»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.
MasterDuke timotimo: what are you working on? 00:02
timotimo debugserver stuff 00:11
have you looked at that at all yet?
MasterDuke i sort of followed your session with Zoffix, haven't used it yet, but plan to 00:12
timotimo OK 00:13
then if i explained, it wouldn't make terribly much sense
but here goes
comborico1611 say ~$/ if "abcdef" ~~ / bc.e / # Is ~ concatenating anything or is it being overloaded to another task?
evalable6 bcde
timotimo whenever moar mentions an object to you, it will allocate a handler, which is just an integer that you can use to refer to the given object
however, you can't tell if two handles are actually the same object underneath
so i'm implementing a command that'll give you a list of lists of handles that point at the same object 00:14
comborico1611: ~ and ~~ are two different operators
tobs comborico1611: it coerces $/ to a string
timotimo oh, you mean ~ in front of $/ there?
yes, what tobs said
comborico1611 timotimo: Yes
MasterDuke timotimo: is that handler ever visible in HLL-land (without using the debugserver)? 00:15
comborico1611 tobs: Thanks!
timotimo nope 00:18
better to call it handle than handler
since it's a passive thing, whereas handler sounds more active
MasterDuke related to MVMFrameHandler? 00:20
timotimo nope 00:23
it's literally just a 64bit integer
associated with any MVMObject that was part of some response from the debugserver 00:24
MasterDuke cool. i'm trying to profile compiling the core setting during a jvm build with visualvm. however, i've never used it before, so i have no idea what i'm doing 00:28
timotimo oof, yeah, that's a lot to learn at once i bet 00:29
the output from "all lex" is much nicer now that handle numbers are coalesced
buggable New CPAN upload: ArrayHash-0.3.tar.gz by HANENKAMP cpan.metacpan.org/authors/id/H/HA/...0.3.tar.gz 01:39
Zoffix 👂📌💧 01:47
Zoffix So quiet :) 01:48
Listen to this woman's beautiful singing: youtu.be/SAB0hlaPgFc?t=1m18s
buggable New CPAN upload: DOM-Tiny-0.4.0.tar.gz by HANENKAMP cpan.metacpan.org/authors/id/H/HA/...4.0.tar.gz 01:49
Geth perl6.org/bs4: 78cdf936be | (Zoffix Znet)++ | templates/home.html.ep
Add "Lazy Evaluation" to slides
 02:44
jmerelo releasable6: status 04:22
releasable6 jmerelo, Next release will happen when it's ready. 0 blockers. 266 out of 266 commits logged
samcv jmerelo: should be releasing MoarVM imminently 04:42
jmerelo samcv: great news and great job. 04:45
samcv thanks :)
lots of stuff in the changelog for MoarVM at least
jmerelo samcv++
samcv probably in rakudo as well though i haven't followed it super closely :)
always so much going on :) 04:46
jmerelo samcv: including pull requests in github/linguist :-)
Geth perl6-pod-to-bigpage: 165791cf05 | (JJ Merelo)++ | t/render.t
Adds test for render refs #14
 05:08
zostay m: class F { ... }; class F:ver<1> { }; say F.^ver 05:10
camelia (Mu)
zostay m: class F:ver<1> { ... }; class F { }; say F.^ver
camelia v1
zostay apparently, it's important to make sure the first reference to a class as the adverbs set, even if it's a stub 05:12
buggable New CPAN upload: HTTP-Headers-0.1.tar.gz by HANENKAMP cpan.metacpan.org/authors/id/H/HA/...0.1.tar.gz 05:19
New CPAN upload: Hash-MultiValue-0.5.tar.gz by HANENKAMP cpan.metacpan.org/authors/id/H/HA/...0.5.tar.gz 05:29
Geth perl6-pod-to-bigpage: JJ++ created pull request #22:
Eliminates anchors for indexed terms		 05:55
jmerelo 177 perl6 & rakudo & parrot users in StackOverflow. t.co/UecHGtGwFf Sigh 06:52
Actually, 179 if we include parrot. 06:53
jmerelo For the sake of comparison, there are 40596 for perl. To be fair, a few of that will be for perl6. I'll add that tag if I find them. Anyway. 06:57
moritz and by "users" you mean "people who asked questions", right? 07:01
'cause I'm a perl 6 user on stackoverflow, but don't show up there
jmerelo moritz: I've tried to find the way to include people to answer. There should be one... 07:02
Hm, there's a Comment table 07:04
moritz jmerelo: it's an OK metric; just want to be clear on what you mean by "users"
El_Che :hi 07:10
moritz \o El_Che 07:17
El_Che mmm, zsh does not set a $SHELL parameter it seems 07:18
I wanted to provide a script to set the path for perl6
not does sh 07:19
nor
donpdonp mycomputer$ echo $SHELL
/usr/bin/zsh
El_Che donpdonp: I suspect it's set by the OS. A Ubuntu container leaves it empty if you just type "zsh" 07:20
maybe it must be the login shell
donpdonp hmm ok
El_Che (however, zsh overrides the $SHELL set by bash, by removing it)
and other way is of course to look what files are in a users home 07:22
Geth doc: fb56305876 | (Jonas Kramer)++ | doc/Type/Any.pod6
Fix link.
