»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
00:00 lizmat joined 00:01 markong left 00:04 lizmat left, w_richard_w joined 00:15 DataLinkDroid joined
Kaiepi there we go 00:35
installing sarasa gothic gets unicode to display properly in vim
00:44 comborico1611 left 00:49 shareable6 left
Kaiepi 400 lines documentation done! \o/ 00:54
01:00 kurahaupo_ joined 01:01 g- joined
Kaiepi since i can't really write editline bindings until i can work out a way to get wchar_t strings to be supported, i think i'll work on getting sec-websocket-protocol support added to the websocket library 01:01
01:03 f_ left
Kaiepi wait do any debug logger cpan modules exist? 01:04
01:05 mcmillhj joined
Kaiepi oh they do 01:05
01:05 kurahaupo left 01:08 raschipi left, raschipi_ joined 01:16 f_ joined
Geth ecosystem: 295da236a2 | (Zoffix Znet)++ (committed using GitHub Web editor) | META.list
Add App::GymScheduler to ecosystem

Randomized schedule maker for your workouts:
  github.com/zoffixznet/perl6-App-GymScheduler
01:29
01:41 rouking joined
rouking Anyone know of some a non-abandoned lib for OpenGL bindings? 01:42
s/some// 01:44
01:45 shareable6 joined, ilbot3 left
AlexDaniel rouking: I know that FROGGS was working on Box2D and SDL bindings 01:52
01:52 mcmillhj left
AlexDaniel rouking: github.com/FROGGS/p6-Box2D 01:52
but I can't find the repo for SDL 01:53
01:57 ilbot3 joined, ChanServ sets mode: +v ilbot3 02:12 HaraldJoerg left 02:14 uzl joined 02:18 uzl_ joined 02:19 pilne left 02:42 lookatme_q joined
lookatme_q :) 02:44
02:44 lookatme_q left
lookatme :) 02:44
haha, I am already online
why the error message is truncated ? 02:47
Type check failed in binding to parameter '$value'; expected NativeCall::Types::Pointer[NativeCall::Types::void] but got NativeCall::Types::Po...
02:54 Zoffix joined
Zoffix lookatme: is your rakudo old? 02:54
What's perl6 -v give?
lookatme Zoffix, seems not, it's 2018.04 02:56
pastebin.com/2yuMBCiU
Can you figure out what's wrong here ?
Zoffix lookatme: is that the full code? 02:57
Yeah, it's a known bug that we fixed twice already. I would've thought it was fixed for good by now
This was the last fix: github.com/rakudo/rakudo/commit/ce...d2edfb22a0 02:58
Landed in .03 02:59
oh maybe that's a slightly different bug 03:00
lookatme It's a part of code 03:01
Zoffix m: say 'NativeCall::Types::Po'
camelia NativeCall::Types::Po
Zoffix m: say 'NativeCall::Types::Po'.chars
camelia 21
Zoffix Yeah, I see where it's being cut.
lookatme oh, you mean the debug message truncated is a bug ? 03:02
But I am not seen what's wrong here 03:03
Zoffix lookatme: do you have a full, runnable program that demonstrates that error?
lookatme It's `Pointer[void]` obviously, why it raised that error
It's a lot of code 03:04
Where can I post it ?
Zoffix gist.github.com/
lookatme :( Can not access that website 03:05
I am on the back of Firewall
03:07 eliasr left 03:10 uzl__ joined, uzl left
Zoffix Just pastbin it then 03:10
lookatme okay, wait
03:11 uzl joined 03:12 uzl_ left
lookatme pastebin.com/gvAAqAQn 03:14
Zoffix, a lot of code
I comment the file name on the top
Zoffix lookatme: how come you aren't using public repos, like say on GitHub or on gitlab.com? 03:16
lookatme Zoffix, I am not ready to put it on the github then
I will :) 03:17
lunch time :)
03:18 uzl__ left, uzl left, uzl__ joined, uzl joined 03:20 w_richard_w left
Zoffix m: class Supercalifragilisticexpialidocious {}; (my $x := class {}.new).^set_name: <Supercalifragilisticexpialidocious>; -> Supercalifragilisticexpialidocious {}($x) 03:22
camelia Type check failed in binding to parameter '<anon>'; expected Supercalifragilisticexpialidocious but got Supercalifragilistice...
in block <unit> at <tmp> line 1
Zoffix works with my local fix
03:33 uzl left, uzl__ is now known as uzl
lookatme Zoffix, how ? 03:36
Zoffix lookatme: github.com/rakudo/rakudo/commit/a7...74431d3d0e 03:39
lookatme: basically, you're getting your error because two objects that have the same .^name are actually different objects, so try to see why they might be different. 03:40
lookatme Zoffix, okay thanks 03:41
Zoffix lookatme: and if you manage to golf it, report it github.com/rakudo/rakudo/issues/new they probably shouldn't have the same names, or at least the names could be improved to make those errors better 03:42
What does "invres" in p6bindattrinvres stand for? I always forget the name of that op... 03:46
03:59 Zoffix left 04:09 wamba joined 04:10 atroxaper joined 04:12 raschipi_ left 04:15 AlexDaniel left 04:25 uzl left 04:29 kaare_ joined 04:47 molaf joined 04:51 Zoffix joined 04:54 Zoffix left 04:56 skids left 05:06 jmerelo joined
Geth doc: acc559a7c8 | (Ronald Schmidt)++ | doc/Language/grammar_tutorial.pod6
Update grammar_totorial.pm6 add clarification on extending grammars with role composition.
