's using that, but still not highlighting it...
07:54 dolmen joined
sena_kun SmokeMachine, no idea, but I see help.github.com/en/articles/using-...thub-pages 07:54
SmokeMachine sena_kun: here it's highlighted: github.com/FCO/Red/blob/master/doc...tSeqSeq.md but here it isn't: fco.github.io/Red/api/Red/ResultSeqSeq
and: raw.githubusercontent.com/FCO/Red/...tSeqSeq.md 07:55
tellable6 SmokeMachine, I'll pass your message to and
SmokeMachine sena_kun: I saw that... but rouge doesn't recognises perl6...
08:03 aborazmeh left 08:04 scimon joined 08:05 sono__ joined 08:06 Black_Ribbon left 08:18 Itaipu joined 08:28 Itaipu_ joined 08:29 Itaipu left 08:40 grumble left 09:01 aborazmeh joined, aborazmeh left, aborazmeh joined, HoboWithAShotgun joined
daxim sena_kun, I can't youtube here. what does it show? 09:04
sena_kun daxim, (apparent) prononciation of "raku", which I found the most matching among other video that try to do it 09:06
daxim what's it in ipa?
sena_kun has no idea how to use ipa. "r" as in "road", "a" as in "slavic", "k" as in "kitchen", "u" as "oo" in "google". 09:09
huf i'm just completely flabbergasted that the pronunciation is an issue
"raku" is about as easy as you can possibly get in english 09:10
daxim I have no idea what kind of english you're having in mind. learn ipa to avoid ambiguities.
El_Che /raku/ probably :) 09:12
huf also, you're probably never going to get everyone to pronounce it the same :)
there's gonna be people who say it like the front part of "raccoon" 09:13
El_Che peetô and databaaasé
something I hear when in France
huf "perl" probably had a bunch of international pronunciation attempts
jnthn huf: hah, that's how I assumed it should be said :P
daxim El_Che: alveolar trill? 09:14
huf jnthn: it's the one that comes to my lips easiest, but that reduced first vowel is not very japanese i think?
El_Che daxim: raku for the French? Sure :) 09:15
lizmat pronunciation has never been an issue for SQL, has it ?
huf not a *real* issue
just an evergreen joke
or possibly nevergreen 09:16
lizmat I'll go for evergreen any time
jnthn huf: Maybe not; my attempts at Japanese extend to asking for a draught beer. :P
lizmat means a long life ahead
El_Che anyway, pronunciation of software has always been tricky
Take Linux
huf and seriously, not even C is pronounced consistently (across the world)
daxim now I'm upgrading that "approximate pronunciation is fine" be included from a suggestion to a strong suggestion
jnthn And probably doing that poorly :P
El_Che some American infidels read it as La-ee-nux
and no one cares, really 09:17
jnthn
.oO( all this talk of IPA makes me thirsty... )
lizmat it *can* be important: in the mini-computer days, a Japanese company called "FAComp" had trouble in the US
"fake-em" "f*ck-em" ? 09:18
huf yes, those things inevitably happen
but they happen within a single country too
penisland, etc
09:18 aborazmeh left
lizmat what? 09:18
huf www.penisland.net/
SmokeMachine easypronunciation.com/en/japanese-...nscription
huf it's about pens.
daxim ファコンプ
sena_kun エフエイコンプ 09:21
El_Che Even the most simpel name of a computer language --Go-- is pronounce differently 09:22
English speakers use of course the hard G, but I've heard it with a soft variant oferen, and for the o there are lots of variants 09:23
and if the name can easily localized to a local pronounciation, it will 09:25
09:26 pat_js joined
Voldenet in Polish "raku" is locative/vocative case of "rak" which means "crayfish"/"cancer" 09:30
sena_kun Voldenet, it holds true for number of slavic languages, e.g. in Ukrainian too, as well as Serbian, was mentioned already in the long-long thread 09:31
El_Che Voldenet: the asociation is not cancer-constellation-crab?