 09:37
synopsebot Link: doc.perl6.org/type/Any
Geth doc: 55210f24c2 | 陈梓立++ (committed using GitHub Web editor) | META6.json
Update META6.json
 10:49
doc: a065a6b1b6 | 陈梓立++ (committed using GitHub Web editor) | META6.json
Update META6.json
Zoffix With people doing constantly doing `release: status` I wonder if we should add a `release` alert level to alerts.perl6.org and post releases to it. That way people can sub to it through RSS 11:06
moritz I see alert.perl6.org mostly as a tool for users
whereas the release status, especially before a planned release, is more of interest to developers 11:07
jmerelo So, somebody answered my question in StackOverflow. What a surprise :-)
Here's the query that includes answers and comments data.stackexchange.com/stackoverfl...er-tag-set There are 459 persons participating in such questions. 11:08
Zoffix moritz: I don't follow. A new release of the compiler is surely of interest to users. Especially since the date of it isn't very predictable.
Zoffix Also, downstream packagers are users. 11:09
jmerelo Zoffix: putting it in p6alerts can't hurt. But don't take away our lovely releasable6 bot from us.
AlexDaniel well, they can also subscribe to the mailing list right now 11:10
Zoffix: but maybe we should only issue alerts for rakudo star? 11:11
or both
moritz those sound like notifications, not alerts
if we say that alerts.perl6.org should serve for general notifications, not just alerts, I'm fine with that 11:12
just saying that this seems like a bit of a departure from the original goals (which might be fine too)
Zoffix moritz: right. good point. 11:13
And fo Rakudo Star there's already an RSS feed from rakudo.org's posts feed
(which isn't linked to from anywhere ATM github.com/perl6/rakudo.org/issues/4 ) 11:14
stmuk "severity" doesn't really match with a release notification (unless the release is a security fix) 11:16
lizmat weekly: data.stackexchange.com/stackoverfl...er-tag-set 11:20
notable6 lizmat, Noted!
AlexDaniel weekly: clear
notable6 AlexDaniel, Moved existing notes to “weekly_2018-04-25T11:20:50Z”
AlexDaniel weekly: data.stackexchange.com/stackoverfl...er-tag-set
notable6 AlexDaniel, Noted!
AlexDaniel (had some notes from the previous week) 11:21
lizmat AlexDaniel++
robertle shameless plug for short blog post: www.linkedin.com/pulse/less-stern-...rt-lemmen/ 11:25
AlexDaniel you can plug it yourself: 11:27
weekly: www.linkedin.com/pulse/less-stern-...rt-lemmen/
notable6 AlexDaniel, Noted!
AlexDaniel :)
robertle: so why didn't you make a perl 6 module? :) 11:29
robertle this isn't javascript, right? ;) 11:30
AlexDaniel what do you mean?
Zoffix robertle: well, we could do the same thing ether did and test how well our ecosystem supports it.
Unless the Hebrew 666 module is still in the ecosystem... 11:31
robertle AlexDaniel: that 3-line modules may not be the way to go. 11:32
AlexDaniel ah, that thing
robertle perhaps if they actually do something useful
AlexDaniel I don't think it's a problem for Acme::*
Zoffix doesn't look like it's still in there 11:34
Lots of easy rakudo tickets to fix: github.com/rakudo/rakudo/issues?q=...abel%3ALHF 11:43
Step right up \o/ Be a hero :)
huggable: zscript 11:44
huggable Zoffix, Helper script for Rakudo Perl 6 core development: github.com/zoffixznet/z
El_Che Ik have been bitten a zillion times by this: 11:46
m: class Result { has $.latest is rw; }; my $r = Result.new; $r.latest("Foo"); say $r;
camelia Too many positionals passed; expected 1 argument but got 2
in method latest at <tmp> line 1
in block <unit> at <tmp> line 1
Zoffix ^ it doesn't know how to handle forks when `z init`; but just run it and afterwards do `git remote remove origin; git remote add origin github.com/YOUR-USER/rakudo` and the same for all the other repos
El_Che m: class Result { has $.latest is rw; }; my $r = Result.new; $r.latest = "Foo"; say $r;
camelia Result.new(latest => "Foo")
El_Che the . notation make the method syntax more natural for me 11:47
Am I talking crazy?
AlexDaniel well, not crazy for sure
but what about 11:48
m: class Result { has $.latest is rw; }; my $r = Result.new(latest => "Foo"); say $r;
camelia Result.new(latest => "Foo")
AlexDaniel m: class Result { has $.latest is rw; }; my $r = Result.new: :latest("Foo"); say $r; 11:48
camelia Result.new(latest => "Foo")
El_Che AlexDaniel: constructors are OK. It mostly for values written after 11:49
AlexDaniel right
Zoffix El_Che: maybe invent some mnemonic to remember it easier? **A**ttributes are **A**ssigned into. Or "you use `$!foo = "bar"` inside the class; and outside it's the same" 11:51
To me it makes more sense to assign into them than call them as a method. Makes the whole "we auto-gen a method for you" thing that much more hidden. 11:52
El_Che yeah, I wish the compiler would change the error to "El_Che, what are you doing, bro?"
