»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.
timotimo alternatively it could be geth just checked out for a while and just came back and reported everything left over form durign the crash 00:02
Geth modules.perl6.org: c518c09135 | (Zoffix Znet)++ | 4 files
[REBUILD] Fix tag UI issues

Fixes #73
  - Move decisions on which tags are weak and should be hidden
   to server side
  - Start with non-weak tags visible by default instead of starting
   them all hidden and showing them with JS after load
  - If weak tag is active, do not hide weak tags either
  - If active tag is clicked, assume user wants to go back to all tags
  - Fix styling for active weak tag		 00:33
Geth modules.perl6.org: f10acaadf5 | (Zoffix Znet)++ | templates/root/index.html.ep
Make Camelia a link to perl6.org

Fixes #72		 00:37
Geth modules.perl6.org: dcd3590c46 | (Zoffix Znet)++ | site-tips.txt
Use HTTPS URLs in site tips
 00:40
Geth modules.perl6.org: 6ee47b79a2 | (Zoffix Znet)++ | site-tips.txt
Add link-to-tag and info on JSON API to site tips
 00:44
IOninja I didn't say I went to sleep. I said I was done for the night :} 01:07
AlexDaniel buggable: tag segv 02:05
buggable AlexDaniel, There are 28 tickets tagged with SEGV; See perl6.fail/t/SEGV for details
AlexDaniel ah-ha 02:07
Geth doc: bcfaf4c8b6 | (Samantha McVey)++ | util/trigger-rebuild.txt
Trigger rebuild to pull in highlighter changes
 10:00
DrForr I'm rewriting the "Unicode Friendly" slides for OSDays. Any suggestions? 11:34
masak just found contributor-covenant.org/ 12:52
...hi, #perl6
DrForr s/contributor/alien/ :) 12:52
masak ;) 12:53
DrForr Any Unicode suggestions? I'm already showing that ½τ == π works (with \c[INVISIBLE TIMES])... 12:54
raschipi masak: please don't try to kill Perl6.
masak raschipi: unexpected reply. hold on, I'll make a note here "try not to kill Perl 6"... :P 12:55
raschipi I'm all in favor of a contributor agreement, just not that one. That one is toxic. 12:56
masak DrForr: maybe mention that any use of invisible characters in source code is automatically sinful :P
raschipi How is one supposed to program without spaces or newlines? 12:57
masak raschipi: I feel like there's some background information here I should know about but don't
DrForr Nearly fell out of my chair. I'm watching CinemaSins.
masak raschipi: those two get some kind of exception
raschipi Let me get a coffee first, then I will tell you the drama. 12:58
masak don't know if I should look forward to the story or not
raschipi Well, it's simple, that contributor agreement was created to ALLOW harassement, not to stop it. As long as it's the politically correct kind of harassment, of course. 12:59
masak I'm no expert, but I hear that kind of argument sometimes at the edges of my Twitter feed. often together with phrases like "social justice warrior" and "ethics in journalism" 13:01
sjn reads masak's link 13:02
raschipi I'm just trying to avoid drama, from any side.
masak raschipi: you have been most civil so far. that is appreciated.
I'm not sure "avoid drama from any side" is the right metric, short-term
to optimize for, I mean 13:03
masak looks forward to hearing what sjn thinks
raschipi Like I said, I'm in favor of a contributor agreement, but one that doesn't bring the king of baggage that one brings.
masak well, it's good to know that according to at least some people, it has baggage
raschipi I have to do something, we will talk more later.
masak I wasn't necessarily thinking of adopting it so much as being inspired by it, perhaps 13:04
sjn Hm. A cursory look at it already shows a few problems.. :-(
DrForr Contributor agreement... sigh.
lucs I'm amazed that all these covenants and other codes of conduct are popping up all over the place. 13:13
It looks like at least a whole generation forgot or was not taught or ignores a simple concept: 13:14
Behave
sjn masak: First impression; it's not horribly bad; maybe little is said about how the document evolves, and it leaves terms like "appropriate", "fair", "offensive" and "inappropriate" to be defined by the project maintainers... :-\
masak lucs: I've come to think that "behave" isn't enough and that an explicit Code of Conduct can be better than none 13:15
sjn: well, take something like the Perl 5 IRC ecosystem
lucs masak: Yes, people appear not to know what behave means.
masak lucs: I don't think it's because we've left some lovely Golden Age, though
lucs: I think this is what happens when communities grow and mature 13:16
jast personally, the only reason I'm not put off by the language in Codes of Conduct is that I don't read them and rely on my common sense being enough
masak sjn: people like mst perform a community service by intervening when someone acts in a trolling way, kicking and banning as needed
DrEeevil I find them very confusing 13:17
masak sjn: it's a job I don't imagine is very fulfilling, but it's needed
DrEeevil most are badly worded and arbitrary, which makes enforcement very ugly
sjn yeah
jast my point is that you can just as easily intervene when there is no code of conduct
masak sjn: I get the impression that "mst and his clique" end up getting a lot of hate/flames because of it
sjn: but I'd rather they keep cleaning up obvious trolls than not
(and rather them than me, too) :/ 13:18
sjn To me the text looks like the type of ad-hoc "legal" document that might curb certain behaviour, but at the cost of taking away attention from just helping eachother with the everyday community work
masak jast: sure, of course
jast at the core is this: put up a document of rules and you end up getting a legal system developed around it, whether you want one or not
masak Perl 6's one venture into the code-of-conduct thing was a document that can be summarized as "just be level-headed and reasonable, OK? give people the benefit of the doubt and don't yell" 13:19
jast that sounds reasonable ;)
masak well
sjn masak: I like the Reasonable Person Principle
masak then the author of that document went and had a big argument with someone on the channel 13:20
and wouldn't calm down or listen to arguments
DrEeevil jast: you might end up with an acute infection of bureaucrats ;)
masak and ended up leaving the community
not sure a different code-of-conduct could've prevented that, but... 13:21
sjn www.irc.perl.org/reasonable_person.html
jast definitely not
masak ...it made me conclude that "just be reasonable" isn't always enough
jast that seems to me like inverse logic
"the document did not prevent the behaviour it was meant to highlight as being bad, therefore it's not sufficient" 13:22
if that's your standard, no document will achieve it
masak more like "the author of the document couldn't live up to the guideline outlined in the document"
jast that's a failure of the author, not of the document 13:23
masak more the former than the latter, yes
anyway, I see what you're saying
I'm still thinking about all these things
still not convinced either way 13:24
jast apart from that, let's look at this from another angle. if you have a document, prominently advertised, that keeps talking about abuse and harassment and insults and trolling, what kind of impression does that create about what are the most central problems to your project?