05:12
doc: e187cf9385 | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | doc/Language/grammar_tutorial.pod6
Merge pull request #2036 from ronaldxs/grammar_totorial-role-clarification

grammar_totorial.pm6 clarification of extending grammars with roles
synopsebot Link: doc.perl6.org/language/grammar_tutorial
05:14 Khisanth left
Geth doc: 47bdb6315a | (JJ Merelo)++ | doc/Language/unicode_entry.pod6
Shortens heading, closes #2037
05:28
doc: 3e2d277967 | (JJ Merelo)++ | doc/Language/unicode_entry.pod6
Capitalizes heading
synopsebot Link: doc.perl6.org/language/unicode_entry
05:29 Khisanth joined 05:32 robertle left 05:34 shareable6 left 05:36 ExtraCrispy left 05:38 wamba left
jmerelo quotable6: use lib 05:40
yoleaux 20 May 2018 20:01Z <AlexDaniel> jmerelo: “better ways so say something” *to* say something? ;)
quotable6 jmerelo, OK, working on it! This may take up to three minutes (4582161 messages to process)
yoleaux 20 May 2018 20:05Z <El_Che> jmerelo: You shoud s/Perl Documentation/Perl 6 Documenation/ in the dev.to article if you don't have to start a civil war
jmerelo AlexDaniel`: probably. Where is that?
quotable6 jmerelo, 4 messages (Inf⌁-Inf): gist.github.com/cbdc1daccb7c85c2da...c453528cb9 05:41
05:51 parv joined, andrzejku joined, parv left 05:56 andrzejku left 05:59 samebchase1 is now known as samebchase 06:09 n0tjack joined 06:21 robertle joined
Geth doc: 9966768306 | (JJ Merelo)++ | 3 files
Adds entry for the decont operator closes #2025
06:22
06:24 jmerelo left 06:25 n0tjack left 06:28 n0tjack joined 06:29 parv joined 06:32 AlexDaniel joined 06:43 n0tjack left 06:47 DataLinkDroid2 joined 06:50 DataLinkDroid left 06:56 darutoko joined 07:00 jmerelo joined 07:07 xtreak joined 07:21 domidumont joined
Geth doc: Kaiepi++ created pull request #2038:
Implement Perl 6 NBSP linter script
07:22
07:27 domidumont left
Geth doc: JJ self-unassigned Use always non-breaking space when talking about Perl 6 and Pod 6 github.com/perl6/doc/issues/1989
JJ assigned to Kaiepi Issue Use always non-breaking space when talking about Perl 6 and Pod 6 github.com/perl6/doc/issues/1989

Add Node.js to Perl 6 nutshell page
Fixes #1968
07:28
07:28 domidumont joined 07:29 rindolf joined 07:42 dogbert11 left 07:47 dakkar joined, diegok joined 07:53 scimon joined
lookatme How can I provide a Null Callable like C NULL function pointer to the function ? 08:02
such as ` void f( void(*fp)(int) = NULL) { }` --> sub f(&fp:(int32)) is native(Str) { } 08:03
sub f(&fp:(int32) == ?) is native(Str) { }
moritz I think a type object is interpreted as Null 08:16
so try Callable
08:21 xtreak left 08:30 wamba joined
lookatme moritz, It's report Callable not have a signature method 08:34
not working
I tried other type object such as Code, Any, Nil
not working too
jmerelo lookatme: Why don't you post it in StackOverflow? It stays longer to be answered there, and next one with the problem can benefit from it. 08:35
lookatme I forget my password and account, jmerelo :) 08:36
jmerelo lookatme: you can log in using Google account. Or create a new one :-)
lookatme: I can post it from you, but you won't be notified of the answers then :-)
lookatme I am on the back of Firewall, can not access Google 08:38
will recovery my account 08:39
when I am off work
08:40 domidumont left 08:45 xtreak joined
jmerelo lookatme: great :-) 08:46
lookatme I always forget that thing
08:54 parv left 08:58 HaraldJoerg joined 09:08 amalia___ joined 09:12 AlexDani` joined 09:13 AlexDaniel left
scimon So I upgraded my Ubuntu install and I've been having issues with libssl (specifically cro not running because "Cannot locate native library 'libssl.so'") has anyone else had a similar issue? 09:17
jnthn scimon: apt install libssl-dev 09:20
El_Che jnthn: would "apt-get install libssl1.0.0" be the answer? 09:41
Ubuntu 18.04 comes with ssl 1.0 and 1.1
jnthn El_Che: No idea. I just suggested the one I know works. :) 09:56
09:56 lizmat joined
El_Che mine works as well 09:56
:)
jnthn Ah, cool. Those are perhaps better :) 09:57
10:01 AlexDani` left
El_Che I wonder if your solution uses ssl 1.0 or 1.1 10:06
if the latter it could be a better solution if something from 1.1 is needed 10:07
scimon jnthn: unfortunately I have that installed and it's still not working. 10:12
I shall try the other one.
Nope :( 10:13
10:15 markong joined
scimon Also tried installing libssl1.0-dev 10:16
jnthn Hm, maybe have to run...is it ldconfig...after? 10:18
El_Che letme spin a continaer for you 10:19
jnthn If it's for cro there already are some containers :) 10:20
scimon Yeah but I kinda wanted to test making them :) 10:21
El_Che docker run -ti --rm rakudo/ubuntu-amd64-18.04:2018.04.1
there you go, a perl6 container with zef
no cro :)
scimon I could just go with that :) Thanks.
El_Che (to debug your problem)
it drops you to the perl6 repl 10:22
so if you want a shell, do this:
docker run -ti --rm --entrypoint=bash rakudo/ubuntu-amd64-18.04:2018.04.1
scimon Cheers :) (I should probably get back to debugging test suites sigh) 10:23
El_Che do apt-get update before installing a package and you're set to go
keep us posted of your findings
:)
10:25 wamba left
scimon Will do 10:25
(One thing, that image doesn't have make installed, Digest::SHA1::Native uses make in it's build process (and it's a dependency of cro) 10:29
We may want to add build-essential to that image. 10:32
10:32 AlexDani` joined, AlexDani` is now known as AlexDaniel
scimon AlexDaniel: Thanks for the patch on the Polyline encoding module :) Sorry it took me a while to notice. 10:34
AlexDaniel scimon: thank you!