daxim also cancer the illness 09:32
huf if it's anything like hungarian, it's the constellation, cancer, crab, crayfish, lobster, shrimp, etc 09:33
all those weird water insects
Voldenet It's mostly used for cancer the illness, in IPA it would probably be written as "raku" simply
09:37 aborazmeh joined, aborazmeh left, aborazmeh joined, domidumont left 09:51 cgfbee left 09:54 pat_js left, pat_js joined 10:02 cgfbee joined 10:24 grumble joined 10:29 [particle]1 left 10:34 aborazmeh left 10:36 sono__ left 10:44 dolmen left 10:48 rurban joined, rurban left 10:59 robertle_ left 11:00 [particle] joined 11:01 robertle_ joined 11:14 HoboWithAShotgun left 11:21 domidumont joined 11:24 sena_kun left 11:34 titsuki left 11:37 aborazmeh joined, aborazmeh left, aborazmeh joined, Guest93 joined 11:41 zakharyas left 11:57 cwilson joined 12:16 domidumont left 12:21 aborazmeh left 12:22 gabiruh left, gabiruh joined
ufobat does anyone have an idea how to obtain the depth of the perl6 callstack (related to github.com/rakudo/rakudo/issues/3148) 12:32
12:35 molaf joined 12:36 grumble is now known as \emph{grumble}
SmokeMachine I've always listen rake as /ɾɐ.ˈku/ 12:37
12:38 dolmen joined
SmokeMachine I mean: /ɾa.'ku/ 12:41
Voldenet ufobat: `+(Backtrace.new.Str ~~ m:g/\n/)` works fine 12:45
SmokeMachine and not /ra.'ku//
Voldenet maybe this is broken, because of github.com/rakudo/rakudo/blob/mast...e.pm6#L316
SmokeMachine /ra.'ku/
Voldenet `do we need more than 100???` ツ 12:46
SmokeMachine: not many people can actually say r
most people read that as ɹ or ɻ 12:47
SmokeMachine in Portuguese we do ɾ 12:48
Voldenet: ^^
and ʁ 12:49
Voldenet: I thought Japanese "r" was more to ɾ than to ɹ, ɻ and r... (but I don't speak Japanese...) 12:51
12:52 HoboWithAShotgun joined
ufobat Voldenet, thats a nice trick! thanks! 12:55
12:55 lucasb joined
Voldenet Let me rephrase, a lot of english-speaking people simply don't pronounce r/ɾ :) 13:05
13:11 domidumont joined 13:17 HoboWithAShotgun left 13:26 zakharyas joined
SmokeMachine Voldenet: yes, but should we only care about English speakers? 13:31
Voldenet I don't think official IPA name matters much, people will not use it anyway 13:33
13:33 esh joined
SmokeMachine Voldenet: as almost every Brazilian say `perl` wrong... (including my self...) 13:34
Voldenet as "pel"? :) 13:36
Well, some people pronounce nginx as "n jinx" even though it should be read as "engine x" 13:37
timotimo i pronounce it "n jinx"
Grinnz "en ginx" here
correctly pronouncing nginx has been very difficult for me :P
Voldenet I pronounced it like that when someone told me I should use the other reading 13:38
Now I still read it like that, but now I feel bad about it
Grinnz "I still do, but I feel bad" apologies to mitch hedberg ;) 13:40
13:48 lizmat left
Voldenet I think "raku" is universal enough that most people will read it correctly if they used english though 13:49
13:52 HoboWithAShotgun joined
Grinnz as an american, I pronounced it correctly by default, probably due to its similarity to ragu :P 13:52
but I am sure "ray ku" will be common as well here 13:53
seems like a far cry from the nginx and sql problems 13:54
jdv79 is it "reh jex" or "reh gex" or "reh gehxp"? 13:59
14:00 Kaiepi joined 14:01 pmurias joined
pmurias . 14:01
[Coke] I say "reh jex" myself. and if you say "en jinx", you're a monster. :) 14:07
Voldenet Welp, sql had no official pronunciation, so mysql is "my es q el", but sql language was literally called "sequel" 14:08