Zoffix I wonder if CommaIDE would be able to detect this sort of stuff
weekly: Lots of easy rakudo tickets to fix: github.com/rakudo/rakudo/issues?q=...abel%3ALHF Talk to devs in #perl6-dev if you need assistance in fixing them: webchat.freenode.net/?channels=#perl6-dev 11:53
notable6 Zoffix, Noted!
Zoffix Thank you, robot.
El_Che Zoffix: maybe the wording on the doc page is confusing: "While there is no such thing as a public (or even protected) attribute, there is a way to have accessor methods generated automatically: replace the ! twigil with the . twigil (the . should remind you of a method call)."
Zoffix botsnack
.botsnack
yoleaux :D
El_Che method all over the place
Zoffix El_Che: you got a commit bit. Fix it! :) github.com/perl6/doc 11:54
Zoffix &
El_Che Fixing is not the issue. I am wondering if others find the wording troublesome 11:54
stmuk El_Che: the sentence is probably too long as for your original $r.latest("Foo") I probably made the same error first time 12:02
El_Che stmuk: the problem is that whenever I am away from perl6 I make the error again 12:05
:)
jkramer What does this mean? MoarVM panic: Internal error: Unwound entire stack and missed handler 12:13
jkramer Ah looks like it's caused by doing a 'return' in a whenever-block 12:15
jnthn Yeah, that wants a better error for sure 12:22
Should probably say the handler is not in dynamic scope 12:23
jnthn Well, when that's the case at least. 12:24
Not quite sure if we should go so far as trying to make it magically DWIM 12:25
At least, in the cases where that makes any sense
Probably not
jkramer I kinda assumed it would implicity do a "done;" and return from the sub the react-block was in, but crashing is also fine :) 12:27
Ven`` o/
jnthn jkramer: The whenever can be running on another thread from the one waiting for the react block, though, and yes, it would need to arrange to make the `done` work 12:28
jnthn jkramer: I think that's all do-able, but what does give me a bit of pause for thought is that react today is implemented in terms of supply, and `return` can never work there, so it'd introduce a difference there 12:29
That may be reasonable, but I'd at least want to think it through a bit more :)
jkramer jnthn: I wasn't sarcastic, I really think crashing is fine in this case :) 12:31
I just had different expectations but it makes sense
Ven`` if magic is involved it could maybe be detected at compile-time 12:32
jnthn Ven``: Indeed, given return works lexically 12:33
In fact, if it is doomed to not work, it can also be detected lexically :)
Ven`` sounds like an interesting thing to try and implement!
jnthn It's not unreasonable to want to `return` out of a `react`, I don't think.
Ven`` so it'd convert return to e.g. my $return-after-react; react { $return-after-react = True; done; }; return if $return-after-react;? 12:35
jnthn Something like that, I guess 12:40
Though return is a control exception so another way is to handle that
Ven`` that sounds quite a bit harder than forbidding it 12:42
.oO( How very Perl 6-y, blood sweat and tears from the impl. to allow things the user to do everything they'd want to )
Ven`` s/things// 12:45
jkramer Is this how it's supposed to be? 13:03
m: sub lol(Str $foo?) { say "lol: --$foo--" if $foo.defined }; say "haha" ~ lol ~ "this is not supposed to be an argument for lol()"
camelia lol: --this is not supposed to be an argument for lol()--
hahaTrue
timotimo it is 13:04
jkramer I guess lol ~ "..." could be interpreted as lol with "..." as argument coerced to Str with ~
timotimo sub calls gobble up until the end of the line if they can
what you want here is a term, not a sub
Geth perl6-most-wanted: 5b9f571cac | (Jonas Kramer)++ (committed using GitHub Web editor) | most-wanted/modules.md
WIP: Proc::InvokeEditor
 13:16
donpdonp my code has "use BlkMeV::Util; BlkMeV::Util::int32Buf(1);" 13:57
how can I redo that so i can call int32Buf(1) without specifying the module part
jkramer Is anyone using mi6 for releasing to CPAN? 13:59
Zoffix donpdonp: just export it in the module. sub int32Buf (...) is export {... } 14:02
jkramer: yes, some people are using it. 14:03
jkramer Zoffix: Any idea what ~/.pause is supposed to look like? Can't find docs for that 14:04
donpdonp Zoffix: ok thx
is it possible to do a partial name "Util::int32Buf(1);" 14:05
Zoffix jkramer: gist.github.com/zoffixznet/60452b3...073b5605c3
jkramer Zoffix: Thanks! 14:06
Zoffix donpdonp: this would be one way of doing it: package BlkMeV { module Util is export { our sub int32Buf (int $x) { Buf.new: $x; } } } 14:10
jkramer Hmpf "599 Internal Exception, pause.perl.org/pause/authenquery" 14:11
donpdonp Zoffix: i think thats just what I wanted, thx.