10:36 ExtraCrispy joined
scimon (So after apt installing libssl-dev and build-essential the zef install cro is failing in IO::Socket::Async::SSL with Cannot locate symbol 'sk_num' in native library 'libssl.so') 10:36
AlexDaniel scimon: I don't know anything about it, but: irclog.perlgeek.de/perl6/2018-05-16#i_16172708 10:37
scimon That looks like it. So this is me trying to install cro on a fresh ubuntu 18.04 with rakudo image El_Che referenced. 10:40
10:41 a3f_ left 10:42 a3f joined, a3f left, a3f joined
AlexDaniel squashable6: status 10:43
squashable6 AlexDaniel, Next SQUASHathon in 10 days and ≈23 hours (2018-06-02 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
scimon I shall investigate further later. 10:48
10:53 shareable6 joined, Kaiepi left 11:00 Kaiepi joined 11:02 araraloren joined
buggable New CPAN upload: P5getservbyname-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5getservbyn...:ELIZABETH 11:02
Kaiepi somehow i managed to cause a kernel panic hastebin.com/yogixopuce.pl 11:05
i probably need a break since this is way off track of what i'm trying to do, but it's not something i would've expected to happen 11:06
11:11 lichtkind joined 11:16 Kaffe left 11:21 jmerelo left 11:22 athenot joined 11:26 raschipi joined, lichtkind left
El_Che scimon: the image is not meant as a cro image but as a base rakudo+zef image and thus to be be as small as possible. Everything needed is an apt-get install away 11:27
11:27 athenot left 11:38 raschipi_ joined 11:39 raschipi left 11:40 lichtkind joined 11:41 raschipi_ is now known as raschipi 11:44 athenot joined 11:54 jmerelo joined 11:57 itaipu joined
El_Che scimon, jnthn: in order to install Cro on ubuntu 18.04 you need: "apt-get install build-essential libssl1.0-dev" 11:59
scimon, jnthn: that the only winning^Wworking combination atm
jnthn: I think it should be documented somewhere
nativecall versions are a major pain. I have seen several modules broken (Readline, DBIish on Debian stable) or non trivial to be installed (Cro, DBIish on Ubuntu) 12:01
moritz what would it take to fix that? 12:03
maybe a list of libraries that nativecall tries to load? 12:04
El_Che moritz: I have been thinking about it as well 12:05
12:06 w4and0er96 joined 12:10 araraloren left, araraloren joined 12:11 araraloren left
El_Che it's kind of a keep-up race and an impossible matrix 12:11
different support versions of the same distro may require an other lib 12:12
eg debian 8, 9 and Ubuntu 18.04
they all use an other mysqlclient lib
12:12 araraloren joined
El_Che with Debian 9 using mariadbclient lib 12:12
a crazy idea: 1 up to date module that has the right pkg name for libs? 12:13
12:14 AlexDaniel left
lizmat El_Che: doesn't sound that crazy to me :-) 12:14
tbrowder_ hi all
El_Che e.g.: running %libs<debian><8><ssl> = libssl1.0-dev
hahainternet there are debian subversions 12:15
El_Che thing is that nativecall is ambibious about versioning
hahainternet 8.x
it's not as simple as it appears
besides, it's libssl-dev, which is a metapackage iirc
El_Che hahainternet: hence the crazy part
hahainternet: no, that will install 1.1, while everything seems to expect 1.0 12:16
hahainternet El_Che: 'everything'?
El_Che hahainternet: yes. packaged 1.1 is recent enough that no modules I have encoutered seem to target 12:17
and they break when 1.1 is installed and 1.0 is not
hahainternet how bizarre, the changes are surely not that significant
there are only 47 packages in stretch that use ssl1.0
El_Che yes, that was the second part of my message
nativecall and versioning
take readline
tbrowder_ jmerelo has highlighted problems with pod and non-breaking ws being transformed to breaking ws. one fix would be to treat all nbws as an ordinary character as part of one word in all cases where words are split or ws is collapsed to one char. thoughts! 12:18
xiaomiao openssl 1.1 is just different enough to break all previous assumptions, it's glorious
El_Che I created a 1 line fork that works on Ubuntu 18.04 and distros that moved to readline 7
12:18 kurahaupo_ is now known as kurahaupo
xiaomiao libressl is similarly almost, but not quite unlike 12:18
El_Che why a (temporarely) and nto a PR?
because readline used to hardcode the version ('readline', v5)
then DrForr removed it to accomodate v6 12:19
('readline')
but that does not find v7, so my patch is ('readline', v7)
whay of course won't work with distros running readline 5 and 6
(and we're not even talking about API/ABI breakage, non here luckily enough) 12:20
timotimo reminded of jnthn's slide "dynamic problem? dynamic solution!"
El_Che timotimo: that is kind of my crazy module idea :)
I have seen 2 solutions on the ecosystem to find libs
they work most of the time until they don't :) 12:21
timotimo at some point we'll have to bite the bullet and implemend ld scripts, too
El_Che like when the distro ships 2 modules
timotimo at least one library we bind uses ldscripts
El_Che I hope my ramblings make sense 12:25
jmerelo tbrowder_: It was thanks to AlexDaniel` . I have been trying to figure out where it happens, but I simply couldn't. I mean, I have no idea where sentendes are broken by any kind of space. 12:27
tbrowder_: candidate was pod2text in Rakudo, but I don't see anything there doing that.
tbrowder_: I have seen no tests for that behavior, or the contrary, so I think it's simply not specced. 12:28
12:35 wamba joined
raschipi El_Che: If people use C, they will have to use their tooling too, makes sense. Obviously, this problem has been solved for C. 12:41
El_Che we may loose the might of "it works out the box" we have now 12:42
(at least when the stars align)
12:42 jmerelo left
araraloren m: use NativeCall; say (my long $ = 42).WHAT 12:42
camelia (Int)
araraloren Is this a bug ? if so is it reported ? 12:43
timotimo not a bug 12:45
you can't call .WHAT on a native int because native ints aren't objects
El_Che a feature
timotimo so you get an object created for you so .WHAT can work at all
araraloren Okay 12:47
timotimo if this weren't the case, you couldn't, for example, do (my int $foo = 42).log
m: say (my int $foo = 42).log
camelia 3.7376696182833684
araraloren sub time(uint32 is rw --> uint32) is native(Str) { * }; say time(my uint32 $ = 42); # How about this ? it report : Native call expected argument that references a native integer, but got P6int
timotimo that's more interesting 12:48
does it change if you put the declaration outside the call?