[Coke] (nginx has a FAQ about that somewhere)
timotimo when i was a teenager, other teenagers said "SQL" when they meant "school"
[Coke] Voldenet: I have always heard "my sequel" in my circles
timotimo well, they probably spelled it "sql"
Voldenet dev.mysql.com/doc/refman/8.0/en/wh...mysql.html 3> The official way to pronounce “MySQL” is “My Ess Que Ell” (not “my sequel”), but we do not mind if you pronounce it as “my sequel” or in some other localized way. 14:09
timotimo for some time, the firefox FAQ said firefox was to be abbreviated "fx" 14:10
kawaii Weekly "where is Rakudo Star?" post
[Coke] kawaii: are you on windows? 14:11
kawaii Linux, tried to build it myself but the install scrips or something have changed in Rakudo itself and the build instructions are therefore outdated 14:15
[Coke] Voldenet: neat, thanks. 14:16
14:25 HoboWithAShotgun left 14:32 sena_kun joined
guifa doesn’t think pronunciation of raku will be a problem 14:40
14:41 wamba left, yqt joined
tadzik rakudo didn't have any problems, so I don't think raku will :) 14:41
14:41 wamba joined
guifa Any variation will be less, no doubt, than the differences between dialects. I’ll say raku with the A in father, but if a Brit says raku with the A in ate, I’ll just assume dialect difference, not “hey doesn’t know how to say it” 14:41
Juerd I've heard numerous different pronunciations of "Perl" too, and that's never been an issue 14:43
Grinnz oh yes, the juxtaposition (or whatever you call it when it's not contrasting) with rakudo probably helped me pronounce raku correctly
guifa ^^ Exactly
Juerd Many Dutch people like to call it "prul", which means "piece of trash"... 14:44
(Trash bin = prullenbak)
guifa Just posted a huge code review on SE, comments welcome: codereview.stackexchange.com/quest...-generator 14:45
14:49 pmurias left
Voldenet m: sub prefix:<#> { $^hex; }; say #242424; 14:50
evalable6 (exit code 1) 04===SORRY!04===
Argument to "say" seems to be malformed
at /t…
Voldenet, Full output: gist.github.com/da136ab29d8069733e...65a757caf7
14:54 HoboWithAShotgun joined
guifa I think doing a # prefix operator could uh… result in problems lol 14:56
jnthn No comment... :)
Ah, actually, comments win :) 14:57
guifa That looks like a beautiful potential for a tweet if I tweeted
Grinnz perlbot: sweval: say q#foo#
perlbot Grinnz: ===SORRY!===␤Could not find warnings at line 1 in:␤ inst#/langs/perl6/share/perl6/site␤ inst#/langs/perl6/share/perl6/vendor␤ inst#/langs/perl6/share/perl6␤ ap#␤ nqp#␤ perl5#␤[Exited 1]
Grinnz oh it defaults to 6 here, right :P
lucasb m: my @a; $#a 14:58
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/wpvaH2yHP2
Unsupported use of $#a variable; in Perl 6 please use @a.end
at /tmp/wpvaH2yHP2:1
------> 03my @a; $#a08⏏04<EOL>
jnthn m: say q#but here?#
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/A3nTmgC1lA
# not allowed as delimiter
at /tmp/A3nTmgC1lA:1
------> 03say q#08⏏04but here?#
jnthn There it knows it's not going to allow it :)
Grinnz spoiler, it's allowed in perl 5 ;)
jnthn Can you put comments inside of a regex using # in perl 5? 14:59
Grinnz with /x, yes
timotimo m: "a" xx 1024 ~~ / (a+)+ b+ | <!> /; say now - INIT now
evalable6 0.0303148
timotimo ^- use the power of NFA to get an otherwise catastrophically backtracking regex to give non-match results instantaneously 15:00
Grinnz i'm not sure if allowing # as delimiter is a better or worse idea than allowing \ as delimiter
hmm... or if those non printable characters still work 15:01