jast putting up a set of rules like that implies: "this is the behaviour we anticipate having to deal with" 13:24
I don't know about you, but to me that kind of implication doesn't actually seem like something that creates positivity 13:25
kurahaupo__ jast: we should anticipate dealing with it, because it happens everywhere. But it could focus more on education and amicable resolution.
DrEeevil even worse: since it defines unacceptable behaviour people will argue that everything else is acceptable, and there's subjective stuff like "being offended" 13:26
b2gills Just looking at the second paragraph saying “… thoughtless use of pronouns, assumptions of gender, …” has put me off of it.
DrEeevil ... now you have more problems ;)
kurahaupo__ no, the last item is "whatever else would not be acceptable in a professional environment"
jast kurahaupo__: definitely anticipate that it's going to happen, but not give it so much focus that it seems like the *expected* thing to happen 13:27
b2gills Has the document that ShimmerFairy created been put into a prominent place yet
DrEeevil kurahaupo__: that is difficult - I've worked in very weird but professional environments
kurahaupo__ It would be nice to add some words like "we hope it doesn't come to this but..."
jast how often do you see stores put up a prominent sign saying, "please do not insult our employees"?
putting up that sign will bias your customers in ways that probably aren't going to help you 13:28
masak there's something about that argument I don't exactly like
can't put my finger on it, but...
b2gills Just go to a construction site, and listen. 13:29
masak using a similar argument, you could never really say something like "we expect people not to be jerks"
jast oh, it's a terrible argument, because nobody likes for this to be going on, but it does
(go on)
masak that's the thing, though
for years, the Perl 6 community has been a source of joy and happiness for me, because people are largely nice
I'd like for it to stay that way
jast quite understandable :) 13:30
masak but I also know that when communities grow and mature, they grow less polite and the discourse deteriorates
jast documents don't fix that, people do
a community needs people who reinforce positive behaviour, and swift action against unreasonable people 13:31
masak ok, that's something I can agree with
b2gills Well reasonable interpersonally, Larry is quite unreasonable, otherwise he wouldn't have come up with Perl 6 … or even Perl in general 13:32
jast unreasonable in terms of conflicts with others, I mean
there's nothing wrong with having unusual ideas and pursuing them, but if you start attacking others over it, something's wrong 13:33
jast I guess it boils down to: differences in opinion/values are good, attacks over differences in opinion/values are not 13:35
masak but that's so generic that virtually anyone would agree 13:36
even trolls
IOninja masak, imo these codes of conduct are the one sure way to create a hostile environment. The safe people won't even bother reading them and the trolls will just use them to troll everyone. The crucial point their near universally miss is: whom to contact in private to resolve serious issue... "This guy keeps messaging me and asking for nudes; what do I do?" "This person keeps closing my Issues because they 13:37
don't like that I'm gay. Who do I talk to?" "This person keeps posting my personal information. How do I stop them?" Those are the real problems and I'm yet to see a CoC that doesn't just stupidly enumerate "unwanted behaviours".
jast I don't think so. many (not all, but this wasn't meant to be the One Golden Rule) trolls are, quite obviously, very much in favour of attacking people
masak I find the argument "codes of conduct are what creates a hostile environment" totally bizarre
jast personally I think that argument overstates things a bit 13:38
masak but I do agree on the importance of addressing those crucial points
DrEeevil masak: you've not seen an infection of bureaucrats and legalistic trolls? very easy to be a pain in the ass while within the rules, while everyone who attacks you oversteps them
jast it's true that some people like doing nasty things with detailed sets of rules
and those people WILL cause you serious issues if you give them that kind of ruleset
IOninja masak, oh yeah? Look up the day when Mojolicious implemented code of conduct and read their channel log. It's nothing but incessant trolling about what does and does not constitute appropriate 13:39
And the rest is silence, because people are utterly confused about whethet they can talk about $X or not 13:40
masak DrEeevil: haven't seen, no. but it does sound like some of Wikipedia's problems.
masak anyway, DrEeevil and IOninja: interesting feedback. thank you. 13:41
DrEeevil masak: augh. yeah, wikipedia is difficult ...
kurahaupo__ I think I like the 3-sentence version: This project runs on good will, so we must all tend and nurture the community. If something you do (or neglect) offends or discomodes another member of the community, it is expected that you will apologize, whether or not you intended it and whether or not you think that taking offense was a reasonable response. And though we sincerely hope it never comes to it, anyone whose activities cause distress may be excl
jast FWIW, here's what the #git channel (avg. ~1250 users) has in terms of a "code of conduct": 13:42
A sense of community thrives on people connecting on a personal level, if even just very slightly. Personally, I like the idea that Git developers and users can be one big chosen family, if you'll forgive the needlessly cheesy metaphor. It doesn't matter if you only visit us for a few minutes, or stick around for years! Of course there will always be wildly different people in that family, and all of us won't hit it off equally well, but that's okay! We're pretty go
full disclosure: I wrote that
IOninja masak, IME it's the regulars who are often the source of hostile tone. They start getting abusive but no one steps up to tell them to stfu because they got ops, or they contributed a billion commits. 13:42
DrEeevil kurahaupo__: that is very badly phrased
jast kurahaupo__: your message got cut off at "distress may be excl" 13:43
masak jast: your message got cut off at "We're pretty go"
jast [...] We're pretty good at disagreeing without killing each other... not even slightly. (We've argued about text editors and we're still here!)