El_Che i suspect it does
araraloren yeah, it working
but inside not 12:49
timotimo i wonder what makes the container disappear there
12:49 xtreak left
araraloren timotimo oh wait, it's not working 12:49
12:50 pierrot left
araraloren It's working if I move the declaration and assignment outside 12:50
12:50 Kaiepi left
timotimo ah, i meant the whole thing, indeed 12:50
araraloren sub time(uint32 is rw --> uint32) is native(Str) { * }; my uint32 $a; say time($a = 42); 12:51
not working too
timotimo so the return value of the assignment is the value rather than the container?
araraloren yeah, I agree 12:52
12:52 pierrot joined
timotimo m: say (my int $foo = 1) = 5 12:52
camelia Cannot modify an immutable Int (1)
in block <unit> at <tmp> line 1
timotimo that could be on purpose 12:53
araraloren m: my int $foo; say ($foo = 1) = 5
camelia Cannot modify an immutable Int (1)
in block <unit> at <tmp> line 1
12:53 ml_ joined
araraloren Is the error report treat it as Int or make a new object too ? 12:54
timotimo no, p6int is native ints
12:55 Kaiepi joined
araraloren I see it is said immutable Int 12:55
timotimo Int is actually a P6Opaque that boxes a p6bigint
12:55 Kaiepi left 12:56 mcmillhj joined, Kaiepi joined
araraloren Anyway I will submit my stackoverflow question first: stackoverflow.com/questions/504494...nativecall 12:57
13:04 atroxaper left 13:07 Kaz joined
Kaiepi i have some code that causes a kernel panic when i try to run it 13:08
13:09 Kaz left, atroxape_ joined
Kaiepi i was screwing around with different ways i could replace spaces with U00A0 for the doc repo and wrote this, which doesn't do all of what i need it to do but was something to test that spaces were getting replaced properly hastebin.com/omiyadirev.pl 13:11
13:11 pierrot left
Kaiepi does the same thing happen on other OSes if you run it in util/perl-nbsp.p6? 13:12
13:12 pierrot joined, atroxape_ left 13:13 athenot left
El_Che are you just launching a huge numbers of perl -e without arguments? 13:13
i misread 13:14
ignore
13:16 aindilis left, athenot joined 13:17 aindilis joined
buggable New CPAN upload: P5hex-0.0.5.tar.gz by ELIZABETH modules.perl6.org/dist/P5hex:cpan:ELIZABETH 13:22
13:28 skids joined 13:31 wamba left
buggable New CPAN upload: P5index-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/P5index:cpan:ELIZABETH 13:32
New CPAN upload: P5lc-0.0.6.tar.gz by ELIZABETH modules.perl6.org/dist/P5lc:cpan:ELIZABETH
New CPAN upload: P5lcfirst-0.0.7.tar.gz by ELIZABETH modules.perl6.org/dist/P5lcfirst:cpan:ELIZABETH 13:33
tbrowder_ jmerlo: i’m going try to add a test to roast (per yr code) to test pod nbsp handling. that will hopefully point to one part of the problem. as i said, we’ll need to get consensus about handling of nbws in pod if we need to change current code. 13:40
buggable New CPAN upload: P5length-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/P5length:cpan:ELIZABETH 13:42
New CPAN upload: P5localtime-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5localtime:...:ELIZABETH
13:44 wamba joined 13:54 lucasb joined 13:57 wamba left 13:58 mcmillhj left 14:00 mcmillhj joined, mcmillhj left 14:01 gabiruh left 14:05 mcmillhj joined 14:11 kaare_ left 14:15 w4and0er96 left 14:16 ml_ left 14:17 ml_ joined 14:18 ml_ left 14:33 ChoHag left, ChoHag joined 14:36 itaipu left
[Coke] there's a git repo somewhere that's all the modules as git submodules or something, yes? 14:38
moritz github.com/moritz/perl6-all-modules 14:39
14:48 jmerelo joined, ChoHag left 14:49 ChoHag joined 14:54 troys joined 14:55 ChoHag left 14:56 ChoHag joined
El_Che .seen drforr 14:59
yoleaux I saw DrForr 13 Apr 2018 14:25Z in #perl6: <DrForr> My head isn't in that space right now otherwise I'd look for possible examples...
jmerelo El_Che: you can probably contact him via Facebook 15:03
El_Che facebook 15:04
I like drforr a lot, but not enough to get a facebook account 15:05
;àµ
;)
jmerelo El_Che: Yep, I know... Twitter also.
tbrowder_ .tell jmerelo i have a test to put in roast that proves p6 is silently changing non-breaking space to a normal space. 15:06
AlexDaniel: are you here now? 15:07
15:08 pmurias joined
pmurias blogs.perl.org/users/pawel_murias/2...uture.html - rakudo.js blog post 15:08
tbrowder_ .tell jmerelo i have a test proving pod nbsp handling is not as expected
yoleaux tbrowder_: I'll pass your message to jmerelo.
15:11 Sgeo_ left
El_Che "Billy, ask dad to pass me the salt" 15:11
aka jmerelo is here
tbrowder_ .tell AlexDaniel i can fix p6 part of pod handling but need some guidance and consensus of how to do it. 15:12
yoleaux tbrowder_: I'll pass your message to AlexDaniel.
jmerelo tbrowder_: great. Thanks!
yoleaux 15:08Z <tbrowder_> jmerelo: i have a test proving pod nbsp handling is not as expected
jmerelo tbrowder_: that will create a cascade of solutions when solved. Great work! 15:13
15:14 Sgeo joined
jmerelo El_Che: you can also reach him by email, probably... 15:15
El_Che jmerelo: I just wanted to check something, nothing urgent 15:16
tbrowder_ question for defining a fix: can we assume, in all pod, that any single nbsp between two words should be treated as part of the adjacent words? if so, i think the fix is fairly easy inside Pod.nqp; otherwise, we will need a more complex solution. 15:17
jmerelo tbrowder_: I guess that's by definition, isn't it? 15:18
tbrowder_ i think so, but i’ll bet someone will come up with an edge case! 15:19
jmerelo hates edge cases
15:20 wamba joined
synopsebot Link: doc.perl6.org/language/js-nutshell
Geth doc: d07dbce627 | (Ben Davies)++ | doc/Language/js-nutshell.pod6
Add operators documentation
doc: 343c1279d1 | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | doc/Language/js-nutshell.pod6
Merge pull request #1975 from Kaiepi/nodejs

WIP: Add Node.js to Perl 6 nutshell page Thanks! Great work.
