»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.
00:04 cpage_ joined, pyrimidine left 00:16 Sgeo joined 00:19 labster left 00:27 baest left
timotimo github.com/timo/SDL2_Raw-p6 - added screenshots of example code into the readme :) 00:29
(and also, the readme is completely new)
IOninja \o/ 00:30
lookin' sweet
timotimo a little bit :3 00:32
i took extra care to make the screenshots as tiny as they can get 00:35
it would have been better if the window didn't have an icon, or if the upper-right controls didn't have a gradient to them 00:36
OSLT
00:37 baest joined 00:40 ggoebel joined 00:53 geekosaur left
timotimo tony-o: are you the right person to ask about modules.zef.pm, or is it ugexe? 00:53
00:54 aborazmeh joined, aborazmeh left, aborazmeh joined 00:55 geekosaur joined
timotimo ah! 00:58
i didn't notice the big "About" button at the top
00:59 labster joined, d^_^b joined 01:00 pyrimidine joined
tony-o i'd be the guy 01:01
timotimo: what's up?
timotimo submitted all the issues just now :) 01:03
01:04 felher joined 01:05 pyrimidine left 01:07 jpl_ joined
timotimo i think i'm done with uploading issues for now 01:07
tony-o timotimo++
i'll take a look later tonight or tomorrow
timotimo cool
tony-o glad to know someone is using modules.zef.pm aside from me 01:08
timotimo right now you can put XSS onto modules.zef.pm, but the site isn't security-critical, so ... :)
tony-o XSS? 01:09
oh it's getting a lot of traffic apparently, i haven't looked at the traffic in a few months 01:10
timotimo someone on hackernews linked to it from the perl6 thread
well, i can put random html into my code and when somebody browses my code, their browser will execute that html
01:10 notostraca is now known as TEttinger
timotimo do you have some kind of monitoring up where i could look at your stats? like a collectd-web or munin or something? 01:11
i haven't looked at collect.p6c.org in a while :(
01:13 jpl__ joined, jpl__ left, jeff_linahan joined, jeff_linahan left 01:14 jpl_ left, jpl_ joined
jpl_ hey guys it's jeff 01:14
gotta find a handle that fits
timotimo o/ 01:15
what is it that you need a handle on? 01:16
jpl_ so I told some people about perl6 they were interested
irc
jeffythedragonslayer doesn't fit
timotimo oh, i see
jeffyslaysdragons
what's the limit on freenode anyway?
jpl_ it cut me off at jeffythedragonsl 01:17
timotimo then it'd cut the s at the end of my suggestion
jpl_ sl is steam locomotive when you type ls backwards
timotimo yes! :D
jpl_ if I'm ever in an action movie there needs to be a scene when I'm riding a dragon and jump off onto moving train 01:18
timotimo hm, a train heist?
jpl_ lol
geekosaur yes, 16 is the limit. also limited character set
TimToady drat, was gonna go for jeffy殺龍 01:19
jpl_ that works :)
when I worked at jpl people thought I had stolen a laptop because my bash prompt was my username: jpl 01:20
timotimo you worked at the jet propulsion laboratory? 01:21
TimToady wow!
jpl_ just an intern
timotimo pretty cool
geekosaur otoh an other channel I'm in has a MarvinPA (as abbrev for "the Paranoid Android")
TimToady was just an extern :)
timotimo heh 01:22
hobbs but not a volatile
TimToady JPL's a cool place, you could go to lectures by visiting guests every day and never get a lick of work done...
timotimo jffydrgnslr
jpl_ they should add an extern "TimToady" to embed perl code in C++
TimToady also saw Maggie Thatcher at JPL 01:23
timotimo the thatcher that's known for destroying all the infrastructure of great britain?
jpl_ that's pretty cool
TimToady well, they didn't let her talk, it was just a walk-through :)
timotimo OK 01:24
jpl_ I saw Bill Nye in pasadena give a speech
TimToady I think Bill Nye wasn't even born yet when I was there :P 01:25
01:25 aborazmeh left
TimToady actually, he's only a year younger than me :) 01:26
jpl_ TimToady: I honestly thought maybe you weren't developing perl anymore when I came across your website a couple years ago 01:27
TimToady it's a bit out of date, but I've been too busy developing perl to update it :) 01:28
jpl_ oh don't get me wrong I love the bare html look 01:29
I just saw wall.org/~larry/perl.html wasn't found and thought you weren't working on it 01:30
tony-o timotimo: i have google analytics, i can add you to the site if you'd like 01:34
timotimo: i have no idea what i would do with the info 01:35
timotimo oh, cool 01:36
i also have no idea :)
but i do like to look at graphs for 10 seconds and then forget i ever did
tony-o 19:54 < ugexe> yeah probably needs to be setup eventually 01:37
oops
i.imgur.com/RfMDfNs.png 01:38
timotimo interesting, 3 japanese sessions were counted 01:39
tony-o strangely i have a lot of japanese/korean followers on social media
jpl_ japanese was actually the first foreign language I got interested in
cause I'm a weeaboo 01:40
timotimo that's fine
tony-o i don't know any japanese or korean
and the german has gone by the wayside because there aren't many in california
jpl_ well you start with my name 殺龍
hobbs tony-o: move to Texas ;) 01:41
tony-o no thanks
just spent 3 of every 5 weeks in houston :-)
hobbs okay, PA then. Much nicer. :) 01:42
tony-o i would definitely go back to the midwest
PA isn't bad
01:42 jpl_ left 01:43 jeffythedragonsl joined, jeffythedragonsl left 01:44 yqt left 01:45 jeffythedragonsl joined 01:48 pyrimidine joined
jeffythedragonsl I'm from PA 01:49
MasterDuke jeffythedragonsl: same here (originally)
TimToady
.oO(Palo Alto)
 01:50
.oO(Panama)
jeffythedragonsl in maryland now
TimToady my son used to live in silver spring 01:51
MasterDuke also in md now
01:56 jeffythedragonsl left 01:57 pyrimidine left
tony-o twinsies 01:59
02:00 jeffythedragonsl joined
jeffythedragonsl freenode not accepting unicode in nicknames *sigh* 02:01
tony-o erroneous 02:02
hobbs jeffythedragonsl: which bit of PA?
jeffythedragonsl collegeville
only college in collegeville is ursinus though 02:03
hobbs but it's a very collegial atmosphere
I used to go down to King of Prussia if I wanted to see a movie REALLY BIG or gawk at stuff I couldn't afford in the Apple Store 02:04
tony-o king of prussia is where i was in PA 02:05
for SAP -
jeffythedragonsl imax is great
02:06 kaare__ left
hobbs lived in Monroe County for 15 years 02:07
jeffythedragonsl i had a friend put a python script that printed "gates>jobs" in a loop or something at that store
they couldn't figure out how to kill it and offered him a job 02:08
tony-o the people at the apple store? 02:13
while [ 1 == 1 ]; do echo 'gates>jobs'; done; 02:15
02:15 baest left
tony-o i'm sure there's a shorter way 02:16
jeffythedragonsl he made it a bit harder to kill than that :)
tony-o oh he put some effort into it 02:17
i used to like renaming the 'recycle bin' at best buy
02:17 kaare__ joined, cyphase joined
hobbs tony-o: "while true", no need to get your test operators out :) 02:18
jeffythedragonsl while true works in zsh too
tony-o got too fancy 02:19
02:20 gdonald left
geekosaur or while : 02:20
02:21 gdonald joined
tony-o funny how much time can be spent focuing on something so trivial 02:23
focusing
02:24 sufrostico left 02:25 labster left
IOninja
.oO( I bet they're sick of sinuses jokes at ursinus college... )
 02:30
02:31 bpmedley joined, baest joined
Geth specs: c6cdaff34d | (Zoffix Znet)++ | v6d.pod
Add reminder to un-LTA log(42, 1) error
 02:32
specs: c112fa54a8 | (Zoffix Znet)++ | v6d.pod
)
 02:33
IOninja So, what's the acronym for making LTA stuff Awesome...? 02:34
ATB... "make error Awesomer Than Before" 02:35
And what's an error that is not LTA?
timotimo PDG? Pretty Damn Good?