kurahaupo__ DrEeevil: well, it's a work in progress
jast: cat>| 13:44
jast and here's #git's message to regulars: jk.gs/git/helping.html
it's not perfectly applicable to a project with a larger scope than just "support", of course
DrEeevil kurahaupo__: "it is expected that you will apologize" ... that's something I wouldn't accept
("I'm offended that you feel offended. Now whose offense is more important?") 13:45
jast in any case, I find it very valuable to have a light-hearted tone in these things, especially if you want light-hearted interactions in the project ;) 13:47
IOninja masak, and IME, regulars start getting abusive when they're burnt out but are still not taking a break and try to field questions. So they start getting annoyed by "stupid questions" or users not responding fast enough. In fact, I think Code of Burn Out would be more effective than any CoC.
jast yeah, that's what my helping.html was about... it focuses on reducing burn out and frustration when dealing with newcomers 13:48
raschipi kurahaupo__: That takes away all the power apologies have.
jast and the thing IOninja mentioned, about core contributors brushing off/rejecting contributions from some contributors, is a problem that needs a rather delicate discussion between that person and other core contributors... it's basically a responsibility all the most visible project members have 13:50
jast avoiding bad blood in a situation like that is not easy at all, and rules won't change that at all 13:51
kurahaupo__ raschipi: maybe you're right; that simply describes how I try to operate for myself, but bring told that you have to apologise does rather kill the authenticity 13:53
being* told
jast an interesting keyword from real life is "informal social control". emphasis on informal. 14:00
raschipi Rules are best made when dealing with concrete cases, I think. 14:02
raschipi Besides, these Codes of Conduct have the goal to change communities, which we don't need. 14:02
[Coke] Amazing the push back on having a code of conduct. 14:03
raschipi [Coke]: It's a politically charged subject.
raschipi It's no wonder people want to stay out of it. 14:05
ugexe the community should be accepting/rejecting whatever behavior naturally - I don't see why it needs to be defined ahead of time once and for all 16:42
tojo hmm. didn't find any clues from the docs so how I can do like this with unicode names: my $animal='SNAIL'; say "Got \c[$animal]"; ? 16:47
IOninja tojo: $animal.parse-names 16:49
m: my $animal = 'SNAIL'; say "Got {parse-names $animal}" 16:50
camelia Got 🐌
IOninja tojo: it's a recent addition though. Not in any release yet. 16:50
tojo IOninja: oh! ok, have to wait for that then :) 16:52
tojo IOninja: but nice, thx! 16:53
IOninja m: use nqp; say nqp::getstrfromname('SNAIL') 16:54
camelia 🐌
IOninja tojo: you can use internals like that ^ in the meantime.
ilmari u: snail 16:56
unicodable6 ilmari, U+1F40C SNAIL [So] (🐌)
tojo IOninja: great! I'll use that for now, thx :)
ilmari u: slug
unicodable6 ilmari, Found nothing!
ilmari discrimination!
geekosaur in several senses@
(lead type :p ) 16:57
AlexDaniel I se you guys are having fun here 17:05
see*
IOninja ? 17:06
AlexDaniel IOninja: backlogging the discussion about … er… something, and cannot even find the start of it 17:07
IOninja I don't even know what brought this on. 17:10
ilmari tojo used snail as an example of a unicode character name, then I wondered if there was a slug character too 17:11
but the unicode consortium emoji working group is clearly shellist
IOninja All I remember is regulars leaving never to be seen again when SoC was implemented in irc.perl.org/#perl; endless trolling and arguing and banning and heavy air in #mojo when *they* implemented SoC. And now... well, I'm here 17:12
IOninja CoCs are bad news. 17:13
tadzik I'm assuming this is not about Systems on Chips 17:18
IOninja Standard/Code of Conduct
tadzik yeah, I was worried about that 17:21
IOninja No one reads them anyway. The common two reasons they get violated are: cultural differences and people just being assholes. And instead of communicating the issues like humans, people throw CoCs at each other and you always have 1 person who acts as a policeman and 1 person who's automatically in the defensive mode. And arguing ensues whether the policeman is applying the written law correctly and whether the
offender broke any written laws at all.
IOninja NeuralAnomaly: status 17:24
lucs IOninja: My impression: s/like humans/like grownups/
NeuralAnomaly IOninja, [✘] Next release is today. Since last release, there are 59 new still-open tickets (4 unreviewed and 0 blockers) and 20 unreviewed commits. See perl6.fail/release/stats for details
tadzik I know it's kind of a weak argument to say “I've never seen anything good come out of a CoC”, but like IOninja I've seen a lot of bad things happen instead. Including being personally attacked for having a nerve to stand on the other side of the discussion 17:27
AlexDaniel oh, now I see
tadzik it tends to spawn horrible emotions from people on both sides, for some reason 17:27
geekosaur once there's a line drawn in the sand, territorialism kicks in 17:28
faraco Hi guys, where can I find the history of Camelia mascot, on why the the butterfly and what it stands for?
IOninja faraco: raw.githubusercontent.com/perl6/mu...amelia.txt
faraco IOninja: ty 17:28
ilmari faraco: click on her on perl6.org/
which takes you to the link IOninja posted
faraco ilmari: oh, I never did that. ty for that info too. 17:29
IOninja I think places where CoCs work aren't so much about the CoCs' content but about people applying them. There's a big difference between "We try to keep an inclusive and professional atmosphere in the channel. Would you mind trying to avoid using coarse language?" and "First warning! No swearing. Don't you see rules in the TOPIC?" 17:32
tadzik honest question, where have you seen the latter? 17:34
AlexDaniel apologizes and corrects himself: “I see you people are having fun here” 17:35
geekosaur I've seen it in a number of IRC channels
tadzik (not literally of course, but this sort of attitude)
geekosaur and, for what it's worth, the Haskell ops are big fans of the former over the latter
tadzik *shudder*
lucs IOninja: I quit hanging out in #perl on both freenode and MAGnet because I was sick of seeing that kind of attitude way too often. 17:36
IOninja tadzik: you could probably find a closer example if you grepped #mojo's log more closely, but how about this conversation: irclog.perlgeek.de/mojo/2015-09-04#i_11171489 17:40
tadzik: I see they now relaxed the "no swearing" policy, but I recall my time in that channel as incessant arguments over what is and isn't a swear word.