Voldenet m: sub prefix:<$_> { }; say $_ 15:03
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/QBImmVbFcK
Prefix …
Voldenet, Full output: gist.github.com/f309af6e1539001644...e27954af75
Voldenet Splendid 15:04
jnthn m: sub prefix:<$_> { $^n + 1 }; say $_ 41 15:05
evalable6 42
jnthn hah :)
15:06 pmurias joined
Voldenet m: sub prefix:<my> { $^a }; sub prefix:<$_> { $^a }; sub prefix:<=> { $^a }; my $_ = 2; 15:08
evalable6 (exit code 1) ===SORRY!===
Cannot find method 'ast' on object of type NQPMu
Voldenet i'd be quite worried if that worked
15:08 yqt left 15:12 pmurias left 15:13 nikita_1 joined, pmurias joined, yqt joined 15:14 ufobat_ joined 15:18 ufobat left
nikita_1 hello, fathers. I'm trying to 'make install' the latest rakudo-star on DragonFlyBSD, but there are a lot of error messages: "Failed to resolve host name 'localhost' with family 0. Error: 'servname not supported for ai_socktype'" where the problem can be? 15:18
15:19 yqt left, robertle_ left 15:23 zakharyas left 15:25 zakharyas joined 15:38 domidumont left
Geth problem-solving/path-to-raku: 4 commits pushed by (Jonathan Worthington)++ 15:40
Kaiepi nikita_1, if you give me a minute, would you be up for running a small c program to see why that could be happening? 15:55
15:56 leont joined 15:57 wamba left 16:01 aborazmeh joined, aborazmeh left, aborazmeh joined
Geth ¦ problem-solving: ajs assigned to AlexDaniel Issue Perl 6, if renamed, needs new filneame extensions github.com/perl6/problem-solving/issues/101 16:03
¦ problem-solving: AlexDaniel assigned to jnthn Issue Perl 6, if renamed, needs new filename extensions github.com/perl6/problem-solving/issues/101 16:04
¦ problem-solving: AlexDaniel unassigned from jnthn Issue Perl 6, if renamed, needs new filename extensions github.com/perl6/problem-solving/issues/101
¦ problem-solving: AlexDaniel self-unassigned Perl 6, if renamed, needs new filename extensions github.com/perl6/problem-solving/issues/101
¦ problem-solving: AlexDaniel assigned to jnthn Issue Perl 6, if renamed, needs new filename extensions github.com/perl6/problem-solving/issues/101 16:05
16:12 lizmat joined 16:13 jmerelo joined 16:20 scimon left 16:24 pmurias left, pat_js left, aborazmeh left 16:26 wamba joined
Geth problem-solving/path-to-raku: 9310222f55 | (Jonathan Worthington)++ | solutions/language/PATH-TO-RAKU.md
Add a draft module author guide to the changes
16:27
tony-o nikita_1: does localhost resolve to 127.0.0.1 ?
problem-solving/path-to-raku: 9285582c0a | (Jonathan Worthington)++ | solutions/language/PATH-TO-RAKU.md
Grammar tweak
tony-o m: module Foo { our $xz = 42; our sub foo is export { say $xz//"undef"; }; }; import Foo; foo()
evalable6 42
tony-o m: module Foo { our $*xz = 42; our sub foo is export { say $*xz//"undef"; }; }; import Foo; foo()
evalable6 undef
16:28 Actualeyes joined
SmokeMachine This script is given this error: www.irccloud.com/pastebin/k8vAJCp6/ 16:31
www.irccloud.com/pastebin/cuW1aVB8/
nikita_1 tony-o: yes, on my machine localhost is 127.0.0.1, I suppose, that ai family is set to zero, not PF_INET. 16:32
16:37 dakkar left 16:42 pmurias joined 16:49 dolmen left
nikita_1 Kaiepi: yes, i'm able to run 16:51
Kaiepi ok nikita_1 run this and tell me what if it prints an error or not hastebin.com/lahefeyexu.cpp 16:53