Zoffix This is another way (cuts down on indents a bit): sub EXPORT { { 'Util' => ::('BlkMeV::Util') } }; unit module BlkMeV::Util; our sub int32Buf (int $x) { Buf.new: $x } 14:12
donpdonp o^O the first way is more reable 14:13
Zoffix :)
donpdonp readable. esp to a p6 newb :) 14:14
Zoffix jkramer: dunno. Site works with the web interface. You could try installing this Perl 5 script and seeing if it is happy with your ~/.pause (e.g. run `pause ls` and see if it errors out the same): metacpan.org/pod/distribution/App-...ript/pause 14:19
jkramer Zoffix: Uploaded it manually now. 599 seems like an internal error code of HTTP::Tinyish and not from the upload server/site, but I can't be bothered to debug it now 14:21
Zoffix FWIW HTTP::Tinyish needs curl installed. Do you have it? 14:25
jkramer Yes
Zoffix shrugs
jkramer BTW how does CPAN keep p5/p6 modules apart? Is there a p6 section on CPAN somewhere or is that modules.perl6.org?
Zoffix jkramer: how did you manually upload? 14:26
('cause if you done it right, it'd kinda answer that question :P)
Did you select "Perl6" directory in the box? 14:27
jkramer Zoffix: Like in the olden times with p5 modules (pause.perl.org/pause/authenquery?A...N=add_uri) just with "Perl6" in that directory field.
As per your instrucations here :) blogs.perl.org/users/zoffix_znet/20...-here.html 14:28
Zoffix Right. So that's how it keeps it apart. All Perl 6 modules are in the user's Perl 6 dir: www.cpan.org/authors/id/J/JK/JKRAMER/Perl6/
And Perl 5 tools ignore that dir 14:29
jkramer Well my question was more about: can I select "perl 6" on search.cpan.org/metacpan to find P6 modules or is it just modules.perl6.org?
buggable New CPAN upload: Proc-Editor-0.0.1.tar.gz by JKRAMER cpan.metacpan.org/authors/id/J/JK/...0.1.tar.gz
scimon So I was thinking of using Concurrent::Queue to build a DBIish Connection Pool module.
Zoffix jkramer: just modules.perl6.org
jkramer Ok :)
Hey, here we go \o/
scimon jkramer: So what does that do? 14:30
Zoffix jkramer: the module schemes are inherently incompatible. We can have multiple authors having the same version name and multiple modules having same version but different APIs, and metacpan/search.cpan.org don't support that
*multiple authors having the same module name 14:31
jkramer scimon: Run an editor, much like Proc::InvokeEditor (was on the most-wanted-modules list so I just gave it a shot)
scimon github.com/Scimon/p6-Proc-InvokeEditor I probably should have updated the most wanted modules list.
jkramer Oops :)
scimon (Also I should probably finish this. There's a few things I'd like it to do it doesn't).
Zoffix jkramer: now that it's in ecosystem, you should remove it from most wanted. The WIP tag is really to avoid multiple authors working on the same thing when the thing isn't listed on the ecosystem 14:32
scimon No problemo, the more the merrier.
Zoffix Yeah, like that ^ ^_^
scimon My fault.
Zoffix hehe
jkramer Zoffix: Yup, was just about to do that :)
Geth perl6-most-wanted: 19c19180d6 | (Jonas Kramer)++ (committed using GitHub Web editor) | most-wanted/modules.md
Removed Proc::InvokeEditor.
 14:33
scimon Mine currently has zip windows support because whilst I could get Rakudo star installed in Windows I couldn't get zef to install anything.
Also it currently only uses hard coded paths, plan was to add PATH checking but I'd not done it yet.
Zoffix scimon: is there a bug report for that zef unable to install anything? 14:34
scimon Well the problem isn't zef it's that zef can't unzip files but then I start hitting the "Why the hell am I developing in Windows?" point and go back to playing games. 14:35
jkramer scimon: Those are things I want to add too. Currently it checks $VISUAL and $EDITOR plus some defaults and just runs them, if running them fails it tries the next one
Zoffix scimon: well that's weird innit, 'cause Windows comes with unzipper...
scimon I'll try running it (probably tomorrow I'm out tonight) and put a note here. I went with rakudo star msi installer. 14:36
Zoffix Hm, looks like it's only in GUI mode, I guess. No built in command-line utility: stackoverflow.com/questions/102155...mmand-line
scimon jkramer: I check the file exists and is executable :) 14:37
Windows Commandline is so painful.