tadzik let me just say "holy shit" while I'm still allowed 17:42
IOninja :) 17:43
tadzik perhaps me not making an immediate connection between "dropping the soap" and "joking about rape" is me being insensitive 17:45
IOninja tadzik: here's a more closer match. A frustrated user says "dammit" and get reprimanded, which is then followed by a discussion on whether it's a swear word and whether it was directed to anyone: irclog.perlgeek.de/mojo/2015-08-31#i_11149797 17:47
IOninja "But a swear word is a swear word." c'mon -_- 17:48
skids Eh, it'll take a while before pushback against prison abuse culture sinks in. Patience required... it's a theme with a lot of inertia behind it. Every once in a while something relatively offensive left rattling around in my head since high school surfaces; I try not to worry about it too much.
tadzik reading this makes me sad, and a part of me believes that there's no way anyone would do that unironically 17:49
geekosaur if only
huf with friends like these, who needs enemies? :) 17:50
tadzik *giggle* 17:51
skids It boils down to people on one side need to learn to apologize gracefully, and people on the other side need to learn how to reprimand gracefully.
tadzik though as IOninja says, this is more of a case of following any ruleset hyperradically rather than an argument against a CoC itself 17:52
skids Yesah I'm not going to read that URL, it's probably not worth the time and cortisol
tadzik I'd say it's more of a case of not everything deserving a reprimand or an apology 17:53
pmurias don't we have a CoC somewhere already? 17:54
tadzik and people disagreeing on whether or not that's true :)
huf *i* think all CoC-s should start by banning terrible genital puns based on "CoC". 17:55
that sort of lampshade hanging should set the mood. 17:56
IOninja SoC
tadzik this channel wouldn't be the same without terrible puns 17:58
IOninja m: class This-Channel {}; say (This-Channel but role Terrible {}.^pun) eqv This-Channel 17:59
camelia Cannot mix in non-composable type Terrible into object of type This-Channel
in block <unit> at <tmp> line 1
IOninja m: class This-Channel {}; say (This-Channel does Terrible {}.^pun) eqv This-Channel
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared name:
Terrible used at line 1
IOninja :(
m: class This-Channel {}; say This-Channel cmp This-Channel without role Terrible {}.^pun 18:02
camelia Same
IOninja >:D
faraco camelia: say so if "IrIX".contains("IX")
m say so if "IrIX".contains("IX")
IOninja that's broken
faraco m: say so if "IrIX".contains("IX")
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
if used at line 1
IOninja m: say "IrIX".contains("IX") 18:03
camelia True
IOninja m: "IrIX".contains("IX") and .not.say
camelia True
IOninja :)
faraco ty 18:04
IOninja faraco: don't use the last one. It was a joke :)
faraco the last one is too 'clever' for me. Take a bit for me to figure that out. 18:05
IOninja m: dd $_ 18:06
camelia Any $_ = Any
IOninja faraco: ^ $_ exists and at this moment it's false. The `and .not.say` bit will run only if "IrIX".contains("IX") is true (and it is) and it's equivalent to $_.not.say and since $_ is false, it's .not is True and .say prints it, making it look as if it printed the result of .contains() when it didn't 18:07
raschipi m: "IrIX".contains("XI") and .not.say 18:08
camelia ( no output )
faraco hmm, so $_ is false, which .not.say make $_ true and both case (contains) will run, correct? 18:12
mst masak: many CoCs appear to be designed more as a political bludgeon than a true CoC - the annoying thing is that can end up being self fulfilling in certain ways 18:13
note that irc.perl.org has an SoC but it's mostly a write-up of what we were doing anyway
IOninja faraco: no, $_ stays false calling .not on it returns True (without modifying it) and calling .say on it prints it
mst which I worked hard on, including with e.g. libertarian regulars, to ensure it didn't seem like a weapon
and then a bunch of people quit irc.perl.org #perl *anyway* because the mere fact we had one was EEEVILLL or something
even though I was 100% confident it didn't forbid anything I wouldn't've banned you for anyway 18:14
IOninja m: dd [ .so, .not, .not.not.not.not.not ] 18:14
camelia [Bool::False, Bool::True, Bool::True]
mst masak: the contributor covenant is a trainwreck though; I've seen it updated by the author *specifically* to be able to try and get a specific person kicked out of a project after I pointed out it didn't actually cover the behaviour they were trying to get the target removed for 18:15
timotimo i thought a big part of CoC is to signal to certain minority groups that they're not going to get harrassed without the harrasser being asked to stop
mst any value judgement about the target's behaviour aside, a willingness to do that made it clear it's a weapon, not just a defence 18:16
mst timotimo: that's where they started off, then some of them are to signal that if you don't hold the correct politics you aren't welcome 18:16
timotimo mhm 18:16
there's a really broad spectrum there, though 18:17
Guest78040 i have perl6.bat from racuda star, but I want to run under msys2 shell
@ "C:\rakudo\bin\moar" --execname="%~dpf0" --libpath="C:\rakudo\share\nqp\lib" --libpath="C:\rakudo\share\nqp\lib" --libpath="C:\rakudo/share/perl6/lib" --libpath="C:\rakudo/share/perl6/runtime" C:\rakudo\share\perl6\runtime\perl6.moarvm %*
mst like, there is a big difference between "misgendering people is bad" and "you must hold the Correct opinions about all LGBT+ issues"
the former will get you banned by me, no questions; the latter - behave politely in project spaces and I don't see it's any of my business 18:18
Guest78040 also running ufo for the first time with perl6.bat in a powershell window, I get a Malformed UTF-8 error message 18:19
timotimo but expecting someone who's a firm believer that queer people are "just sick" to not misgender (or similar things) is differnet from asking them to hold different opinions?