16:57 aborazmeh joined, aborazmeh left, aborazmeh joined 17:01 molaf left 17:12 sno joined, sno left 17:20 MasterDuke joined 17:22 pmurias left 17:26 gabiruh_ joined 17:27 gabiruh left 17:28 wildtrees joined 17:33 aborazmeh left 17:42 zakharyas left 17:51 ravenousmoose joined 18:00 Sgeo__ joined 18:03 Sgeo_ left 18:04 pmurias joined 18:09 |oLa| joined
Geth ¦ problem-solving: vrurg assigned to AlexDaniel Issue Use of cmake for build subsystem github.com/perl6/problem-solving/issues/102 18:13
18:18 sno joined 18:21 pmurias left 18:22 pmurias joined 18:23 sauvin left 18:41 Guest93 left
Kaiepi i haven't used :D, :U, or :_ very much for typing stuff even though i type everything possible in my code, but i'm catching a lot more bugs in my code when using it now 18:42
guifa Kaiepi: ikr? I’m generally forced to use it because I do so much terminal output. I just go ahead and assume gist needs to be multi based on definedness 18:43
What does Incompatible MROs in P6opaque rebless for types Rat and MyEnumClass mean? 18:44
Seems like you can’t use non-integers for enum vals I guess 18:45
tony-o that would be my guess, you might be able to wrap p6opaque in something to make it work but that's a wag 18:47
Geth ¦ problem-solving: vrurg assigned to jnthn Issue Roles, submethods, constructors, destructors. github.com/perl6/problem-solving/issues/103 18:54
18:54 krychu joined 18:56 |oLa| left 19:00 Kaypie joined, jmerelo left, Kaiepi left 19:05 ravenousmoose left
guifa Is it possible to use regex for operator subs? Like postfix:(/f o*/) to capture 1f 2foo 3fooo 2fooooooo ? 19:08
19:09 reach_satori left, reach_satori_ joined, Kaypie is now known as Kaiepi
Kaiepi m: subset Foo where Int; my Foo:U $foo; say $foo 19:10
evalable6 (Foo)
Kaiepi m: subset Foo where Int; my Foo:D $foo; say $foo
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/p4QbIPNpyh
Variabl…
Kaiepi, Full output: gist.github.com/3dc8147a321e2f755d...349084c490
Kaiepi m: subset Foo where Int; my Foo:D $foo = 1; say $foo
evalable6 1
19:10 AlexDaniel left 19:11 AlexDaniel joined, AlexDaniel left, AlexDaniel joined 19:18 __jrjsmrtn__ joined 19:20 _jrjsmrtn left 19:21 aborazmeh joined, aborazmeh left, aborazmeh joined 19:28 yqt joined 19:30 Ven`` joined 19:41 AlexDaniel left, AlexDaniel joined, AlexDaniel left, AlexDaniel joined
tony-o m: multi sub x (Str:D $str where * ~~ /^'auth:'/) { say "yes: $str"; }; multi sub x (Str:D $str) { say "no: $str"; }; x('str ~~ /^auth:/'); x('other x'); 19:42
evalable6 no: str ~~ /^auth:/
no: other x
tony-o m: multi sub x (Str:D $str where * ~~ /^'auth:'/) { say "yes: $str"; }; multi sub x (Str:D $str) { say "no: $str"; }; x('auth:test'); x('other x');
evalable6 yes: auth:test
no: other x
19:42 molaf joined
tony-o guifa: ^ 19:42
guifa tony-o: I Mean with postfix operator subs :-) 19:44
multi postfix:<foo> ($bar) { … } works
tony-o oh, i see - i'd imagine that only takes a string but likely doable in nqp 19:45
i haven't dug around in those guys in a number of years though 19:46
guts*
guifa I have no plans to use it haha but was answering a code review question and thought there was potential for a good use for it
19:48 marcusal1987 joined
guifa codereview.stackexchange.com/quest...804#227804 19:49
19:50 aborazmeh left
tony-o hm, there's no $?ROUTINE in postfix: subs either 19:52
m: sub postfix:<foo> (Str:D $a) { $?ROUTINE.name.say; }; 'a'foo;
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/Nn1KPD9ElD
Variabl…
tony-o, Full output: gist.github.com/affa1bcf6e74382ee1...99315144f5
nikita_1 Kaiepi: nikita@dfly  ~  ./test2 19:53
Error resolving localhost:8000: servname not supported for ai_socktype