15:23 kaare_ joined 15:27 shareable6 left 15:30 araraloren left 15:34 Sgeo left
raschipi tbrowder_: do you have a patch already? I think that the easiest way to find the edge cases is to run roast and toaster after the changes... 15:38
15:39 Sgeo joined
Geth doc: 8d11b5f8a3 | (JJ Merelo)++ | doc/Language/list.pod6
Changes phrasing for description of binding

It limits itself to the problem pointed out in the OP, so this should close #1378.
The terminology suggested by @smls is great, but it might be better to open some issues to pages where it's not well expressed.
15:45
synopsebot Link: doc.perl6.org/language/list
15:46 uzl joined 15:48 athenot left
tbrowder_ raschipi: no fix yet, but that’s a good idea; i’m about to add a todo test for the intended fix to roast in just a moment... 15:50
jmerelo tbrowder_++
raschipi tbrowder++ 15:51
[Coke] moritz: Danke. 15:54
[Coke] just skips review.
15:57 uzl_ joined 15:59 uzl left 16:01 sftp left
Geth doc/master: 4 commits pushed by (Will "Coke" Coleda)++ 16:16
16:17 athenot joined
El_Che notable6: weekly blogs.perl.org/users/pawel_murias/2...uture.html 16:18
notable6 El_Che, Noted!
El_Che pmurias++
16:24 shareable6 joined 16:25 DataLinkDroid3 joined 16:27 DataLinkDroid2 left 16:28 MilkmanDan left, MilkmanDan joined 16:31 sjoshi joined, scimon left 16:32 ChoHag_ joined, ChoHag left, mcmillhj left 16:34 benjikun left 16:36 benjikun joined, dakkar left, uzl_ is now known as uzl 16:42 zachk joined 16:44 zachk left, zachk joined
uzl Naive question: when changing a file in my local repo of rakudo, should I `git add` it before making any changes to it? 16:45
16:45 mcmillhj joined, mcmillhj left
mst if it hasn't been changed -yet- then 'git add' is a no-op 16:45
jmerelo uzl: adding does not matter much, except when you do a commit.
uzl: and when you commit, it will be the changes in that moment the ones that are committed. So no need to add it before making changes. 16:46
uzl: git add is a bit like "Hey, here's this file, I'm gonna do stuff on it, only not right now. Keep it in mind for latter" 16:47
uzl: and then, later, when commiting, git is like "Hey, you told me you cared about this file, so I'm registering changes and putting them into the local repo" 16:48
uzl: technically, git add adds the file to the index. Nice explanation and all the stuff, as usual, in SO stackoverflow.com/questions/368983...dex-in-git 16:49
uzl jmerelo: So what I happened is that I made a change to nqp/src/HLL/Compiler.nqp, then added, commited and pushed it to github
16:50 mcmillhj joined
uzl but when I went to do a pr, it didn't show that I made a simple addition but the whole file. 16:50
jmerelo: I'll take a look at that link. Thanks! 16:51
jmerelo uzl: that might indicate a change in encoding.
uzl: good luck, anyway.
uzl jmerelo: Thank you.
16:54 sjoshi left 17:01 cpage joined 17:04 DeepIO joined, DeepIO left, DeepIO joined
El_Che uzl: if you're not addidng files, all you need is "git commit -am "your message" && git push 17:09
uzl: the -a commit all changes on all files you changed
if you you want to commit something specific: "git commit -m "msg" filename"
jmerelo El_Che: actually, -a adds all files that have been changed. 17:10
timotimo i like putting a -v there so it will put the complete diff in the editor
jmerelo El_Che: if you don't want that, you have to add every file you want in the commit one by one
El_Che jmerelo: that's what I said 17:11
jmerelo El_Che: and the first message was written before your second message (sorry)
timotimo you don't have to do it one-by-one
jmerelo El_Che: (and this message before your 4th message)
timotimo just use -p and it'll ask you for every bit of change
-p is available for "git add", "git commit", "git stash" and "git checkout" at the very least
El_Che on my own code I tend to always use a branch, commit a lot and work mostly one file per commit 17:12
17:12 stmuk joined
El_Che so the -a is a good match in this workflow 17:12
17:14 stmuk_ left
uzl El_Che: Well, what happened is that I added two lines of code to /nqp/src/HLL/Compiler, commited the change to remote and realized that it says I've made 17:16
roughly 750 additions when I just added two lines to the file.
jmerelo uzl: you can always revert the change. Easy if it's the last one on the repo, not so easy if it isn't, if it's a PR it's no big deal.
El_Che uzl: your editor does probably some automatic formatting 17:17
17:19 domidumont joined
uzl El_Che: I'm using vim with some configurations. jmerelo said it might be some change in encoding. 17:19
jmerelo uzl: if you have only changed the carriage return, that might account for a change in all lines. Or encoding, but this is subtler... 17:20
17:20 shareable6 left
timotimo git handles changes in newline modes, though 17:21
El_Che git diff origin/master
will show the changes with master 17:22
exactly what changed
change origin/master with the name of the non changed source
I have a origin (my fork) and upstream (rakudo) remote in my rakudo repo 17:23
in my case so see the changes against rakudo master, I do "git diff upstream/master"
uzl El_Che: Alright. I'll try to do that. 17:24
17:35 f_ left
tobs m: say ('a'..'z').rotor(1, 1, *+* ... *, :partial) 17:41
camelia ((a) (b) (c d) (e f g) (h i j k l) (m n o p q r s t))
tobs is the :partial eaten by the sequence operator? 17:42
timotimo m: say ('a'..'z').rotor(:partial, 1, 1, *+* ... *) 17:43
camelia Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5partial' (indicated by ⏏)
in block <unit> at <tmp> line 1
timotimo m: say ('a'..'z').rotor(:partial, |(1, 1, *+* ... *))
camelia MoarVM panic: Memory allocation failed; could not allocate 920 bytes
timotimo ah, heh.