Zoffix yeah
scimon Anyway, two different modules doing very similar things... isn't that what all the cool languages do? We should probably have given them dumb names. 14:42
stmuk scimon: I usually test zef on windows before shipping the MSI. Which module(s) couldn't you install? 14:46
scimon stmuk: I can't remember it was a few weeks ago and I've paged the memory to /dev/null. I'll try again tomorrow evening and actually ask questions here. (And possibly on Stack Overflow (before anyone says anything)) 14:54
Geth doc: zoffixznet++ created pull request #1961:
Post release 2018.04
doc/master: 10 commits pushed by (Zoffix Znet)++
review: github.com/perl6/doc/compare/a065a...59efe2f74d
scimon Like I say I don't think the issue is zef it's more you need some stuff setup in windows in order to use it that don't come as standard. 14:55
(I think I had to install curl for instance. I managed that.)
stmuk scimon: well you will need windows git if the module isn't on CPAN and Strawberry's Perl gcc if its nativecall 14:56
scimon Yeah I had to install git.
(My Windows box is mostly for games, though it recently got Unity because I felt that way inclined). 14:57
stmuk scimon: if you are talking about the last release of Star then there was an issue in the version of zef (recommended to upgrade on all platforms)
scimon Well I'll try when I get home (in the office on a real computer :) )
jkramer scimon: Not having dumb names is one of the best things in the perl ecosystem IMO. :) If you need a module that does X you just search for X and you'll find a bunch, ie. "template" gives you all the Template:: stuff. 15:05
jmerelo O/ 15:06
jkramer I remember once I had to use Java. How do you find a templating framework called "velocity"?! :)
El_Che jkramer: it's slow as hell :) 15:06
jkramer El_Che: Yep, and annoying to use too AFAIR (it's been a long time since then) 15:07
El_Che jkramer: it's slow as hell :)
oops
:)
scimon jkramer: I was (as is often my way) of course joking. Ruby and Node modules have the most bonkers names ever. 15:08
jkramer scimon: Yeah I got that, I just wanted to use the opportunity to praise the perl ecosystem :) 15:09
scimon Indeed :)
TEttinger jkramer: you find it like you live in the 21st century. www.google.com/search?hl=en&q=...0framework 15:12
TEttinger how vulnerable is cpan or whatever cpan equivalent is used for perl6 to leftpad-like situations? I doubt that perl authors are doing things as stupid as leftpad dependencies (and dependencies on "libs" like is-even) just because that's an unusually reckless thing for any ecosystem to do, but could a popular lib cease to exist? 15:19
jkramer TEttinger: That's kinda cheating and it's still difficult to get a good overview of available libs :) 15:20
FROGGS o7 15:20
TEttinger if google is cheating you need to reexamine the world around you 15:21
Zoffix TEttinger: not a lot. We have a regularly cloned version of all github-hosted repos and CPAN-hosted dists are backed by Backpan
jkramer TEttinger: I mean that you can't compare a google search with a directory like CPAN 15:22
TEttinger jkramer: Template is also an overloaded term. if I just blindly added a dependency to "template" I could get templating engines, C++ template library interop, a blank template for an app... 15:23
jkramer How do you know you're not missing an awesome framework because its authors are not good at/don't care about SEO?
TEttinger well mostly because as you said, velocity isn't awesome 15:24
jkramer What I meant is that it's hard to find with that name, compared to modules organized in namespaces in a directory like CPAN 15:25
TEttinger in that google search about half of the results were direct links to projects for me, the other half were suggestions/ratings for libs 15:26
AlexDaniel TEttinger: there was a case when one guy pulled out most of his repos, which created a tiny chain reaction with a few other modules. I simply restored removed modules by putting them on github.com/perl6-community-modules
TEttinger nice AlexDaniel 15:27
Geth doc: 21e290b5c3 | (JJ Merelo)++ | doc/Language/syntax.pod6
Adds example of multi-line comments closes #1944
synopsebot Link: doc.perl6.org/language/syntax
AlexDaniel Zoffix: by regularly cloned version of github-hoster repos you mean perl6-all-modules? That thing includes CPAN also! 15:28
d*
TEttinger also hilariously I just realized the start of the convo regarding poor names was about a project named "zef" 15:30
AlexDaniel TEttinger: what convo? 15:30
ah, here 15:31
TEttinger zef wasn't even remarked as bad
but how would I even know what it is from that name...