02:36 jeffythedragonsl left
hobbs MTLTA 02:37
IOninja I like PDG :D 02:38
02:39 atweiden-air joined
samcv hmm so 9482 in base -10 equals -8678 02:42
i got negative bases from -1 to -10 working
hobbs balanced ternary or nothing at all
samcv heh 02:43
hobbs, what is that, 0,1,2 where 1 is 1 and 2 is -1? 02:44
02:44 baest left
hobbs pretty much 02:44
02:44 baest joined
samcv cool 02:44
hobbs usually you write the -1 as an upside-down or crossed-out 1 02:45
it has a few interesting mathematical properties 02:46
samcv u: upside down number one
unicodable6 samcv, Found nothing!
samcv u: 1 02:47
unicodable6 samcv, U+0001 <control-0001> [Cc] (control character)
samcv, U+0010 <control-0010> [Cc] (control character)
samcv wtf
not what i wanted
m: '1'.uniname.say
camelia rakudo-moar f569c4: OUTPUT«DIGIT ONE␤»
samcv u: digit one
02:47 ilbot3 joined, ChanServ sets mode: +v ilbot3
hobbs unicode usually uses "turned" for something that's been rotated 180 02:48
timotimo also for the undead
hobbs but only seems to have TURNED DIGIT 2 and TURNED DIGIT 3
no 1 :)
unicodable6 samcv, U+0031 DIGIT ONE [Nd] (1)
samcv, U+0661 ARABIC-INDIC DIGIT ONE [Nd] (١)
samcv, gist.github.com/56b630e8a3e84dbc7f...6677bf25a3 02:49
timotimo it probably used to have only ^2 and ^3
and then the turned variants were introduced for some reason. and then the rest of ^1, ^4, ...
timotimo maximum shrug
unicode consortium giveth, unicode consortium taketh away 02:50
hobbs mostly it giveth emoji
unicodable6 samcv, gist.github.com/ae64845843cde54a29...20243796b6
samcv what it taketh away?
timotimo dunno
TimToady I usually write it with +, 0, and -
timotimo i just accept my fate kneeling down
hobbs TimToady: sensible! 02:51
02:51 cdg_ left 02:52 cdg joined 02:53 pyrimidine joined 02:56 baest left 02:59 pyrimidine left 03:01 baest joined 03:03 jeffythedragonsl joined 03:04 xtreak joined 03:14 labster joined 03:29 cdg left 03:33 jeffythedragonsl left 03:41 noganex joined 03:53 Eddward joined
Eddward samcv: Hi. Are you able to see an attachment on issue 525? 03:53
yoleaux 8 Feb 2017 10:35Z <samcv> Eddward: I am told if you open the file with :enc<utf8-c8>, or you can rename the file, ($file).copy($file.Str.encode('utf8-c8').decode('utf8')) where $file is a IO::Path obj
samcv hi Eddward
Eddward I tried to upload a recreate script. I don't think it worked. 03:54
samcv nope i can't. how big is the file
and is it just one or multiple
or is it just a text file? 03:55
Eddward 50k
It's a zip with 4 files.
It wouldn't let me just upload the script. It only like certain file types.
samcv try uploading again 03:56
or rename from like foo.p6 to foo.p6.txt
but i think zips work
Eddward It created a link, but it just links back to the issue.
I'll try again.
samcv yeah the link got messed up
edit the post and try uploading again
Eddward ugh. I'm getting the "something went really wrong" thing again. 03:57
I'm trying to find a pastebin for binaries. 03:59
Does this work for you? filebin.ca/3Bs49AZXK4GX/recreate.zip 04:02
samcv yep
04:03 atweiden-air left
Eddward cool. I added the link to the issue. 04:03
04:03 atweiden joined
Eddward oh. I should have added :bin to the slurp()s I bet. 04:05
samcv no, just :enc('utf8-c8')
so it doesn't put the unicode into NFC
Eddward Will that force everything to be utf8-c8? 04:06
I'm afraid I don't really even know what utf8-c8 or NFC really are. I'm just worried I'll fix this file and potentially break others. 04:07
samcv yeah. ok well let me try it
atweiden i've narrowed down a role-related error message that i don't comprehend at github.com/atweiden/hxc 04:08
Eddward Like if I had a file with the 3 byte u and the 2 byte u, would it handle both?
atweiden Method 'bullet-point:sym<->' must be resolved by class HXC::Parser::Grammar because it exists in multiple roles (HXC::Parser::Grammar::Paragraph, HXC::Parser::Grammar::ListItem, HXC::Parser::Grammar::ListBlock)
what's the proper way to fix this?
samcv Eddward, it will retain all the data 04:09
and if you try and open that file it will work too if the file's not in NFC form
NFC is canonical unicode representation, lets say it's official Correct Way™ to store unicode
but some characters like ŭ can be stored as two characters, but the NFC form is only one codepoint 04:10
stored as two codepoints i mean
i'm installing the modules you are using Eddward
04:10 labster left
Eddward ok. I'm using star from January 04:11
like this? slurp "$directory/expected.txt", :enc('utf8-c8') 04:15
samcv ok what is this program supposed to do
yep
Eddward It's still writing a 2 byte U instead of the 3 byte. 04:16
samcv how are you writing the file
oh you need spurt :enc('utf8-c8'
Eddward oh. I did the slurp, not the spurt. 04:17
skids atweiden: if it is not a diamond problem, adding an "is default" to one of the rules might work.
Eddward same thing: spurt "$directory/found.txt", $unescaped-data, :enc('utf8-c8'); 04:18
samcv ok but what are you doing
what is what you want the result to be
how do i check what you're checking to determine this 04:19
do you want expected.txt and found.txt to be the same?
Eddward I cd into recreate-525/ and run ./recreate.pl6
Then I open found.txt and expect.txt in a hex editor.
samcv ok
you want found.txt and expect.txt to be the same contents 04:20
sec now that i know what you want the result to be i can do this lol
Eddward The script will print good or bad if it thinks it worked or not. I just just with the hex editory because I'm not sure I trust what I'm doing in the script to correctly compare them.
Yes.
expected.txt is the output of ls with file:// added and the newline removed. 04:21
I can get the same thing in perl5.
samcv ok so you are using DBIish to get the entry?
Eddward Yes.
samcv ok. so it's DBIish that's normalizing the text 04:22
Eddward sample.db is a pruned banshee.db.
atweiden skids: diamond problem?
samcv sec
Eddward I guess. DBI has the the string URI escaped hex codes.
skids m: role B does A { method foo { } }; role C does A { method foo { } }; role D does C does B { }; D.new; # long-standing problem with roles
camelia rakudo-moar 301bcf: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Invalid typename 'A'␤at <tmp>:1␤------> 3role B does A7⏏5 { method foo { } }; role C does A { met␤»
04:22 zakharyas joined
skids atweiden: ^^ 04:23
oops
Eddward I think uri_unescape() is creating the normalized u.
samcv hm k
second
ok so that makes it easier then 04:24
skids m: role A { method foo { } }; role B does A { method foo { } }; role C does A { method foo { } }; role D does C does B { }; D.new; # long-standing problem with roles
camelia rakudo-moar 301bcf: OUTPUT«Method 'foo' must be resolved by class D because it exists in multiple roles (C, B)␤ in block <unit> at <tmp> line 1␤␤»
samcv if it's just URI::Escape
skids atweiden: It isn't supposed to be that way.
Eddward That's if I'm guessing right. I'm still not sure I understand what's happening. 04:25
atweiden skids: where would one add the `is default` ?
skids Actually I tried it doesn't help. 04:26
Neither does my old workaround anymore.
:-(
m: role A { multi method foo ($self: :$diamond? where {True}) { } }; role B does A { }; role C does A { }; role D does C does B { }; D.new; # old workaround
camelia rakudo-moar 301bcf: OUTPUT«Multi method 'foo' with signature :($self: :$diamond where { ... }, *%_) must be resolved by class D because it exists in multiple roles (C, B)␤ in block <unit> at <tmp> line 1␤␤»
samcv hmm trying to find source of uri escape 04:27
Eddward On a side note, I might be safer if you default $dbfile to "$directory/sample.db". I forgot that.
Then you don't have to be in the directory.
rakudo-star-2017.01/modules/uri/lib/URI/Escape.pm 04:28
That's where I find it in the star directory.
atweiden skids: does something need doing in rakudo? 04:29
Eddward github.com/moritz/perl6-all-module.../Escape.pm
samcv well i have 175 vs 176 now
but i maybe messed it up
give me a few mins ok
skids atweiden: Yeah, it's a skip in roast. RT#12474 04:31
Eddward yep
skids oops wrong paste
RT#124749
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=124749
skids (and it's been hard to get a firm answer on the same problem for attributes) 04:32
04:32 labster joined
samcv ok i fixed it Eddward 04:32
Eddward cool!
What's the change?
I'll try it on the larger library. 04:33
uri-unescape()?
samcv wait hold on
yeah i'm working on it
Eddward ok 04:34
skids atweiden: though I'm not sure about the second test in that roast block WRT punning/inheritence.