Guest78040: ufo is a tool that hasn't been useful for a long time; where did you see the recommendation for it? we should probably remove that
faraco IOninja: oh, sounds like it changes the 'copy'?
mst timotimo: yes. 18:20
IOninja faraco: I guess... It makes a new value for you, without affecting the old one.
faraco urm, I need to learn more Perl 6, haha. The last one I did any was around a month ago.
Guest78040 ufo for building the JSON::Tiny example
mst timotimo: just like I expect somebody to be capable of being an atheist without talking about 'invisible sky father' and similar pointless bullshite
IOninja Guest78040: it's probably just an outdated README. People use `zef` these days.
buggable: eco zef
buggable IOninja, zef 'It's like [cpanm] wearing high heels with a tracksuit': github.com/ugexe/zef
timotimo that makes sense 18:21
IOninja Guest78040: I think it's included with latest Rakudo Star, so you may already have it. If not, ^ that link has bootstrapping instructions
Coleoid m: class Channel { }; my $p6-channy = Channel.new(); role Terrible-Pun { method pun() { "Our zoo is so small--just a one-lioner." } }; my $p6-channy-puns = Channel.new() does Terrible-Pun; say $p6-channy-puns.pun; say $p6-channy eqv $p6-channy-puns;
camelia Our zoo is so small--just a one-lioner.
False
Coleoid WEEE
are the channy-puns 18:22
Guest78040 zef gives me an error when running msys shell because no script file is there. zef works under powershell because perl6.bat is in path
IOninja No idea, sorry. I've never even heard of msys shell
Guest78040 msys2 shell 18:23
basically msys2 is mingw bash port for windows 7 & 10
IOninja Is that the "Ubuntu on Windows" stuff? 18:24
jonadab No.
IOninja Ok
Guest78040 trying not to need a virtual machine all the time, like Ubunto on Windows
raschipi timotimo: I have seen those people that insist transgender people are sick to actually cooperate a lot with them sometimes. Transgender activists want medical care to be paid by the government to help transgender kids just kicked out of home. They are actually closer than one would suppose at first.
jonadab msys has been around a while (not as long as Cygwin). 18:25
jonadab The subsystem for Ten is superior in most respects but requires that you be running Ten, which a lot of folks don't want to do because of the UI changes. 18:26
Guest78040 transgender kids are high risk of suicide, that is one good reason they need a lot of help. without a lot of help, they might not survive.
timotimo that makes sense; i was thinking of "just mentally ill", though. people who say that would also tell them to "just get over it" or it's "just in your head", though. at least that's my expectation
no doubt about them needing help 18:27
unclechu `rakudo-star` package supposed to provide `zef` binary, right?
IOninja unclechu: not sure. What sort of package? (what os?) 18:28
unclechu IOninja: fedora 25 workstation
pmurias timotimo: I don't minding using whatever gender a legally mentally ill person requests
unclechu sent a long message: unclechu_2017-03-17_18:28:49.txt - matrix.org/_matrix/media/v1/downlo...uWBhldfrQp
timotimo rakudo-star on fedora25 is currently at version 2017.01-2
Guest78040 zef is there but wants to run perl6 which defaults to perl6.bat in a windows shell 18:29
mst excuse me, everybody 18:29
timotimo unclechu: you would have gotten a more helpful output from "dnf list" rather than "dnf search"
mst allow me to make it really fucking clear that my "you may hold whatever opinions you like on trans issues elsewhere provided you're civilised here" policy requires you to keep that particular discussion *elsewhere* 18:29
Guest78040 but msys2 cant find perl6 in path because it is a .bat file. I need to find a script without the .bat and written in bash not microsoft 18:30
IOninja unclechu: I think it was included only in the latest Rakudo Star 2017.01
timotimo fair enough
IOninja mst: where do we stand on invisible sky people, though? :)
pmurias sorry
mst and I will happily mallet anybody who tries to create an argument about it in here no matter whether I agree with their attitude to it or not
unclechu sent a long message: unclechu_2017-03-17_18:31:02.txt - matrix.org/_matrix/media/v1/downlo...AYuNkYtvCy 18:31
IOninja: as i can see i have it
IOninja Yup, Looks it 18:31
raschipi Sorry for bringing the political position here, it was just an example to show it's more complicated than people suppose at first.
mst IOninja: I'm an atheist. larry's a christian. chip salzenberg is a pope of discordianism. let's not.
IOninja :) 18:32
andrzejku huh I need to say that vagrant is so slow
mst raschipi: the problem fundamentally is that particular topic tends to rapidly attract the stupidest sort of rightist *and* the stupidest sort of leftist, neither of which groups are actually basing their opinions on scientific/medical reality, and it always ends up as a complete disaster 18:34
Guest78040 its the christian thing to help them. not sure how to help, but they are a severe sucicide risk population
mst Guest78040: I SAID WE ARE NOT DISCUSSING THIS
next person to violate that gets a free 24h +q for the sake of #perl6's ability to discuss perl6
and I'mma shut up about it too 18:35
IOninja \o/
mst raschipi: you're welcome to reply to my last comment via /msg if you wish
timotimo right. there is no discussion here. i'm pretty sure everybody agrees that trans people are very allowed to get help
so yeah. no discussion to be had
timotimo also shuts up
[Coke] "here, let me keep talking while I'm shutting up." :P 18:36
timotimo andrzejku: what kind of slow are you experiencing?
maybe you need to turn hardware virtualization on in your BIOS, for example
or maybe the slowness is just downloading installation media or prebuilt images or what 18:37
andrzejku timotimo no it is like i pass vagrant command
and i even need to wait for help output 18:38
timotimo oh, huh
what does "time" say? is there a lot of time spent in "system", or is it all "user"?
andrzejku timotimo wait I will try to analyze it
timotimo i've actually got to go AFK for a bit anyway. i hope you can figure it out! 18:39
andrzejku it is like 1 second
to wait
timotimo it's 0.54s on my machine, 0.04s system time 18:40
so yeah 18:41
i guess that's expected? maybe it's the language it's written in. ruby, right?
andrzejku yep
timotimo however as I know perl6 is slower than ruby
so I shouldn't complain for that
timotimo a bit
i suppose it depends on what vagrant does, exactly. for example, does it have a mechanism that allows plugins to put new subcommands in? 18:42
andrzejku nevermind 18:43
timotimo that'd mean it'd have to look through whatever module installation/registration mechanism they have
andrzejku do you use emacs?
anyone
AlexDaniel yes 18:44
andrzejku AlexDaniel for which language?