Zoffix
.oO( pointless arguments )
AlexDaniel btw whateverables are called like this for a similar reason 15:31
TEttinger the able thing is good 15:32
AlexDaniel there was a discussion a few years ago and someone suggested that bots be named as *able depending on their function
TEttinger I like that better than the IWhatever convention in C#/.NET, or the complete chaos of the JVM
AlexDaniel I loved the idea very much, and it's definitely good looking now at 17 existing bots in the suite :) 15:33
El_Che rakudo i386 fails 2 tests :( 15:40
pmurias TEttinger: meaningless names obviously suck but for a things like webframeworks they sort of make sense because there isn't a obvious way to call them 15:41
timotimo how come we notice that an hour after the release? :\ 15:42
AlexDaniel El_Che: ok, which tests? 15:44
lizmat [17:39:19] <El_Che>t/hll/06-sprintf.t ..................... Failed 2/286 subtests 15:45
[17:39:33] <El_Che>Failed tests: 111-112
AlexDaniel timotimo: not enough people testing on different platforms, especially those that are used less frequently
e.g. I wouldn't be surprised if somebody came and said that something doesn't work on arm or whatever
i386 is a bit more unfortunate though 15:46
timotimo these tests are already automated, though
AlexDaniel yeah but to notice that people creating packages should test non-release versions, I can't blame them for not doing that 15:48
buggable New CPAN upload: IO-Glob-0.4.tar.gz by HANENKAMP cpan.metacpan.org/authors/id/H/HA/...0.4.tar.gz 15:49
Geth doc: c5191f6e29 | (Timo Paulssen)++ | doc/Type/Promise.pod6
Add factory methods "kept" and "broken" to Promise
 15:53
synopsebot Link: doc.perl6.org/type/Promise
jmerelo timotimo: there is a release? I was afraid to ask releasable6 again... 15:58
timotimo like an hour ago
lizmat jmerelo: yeah, there's a release :-)
and some post release merges already :-) 15:59
buggable New CPAN upload: Template-Anti-0.4.3.tar.gz by HANENKAMP cpan.metacpan.org/authors/id/H/HA/...4.3.tar.gz
jmerelo lizmat: The rakudo star not yet, right? perl6.org banner still flaunts 2018.01 16:04
lizmat jmerelo: just the compiler release at this time 16:05
jmerelo Hum. Who has @perl6 in Twitter? 16:06
comborico1611 El_Che: You around? Looking for that link to new Rakudo README. I'm on a different computer, and Github is not showing my comment I left. So I can't find it. 16:07
jmerelo Well, humility is a great virtue. But really, I can't find an announcement of the new Rakudo version in Twitter. However, Perl Group is announcing Open House tonight! In Brooklyn! 16:08
AlexDaniel jmerelo: from what I can see there will be a point release also :) 16:11
comborico1611 El_Che: Nevermind. Found it 16:13
jmerelo AlexDaniel: but something should be said about the new version. Somewhere. Really. 16:15
AlexDaniel jmerelo: well, it does say here: www.nntp.perl.org/group/perl.perl6.compiler/ 16:16
and you can subscribe to the mailing list: perl6.org/archive/lists/
(btw the mail is not rendered properly because that thing doesn't understand signed messages or something)
jmerelo: but generally we tend to have slighly louder announcements for rakudo Star, which is what we expect most users to use 16:17
jmerelo AlexDaniel: OK, I'll put it in Twitter. 16:18
AlexDaniel yea that's ok
jmerelo Here you go twitter.com/jjmerelo/status/989177823737843718 16:24
buggable New CPAN upload: Hash-Util-0.0.1.tar.gz by ELIZABETH cpan.metacpan.org/authors/id/E/EL/...0.1.tar.gz 16:29
New CPAN upload: Bailador-Plugin-NamedQueries-0.1001.tar.gz by MZIESCHA modules.perl6.org/dist/Bailador::Pl...n:MZIESCHA
lizmat and another CPAN Butterfly Plan module hits the Net: Hash::Util 16:30
giving you restricted hashes, either for the whole hash (a la Map), or for a set of keys, or just the values 16:31
jmerelo lizmat++
jkramer lizmat: Is there a list of butterfly plan modules that need to be implemented? 16:33
I might like to give it a go as practice :) 16:34
lizmat jkramer: well, there's the Most Wanted list
but what I do is target modules that have a large number of reverse dependencies on metacpan.org
I don't have a specific list 16:35
jkramer Ah, good idea :)
lizmat I also target what I think is doable within a few days
Geth doc: 6db737d6ab | (JJ Merelo)++ | 2 files
De-dups compunit and increment, refs #1912
 16:36
doc: 93ace4c770 | (JJ Merelo)++ | doc/Language/control.pod6
deduped control ops refs #1912
synopsebot Link: doc.perl6.org/language/control
jkramer Already went through the most-wanted list and most of the wanted modules don't seem particularly useful/interesting to me