04:37 aborazmeh joined, aborazmeh left, aborazmeh joined, Cabanoss- joined 04:41 agentzh left, Cabanoss- is now known as Cabanossi
samcv ok i see what's happening 04:45
the unescaper is unescaping one character at a time i think
so replaces the character in the string, then goes to the next char
atweiden there is more than one way to do it, thanks anyway skids
samcv and it ends up getting normalized because it's not doing it at once
.subst(:g, / [ '%' (<.xdigit> ** 2 ) ]+ /, -> $/ { 04:46
Buf.new($0.flatmap({ :16(~$_) })).decode($enc)
})
this is what i'm doing now, and $enc is 'utf8-c8'
Eddward ok. I'll try that in a minute.... 04:47
samcv well it doesn't work :) i'm working on a solution
ah ok i see 04:48
because the character before the diacritic is not encoded with the uri escaping, we add the diacritic to it, and perl 6 does NFC 04:49
04:49 pyrimidine joined
samcv so even decoding the buff with utf8-c8 doesn't work. because a lone diacritic is the same in NFC form, it's only attached to something that it's different 04:49
Eddward Well, you understand it. A solution doesn't have to come today. 04:50
samcv ok. i will eventually figure it out :)
and then hopefully we can update uri-unescape so that it can do utf8-c8 04:51
04:51 AlexDaniel left
Eddward cool. Thanks for looking at it. Sorry to be such a pest. 04:52
samcv no problem. it's a real world problem that needs solutions :)
Eddward I'm going to drop for a while. I have some errands to run before midnight. 04:54
samcv kk
04:54 Eddward left, pyrimidi_ joined, pyrimidine left 04:56 BenGoldberg left, AlexDaniel joined 04:57 atweiden left 04:59 pyrimidi_ left 05:01 AlexDaniel left 05:03 abruanese left 05:05 eroux joined, gdonald left 05:06 gdonald joined, abruanese joined, dogbert11 joined 05:07 Cabanossi left, Cabanossi joined 05:09 Actualeyes joined, wamba joined 05:11 atweiden joined 05:12 xtreak left 05:13 xtreak joined 05:18 atweiden left 05:19 xtreak left 05:24 xtreak joined 05:30 eroux left 05:31 eroux joined 05:35 aborazmeh left 05:36 curan joined 05:37 circ-user-DyWuw joined
samcv well i actually got it. heh 05:41
pretty hacky
05:48 eroux left 05:50 pyrimidine joined 05:54 eroux joined 05:55 pyrimidine left, cibs joined 06:05 labster left 06:13 eroux left 06:22 agentzh joined 06:29 aborazmeh joined, aborazmeh left, aborazmeh joined 06:31 darutoko joined 06:32 agentzh left 06:33 agentzh joined 06:37 aborazmeh left 06:48 pyrimidine joined 06:51 RabidGravy joined 06:53 pyrimidine left 06:54 pyrimidine joined 06:55 pyrimidi_ joined, pyrimidine left, abraxxa joined 06:59 pyrimidi_ left 07:02 itaipu left 07:04 brrt joined 07:06 abraxxa left 07:16 nebuchadnezzar joined 07:21 brrt left, eiro joined
eiro hello everyone 07:21
masak heiro! \o 07:28
07:29 abraxxa joined 07:36 cpage_ left 07:40 wamba left 07:44 faraco joined 07:46 circ-user-DyWuw left 07:56 pyrimidine joined 08:00 bjz joined 08:01 pyrimidine left 08:03 brrt joined 08:09 ufobat joined 08:12 labster joined 08:17 bjz left 08:25 rindolf joined 08:26 circ-user-DyWuw joined 08:27 wamba joined 08:28 cpage_ joined 08:31 circ-user-DyWuw left, abraxxa left 08:33 bjz joined 08:46 xtreak left 08:47 xtreak joined 08:48 xtreak left 08:52 cgfbee joined 08:58 pyrimidine joined 09:03 pyrimidine left 09:05 donaldh joined 09:13 abraxxa joined 09:16 pyrimidine joined 09:21 pyrimidine left 09:25 [particle] joined 09:33 xtreak joined 09:38 Xliff joined
Xliff \o 09:38
faraco o/ 09:39
Xliff Does anyone know if Perl6 can read P5's Storable files?
DrForr Through Inline::Perl5, yes.
Xliff Kinda defeats the purpose of porting a Perl5 module, yes? 09:40
But I was afraid that was to be the case.
May want to convert those datafiles to JSON or something.
DrForr: You aren't aware of a native Perl6 solution? 09:41
DrForr It's entirely possible for Perl 6 to read Perl 5 Storable format files, yes.
I don't think anyone's written a module to do that yet, however.
Xliff Ah.
DrForr There's PerlStore, which is a Perl 6 format. 09:42
Xliff Right, but I am trying to port a Perl 5 module to Perl 6 which has data in Perl 5's Storable format.
DrForr Okay, cool. 09:43
Xliff So I will look into PerlStore. 09:45
Because looking at Storable's code made my head hurt.
DrForr My followup question would be: How often are you going to want to migrate Perl 5 data to Perl 6? And wouldn't a converter for that data that uses Inline::Perl5, reads the storable format and then dumps out Perl 6 format be easier than doing bit-for-bit compatible Storable code?
Xliff Dead ended at what I expect is an XS routine.
DrForr Yes, Storable isn't pure Perl 5. 09:46
Xliff LOL! Your followup whas what I just decided to do.
DrForr use Storable:from<Perl5>; my $x = Storable.new($filename); "file".put($x.perl); # or something similar. 09:47
Xliff Yes. That's the nutshell.
09:49 bjz left 09:50 Ven joined, zakharyas left 09:51 zakharyas joined
Xliff Well.... "use Storable:from<Perl5>" chokes 09:53
$ perl6 -e 'use Storable:from<Perl5>;' 09:54
===SORRY!===
Type check failed in binding to $handle; expected CompUnit::Handle but got Sub (sub EXPORT (*@args) {...)
Bug?
Lemme try something else;
09:56 zakharyas left 09:57 gregf_ joined
faraco star: use Storable:from<Perl5> 09:58
camelia star-m 2016.10: OUTPUT«===SORRY!===␤Please install Inline::Perl5 for Perl 5 support. ␤»
faraco star: use LWP::Useragent 09:59
camelia star-m 2016.10: OUTPUT«===SORRY!===␤Could not find LWP::Useragent at line 1 in:␤ /home/camelia/.perl6␤ /home/camelia/star-2016.10/share/perl6/site␤ /home/camelia/star-2016.10/share/perl6/vendor␤ /home/camelia/star-2016.10/share/perl6␤ CompUnit::Repository:…»
09:59 ocbtec joined 10:04 donaldh left
Xliff I may have to go back to the original author for the raw data. 10:06
Recursion limit of 1000 exceeded at /usr/lib/i386-linux-gnu/perl/5.24/Data/Dumper.pm line 233 10:07
10:07 wamba left 10:08 xtreak_ joined 10:09 pyrimidine joined 10:10 xtreak left 10:14 pyrimidine left
Xliff Well, I can load the data in Perl5, but it's bloody recursive. 10:18
10:19 ccntrq joined 10:21 donaldh joined 10:22 bjz joined 10:35 Ven left 10:37 wamba joined, bjz left
Xliff OK. Too tired to suss this out any further. I know I can load the data via "use Inline::Perl5; my $p5 = Inline::Perl5.new; $p5.call("use Storable; my $h = retrieve($filename)") 10:40
The problem is figuring out the bloody structure because it looks like endless arrays.
10:44 jjido joined, TEttinger left 10:48 jjido left
DrForr Xliff: use Inline::Perl5; use Storable:from<Perl5>; my $s = Storable.new(...); 10:57
github.com/niner/Inline-Perl5 # The SYNOPSIS is pretty clear. 11:00
use Storable:from<Perl5> <retrieve>; # import retrieve() from Storable.
11:02 patrickz joined 11:04 andrzejku joined
Xliff DrForr: I mentioned earlier that method was a no go: 11:09
$ perl6 -e 'use Storable:from<Perl5> <retrieve>; my $h = retrieve("perl5/english")'
===SORRY!===
Type check failed in binding to $handle; expected CompUnit::Handle but got Sub (sub EXPORT (*@args) {...)
11:09 donaldh left
Xliff As a matter of fact, I can boil that down to "use Storable:from<Perl5>" 11:10
Is that a bug?
11:11 pyrimidine joined 11:16 pyrimidine left 11:23 xnhp0320 joined, xnhp0320 left, cpage_ left 11:24 donaldh joined 11:31 espadrine joined 11:36 llfourn joined 11:40 donaldh left 11:42 xtreak_ left
DrForr Looks that way, though I'd check the test suite to make sure the documentation hasn't come ad rift from the actual code. 11:43
11:48 faraco left 11:52 pyrimidine joined 11:54 patrickz left 11:56 labster left, pyrimidine left, pyrimidine joined 12:01 cpage_ joined, pyrimidine left 12:12 sufrostico joined 12:15 thundergnat joined 12:16 Ven joined
thundergnat .tell samcv Take a look at rosettacode.org/wiki/Negative_base_...ers#Perl_6 for negative base number for bases -2..-36. 12:17
yoleaux thundergnat: I'll pass your message to samcv.