AlexDaniel andrzejku: for almost everything, actually 18:45
andrzejku AlexDaniel nice I know only a bit emacs 18:46
AlexDaniel but I think to use it more, for C++
AlexDaniel would you lie to stay my emacs mentor? 18:48
AlexDaniel andrewalker: I don't think so. A lot of things I do are based on my highly customized keyboard layout
oops.
andrzejku: ↑
everyone has to find his own path, I think :) 18:49
andrzejku AlexDaniel :) 18:51
cale2 Now this is a great book cover: www.nostarch.com/Rust 18:53
andrzejku cale2 how is Rust? 18:54
cale2 andrzejku: It's not for the impatient :) I want to build a project in either Rust or Golang. Maybe I'll do the same project in both and compare my experiences at that point. Then I'll get back to you lol. 18:59
andrzejku cale2 nice :)
SmokeMachine m: class C {proto method xxx(|) {*}; multi method xxx(Str(Cool:D) $n) {1}; multi method xxx(Str(Cool:D) $n, Int:D $p) {2}}; say C.new.xxx("42") # <- is that expected? 19:03
camelia Ambiguous call to 'xxx'; these signatures all match:
:(C $: Cool:D $n, *%_)
:(C $: Cool:D $n, *%_)
in block <unit> at <tmp> line 1
SmokeMachine why its ignoring the Int:D $p?
RabidGravy boom!
AlexDaniel m: class C {proto method xxx(|) {*}; multi method xxx(Str(Cool:D) $n) is default {1}; multi method xxx(Str(Cool:D) $n, Int:D $p) {2}}; say C.new.xxx("42") 19:10
camelia Ambiguous call to 'xxx'; these signatures all match:
:(C $: Cool:D $n, *%_)
:(C $: Cool:D $n, *%_)
in block <unit> at <tmp> line 1
jnthn m: say Str ~~ Cool:D; say Cool:D ~~ Str 19:14
camelia False
False
jnthn A multi with a coercion type gets entered into the dispatch table twice 19:15
jnthn It seems that thanks to the :D that entry then ends up conflicting with itself 19:15
Potentially because at the point it does this double insert it doesn't recognize that Cool:D is a definiteness type 19:16
So the result is they end up tied in the candiate sort 19:17
Worth an RT so it can be looked at further.
IOninja m: class C {multi method xxx(Str(Cool:D) $z) {1}; }.new.xxx: 'z' 19:18
camelia Ambiguous call to 'xxx'; these signatures all match:
:(C $: Cool:D $z, *%_)
:(C $: Cool:D $z, *%_)
in block <unit> at <tmp> line 1
IOninja m: class C {multi method xxx(Str:D(Cool:D) $z) {1}; }.new.xxx: 'z'
camelia ( no output )
SmokeMachine jnthn: ok, I'll RT 19:36
hankache hola #perl6 19:39
raschipi oi! 19:40
SmokeMachine IOninja: I was trying to solve the problem (rt.perl.org/SelfService/Display.ht...n-1452828) of index this way: www.irccloud.com/pastebin/J2IFVG2m/ 20:14
IOninja: but I found this "multi bug"... what's the way you suggest me to fix that? 20:15
IOninja I won't have time to look in the next 90 hours. Perhaps someone else could help you? 20:16
SmokeMachine IOninja: ok, thanks! sorry to bother...
Geth doc: 1ba43da5a4 | (Samantha McVey)++ | doc/Type/Str.pod6
Remove sprintf example with $^V which does not exist in P6
 20:32
samcv not sure why we had $^V in an example but hey. it's gone now 20:34
must have been written forever ago
samcv Tom Browder 2016-08-31 06:27:16 -0500 557) NYI sprintf "version is v%vd\n", $^V; # Perl 6's version 20:35
oh actually not that old. maybe just a mistake
geekosaur looks to me like it was intended to be a placeholder? because that code is perl5 20:36
so intended to mean "we need the perl 6 version of this chunk of perl 5 here"
samcv i gueqss 20:37
IOninja m: { $^V.say }(42) 20:38
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of $^V variable; in Perl 6 please use $*PERL.version or $*PERL.compiler.version
at <tmp>:1
------> 3{ $^V7⏏5.say }(42)
samcv the example above it was fine but there's no $^V
IOninja
.oO( *sigh* )
samcv and using $*VERSION wouldn't work because we have letters
it would look really weird
timotimo yeah, it's sometimes pretty annoying to have these perl5-warnings 20:39
samcv is it?
raschipi When it's valid P6 code, it can get in the way. 20:40
timotimo yup
samcv ah
samcv it won't let you use any of the capital $^Letters i think 20:42
m: { $^b.say; $^y.say}(42,11) 20:43
camelia 42
11
samcv hmm didn't know i could name them anything i wanted. always had done a,b, etc in order 20:44
well or swapped if i wanted it reverse
IOninja m: { say $^ッ, $^ツ, $^ヅ }(1, 2, 3) 20:46
camelia 123
AlexDaniel m: { say $^ッ, $ツ, $ヅ }(1, 2, 3) 20:47
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$ツ' is not declared
at <tmp>:1
------> 3{ say $^ッ, 7⏏5$ツ, $ヅ }(1, 2, 3)
AlexDaniel ah
IOninja :/
:)
AlexDaniel every character is different…
IOninja m: say ('A'..'Z').grep({ (try "\$ = \$^$_.so; 1".EVAL) ?? $_ !! "-"}).join 20:48
camelia ABCDEFGHIJKLMNOPQRSTUVWXYZ
IOninja orly
samcv yeap 20:51
m: { $^a.say; $^y.say}(42,11)
camelia 42
11
SmokeMachine m: use Test; like Blob.new(<1 2 3>).WHICH, /^ "Blob"/
camelia Cannot resolve caller like(ObjAt, Regex); none of these signatures match:
(Str $got, Regex $expected, $desc = "")
in block <unit> at <tmp> line 1
SmokeMachine bisectable6: use Test; like Blob.new(<1 2 3>).WHICH, /^ "Blob"/ 20:52
bisectable6 SmokeMachine, Bisecting by output (old=2015.12 new=79f2681) because on both starting points the exit code is 1
SmokeMachine, bisect log: gist.github.com/50a14d13c539fc8fd3...79c3ffce74
SmokeMachine, (2016-03-04) github.com/rakudo/rakudo/commit/ea...341e986a91
IOninja oh. I need to use map not grep -_-
IOninja needs to get some medicine -_-
m: say ('A'..'Z').map({ (try "\$ = \$^$_.so; 1".EVAL) ?? $_ !! "-"}).join 20:53
camelia --------------------------
IOninja That's pretty heavy-handed since not all of them are magic vars :/
Oh maybe my eval is broken 20:55
nope all of them are banned 20:57
SmokeMachine I think it ^^ is breaking the S32-container/buf.t...