lizmat also: a module that takes adds some missing functionality to Perl 6 would also be appreciated 16:37
which is one of the reasons I took on Hash::Util
buggable New CPAN upload: Proc-Editor-0.0.2.tar.gz by JKRAMER modules.perl6.org/dist/Proc::Editor:cpan:JKRAMER 16:39
lizmat is looking forward to a long list of module updates in the P6W next week :-) 16:40
jkramer lizmat: Is that reverse dependency list somewhere on metacpan or did you make it yourself using the API or something? 16:43
lizmat it's listed on a right hand side of a module, if there are any
e.g. metacpan.org/pod/Hash::Util 16:44
follow: Reverse dependencies
you wind up at: metacpan.org/requires/module/Hash:...rt=[[2,1]]
jkramer Oh so you're just going threw modules manually? I thought there was some auto-generated list sorted by reverse dependencies :) 16:45
lizmat I asked the question on #metacpan, but apparently not so simple 16:48
[10:40:53] <mickey>lizmat: this is a hard one :) even for a single distribution it involves 3 separate ES queries as we don't (atm) have an index of reverse dependencies to pre-process this data. having said that, you could query a map of *all* (latest) releases to their direct dependencies in a single ES query and build the rev-dep map in your script. 16:49
[10:41:02] <haarg>curl 'fastapi.metacpan.org/v1/distributi...p;size=20'
[10:41:46] <haarg>or s/total/immediate/
buggable New CPAN upload: P5caller-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5caller:cpan:ELIZABETH
lizmat jkramer: if you would want to set your teeth in that, that would be brill :-) 16:50
jkramer lizmat: I might just do that :) 16:53
buggable New CPAN upload: P5each-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5each:cpan:ELIZABETH 16:59
New CPAN upload: P5fc-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5fc:cpan:ELIZABETH
ufobat_ jkramer, i think there is no sereal for perl6 :p if you wanna practis this :p 17:13
El_Che jkramer: I would like to have App::Fatpacker 17:17
buggable New CPAN upload: P5lc-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/P5lc:cpan:ELIZABETH 17:19
New CPAN upload: P5lcfirst-0.0.5.tar.gz by ELIZABETH modules.perl6.org/dist/P5lcfirst:cpan:ELIZABETH
New CPAN upload: P5reverse-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5reverse:cpan:ELIZABETH
New CPAN upload: List-AllUtils-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/List::AllUti...:ELIZABETH
New CPAN upload: List-MoreUtils-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/List::MoreUt...:ELIZABETH
New CPAN upload: List-SomeUtils-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/List::SomeUt...:ELIZABETH
jkramer lizmat: curl -sS 'fastapi.metacpan.org/v1/release/_search' -d '{ "aggs": { "dependency": { "terms": { "field": "dependency.module", "size": 0 } } }, "size": 0 }' | jq -r '.aggregations.dependency.buckets | map((.doc_count | tostring) + " " + .key) | join("\n")' 17:22
(Or just look at the result here: gist.github.com/jkramer/bca08ec743...9b2563d4c)
:) 17:23
lizmat whee!
jkramer It only coveres direct dependencies, but I guess it's good enough :)
lizmat yeah, that's a great start :-) 17:24
jkramer That's enough for today, I'll go through the list tomorrow and see if there's something fun to implement for me :) 17:25
lizmat jkramer++ :-)
jkramer Have a nice evening everyone o/
lizmat you too!
buggable New CPAN upload: List-Util-0.0.6.tar.gz by ELIZABETH modules.perl6.org/dist/List::Util:cpan:ELIZABETH 17:29
buggable New CPAN upload: List-UtilsBy-0.0.2.tar.gz by ELIZABETH modules.perl6.org/dist/List::UtilsB...:ELIZABETH 17:39
New CPAN upload: Scalar-Util-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/Scalar::Util...:ELIZABETH
New CPAN upload: Sub-Name-0.0.6.tar.gz by ELIZABETH modules.perl6.org/dist/Sub::Name:cpan:ELIZABETH
New CPAN upload: Sub-Util-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/Sub::Util:cpan:ELIZABETH
buggable New CPAN upload: Sys-Hostname-0.0.5.tar.gz by ELIZABETH modules.perl6.org/dist/Sys::Hostnam...:ELIZABETH 17:49
jmerelo Wow, lizmat++ You're hyper-productive. 18:04
lizmat jmerelo: no, not really, just fixing copy-pasto's basically :-( 18:33
ufobat_ how would a good seral implementation for perl6 look like? for me the benefit of sereal over json::xs (in perl5) is, that its faster 18:43
lizmat ufobat_: a Sereal implementation would be high on my list 18:44
it would help getting Perl 6 adopted in some Perl 5 houses :-)
ufobat_ the perl5 sereal impelmentation is partially in XS, right?