17:43 athenot left
timotimo m: say ('a'..'z').rotor(:partial, (1, 1, *+* ... *)) 17:43
camelia ((a) (b) (c d) (e f g) (h i j k l) (m n o p q r s t) (u v w x y z))
timotimo seems to be the case, yes
17:45 athenot joined
tobs thanks, timotimo. The parentheses are good enough to disambiguate 17:46
timotimo yes, the named argument doesn't have to be in front 17:47
17:47 Ven`` joined 17:52 Ven` joined, Ven`` left 17:58 uzl_ joined, uzl left, isBEKaml joined 18:02 ChoHag_ is now known as ChoHag
Summertime is there a reason why (1, * ... *) doesn't produce a seq of repeated 1s? 18:04
I'm guessing the whatever doesn't get made into a functionish thing unless its used (e.g. +* ) in some way? 18:05
18:05 ttkp6 joined, epony left
timotimo that's right, in many places you can pass Whatever as an actual value 18:06
like for the end points of a range 18:07
m: say *..10
camelia -Inf..10
Summertime ah yeah, just reading the documentation for Whatever now, interesting 18:08
tbrowder_ jmerelo: fix for rakudo GH #1852 is in my PR #1854 awaiting review by someone 18:10
lizmat just merged it! 18:12
raschipi m: say (1, +* ... *)[^10]
camelia (1 1 1 1 1 1 1 1 1 1)
timotimo m: say (1 xx *)[^10]
camelia (1 1 1 1 1 1 1 1 1 1)
lizmat tbrowder_: shouldn't I have ?
18:12 MilkmanDan left 18:13 DataLinkDroid4 joined, MilkmanDan joined
tbrowder_ i was just waiting for someone to check it—i don’t want to abuse my new super powers—thanks! 18:14
Summertime and ({1} ... *) too, xx is probably the most cleanest though, keep forgetting about it
18:16 DataLinkDroid3 left 18:18 shareable6 joined
jmerelo tbrowder_: great! Thanks! 18:21
lizmat tbrowder_: ah, hadn't realized you already have a commit bit :-)
congratulationns!
18:22 Bocaneri left
tbrowder_ thanks! i’m honored to join such esteemed company. 18:22
jmerelo tbrowder_: so \s matches non-breaking space? 18:23
tbrowder_ yes
18:24 mcmillhj left
jmerelo tbrowder_: great job. I couldn't for the world get to those depths to know what was going on... 18:25
tbrowder_: will wait until the next release to fix the issues. :-)
18:25 sjoshi joined
El_Che releasable6: status 18:26
releasable6 El_Che, Next release will happen when it's ready. 0 blockers. 115 out of 117 commits logged
El_Che, Details: gist.github.com/15f80c98a840bad6d3...2dab8ea371
jmerelo El_Che: right, we are approaching that day...
18:27 mcmillhj joined 18:28 sjoshi left 18:30 comborico1611 joined 18:31 mcmillhj left 18:32 domidumont left 18:33 pilne joined
tbrowder_ jmerelo: thnx for that: but slogging in pod code is much easier for me than fooling with the grammar, it fits my single-cpu brain better 18:34
18:37 amalia___ left 18:38 eliasr joined, mcmillhj joined
comborico1611 tbrowder_: I'm right there with you on having a single-cpu brain. 18:43
buggable New CPAN upload: P5opendir-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5opendir:cpan:ELIZABETH 18:52
Geth doc: coke assigned to JJ Issue xt/links-not-links false positive? github.com/perl6/doc/issues/2039
ad5ae59a57 | (JJ Merelo)++ | xt/links-not-links.t

In this case, it was counting the number of `http`, and it was getting confused due to the double https in that URL. I have changed it to count them only if they follow the `|` in an `L` sentence. Still hacky, but no fake positives now. Thanks @coke for running this, closes #2039.
18:53
18:55 AlexDaniel joined, darutoko left 18:58 Ven` left 18:59 Ven`` joined, mcmillhj left 19:02 mcmillhj joined, mcmillhj left
buggable New CPAN upload: P5pack-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/P5pack:cpan:ELIZABETH 19:03
New CPAN upload: P5print-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5print:cpan:ELIZABETH
19:05 sjoshi joined 19:07 mcmillhj joined 19:11 sjoshi left
buggable New CPAN upload: P5push-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/P5push:cpan:ELIZABETH
New CPAN upload: P5quotemeta-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5quotemeta:...:ELIZABETH
tbrowder_ comborico1611: and as i age my memory is more like a fifo queue every day 19:20
buggable New CPAN upload: P5rand-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5rand:cpan:ELIZABETH 19:22
19:24 sftp joined
comborico1611 tbrowder_: Heh. All mankind is brought to subjection one way or another. 19:25
[Coke] O_o; 19:27
19:27 espadrine_ joined
buggable New CPAN upload: P5readlink-0.0.2.tar.gz by ELIZABETH modules.perl6.org/dist/P5readlink:cpan:ELIZABETH 19:33
New CPAN upload: P5reset-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5reset:cpan:ELIZABETH
19:36 epony joined
Geth doc: ef6e19e3cc | (JJ Merelo)++ | doc/Language/regexes.pod6
Adds info on :my; refs #1475
19:40
doc: 491f70305c | (JJ Merelo)++ | doc/Language/regexes.pod6
Document :our

Also adds :my and :our to index. Examples provided by the OP (thanks,
  @lefth) and
  [roast](github.com/perl6/roast/blob/d1baf2...ier/my.t). There
does not seem to be many other uses of :our in the wild. Anyway, this closes #1475.
synopsebot Link: doc.perl6.org/language/regexes
buggable New CPAN upload: P5reverse-0.0.5.tar.gz by ELIZABETH modules.perl6.org/dist/P5reverse:cpan:ELIZABETH 19:42
New CPAN upload: P5seek-0.0.2.tar.gz by ELIZABETH modules.perl6.org/dist/P5seek:cpan:ELIZABETH
torbjorn how can I flatten an array of one kv pair hashes? in perl5 I'd use context to take care of it, ie my %foo = map { %$_ } @my_array; 19:44
(these are key => value pairs im gobling up in a Grammar) 19:45
jmerelo torbjorn: Just use hash
p6: <a b c d e f>.Hash
camelia ( no output )
jmerelo p6: say <a b c d e f>.Hash
camelia {a => b, c => d, e => f}
skids or pairup.