12:17 thundergnat left 12:18 curan left
DrForr contemplates negabinary, balanced ternary and imaginary bases. 12:19
12:27 gdonald left, gdonald joined 12:28 Ven left, infrabyte joined, infrabyte left 12:29 sena_kun joined, wamba left 12:30 brrt left 12:46 zakharyas joined 12:47 salv0 joined 12:50 abraxxa left 12:52 faraco joined, user9 joined 12:57 pyrimidine joined 12:58 AlexDaniel joined 13:02 pyrimidine left 13:12 Geth joined, ChanServ sets mode: +v Geth
timotimo Xliff: i think you just have to gte a fresher Inline::Perl5 and/or rakudo 13:16
13:26 zakharyas left
IOninja Are continuations still "somewhat experimental and JVM-only NQP feature"? I'm getting a crash in nqp::continuationreset(PROMPT, &!resumption) but on moar.... 13:28
timotimo rakudo uses continuations to implement gather/take (but sometimes gather/take is optimized to use something simpler, i think when all elements are requested?) 13:30
jnthn Not at all; they're used by gather/take, and also in 6.d await
Don't suppose you're using rotor? :)
jnthn filed a regression ticket about that in RT and it blew up inside the gather impl 13:31
Though I suspect that it's because something downstream is mis-using the iterator
Like pull-one'ing it after it already replied with IterationEnd or similar
IOninja Yeah, I'm debugging that one ATM
jnthn Ah, cool :) 13:32
I suspect if there was a problem in gather/take we'd probably notice it all over the place.
So my guess would be that it's in the rotor/batch impl
13:39 kyan left 13:43 wamba joined, pyrimidine joined 13:48 wcummings joined
IOninja Are you meant to be able to call .pull-one any number of times without anything crashing? 13:48
wcummings whats the state of perl6 in emacs? is henrik's perl6-mode the best thing? hasnt been updated in a while
IOninja wcummings: I see 34 commits on top of that in our fork: github.com/perl6/perl6-mode 13:49
wcummings ahh the fork didnt come up in my search results, thanks
timotimo IOninja: i guess after IterationEnd has been returned, you're not supposed to do it any more? 13:50
IOninja Basically, it looks like R::I.Batch pulls one-too-many times. This already caused a bug with IO::ArgFiles.lines, but there we made its .pull-one not crash when it's called too many times.
13:50 wamba left
IOninja The same issue exists with this .rotor bug which is due to .pull-one being called too many times: 13:51
timotimo that'd probably be the fix, then
IOninja m: given (gather do for ^2 { take "x" }).iterator -> $it { dd do for ^10 { $it.pull-one } }
camelia rakudo-moar 301bcf: OUTPUT«Cannot invoke this object (REPR: Uninstantiable; Callable)␤ in code at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
IOninja timotimo: that being what? :) Make .pull-one not crash after stuff's been exhausted or do we say that there's no guarantees that you can still call .pull-one after you got IterationEnd? 13:52
Xliff timotimo: Thanks for the suggestion. I will refresh Inline::Perl5
I rakudobrew daily
timotimo i don't think we make that guarantee
jnthn called that mis-using 13:53
so i'd be inclined to say "don't do that"
IOninja OK. Then I'll figure out what the RI::Batch is doing and then we'll need to revert the IO::ArgFiles thing (it's introducing X% slowdown) and toss its test, as the test tests for ability to call .pull-one till you're blue in the face 13:54
timotimo mhm
Xliff And that removed the issue from "perl6 -e 'use Storable:from<Perl5>'
timotimo may wait for jnthn to actually confirm that's what he meant
Xliff I keep forgetting that Inline::Perl5 is a bit volatile.
IOninja jnthn: is it acceptable for .pull-one to not be callable after it returned IterationEnd? 13:55
Xliff So....
What's the best way to convert a Perl5Array to a proper Perl 6 Array. Especially when the Perl5Array may be using circular array references. 13:56
???
And is that something that .perl or PerlStore can handle?
jnthn IOninja: You're not meant to use an Iterator again after it gave back IterationEnd
IOninja: Doing so is undefined behavior
IOninja Perfect! 13:57
timotimo it'd be interesting if we could mix in a role that wraps all methods of all iterators to check for protocol adherence
13:58 pyrimidine left, pyrimidine joined
Xliff Does anyone know of a way to read Storable data into Perl 6 without using Inline::Perl5? 14:00
IOninja heh
Well, it's open source. Replicate the code in Perl 6
[Coke] Xliff: no new intel has surfaced since you asked in backscroll, probably. :) 14:01
IOninja But IIRC there are issues reading its data even between different versions. I always viewed it as here-be-dragons type of storage.
[Coke] Did you upgrade your copies of inline::perl5 and rakudo ? (or report which versions you were using?)
IOninja: ah, good point. 14:02
also, do you have a gist showing your current issue? 14:03
14:05 matiaslina joined
Xliff [Coke]: Updated, yes. 14:06
And I am able to get the data in via Inline::Perl5. However that structure is not conducive to .perl or .gist because it contains circular references. 14:07
14:09 ocbtec left
[Coke] What are you trying to get from the data? 14:10
14:12 bitmap joined
Xliff Will PerlStore or .gist handle circular references? 14:14
IOninja .gist will
Xliff The end goal is to remove the dependency of Inline::Perl5 in reading the data.
So I need to take the Perl5 files and convert them into something suitable for Perl6.
[Coke] .gist and .perl are not great about circular references, no. but that still doesn't answer the question - what are you trying to do ?
IOninja m: my @a = [@a]; say @a
camelia rakudo-moar 301bcf: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Cannot use variable @a in declaration to initialize itself␤at <tmp>:1␤------> 3my @a = [@7⏏5a]; say @a␤ expecting any of:␤ term␤»
[Coke] "something suitable"
?
IOninja m: my @a; @a[0] = @a; say @a
camelia rakudo-moar 301bcf: OUTPUT«(\Array_74318568 = [Array_74318568])␤»
[Coke] so, you want Storable6? 14:15
faraco howdy
Xliff m: my @a; @a[0] = 1; @a.push: @a[0] xx 8; say @a; 14:16
camelia rakudo-moar 301bcf: OUTPUT«[1 (1 1 1 1 1 1 1 1)]␤»
[Coke] PerlStore isn't a core thing.
Xliff [Coke]: Storable6. Yes.
[Coke] So, right now, there's no core P6 Storable, (and .perl and .gist weren't meant to fill that niche), and you'll have to talk to teodozjan about PerlStore.
(so, I would consider this a conversation starter, but don't expect anything at the moment to fill this particular niche) 14:17
Xliff my @b; my @c; my @a = (1, @b, @c); @b = (2, @c, @a); @c = (3, @b, @a); say @a;
m: my @b; my @c; my @a = (1, @b, @c); @b = (2, @c, @a); @c = (3, @b, @a); say @a; 14:18
camelia rakudo-moar 301bcf: OUTPUT«(\Array_55036336 = [1 (\Array_55037008 = [2 [3 Array_55037008 Array_55036336] Array_55036336]) (\Array_55037064 = [3 [2 Array_55037064 Array_55036336] Array_55036336])])␤»
IOninja Xliff: .gist won't let you roundrip data, even if it can handle circular structures.
Xliff m: my @b; my @c; my @a = (1, @b, @c); @b = (2, @c, @a); @c = (3, @b, @a); say @a.gist;
IOninja m: say <1/2>
camelia rakudo-moar 301bcf: OUTPUT«(\Array_52104392 = [1 (\Array_52105064 = [2 [3 Array_52105064 Array_52104392] Array_52104392]) (\Array_52105120 = [3 [2 Array_52105120 Array_52104392] Array_52104392])])␤»
rakudo-moar 301bcf: OUTPUT«0.5␤»
IOninja m: say "5"
camelia rakudo-moar 301bcf: OUTPUT«5␤»
Xliff I'm onlyh worried about the circular structures at this point. It was a GOTCHA I didn't forsee. 14:19
14:20 pyrimidine left
[Coke] Looking at FIleStore, it is just using .perl, so that's not helpful. 14:20
14:20 pyrimidine joined
Xliff Screw it. If I need to I can just use .perl or .gist and gzip the resulting file. 14:21
[Coke] I think rakudo has all the pieces that you could render it out as a bytecode file (but that wouldn't be portable between backends)
Xliff Or now.
s/now/not/
I think using compression on the resulting .perl or .gist output would be bytecode enough.
Not efficient, but it should work. 14:22
14:23 Ven joined
Xliff Although I get nervous when I do a .perl or a .gist and the program just hangs. 14:23
DrForr .perl .. still.. can't dump circular structures? I'd swear I filed that bug at least a year ago. 14:24
timotimo it can
IOninja It can't dump a lot of things.
timotimo there's $*PERLSEEN or what is it called? 14:25
IOninja m: (1...*).perl
camelia rakudo-moar 301bcf: OUTPUT«Cannot .elems a lazy list␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
IOninja .gistseen, but it's used only for .gist, innit?
m: my @a; @a = [@a]; @a.perl.say
camelia rakudo-moar 301bcf: OUTPUT«[]␤»
IOninja m: my @a; @a[0] = @a; @a.perl.say
camelia rakudo-moar 301bcf: OUTPUT«(my \Array_74464672 = [Array_74464672,])␤»
DrForr Well, there are lots of things I wouldn't expect it to be able to dump, but it's good to know that the circular case is taken care of.