lizmat afaik, completely in XS 18:45
timotimo .o( i like messagepack )
ufobat_ would a perl6 implementation be written in nqp? 18:47
lizmat I would go for a PP6 implementation first
ufobat_ if you considder starting, lizmat, i'd love to help and contribute! 18:48
lizmat Sereal is deeply embedded in the P5 way of representing values (SV NV IV etc)
and I'm not sure how that would work in a Perl 6 world :-)
ufobat_ exactly1
lizmat I guess now that we have things like IntStr it should be better possible
ufobat_ i'd love to have a rust implementation as well. but well... 18:49
buggable New CPAN upload: Sys-Hostname-0.0.6.tar.gz by ELIZABETH modules.perl6.org/dist/Sys::Hostnam...:ELIZABETH 18:49
New CPAN upload: Tie-Array-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/Tie::Array:cpan:ELIZABETH
New CPAN upload: Tie-Hash-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/Tie::Hash:cpan:ELIZABETH
New CPAN upload: Tie-StdArray-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/Tie::StdArra...:ELIZABETH
New CPAN upload: Tie-StdHash-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/Tie::StdHash...:ELIZABETH
Kuraz Hello
timotimo greetings 18:50
lizmat ufobat_: I guess way would be familiarize yourself with github.com/Sereal/Sereal/blob/mast...l_spec.pod
ufobat_ jup, i've read throuh it 18:52
Kuraz The error is normal? "Variable `'@ARGV' is not declared"` when I'm running `.say for @ARGV;` :/
jmerelo Kuraz: Yep. That would be perl5. @*ARGS in perl6 18:53
p6: .say for @*ARGS
camelia ( no output )
Kuraz Oh so much for me, thanks 18:54
jmerelo (at least it does not say it's not declared) :-)
Kuraz: no problem. If you also ask the question in stackoverflow, much appreciated
jmerelo lizmat: still. Lot of (great) work. 18:59
Geth doc: ronaldxs assigned to moritz Issue Document S05 treatment of alternations as atoms for ratchet github.com/perl6/doc/issues/1962
ronaldxs unassigned from moritz Issue Document S05 treatment of alternations as atoms for ratchet github.com/perl6/doc/issues/1962

Regexes: backtracking vs alternation
for #1962		 19:41
timotimo the docs should probably not say "creates a promise that runs the given code object" 19:51
on the promise page
buggable New CPAN upload: JSON-Pointer-1.0.tar.gz by JNTHN modules.perl6.org/dist/JSON::Pointer:cpan:JNTHN 19:56
New CPAN upload: P5fc-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/P5fc:cpan:ELIZABETH
synopsebot Link: doc.perl6.org/language/regexes
buggable New CPAN upload: OpenAPI-Model-1.0.tar.gz by JNTHN cpan.metacpan.org/authors/id/J/JN/...1.0.tar.gz 20:46
buggable New CPAN upload: OpenAPI-Schema-Validate-1.0.tar.gz by JNTHN cpan.metacpan.org/authors/id/J/JN/...1.0.tar.gz 21:06
buggable New CPAN upload: P5chdir-0.0.1.tar.gz by ELIZABETH cpan.metacpan.org/authors/id/E/EL/...0.1.tar.gz 21:26
lizmat specifically chdir() is different in P5 from Perl 6 21:27
buggable New CPAN upload: P5built-ins-0.0.8.tar.gz by ELIZABETH cpan.metacpan.org/authors/id/E/EL/...0.8.tar.gz 21:36
New CPAN upload: Cro-OpenAPI-RoutesFromDefinition-1.0.tar.gz by JNTHN cpan.metacpan.org/authors/id/J/JN/...1.0.tar.gz 22:06
Herby_ o/ 22:59
timotimo \o 23:00
comborico1611 Question: say "yup $/" if "A12B34D50" ~~ / (\D) {say ~$0} \d \d$/ # Wondering why parens needed for \D. 23:07
timotimo if you don't capture with parenthesis, no $0 exists 23:08
comborico1611 Thanks! 23:10
Zoffix lizmat: FWIW we also have &*chdir that actually changes the working directory rather than changing $*CWD to a new value 23:12
(dunno which one is more appropriate for p5's chdirt)
s/.$//;
)
m: use NativeCall; sub system (Str) is native {}; system 'pwd'; chdir '/tmp'; system 'pwd'; &*chdir('/tmp'); system 'pwd'; 23:14
camelia /home/camelia
/home/camelia
/tmp
kt_ in testing async sockets the client is getting this error "Unhandled exception in code scheduled on thread 7, No such method 'throw' for invocant of type 'Str'", then the server stops acceptig connections. How can I trap more details on the error? 23:27
Zoffix kt_: don't know if it'd help, but you could add --ll-exception flag to perl6 executable. perl6 --ll-exception your-program.p6 23:31
kt_ Unhandled exception in code scheduled on thread 4 No such method 'throw' for invocant of type 'Str' in method throw at SETTING::src/core/Exception.pm line 57 in block at SETTING::src/core/Supply.pm line 90 in method run-catch at SETTING::src/core/Supply.pm line 1890 in any at /opt/rakudo-star-2018.01/share/perl6/runtime/CORE.setting.moarvm line 1 in method throw at SETTING::src/core/Exception.pm line 54 in block at SETTING 23:34
last two lines are in method run-one at SETTING::src/core/ThreadPoolScheduler.pm line 230 in block at SETTING::src/core/ThreadPoolScheduler.pm line 293 23:38
Zoffix kt_: what perl6 version are you on? perl6 -v 23:43
Also, are you able to share the code that throws that error?
kt_ perl6 -v This is Rakudo Star version 2018.01 built on MoarVM version 2018.01 implementing Perl 6.c. 23:46
yes, I can share the code, where would be a good place? 23:47
Zoffix kt_: gist.github.com/ is good
or anywhere really
kt_ it's the chatserver sample code and a client that open 100's of concurrent connections 23:48
kt_ here is the gist: gist.github.com/ktown007/2af8aa40f...c6fbc7337a 23:56
Zoffix looks
kt_ If I start the client side two or three times it will eventually die. 23:57

Powered by Cro and the Raku® Programming Language.

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

Thank you!