IOninja m: say ('A'..'ZZ').map({ try { ('{ $^' ~ $_ ~ ' }(1); 1').EVAL } ?? '' !! $!.^name }).Bag 20:58
camelia bag((676), X::Syntax::Perl5Var(26))
IOninja m: { say $^・, $^ー, $^゠}(3, 2, 1) 20:59
camelia 5===SORRY!5=== Error while compiling <tmp>
Missing required term after infix
at <tmp>:1
------> 3{ say $^7⏏5・, $^ー, $^゠}(3, 2, 1)
expecting any of:
prefix
term
samcv huh 21:00
heh 21:01
AlexDaniel bisectable6: old=2016.04 use Test; like Blob.new(<1 2 3>).WHICH, /^ "Blob"/
bisectable6 AlexDaniel, Bisecting by exit code (old=2016.04 new=79f2681). Old exit code: 0
AlexDaniel, bisect log: gist.github.com/c084de2afffd378b17...c705c3c208
AlexDaniel, (2017-03-16) github.com/rakudo/rakudo/commit/65...db2f276ac7
AlexDaniel SmokeMachine: are you sure this ↑ is not what you were looking for? 21:02
SmokeMachine AlexDaniel: this may be the commit... 21:02
AlexDaniel SmokeMachine: big picture here: gist.github.com/Whateverable/d1712...ac0c4be4e5 ;
;)
cognominal hi, is there any translator from a yacc grammar to a Perl 6 one ? 21:09
samcv NeuralAnomaly, release 21:10
AlexDaniel cognominal: I only know about one for ANTLR (github.com/drforr/perl6-ANTLR4)
samcv NeuralAnomaly, status
AlexDaniel NeuralAnomaly: status
NeuralAnomaly samcv, [✘] Next release is today. Since last release, there are 60 new still-open tickets (5 unreviewed and 0 blockers) and 20 unreviewed commits. See perl6.fail/release/stats for details
AlexDaniel, [✘] Next release is today. Since last release, there are 60 new still-open tickets (5 unreviewed and 0 blockers) and 20 unreviewed commits. See perl6.fail/release/stats for details
timotimo i know we have antlr-to-perl6, but i know nothing about yacc
cognominal AlexDaniel, thx 21:11
AlexDaniel cognominal: I wonder what kind of mess you would get if you go yacc→ANTLR→perl6, but maybe worth a try? :)
if that's possible at all 21:12
cognominal I don´t know antlr so I can´t say 21:13
IOninja samcv: s/today/tomorrow/; 21:16
I wanna first smoke the ecosystem tonight and will release tomorrow morning, EST
cognominal AlexDaniel ANTLR is LL(*) while yacc is LALR and I can´t tell from memory the difference. But wp has a say : en.wikipedia.org/wiki/LALR_parser#LL_parsers 21:18
IOninja Unless I'm still wide awake when I finish with ecosystem :) 21:18
bazzaar o/ perl6 21:19
cognominal AlexDaniel, it seems that LL is more intuitive than LR but more powerful 21:20
AlexDaniel casts TimToady :) 21:21
sena_kun second "more" should be less, right?
bazzaar Hi, I'm having fun playing with Grammar / Actions, and got an unexpected result (pastebin.com/wAH8VX4c) 21:22
sena_kun LR parsers are hard to deal with by humans, afaik. LL is much easier to understand, though less efficient. Hence yacc/bison/etc.
bazzaar I'd appreciate if someone could take a quick look and say why 'Cash' literal seems to have the observed effect 21:24
AlexDaniel bazzaar: ok, so what's exactly unexpected? 21:26
ah 21:27
bazzaar AlexDaniel: the result is down at the bottom of the pastebin , :Cash(....) is unexpected (for me) 21:28
AlexDaniel bazzaar: interesting. Seems like it's just a pair that's .perl-ified differently
m: say ("Cash Flow" => ["48.070000", "£1.000000", "£48.07", "1.69\%"]).perl 21:29
camelia "Cash Flow" => ["48.070000", "£1.000000", "£48.07", "1.69\%"]
AlexDaniel m: say ("Cash-Flow" => ["48.070000", "£1.000000", "£48.07", "1.69\%"]).perl
camelia :Cash-Flow(["48.070000", "£1.000000", "£48.07", "1.69\%"])
AlexDaniel bazzaar: ↑
bazzaar I wondered if Cash is a reserved word?
AlexDaniel no
AlexDaniel bazzaar: it's just that .perl is smart enough in this case to dump it slightly differently 21:29
m: say (hello => world).perl 21:30
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
world used at line 1. Did you mean 'words'?