IOninja Oh, nevermind
14:26 pyrimidine left
IOninja m: my @a; @a[0] = @a; my $z = @a.perl.EVAL; say $z.perl 14:26
camelia rakudo-moar 301bcf: OUTPUT«$[Mu]␤»
Xliff Rut roh 14:27
IOninja seems broke
Xliff m: my @a; @a[0] = @a; my $z = @a.gist.EVAL; say $z.perl
camelia rakudo-moar 301bcf: OUTPUT«5===SORRY!5=== Error while compiling /home/camelia/EVAL_0␤Preceding context expects a term, but found infix = instead␤at /home/camelia/EVAL_0:1␤------> 3(\Array_80645888 =7⏏5 [Array_80645888])␤»
timotimo huh, misses the my?
IOninja .gist isn't meant to be evalable
timotimo oh, right
Xliff Oh. I thought it was the other way around. LOL 14:28
m: my @a; @a[0] = @a; my $z = @a.perl.EVAL; say $z.perl
camelia rakudo-moar 301bcf: OUTPUT«$[Mu]␤»
Xliff m: my @a; @a[0] = @a; my @z = @a.perl.EVAL; say @z.perl
camelia rakudo-moar 301bcf: OUTPUT«[Mu]␤»
Xliff My .gist still hasn't done anything.
Generally a sign that something is stuck down the rabbit hole, somewhere. 14:29
Geth doc: 84a1c452de | (Will "Coke" Coleda)++ | doc/Language/functions.pod6
fix typo
 14:47
14:49 Sound joined 14:52 llfourn left 15:04 ocbtec joined 15:06 matiaslina left 15:14 patrickz joined 15:15 user10 joined, user9 left, user10 is now known as user9
tbrowder hi #perl6 15:18
IOninja \o 15:19
15:19 wamba joined
tbrowder continuing suggestions for website, and IOninja's comments about using p6 because of memory use: maybe a dumb question, but would it help )if possible) p6 to make it a daemon process somehow so the startup time is once? 15:20
15:21 llfourn joined
timotimo the way he's suggesting you use it is already to make it a daemon process 15:22
15:22 araraloren joined
timotimo by using Bailador or something similar 15:22
tbrowder any chance of bailador being made multi-threaded any time soon?
timotimo well, there's a bunch of stability improvements still being made to multithreaded stuff 15:23
i think there's already multi-thread-capable web frameworks
but i haven't had to build any web apps in a long time with perl6
so i don't know which. maybe it's something to do with "crust"?
tbrowder would bailador only block while responding to a request?
timotimo it won't take in or respond to more than one request at a time 15:24
15:24 Ven left
tbrowder so a lightly used site wouldn't be too painful. i mean look at the SLOW dynamic sites out there now! 15:25
timotimo you're free to kick off threads as much as you want, but it won't allow you to respond to a request later than a new request coming in
aye, and if you serve static files from your web server itself (i.e. apache/nginx/cherokee/iis/...) there will only be load for the truly dynamic pages
tbrowder hm, i'll look at crust again... 15:26
timotimo ah, crust is really a thing that a web framework would use 15:27
so you want something that depends on crust rather than crust itself
tbrowder another question about the proxying for bailador: incoming is https, proxy is http, what happens to the response from the proxied server, back out via https?
IOninja yes 15:28
tbrowder cool--thanks IOninja and timotimo!
timotimo it would be a pretty amazing feat if the server would actually somehow make the browser open an HTTP connection so it could send the response back
and somehow make it properly recognize what request that was the response to 15:29
"Frinfon" is a minimal sinatra clone, that's a web framework
that's apparently the only one i can find in perl6-all-modules that depends on Crust and is a web framework like thing 15:30
i don't even know if crust has multithreaded capabilities :) :)
15:31 patrickz left 15:33 patrickz joined 15:36 faraco left 15:38 llfourn left 15:49 xtreak joined 15:52 Ven joined, patrickz left 16:03 ssotka joined 16:14 khw joined 16:15 Ven left
sena_kun [Coke], ping. Any info about github.com/perl6/doc/pull/1181? 16:16
RabidGravy timotimo, Crust relies on the underlying server engine for the multithreaded capabilities 16:17
16:18 cdg joined 16:19 patrickz joined
timotimo OK, so if you put it on top of HTTP::Server::Threaded or HTTP::Server::Async it'd work well? 16:20
RabidGravy well assuming they do that sort of PSGI thing
HTTP::Server::Tiny works
timotimo is Tiny multi-thread-capable? 16:21
tony-o Async is stable with moar-blead 16:22
16:22 Ven joined
tony-o and you could build a PSGI type interface on top of it 16:22
RabidGravy I keep meaning to make that 16:23
timotimo it's stable? cool!
RabidGravy: for lumberjack you used crust directly, right?
so you haven't tested it with async or threaded servers?
RabidGravy no
[Coke] sena_kun: sorry, very busy with $DAYJOB, haven't had a chance to review yet. 16:24
sena_kun [Coke], don't worry, just checking is it actual.
RabidGravy anyhow I visited the warehouse of the company I am working for today 16:25
some really cool robotic stuff
IOninja :o
timotimo neat
ugexe pass the butter
timotimo do they also have these neat little robo buddies that skittle along the floor and move shelves around? 16:26
16:26 user9 left, user9 joined
RabidGravy no, unfortunately not, but the massive picking thing like a huge tap library robot was quite awe inspiring 16:27
IOninja
.oO( tap library..? )
 16:28
RabidGravy tape
IOninja Ah
RabidGravy really whizzy for something two storeys tall and as long as a tennis court 16:30
timotimo whoa. 16:32
our local hackspace is going to get a small pick&place machine
it takes up an area of 1.6m x 1.6m
dunno how tall it is 16:33
but damn, p&p is so fancy
huf i hope you use it to automate something absolutely silly
timotimo we're actually going to produce badges for our upcoming event 16:34
16:35 llfourn joined
tony-o where do you work? 16:40
16:40 user9 left, user9 joined 16:43 llfourn left
RabidGravy who? 16:44
timotimo the babe with the power
RabidGravy I am working for an "online luxury fashion retailer" right now 16:45
16:45 user9 left 16:48 pyrimidine joined
sena_kun Am I wrong or we had a module for creating skeletons for new modules? Not META6-bin, but something else with lib/, t/, etc. 16:49
timotimo mi6 16:50
IOninja buggable: eco Ddt
buggable IOninja, Ddt 'Distribution Development Tool a replacement for mi6': github.com/kalkin/Ddt
timotimo interesting 16:51
sena_kun perfect! thanks. 16:52
16:53 user9 joined 16:54 user9 left
RabidGravy I keep meaning to actually write the one that Oyatul was supposed to be a part of 16:59
timotimo ouya tool?
RabidGravy Booyah! 17:00
IOninja wonders if "Sam S" and "smsl" are the same person
Show up as separate in rakudo's contributor tool
timotimo they are not
RabidGravy well I actually made it for a family of things
IOninja Ah, OK
timotimo github.com/ssutch - this is sam s
IOninja cool
timotimo at least i think so
IOninja no contrib activity 17:01
RabidGravy anything that wants some files laid out in a particular way
IOninja Yet this person apparently submitted one commit to doc, nqp, MoarVM, roast, or rakudo repos
don't matter much I guess 17:02
timotimo BBIAB
IOninja oh, it *is* the same person: github.com/perl6/doc/blob/master/C...#L253-L269 17:04
.ask smsl you're listed twice in contributors under "Sam S" and "smsl". Which should be used? Both in docs credits: github.com/perl6/doc/blob/master/C...#L253-L269 and `smsl` in Rakudo's credits: github.com/rakudo/rakudo/blob/nom/...#L717-L718 17:05
yoleaux IOninja: I'll pass your message to smsl.
17:05 user9 joined 17:10 patrickz left 17:16 patrickz joined 17:17 Ven left
samcv hmm would be cool if Pod::To::Markdown would use ```perl6 ``` for quoting perl 6 code 17:25
yoleaux 12:17Z <thundergnat> samcv: Take a look at rosettacode.org/wiki/Negative_base_...ers#Perl_6 for negative base number for bases -2..-36.
samcv nice. i already did bases -2 to -10 at least. but that is good so i can extended it further :P
to add more of the most useful bases
IOninja IIRC that's just a GitHub extension and would mess up proper Markdown 17:28
Oh, I think even the ``` bit is GitHub extension. I have a hack to reformat it on perl6.party that uses a Multimarkdown renderer: github.com/zoffixznet/perl6.party/...sts.pm#L34 17:29
samcv does ```perl6 mess up other markdown? 17:30
i thought other ones would just ignore it if they can't syntax highlight it?