tony-o nikita_1: what is the output of getent hosts 0 19:55
can you telnet 0 8000 ? 19:56
MasterDuke m: sub postfix:<foo> (Str:D $a) { &?ROUTINE.name.say; }; 'a'foo;
evalable6 postfix:<foo>
tony-o MasterDuke: ty, i am a clown
19:57 lizmat left
AlexDaniel 🤡 19:57
nikita_1 nikita@dfly  ~  getent hosts 0
Grinnz oh god. why did they have to make that an emoji. why
AlexDaniel that emoji is really scary :D
nikita_1 0.0.0.0 0
tony-o guifa: you could abuse a `where` to match the postfix ? 19:58
19:58 pmurias left
guifa hmmm, you mean like sub postfix (Any $a) { #`( Check value based on &?ROUTINE )`} if not redispatch? 19:59
I wonder if that’s possible
tony-o that's way too complicated though
19:59 lizmat joined, lizmat left
AlexDaniel Grinnz: well, I mean… 𓂺 20:00
Grinnz that one's not an emoji yet technically :P
20:00 pmurias joined
Kaiepi nikita_1, try changing ai_flags from AI_PASSIVE to AI_PASSIVE | AI_NUMERICSERV and run it again to see if that works 20:00
AlexDaniel Grinnz: yeah, that part of unicode is not fleshed out properly 20:01
guifa tony-o: Doesn’t work anyways haha I just tried. Looks like you gotta have a string.
nikita_1 Kaiepi: the same error. specifying ai_socktype with SOCK_STREAM (f.e.) solves the problem 20:03
but, i thinks, it's not the way
AlexDaniel propdump: 🤡𓂺
unicodable6 AlexDaniel, gist.github.com/c245d2c4b935850561...1f32f1758f
AlexDaniel indeed, not emoji 20:04
m: say ‘#’.uniprop(‘Emoji’) 20:05
evalable6 True
AlexDaniel I never really understood that part :)
nikita_1 Kaiepi: here is the output: nikita@dfly  ~  ./test2
Successfully resolved localhost:8000!
Family: 28
Socket Type: 1
Protocol: 6
Kaiepi ok nikita_1 a pullreq i have for improving hostname resolution fixes this then 20:06
i checked dragonflybsd's kernel earlier to see when getaddrinfo is supposed to be returning that error and it does so when both ai_socktype and ai_protocol are 0 and AI_NUMERICSERV isn't passed as a flag. at the moment, moarvm does both, but whenever this gets merged it will pick an appropriate socket type and protocol and pass AI_NUMERICSERV since it's never passing a real service name 20:08
20:11 yqt left
nikita_1 Kaiepi: is this code only for hostname resolve? 20:13
20:13 Black_Ribbon joined
Kaiepi yeah 20:13
nikita_1 is gethostbyname simpler? 20:15
Grinnz gethostbyname should generally not be used anymore
20:23 aborazmeh joined, aborazmeh left, aborazmeh joined 20:25 marcusanthl1987 joined 20:28 aborazmeh left
Kaiepi struct hostent doesn't contain any struct sockaddr members, which moarvm needs to use when connecting or binding sockets, so idt gethostbyname2 would be appropriate to use here 20:28
20:29 sena_kun left, marcusal1987 left 20:38 dwarring joined 20:52 pmurias left
Kaiepi is there anything on how to use cro pipelines online besides cro's documentation? they look like something that might make a bot i'm writing simpler 20:53
21:03 marcus1987al joined, pmurias joined 21:06 marcusanthl1987 left
timotimo there's the test suite that tests the pipelines, i'm sure that shows a lot of interesting stuff 21:06
21:12 Xliff joined
Xliff m: %?RESOURCES.say 21:13
evalable6 Nil
21:13 marcusanthl1987 joined
Xliff m: %*DISTRIBUTION.keys.say 21:13
evalable6 (exit code 1) Dynamic variable %*DISTRIBUTION not found
in block <unit> at /tmp/tIXgUoNli6 line 1
Xliff m: $?DISTRIBUTION.keys.say
evalable6 ()
Xliff Are %?RESOURCES and $?DISTRIBUTION the same, now? 21:14
And does anyone know of a decent write up on how to use them?