AlexDaniel m: say (hello => ‘world’).perl
camelia :hello("world")
AlexDaniel m: say (‘hello’ => ‘world’).perl
camelia :hello("world")
AlexDaniel m: say (‘hello with a space’ => ‘world’).perl
camelia "hello with a space" => "world"
cognominal indeed 21:31
AlexDaniel s: :42, 'pair'
SourceBaby AlexDaniel, Something's wrong: ␤ERR: ===SORRY!=== Error while compiling -e␤Malformed radix number␤at -e:6␤------> put sourcery( :42<HERE>, 'pair' )[1];␤ expecting any of:␤ number in radix notation␤
AlexDaniel s: (:42), 'pair'
SourceBaby AlexDaniel, Something's wrong: ␤ERR: ===SORRY!=== Error while compiling -e␤Malformed radix number␤at -e:6␤------> put sourcery( (:42<HERE>), 'pair' )[1];␤ expecting any of:␤ number in radix notation␤
AlexDaniel s: (:42abc), 'pair'
SourceBaby AlexDaniel, Something's wrong: ␤ERR: Type check failed in binding to '&code'; expected Callable but got Nil (Nil)␤ in sub do-sourcery at /home/zoffix/services/lib/CoreHackers-Sourcery/lib/CoreHackers/Sourcery.pm6 (CoreHackers::Sourcery) line 42␤ in sub sourcery at /home/zoffix/services/lib/CoreHackers-Sourcery/lib/CoreHackers/Sourcery.pm6 (CoreHackers::Sourcery) line 33␤ in block <unit> at -e line 6␤␤ 21:32
AlexDaniel s: (:42abc), 'perl'
SourceBaby AlexDaniel, Sauce is at github.com/rakudo/rakudo/blob/79f2...Mu.pm#L542
AlexDaniel phew…
sena_kun well, now I didn't get it. What is the difference between those two lines? They seem completely the same for me, though diff shows that `("Cash-Flow"` part is different. 21:32
bazzaar AlexDaniel: aah, ok ... so false alarm right, just different notations for <pair> for keys with or without a space? 21:33
AlexDaniel sena_kun: the difference is that ｢:hello foo("world")｣ is not a valid perl6 syntax 21:33
sena_kun ah, I'm stupid, nevermind.
AlexDaniel, I've noticed the space now, right. Thanks.
AlexDaniel bazzaar: yes
bazzaar: there is a list of different ways to write pairs here: docs.perl6.org/type/Pair 21:34
bazzaar: in theory, .perl can use any of them, as long as it is valid code :)
bazzaar AlexDaniel: thanks for clearing that up, I'd have been staring at that for the next month of Sundays :)
AlexDaniel bazzaar: if want to output this for users, it is probably better to stringify it yourself the way you want 21:35
bazzaar: I don't think we even have any guarantees as to what .perl will give exactly, so it may change
bazzaar AlexDaniel: thanks for the advice, I will look into doing just that :) 21:37
IOninja ugexe: you around? I forget does zef need `git` on Windows? Someone's having "no fetching backends" issue: www.reddit.com/r/perl6/comments/60...using_zef/ 22:27
AlexDaniel u: INVISIBLE 22:44
unicodable6 AlexDaniel, U+2062 INVISIBLE TIMES [Cf] (⁢)
AlexDaniel, U+2063 INVISIBLE SEPARATOR [Cf] (⁣)
AlexDaniel, U+2064 INVISIBLE PLUS [Cf] (⁤)
AlexDaniel okay
hobbs Invisible Times was never as popular as Medieval Times 22:46
timotimo %) 22:47
it's called "the dark age" 22:48
IOninja m: (2.5..5.5).int-bounds 22:49
camelia Cannot determine integer bounds
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
IOninja why not, I wonder. 22:50
(there's a check for $!min.floor == $!min)
ugexe IOninja: zef does not need git on windows technically. But the ecosystem lists all source-url as .git links, so in practice yes you do 22:54
if they edited the package list to change the github urls to github tar.gz urls then it would work without git 22:55
IOninja ugexe: not all, it normalizes URLs but doesn't change protocol. 22:56
ugexe: it wouldn't need git if they were https:// URLs, correct?
ugexe it still uses git for those as well
IOninja Oh, OK.
hobbs if they're https urls of git repos, it does :) 22:57
timotimo does our meta.json format have some way to say "use any of these URLs to fetch the source"?
IOninja ugexe: It could just download .zip file instead, no?
github.com/zoffixznet/perl6-WWW/ar...master.zip 22:58
ugexe IOninja: yes, but if zef is supposed to guess that stuff then it needs logic on a per-git-server basis
IOninja github.com/zoffixznet/perl6-WWW/ar...5db475.zip
ugexe e.g. i doubt mangling the urls to the archive is the same got bitbucket, gitlab, etc
IOninja Seems a branch/SHA can be used.
ugexe yeah, thats how tbrowder versions his modules in the ecosystem 22:59
timotimo hm? how? 22:59
putting a link to the tagged zipfile into the source url?
ugexe you link the source-url in your meta file to the archive .tar.gz, and then link the meta6.json from that commit to the ecosystem alongside other versions
IOninja ugexe: are you open to people implementing these URL manglers and submitting a PR? 23:00
Feels like even if just GitHub were implemented that'd cover like 95% of our ecosystem
ugexe yeah I would, but im not sure where that logic can be cleanly used 23:02
IOninja Cool.
ugexe but its easier to just mangle an ecosystem packagelist.json
IOninja
.oO( make ecosystem figure this out.... )
 23:03
Cool. I'll add this to my todo list to take a crack at then :) In May, unless someone beats me to it.
(I mean the ecosystem mangling packagelist.json)
timotimo will we be accepting multiple URLs for downloading stuff in META6.json? 23:14
ugexe thats a per-ecosystem policy... source-url is not meta6 spec for instance 23:15
timotimo oh, ok 23:15
ugexe so there is nothing stopping that option other than adaption 23:15
ugexe `Fetching backends [git,wget,curl] not available to handle XXX://github.com/ugexe/Perl6-ecosystems.XXX` 23:27
anyone have a better idea for the error I should display to the user?
ugexe (on windows that would also have `,pswebrequest]` for instance) 23:29
although I think zef invokes wget and curl in a way that their powershell aliases work
timotimo but "shell" won't directly go through to the powershell, rigth? 23:33
ugexe no, it has to be invoked 23:42
timotimo right 23:43

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!