RabidGravy I dunno, but other things definitely support the ```
samcv ^
and all the ones i have seen, if they don't have syntax highlighter installed or don't support syntax highlighting just ignore what's after the ``` 17:31
until a newline
github.com/samcv/perl6-number-base did i mess up how i did the Pod? 17:32
cause this looks ugly
this is what I did github.com/samcv/perl6-number-base...ase.pm6#L3
IOninja No mention of ``` in standard: daringfireball.net/projects/markdown/syntax 17:33
The "other things" must be all the gazillion variations
samcv ``` is pretty common though 17:34
m: say 10.base(-10)
camelia rakudo-moar 02815c: OUTPUT«base argument to base out of range. Is: -10, should be in 2..36␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
samcv m: say "e92".base(10) 17:35
camelia rakudo-moar 02815c: OUTPUT«No such method 'base' for invocant of type 'Str'␤ in block <unit> at <tmp> line 1␤␤»
samcv m: say "e92".parse-base(10)
camelia rakudo-moar 02815c: OUTPUT«Cannot convert string to number: malformed base-10 number in '3⏏5e92' (indicated by ⏏)␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
samcv m: say "e92".parse-base(-10)
camelia rakudo-moar 02815c: OUTPUT«Radix -10 out of range (allowed: 2..36)␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
samcv maybe that should be changed to "malformed base 10 number"
cause base-10 looks like base -10
what do you think IOninja 17:36
AlexDaniel that's a bit too much
samcv hmm? 17:37
AlexDaniel current message is alright
samcv well since base -10 exists, i think the error message is not as good as it can be
to put a hyphen-minus in front of the 10
AlexDaniel wait… base -10?
samcv yes. my module supports it :P
it's a real base
diakopter according to rosetta code: "Negative base numbers are an alternatively way to encode numbers without the need for a minus sign. " 17:38
samcv yep
like base -2
diakopter those alternatively ways
AlexDaniel wait, is it this balanced thingy?
samcv is-deeply from-base('101', -2), 5, "from-base -1 for '101' is 5";
here is one test i have in my code 17:39
IOninja samcv: message looks fine to me
samcv 101 in base -2 is 5
AlexDaniel hm… but still, the error message is fine
17:39 llfourn joined
AlexDaniel it says it's out of range, only 2..36 is allowed 17:39
17:39 cibs left
samcv that one is fine 17:40
i'm talking about the other one
m: say "ae".parse-base(10)
camelia rakudo-moar 02815c: OUTPUT«Cannot convert string to number: malformed base-10 number in '3⏏5ae' (indicated by ⏏)␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
samcv would be better if it just said "base 10" or something
17:41 sufrostico left, cibs joined
samcv m: say (-2)**$_ for ^5 17:42
camelia rakudo-moar 02815c: OUTPUT«1␤-2␤4␤-8␤16␤»
samcv AlexDaniel, that's how negative bases work
1010 in base -2 is -10 for example 17:43
has a 1 in the -2 place, and a 1 in the -8's place
17:43 sufrostico joined
AlexDaniel but it's different from balanced ternary, right? 17:43
samcv yes 17:44
AlexDaniel hmh
samcv ternary is 0, 1, or 2, where the 2 == -1
though they often denote it with some other symbol that isn't a 2
17:44 llfourn left
samcv hmm think i will try and get to the backlog of atom-language-perl6 issues this morning 17:46
not that many. but always easier to hunt and find them all in the same session
17:48 sufrostico left, itaipu joined 17:51 Coleoid_n left 17:56 bwisti joined 18:13 sufrostico joined 18:16 TimToady joined
samcv argh on the docs i can't even search for Regex and get the Language/Regexes page 18:29
we should give the Language pages more weighting
IOninja m: class { has $a = 42; has $.b where $!a }.new: :42b
camelia rakudo-moar 02815c: OUTPUT«Cannot look up attributes in a VMNull type object␤ in block <unit> at <tmp> line 1␤␤»
IOninja Is this supposed to work at all?
AlexDaniel c: 2015.01.1 class { has $a = 42; has $.b where $!a }.new: :42b 18:31
committable6 AlexDaniel, ¦«2015.01.1»:
IOninja Ah.. right :)
AlexDaniel it worked for 3 months, so… 🤷 18:32
IOninja oh, it's pre-Christmas... hmm
TimToady we make sure the = bits happen in the right order, but I dunno about when 'where' runs; maybe 'where' wasn't working at all back then
IOninja c: 2015.01.1 my $z = class { has $a = 42; has $.b where $!a }.new: :42b; say $z.b 18:33
committable6 IOninja, ¦«2015.01.1»: 42
IOninja c: 2015.01.1 my $z = class { has $a = 42; has $.b where $!a }.new: :b("lulz"; say $z.b
committable6 IOninja, gist.github.com/c381503668e3c64112...76b918554f
IOninja c: 2015.01.1 my $z = class { has $a = 42; has $.b where $!a }.new: :b("lulz"); say $z.b
committable6 IOninja, ¦«2015.01.1»: lulz
AlexDaniel oh
Geth doc: 8cd3001dbf | (Samantha McVey)++ | assets/sass/_highlights.scss
Try and highlight the m and rx in m// and rx// differently
AlexDaniel well :)
IOninja c: 2015.01.1 my $z = class { has $a = 42; has $.b where {$!a == 42} }.new: :b("lulz"); say $z.b
committable6 IOninja, ¦«2015.01.1»: lulz
IOninja Yeah, it's just the where ain't working
oh wait 18:34
c: 2015.01.1 my $z = class { has $a = 42; has $.b where {$!a == 41} }.new: :b("lulz"); say $z.b
committable6 IOninja, ¦«2015.01.1»: lulz
IOninja Yeah
c: 2015.01.1 my $z = class { has $a = 42; has $.b where {dd $!a; $!a == 41} }.new: :b("lulz"); say $z.b
committable6 IOninja, ¦«2015.01.1»: lulz
IOninja puts this ticket back into "hard" pile and moves on
18:35 pyrimidine left, llfourn joined, pyrimidine joined 18:40 llfourn left, pyrimidine left
Geth doc: 54fa9b4324 | (Samantha McVey)++ | htmlify.p6
Fix build for macosx
 18:43
18:54 sufrostico left
samcv what is with travis being super slow 18:55
IOninja gifted horse.... 18:56
geekosaur ^ we're freeloading on a commercial service, commercial users come first then they handle any free users as resources are available 18:57
18:58 labster joined
wcummings if i install rakudo from source, is that supposed to include zef? 19:01
sena_kun wcummings, nope.
19:01 xtreak left
sena_kun afaik, at least. 19:02
IOninja wcummings: if by "from source" you mean compiler's source, then no. There are bootstrapping instructions in zef's readme tho
buggable: eco zef
buggable IOninja, zef 'It's like [cpanm] wearing high heels with a tracksuit': github.com/ugexe/zef
IOninja huh... GitHub uglified their top panel? 19:03
samcv omg whyyy
sena_kun seems bad. :|
samcv and the rest of the page is white?
sena_kun, be happy it doesn't follow you as you scroll down the page 19:04
just wait
sena_kun samcv, following by scrolling is not so bad as it seems and can be useful depending on... different stuff. But new top looks like someone accidently inverted colors and pushed it without testing. 19:05
but "we're freeloading on a commercial service"-argument works here too. 19:06
IOninja
.oO( there're always user styles... )
 19:07
19:08 cdg left
AlexDaniel u: subscript c 19:09
unicodable6 AlexDaniel, U+0656 ARABIC SUBSCRIPT ALEF [Mn] (◌ٖ)
AlexDaniel, U+1D62 LATIN SUBSCRIPT SMALL LETTER I [Lm] (ᵢ)
AlexDaniel, gist.github.com/4e5daff0b4555e33aa...1187dd6268
19:09 mcmillhj joined 19:12 mcmillhj left, espadrine left
samcv sena_kun, now when i have a tiny screen 19:12
sena_kun samcv, fullscreen browser mode? 19:14
samcv nope 19:15
IOninja do we have something other than nqp to test whether a thing is containerized?
m: use Test; is-deeply [], [].item.perl.EVAL
camelia rakudo-moar 02815c: OUTPUT«ok 1 - ␤»
IOninja I need something like that ^ except it'd also verify the itemization is there
.. without poking at .perl string
sena_kun samcv, modern web is young ==> all about violence 19:16
samcv heh 19:17
is this for the Test module? 19:18
IOninja my Q? No, I'm just writing a roast test.