torbjorn jmerelo: each element of array is a hash 19:46
skids Oh. Is it an Array or a list?
lucasb m: [{a=>1},{b=>2}].flatmap(*.kv).say 19:47
camelia (a 1 b 2)
jmerelo p6: say [{:42a},{:42b}].Hash
camelia {a 42 => {b => 42}}
jmerelo Hum, not that...
skids m: [{:a},{:b}].List.flat.say
camelia (a => True b => True)
skids Or just .flat for lists. 19:48
But it'll be a Seq so you may want to .cache or .list after that.
jmerelo m: [{:a},{:b}].List.flat.Hash.say
camelia {a => True, b => True}
jmerelo m: [{:a},{:b}].flat.Hash.say
camelia {a True => {b => True}}
jmerelo Hum, not that. 19:49
skids Arrays have an implicit $ on ech of their elements that prevents them from being flattened
Array.List defuses that. 19:50
raschipi beats having to deal with references
skids (And Array.list does not)
torbjorn .flatmap(*.kv).Hash seems right?
(whats the '*' for there?) 19:51
skids It's a way to avoid {} when you want to make a closure. 19:52
raschipi The splat is there to create a lambda fr flatmap to call
for flatmap*
skids called WhateverCode
buggable New CPAN upload: P5shift-0.0.4.tar.gz by ELIZABETH modules.perl6.org/dist/P5shift:cpan:ELIZABETH
New CPAN upload: P5sleep-0.0.7.tar.gz by ELIZABETH modules.perl6.org/dist/P5sleep:cpan:ELIZABETH
New CPAN upload: P5study-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5study:cpan:ELIZABETH 19:53
torbjorn huh
jmerelo torbjorn: just a placeholder for the implicit loop variable.
torbjorn: kinda like $_ in perl 5
lizmat torbjorn: basically *.kv is short for { $_.kv }
raschipi m: [{:a},{:b}].flatmap({$_.kv}).Hash
camelia ( no output )
torbjorn I see. and I agree.
raschipi m: say [{:a},{:b}].flatmap({$_.kv}).Hash 19:54
camelia {a => True, b => True}
torbjorn just need to unbedaffel myself from reading the Whatever doc
jmerelo torbjorn: would you mind to ask that question (and any other you might have) in StackOverflow too? It's better structured for search and so no 19:55
torbjorn In principle I would not, yet now I will, but thats good advice, Ill make it a habit to ask double
(just busy now)
jmerelo torbjorn: re: the doc, if you find some problem there, don't hesitate to write an issue.
torbjorn sure 19:56
can one say that the flexibility and simplicity of the perl5 list context has been replaced by a simplificy of defining lambdas in perl6?
.. which then also leads to good things elsewhere too 19:57
jmerelo torbjorn: yep, kinda. Defining lambda and contextualizing.
19:58 jmerelo left, uzl_ left 19:59 uzl_ joined 20:00 Kaiepi left 20:01 Kaiepi joined
buggable New CPAN upload: P5substr-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5substr:cpan:ELIZABETH 20:03
New CPAN upload: P5tie-0.0.9.tar.gz by ELIZABETH modules.perl6.org/dist/P5tie:cpan:ELIZABETH
raschipi torbjorn: Also, keeping structure is much easier. Don't need to fiddle with references. But to get rid of structure, you need to be explicit. 20:06
buggable New CPAN upload: P5times-0.0.3.tar.gz by ELIZABETH modules.perl6.org/dist/P5times:cpan:ELIZABETH 20:12
lucasb m: 10 ~~ regex { :state $x = 42; { say $x } } 20:21
camelia 42
lucasb m: 10 ~~ rx/ :state $x = 42; { say $x } /
camelia (Any)
20:24 Ven`` left
Geth doc: chsanch++ created pull request #2040:
Change "file handle" to "filehandle"
20:24
lucasb ^^ do I have any wrong expectation or should 'state' work the same with both rx and regex? 20:27
20:33 uzl_ left
Kaiepi m: say "\x00A0".match(/(\s)/) 20:33
camelia 「 」
0 => 「 」
20:34 zachk left
Kaiepi ffs no wonder i wasn't matching only regular spaces 20:34
raschipi u:   20:36
unicodable6 raschipi, U+00A0 NO-BREAK SPACE [Zs] ( )
Kaiepi huh 20:37
i'm used to \s only matching regular spaces 20:38
raschipi \s matches any tipe of whitespace
Kaiepi oh, i was mistaken 20:39
raschipi m: say " " ~~ / \s / 20:40
camelia 「 」
raschipi u:  
unicodable6 raschipi, U+1680 OGHAM SPACE MARK [Zs] ( )
Geth doc: c4a2beb53f | (Christian Sánchez)++ | 11 files
Change "file handle" to "filehandle"

This is part of the fixes suggested in #2015
doc: 12ca2afcb5 | nxadm++ (committed using GitHub Web editor) | 11 files
Merge pull request #2040 from chsanch/term-norm

Change "file handle" to "filehandle"
Kaiepi \s in node is hti and miss with which spaces it matches iirc
u: ᅠ 20:41
unicodable6 Kaiepi, U+FFA0 HALFWIDTH HANGUL FILLER [Lo] (ᅠ)
Kaiepi m: "\xffa0" ~~ /\s/
camelia ( no output )
raschipi That's Lo, not Zs
Kaiepi ah 20:42
20:43 slayergames joined
raschipi Lo is "Letter, other". Zs is "Separator, Space" 20:43
20:43 slayergames left
Kaiepi anyway the question i was leading up to was how to replace a space in the middle of a string 20:44
raschipi Here is the list of what \s matches: www.fileformat.info/info/unicode/c...s/list.htm
There's a discussion going on about it matching U+00A0 NO-BREAK SPACE 20:45
20:46 shareable6 left
Kaiepi i have a regex that looks like s:g/Perl\x[0020](<[5..6]>)/Perl\x[00A0]$0/ that i'm trying to use to replace the space in the middle with a no break space 20:47
i've been working at it for a while but i haven't worked out the syntax to use to make it work 20:48
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2018/05/21/...ng-denial/ 20:50
raschipi lizmat++
Kaiepi how would i go about replacing the space?