m: say %*DISTRO.perl 21:15
evalable6 Failure.new(exception => X::Dynamic::NotFound.new(name => "\%*DISTRO"), backtrace => Backtrace.new)
Xliff m: say $*DISTRO.perl
evalable6 Distro.new(release => "9", is-win => Bool::False, path-sep => ":", name => "debian", auth => "www.debian.org/", version => v9.stretch, signature => Blob, desc => "Debian GNU/Linux 9 (stretch)")
Xliff Why is path-sep ":"? 21:16
moritz because PATH elements are separated by : 21:17
PATH=/bin:/usr/bin:/sbin
21:17 marcus1987al left
Xliff o 21:17
Right
$*SPEC.perl.say 21:18
evalable6 IO::Spec::Unix
Xliff $*SPEC.path-sep.say
m: $*SPEC.path-sep.say
evalable6 (exit code 1) No such method 'path-sep' for invocant of type 'IO::Spec::Unix'
in block <unit> at /tmp/mTSQr5CKUA line 1
Xliff m: $*SPEC.dir-sep.say
evalable6 /
SmokeMachine why is this taking so long? www.irccloud.com/pastebin/3FRW8JEo/ 21:21
21:23 gabiruh_ is now known as gabiruh 21:28 krychu left
jnthn SmokeMachine: Dunno, but 1) what is that .max actually applied to, 2) how much work is there, because start/await innevitably has some overhead, and if you're breaking the problem down into pieces that are way too small to overcome that, it won't help much 21:39
SmokeMachine jnthn: `.max` is being applied to the return of `do for` a list Seq of Lists... 21:42
Xliff \o 21:53
SmokeMachine: Did you get my last bit of info?
SmokeMachine Xliff: Sorry, I think I didn't... :( 21:54
Xliff After setting up the columns, I'm still getting the same error.
SmokeMachine: paste.fedoraproject.org/paste/zbrl...mi6CZZwKaQ 21:57
SmokeMachine Xliff: How is your driver? 21:59
Xliff: I think I've read somewhere that `.prepare` wan't working on that module, is that true? 22:00
22:02 dpk left, dpk joined 22:06 lizmat joined 22:08 Ven`` left
Xliff SmokeMachine: Yeah. I think so. 22:18
SmokeMachine Red's drivers (at least currently) uses prepare to run queries...
Xliff Any query? 22:19
Well, that makes sense.
How can I set up a constant or scalar to use %?RESOURCES in an "is native" trait? 22:20
Or can I just use it directly?
22:24 leont left 22:31 abraxxa left
jnthn SmokeMachine: Hm, then I'm confused, since the loop gives back `Promise`s, and I'd have thought the .max should apply to the result of `await`? 22:33
SmokeMachine jnthn: makes sense... I'm testing... 22:35
m: await do for ^5 { $_ }.^name 22:36
evalable6 (exit code 1) An operation first awaited:
in block <unit> at /tmp/Jy_9YcbAAa line 1

Died with the exception:
Can only specify Awaitable objects to await (got a Str)
in block <unit> at /tmp/Jy_9YcbAAa line 1
SmokeMachine m: await do for ^5 { start { $_ } }.^name
evalable6 (exit code 1) An operation first awaited:
in block <unit> at /tmp/DMAFMgySbY line 1

Died with the exception:
Can only specify Awaitable objects to await (got a Str)
in block <unit> at /tmp/DMAFMgySbY line 1
SmokeMachine m: await(do for ^5 { start { $_ } }).^name
evalable6 22:37
SmokeMachine nothing?
www.irccloud.com/pastebin/fDx8mdmJ/ 22:39
MasterDuke await(do for ^5 { start { $_ } }).^name.say 22:40
evalable6 List
SmokeMachine MasterDuke: sure! thanks! sorry!
22:48 kensanata joined 23:05 lucasb left 23:09 holyghost left 23:18 Xliff left 23:21 wamba left 23:23 marcusal1987 joined, kensanata left 23:24 yqt joined 23:25 marcusanthl1987 left 23:27 rindolf left 23:35 pmurias left 23:39 kst left, yqt left