samcv ah i see
you can see if it's in a Scalar
m: my $var; say $var.VAR.^name 19:19
camelia rakudo-moar 02815c: OUTPUT«Scalar␤»
IOninja m: $().VAR.^name.say 19:21
camelia rakudo-moar 02815c: OUTPUT«Use of Nil in string context␤ in block <unit> at <tmp> line 1␤Str␤»
IOninja m: ().VAR.^name.say
camelia rakudo-moar 02815c: OUTPUT«List␤»
19:23 itaipu left
tony-o m: my \x = $(); x.VAR.^name.say; 19:23
camelia rakudo-moar 02815c: OUTPUT«Use of Nil in string context␤ in block <unit> at <tmp> line 1␤Str␤»
tony-o m: my \x = $(); x = 5; x.VAR.^name.say;
camelia rakudo-moar 02815c: OUTPUT«Use of Nil in string context␤ in block <unit> at <tmp> line 1␤Cannot modify an immutable Str␤ in block <unit> at <tmp> line 1␤␤»
19:23 itaipu joined
tony-o m: my $y = 5; my \x = $y; x = 5; x.VAR.^name.say; 19:23
camelia rakudo-moar 02815c: OUTPUT«Scalar␤»
tony-o m: my $y = 5; my \x = $y; x = 'hello world'; x.VAR.^name.say;
camelia rakudo-moar 02815c: OUTPUT«Scalar␤»
tony-o m: my $y = 5; my \x = $y; x = %( x => 5 ); x.VAR.^name.say; 19:24
camelia rakudo-moar 02815c: OUTPUT«Scalar␤»
samcv m: my $var; say $var.VAR.^methods
camelia rakudo-moar 02815c: OUTPUT«(<anon> <anon> new name of default dynamic WHICH)␤»
samcv m: my \var; say var.VAR.^name
camelia rakudo-moar 02815c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Term definition requires an initializer␤at <tmp>:1␤------> 3my \var7⏏5; say var.VAR.^name␤»
samcv m: my \var = 0; say var.VAR.^name
camelia rakudo-moar 02815c: OUTPUT«Int␤»
samcv that seems to work fine
m: my $var = 0; say $var.VAR.^name 19:25
camelia rakudo-moar 02815c: OUTPUT«Scalar␤»
samcv yep
19:25 darutoko left
IOninja m: say $().VAR.^name 19:25
camelia rakudo-moar 02815c: OUTPUT«Use of Nil in string context␤ in block <unit> at <tmp> line 1␤Str␤»
IOninja m: say ($()).VAR.^name 19:26
camelia rakudo-moar 02815c: OUTPUT«Use of Nil in string context␤ in block <unit> at <tmp> line 1␤Str␤»
IOninja m: say ().item.VAR.^name
camelia rakudo-moar 02815c: OUTPUT«Scalar␤»
IOninja weird
: $[].perl.say 19:27
,: $[].perl.say
m: $[].perl.say
camelia rakudo-moar 02815c: OUTPUT«$[]␤»
IOninja m: [].item.perl.say
camelia rakudo-moar 02815c: OUTPUT«$[]␤»
IOninja m: say quietly $[].VAR eqv [].item.VAR
camelia rakudo-moar 02815c: OUTPUT«True␤»
IOninja m: say quietly $[].VAR.^name
camelia rakudo-moar 02815c: OUTPUT«Scalar␤»
IOninja m: say ($(),).VAR.^name 19:28
camelia rakudo-moar 02815c: OUTPUT«Use of Nil in string context␤ in block <unit> at <tmp> line 1␤List␤»
IOninja Oh, dammit, tripped on the very bug I'm fixing!
m: say $( ).VAR.^name
camelia rakudo-moar 02815c: OUTPUT«Scalar␤»
IOninja samcv++ thanks
19:29 raiph joined 19:32 espadrine joined
IOninja FWIW, grepping the doc repo for `$()` gives nothing, but it's a shorthand for $/.made ?? $/.made !! $/.Str 19:32
RabidGravy :-O 19:34
gfldex IOninja: does roast agree? 19:36
19:36 llfourn joined 19:40 cpage_ left
IOninja Ah, always forget that step :P 19:41
19:41 llfourn left
IOninja gfldex: this seems to be it: github.com/perl6/roast/blob/master....t#L16-L22 19:42
Though I don't see it testing that it uses $/.made ?? as opposed to say $/.made.defined ?? for example 19:43
And personally, I'd expect .defined thing, probably. Never used $() tho
19:44 TEttinger joined
IOninja m: 'foo' ~~ /o+ {make 42}/; say $() 19:46
camelia rakudo-moar 02815c: OUTPUT«42␤»
IOninja m: 'foo' ~~ /o+ {make 0}/; say $()
camelia rakudo-moar 02815c: OUTPUT«oo␤»
IOninja .DEFINITE even 19:47
m: 'foo' ~~ /o+ {make 0 but True}/; say $() 19:48
camelia rakudo-moar 02815c: OUTPUT«0␤»
IOninja hehe
[Coke] yawns. 19:49
19:50 hartenfels joined
ufobat i need some help pls, following line of code produces this output, and i dont understand why 19:51
say "r = $r ceiling -> ", $r.ceiling, " ", $r.perl, " what:", $r.WHAT;
r = 2 ceiling -> 3 <2/1> what:(Rat) 19:52
m: my $a = <2/1>; say $a.perl, $a.ceiling, $a; 19:53
camelia rakudo-moar a148c7: OUTPUT«2.022␤»
IOninja tries to parse that 19:54
I guess r = 2 ceiling -> 3 <2/1> what:(Rat) isn't perl code?
ufobat no thats the output of the say statement
IOninja Ah :D
m: my $a = <2/1>; say [ $a.perl, $a.ceiling, $a ]
camelia rakudo-moar a148c7: OUTPUT«[2.0 2 2]␤»
IOninja Looks good to me? 19:55
ufobat: how do you make $r?
I'm guessing there's floating point noise 19:56
ummm
ufobat but shouln't i see this in $a.perl ? which is 2/1
IOninja ufobat: well, what's $r? how do you make it? 19:57
[Coke] say does a gist.
IOninja So far, I've not seen a reproduction of your issue
ufobat i wasn't able to track it down, i dont know, its part of my xpath evaluation
[Coke], but it is a say $a.perl, so no gist, right? 19:58
IOninja s: 2.0, 'ceiling', \()
SourceBaby IOninja, Sauce is at github.com/rakudo/rakudo/blob/0281...nal.pm#L60
El_Che samcv: I am looking how to update the syntax checking support for perl6 on vim (with JSON error output). How does that work on atom? Does it call an external program (eg "perl6 -c")?
IOninja s: 2.0, 'perl', \() 19:59
SourceBaby IOninja, Sauce is at github.com/rakudo/rakudo/blob/0281.../Rat.pm#L5
IOninja ufobat: what version are you on?
ufobat 2017.01
IOninja aha, well here's the first clue: github.com/rakudo/rakudo/blob/0281...t.pm#L6-L8
Since it's not ending up as 2.0 in .perl, then denominator isn't exactly 1 20:00
And I think there's a ticket for this issue that .perl isn't accurate
(or did I supposedly fix that already?)
ufobat right i get <2/1
> instead of 2.0 20:01
IOninja Yeah, suppoosed to be fixed: github.com/rakudo/rakudo/commit/b5...a15618151d
Ummm, wait a sec. 20:02
nu/de are Ints :S there's no fp noise :S
m: <2/1>.perl.say
camelia rakudo-moar a148c7: OUTPUT«2.0␤»
IOninja ufobat: well, I'll be very interested in a repo case. 20:03
ufobat IOninja, sorry? what, you mean you'd like to see my code?
IOninja ufobat: I'd like to see code that I can run and reproduce that problem 20:04
[Coke] m: <2/1>.perl.put
camelia rakudo-moar a148c7: OUTPUT«2.0␤»
ufobat IOninja, yeah but it isnt tracked down:
IOninja don't matter
I can stick debug prints into rakudo at least and dump stuff 20:05
ufobat give me a sec
github.com/ufobat/p6-XML-XPath/blo...ions.t#L32
github.com/ufobat/p6-XML-XPath/blo...l.pm6#L106 20:06
20:07 cale2 joined, cpage_ joined
IOninja Could not find XML::XPath::ExprOperator::And 20:08
What dist is that in?
ufobat i am sorry, it is pushed now
cale2 hello 20:10
20:10 cpage__ joined
IOninja lol 20:11
I stuck dd [$r, "X42"]; before the buggy line and the bug vanished
20:12 cpage_ left, cpage__ is now known as cpage_, dj_goku joined
ufobat okay, so we just keep the dd there and everthing will be fine :D 20:12
IOninja Schrödinger's bug... I can ask what type it is and all, but as soon as I try to peek at its nu/de, bug goes away :) 20:13
IOninja digs furtehr
20:14 Pattux joined
ufobat gives IOninja a box of cookies and a beer 20:15
IOninja beer? gross
Pattux exit 20:16
ufobat what!? am i not the only one that finds beer disgusting?