Juerd lizmat++
20:51 athenot left
Kaiepi oh, lizmat++ 20:53
geekosaur seems to work here, modulo encoding 20:55
pyanfar Z$ echo 'Perl 6' | 6 'my $x = $*IN.get; $x ~~ s:g/Perl\x[0020](<[5..6]>)/Perl\x[00A0]$0/; say $x' | od -c 20:57
0000000 P e r l 302 240 6 \n
(302 240 is correct for UTF8 encoded)
tbrowder_ lizmat: i’m blushing, thanks for the kind words! 20:58
21:01 dct joined, isBEKaml left
tbrowder_ Kaiepi: whats the char after the $ for! 21:02
geekosaur ???
evalable6 Stub code executed
in block <unit> at /tmp/ZMwjBKuaMc line 1
tbrowder_ s/!/?/
Kaiepi to give a place for what's matched by <[5..6]> 21:03
geekosaur it's $0 (first match) in the replacement
tbrowder_ oh, ok 21:07
Kaiepi m: my $a = 'a b c'; $a ~~ s:g/a\x[0020](<[b..d]>)\x[0020]c/d\x[00A0]$1\x[00A0]f/; say $/ 21:08
camelia Use of Nil in string context
(「a b c」
0 => 「b」)
in code at <tmp> line 1
geekosaur hm. I haven't rebuilt recently, wonder if this is a regressioon
Kaiepi i just rebuilt earlier today 21:09
21:09 raschipi left
Kaiepi This is Rakudo version 2018.04.1-115-g791c3e350 built on MoarVM version 2018.04.1-121-gd634d24cf 21:09
lucasb s/$1/$0/ ? 21:10
geekosaur pulls, rebuilds
Kaiepi Use of uninitialized value of type Any in string context. 21:11
geekosaur fff, not this one again
Kaiepi regression? 21:12
21:12 mcmillhj left
geekosaur no, there's some insanity somewhere that leads to "Couldn't determine correct 'make' program" 21:13
which is what I get for using rakudobrew, more or less
Kaiepi oh 21:14
Configure.pl's kinda screwy with make on my system as well
though not in that way 21:16
tbrowder_ Kaiepi: what was result of your attempt with the Perl regex? 21:17
21:18 rindolf left
geekosaur ok, so rebuilt with moar-blead and it still works 21:18
oh 21:19
m: my $a = 'a b c'; $a ~~ s:g/a\x[0020](<[b..d]>)\x[0020]c/d\x[00A0]$0\x[00A0]f/; say $/
camelia (「a b c」
0 => 「b」)
21:23 robertle left
tbrowder_ hm, i can’t find the \x[] for unicode in the regex doc... 21:23
i think you need \c[] instead 21:24
geekosaur it's not regex syntax, it's double quoted string syntax
tbrowder_ Kaiepi: ^ 21:25
Kaiepi m: "\x00A0" ~~ /\x[00A0]/
camelia ( no output )
Kaiepi m: say "\x00A0" ~~ /\x[00A0]/
camelia 「 」
Kaiepi hold on throwder, it takes a minute for the script to run 21:26
* tbrowder_
21:27 skids left
geekosaur tbrowder_, docs.perl6.org/language/quoting look for "To enter unicode sequences" 21:27
tbrowder_ never mind, my mistake \x is kind of hidden... 21:29
m: say “\x[00a0]”.ord.base(16) 21:40
camelia A0
21:44 HaraldJoerg left 21:48 HaraldJoerg1 joined 22:09 markong left 22:11 pmmmm joined 22:12 lucasb left, markong joined 22:18 slayergames joined 22:21 dct left
Kaiepi the problem ended up being an issue with my logic somewhere else in my code 22:23
22:24 dct joined 22:26 ChoHag left
tbrowder_ that’s good to know: your regex is working as advertised! 22:26
22:27 pmurias left 22:28 espadrine_ left 22:30 slayergames left 22:35 raschipi joined 22:43 ChoHag joined 22:46 mcmillhj joined 22:51 nebuchadnezzar left, dct left, nebuchadnezzar joined 22:58 skids joined, ccntrq joined 23:01 mcmillhj left 23:03 epony left 23:04 epony joined 23:12 ChoHag left 23:16 HaraldJoerg1 left
Kaiepi m: if ("Perl 6" ~~ m/Perl<[\s]>(<5..6]>)/ && $1) { say $0 } 23:21
camelia 5===SORRY!5===
Unrecognized regex metacharacter < (must be quoted to match literally)
at <tmp>:1
------> 3if ("Perl 6" ~~ m/Perl<[\s]>(<7⏏055..6]>)/ && $1) { say $0 }
Regex not terminated.
at <tmp>:1
------> 3if ("Perl 6" ~~ m/P…
Kaiepi m: if ("Perl 6" ~~ m/Perl<[\s]>(<[5..6]>)/ && $1) { say $0 } 23:22
camelia ( no output )
Kaiepi i'll need to check again but this was segfaulting on me
might have to do with the seq of promises using the regex for io in parallel 23:24
timotimo might be able to use race/hyper for that and have less trouble 23:31
Kaiepi that's what i'm using
timotimo oh, ok
and i assume latest rakudo, and not mutating hashes or arrays from multiple threads at once?
Kaiepi ohhh 23:32
i was dumb and mutating an array of promises
timotimo that could be bad, but usually only if you resize it
Kaiepi hm 23:33
what about I/O?
23:33 ChoHag joined
timotimo I/O is fine, there's locks on handles 23:34
annoyingly, reading and writing is one lock, so you can easily run into trouble
Kaiepi i have promises that read a file, modifies its data, and writes it, but there shouldn't be any promises trying to access the same files 23:35
timotimo we do locks per handle, not per file :) 23:36
Kaiepi there's only one file handle per promise 23:37
timotimo that should be fine, then, i guess
23:43 w_richard_w joined 23:55 wamba left