IOninja m: say 4 div 2 + 1
camelia rakudo-moar a148c7: OUTPUT«3␤»
RabidGravy weirdos
IOninja ahah
20:17 Pattux left
cale2 I don't get the obsession with craft beer, but beer is fine in general 20:17
IOninja prepares sandwich bread
Bug sammiches for dinner
RabidGravy "craft beer" is just a term for marketing to hipsters 20:18
20:18 dj_goku left
RabidGravy I buy beer from a local brewery whenever I can 20:18
20:19 bjz joined
tony-o cale2: mostly a hobby for some people, finding out what's out there, trying obscure things 20:19
IOninja ufobat: compiling a fix
ufobat IOninja, yay :D 20:20
cale2 Yeah, I like buying beer from local breweries too. But I have more of a pedestrian palette for beer. Can't do nitro stout or christmas ale 20:21
tony-o i don't care for the seasonal stuff either
ufobat IOninja, what was it?
IOninja ufobat: still thinking about it. 'cause I can't obtain the Rat you got by hand 20:24
oh, maybe some math operation...
hm 20:25
IOninja tries to trace how $r is created
ufobat: where does $r come from? 20:26
where is it made
ufobat well it starts with fn-last()
IOninja Basically, somehow the $r you got is <4/2> 20:28
m: use nqp; my $r = <4/2>; say [ nqp::getattr(nqp::decont($r), Rat, '$!numerator'), nqp::getattr(nqp::decont($r), Rat, '$!denominator') ]
camelia rakudo-moar a148c7: OUTPUT«[2 1]␤»
IOninja ^ but if I make one manually, it gets reduced in .new
20:28 pyrimidine joined
ufobat the $of in fn-last is the $start.elems from Line 12 of Predicates.pm6 20:28
IOninja So there's some step that's missing that reduction step. Right now I stuck it into .ceiling, but feels like it's missing elsewhere
20:29 sena_kun left, r3m joined 20:30 ocbtec left
ufobat the rest is XML::XPath::ExprOperator::* ; *::Div and *::Plus 20:31
if you see a 4/2 its probably the div operator 20:33
20:33 pyrimidine left
ufobat not sure what XML does, but i'd expect its more a less a "4" / "2" 20:34
IOninja ufobat: well, I've just pushed a temporary fix. If you build it you can resume your work. I'll try hunting where the <4/2> Rat gets made and see if a better fix can be made 20:35
ufobat IOninja, awesome :) thanks for your help! 20:36
IOninja Thanks for the bug!
IOninja bites into the sandwich
Yummy
20:37 llfourn joined 20:42 robertle joined, llfourn left 20:49 pyrimidine joined
RabidGravy not digging this new style on github 20:50
20:50 sena_kun joined
IOninja yeah, it's pretty gross 20:51
gfldex it does provide a clear separation between github stuff and a repo 20:52
20:53 yqt joined 20:54 pyrimidine left
geekosaur userstyles.org/styles/37035 ftw 20:54
also userstyles.org/styles/121315 20:55
21:05 cdg joined 21:07 cdg_ joined, bjz left 21:11 cdg left, cdg__ joined 21:14 cdg_ left 21:16 cdg joined
[Coke] guesses he's on the other side of an A/B split on github. 21:17
21:18 itaipu left
geekosaur mm, possibly I am too, in which case I presume my style is about to break :/ 21:19
21:19 cdg__ left
geekosaur thought loading it in an incognito window looked different from what he recalled of unstyled though 21:19
21:21 pyrimidine joined
[Coke] the top menu bar changed color if I'm not logged in, that seems to be it. 21:23
21:25 Sgeo left 21:27 Tonik joined 21:32 pyrimidine left 21:33 pyrimidine joined 21:34 ufobat left 21:35 patrickz left 21:38 pyrimidine left 21:39 llfourn joined 21:44 llfourn left 21:49 sena_kun left 21:54 rindolf left, Tonik left 22:00 dj_goku joined 22:06 dj_goku left, labster left 22:07 cdg_ joined 22:08 labster joined 22:09 bjz joined, RabidGravy left, itcharlie1 joined 22:10 cdg left 22:12 cdg joined 22:15 cdg_ left, Sound left 22:20 bjz left 22:21 travis-ci joined
travis-ci Doc build errored. Samantha McVey 'Try and highlight the m and rx in m// and rx// differently' 22:21
travis-ci.org/perl6/doc/builds/200444907 github.com/perl6/doc/compare/84a1c...d3001dbf5a
22:21 travis-ci left
samcv looks like it just timed out 22:22
22:23 cdg_ joined
timotimo travis has been acting a bit strange lately 22:25
22:25 cdg left 22:29 torbjorn joined 22:36 bjz joined 22:38 cdg joined 22:40 llfourn joined 22:41 cdg_ left
El_Che \o/ my fist implementation of syntax checking for perl6 in vim using rakudo JSON error output is kind of workingç 22:43
!
jnthn :)
Cool!
.oO( mighta been quicker to use your fingers, but whatever works... )
 22:44
El_Che previous version used regex, but this is way cooler
jnthn *nod* 22:45
Yeah, a bunch more robust
22:45 llfourn left
El_Che I wouldn't call it robust there, but it will get there :) 22:46
perlpilot "rakudo JSON error output"? 22:48
22:49 vktec joined
timotimo yep 22:51
El_Che perlpilot: RAKUDO_EXCEPTIONS_HANDLER=1
timotimo um, 1? not JSON?
El_Che JSON 22:52
sorry
vktec I'm considering learning Perl6. What are the downsides of it? 22:53
El_Che If anyone is interested, this branch works: github.com/nxadm/syntastic-perl6/tree/json
timotimo performance isn't very good, that's definitely a downside of perl6 at the moment
El_Che It needs a lot of cleanup and some todo (e.g., highlight error instead of margin)
perlpilot vktec: The main downsides are its newness and performance. But it's otherwise very nice. 22:54
El_Che vktec: new language, needs library (you change to be famous)
timotimo your chance*
El_Che f*ck, too tired
timotimo though you can use libraries from python and perl5 freely
perlpilot :q!
oops
vktec Okay. How bad is the performance issue? 22:55
El_Che vktec: nice enough to force me to learn a little vimscript to have support for it
IOninja vktec: after you learn it, it'll be hard to learn other languages :)
vktec El_Che: Whoa, that's impressive! :P
IOninja: Hahaha
perlpilot vktec: It's not too bad for most things that I've done with it.
vktec Okay. 22:56
perlpilot vktec: Also though, there are "escape hatches" for performance ... you can use Inline::Perl5 to talk to Perl 5 if you need to or with NativeCall, write the slow bits in C
vktec Cool
perlpilot (there are other Inline:: modules to talk to other languages too)
vktec That's a very neat feature
IOninja w00t. nanis's PR successfully fixes RT#130736 on my box. 22:58
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130736
jnthn The concurrency/parallelism stuff can also help somewhat on performance (if it's applicable to what you're doing)
IOninja umm, wtf now it fails :S 23:02
El_Che vktec: for me, I like how expressive it is (a lot of common functions/methods are built so code can be short), modern (unique OO system, gradual typing, concurrent) and dead easy C-binding. The community is very accessible. Downsize is age (needs more libs) and speed for sometype of applications (although it's getting a lot faster). 23:04
vktec Okay, last question before I go start learning Perl6: Any good web frameworks for it yet or should I call back into Perl5 and use something like Dancer2?
El_Che: Okay. The performance issue isn't really a problem for me (if performace mattered, I'd be using C), and age won't be an issue for all that long, I imagine 23:05
vktec goes off to start learning
IOninja vktec: there are a few in the ecosystem ( modules.perl6.org/ ), but AFAIK none of them are really mature yet. I had reasonable success using Mojolicious via Inline::Perl5 tho 23:06
vktec Gotcha. I'll probably stick with Python for web stuff in that case, and wait for things to become a bit more mature 23:07
El_Che vktec: that's a good plan
nine gave a talk at fosdem about web + perl6 but not there yet 23:08
(some audio out of sync issue)
vktec Cool. Well, I'm gonna go learn Perl6! This a good place to start? perl6intro.com/ 23:11
23:11 hartenfels left 23:16 itcharlie1 left
IOninja vktec: yeah. If you already know programming learnxinyminutes.com/docs/perl6/ could be handy. More stuff listed on this page in "For newcomers" section: perl6.org/resources/ 23:18
gfldex vktec: we got blogs too! pl6anet.org/
vktec IOninja: I'll check them out. Thanks! 23:19
gfldex: Ooh, I'll keep an eye on that
23:23 espadrine left 23:26 cibs left 23:28 cibs joined 23:30 lep-delete left 23:35 pyrimidine joined, lep-delete joined 23:37 dj_goku joined 23:38 bjz left, raiph left 23:39 pyrimidine left 23:42 llfourn joined 23:47 llfourn left 23:51 cale2 left 23:58 cibs left

Powered by Cro and the Raku® Programming Language.

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

Thank you!