»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.
AlexDaniel samcv: generally, this is what makes it switch to it: setxkbmap -compat 'complete+ef+ledcaps(group_lock)' 'ef,ef(phonetic)' -option 'grp:sclk_toggle,grp_led:caps 00:00
(missing ' on the end)
samcv kk
AlexDaniel samcv: you can take any other existing layout as an example 00:01
samcv: if I recall correctly, -compat thing is needed to work around some issue with japanese keyboards 00:03
samcv: ( bugs.freedesktop.org/show_bug.cgi?id=91042 )
samcv: so if you're just starting, you probably don't need it
samcv thank you kindly 00:04
AlexDaniel samcv: sclk_toggle is to make scroll lock switch between layouts, again it's something you probably don't need at all
so it is simply setxkbmap 'yourlayout'
00:04 mcmillhj joined 00:06 mst left, mst joined
AlexDaniel samcv: at the time I found this to be very helpful: medium.com/@damko/a-simple-humble-...1ad5e13450 00:07
00:09 Cabanossi left 00:10 Cabanossi joined, mcmillhj left 00:11 cyphase left
AlexDaniel samcv: if you get into problems where some of your custom keys are not working in some software, feel free to ping me. There are a couple of tricks I know 00:11
samcv thank you 00:13
00:13 mcmillhj joined 00:15 cyphase joined 00:16 rpburkholder left 00:18 mcmillhj left 00:23 cyphase left 00:24 labster left 00:25 Actualeyes left 00:28 cyphase joined, Actualeyes joined 00:33 mcmillhj joined 00:34 Actualeyes left 00:38 mcmillhj left 00:39 labster joined, cyphase left 00:43 mcmillhj joined 00:44 cyphase joined 00:48 mcmillhj left 00:52 cyphase left 00:57 cyphase joined 00:58 mcmillhj joined 01:00 sftp left, sftp joined 01:03 mcmillhj left 01:05 cyphase left, aborazmeh joined, aborazmeh left, aborazmeh joined 01:10 cyphase joined 01:15 mcmillhj joined 01:17 ChoHag joined 01:18 cyphase left 01:23 cyphase joined 01:24 mcmillhj left 01:29 mcmillhj joined 01:34 mcmillhj left 01:39 mcmillhj joined 01:43 mcmillhj left 01:45 ilbot3 left 01:48 ilbot3 joined, ChanServ sets mode: +v ilbot3 01:49 mcmillhj joined 01:53 Cabanossi left 01:54 mcmillhj left 01:55 Cabanossi joined 01:58 mcmillhj joined 02:03 mcmillhj left 02:08 mcmillhj joined 02:13 mcmillhj left 02:22 mcmillhj joined 02:30 mcmillhj left 02:31 alimon left 02:34 alimon joined 02:38 noganex_ joined 02:40 mcmillhj joined, noganex left 02:46 araujo left, araujo joined 02:47 mcmillhj left 02:52 mcmillhj joined, xtreak joined 02:53 Cabanossi left 02:55 Cabanossi joined, kurahaupo__ joined 02:56 llfourn_ joined, mcmillhj left 02:59 llfourn left 03:01 mcmillhj joined 03:05 mcmillhj left 03:07 Cabanossi left 03:09 Cabanossi joined 03:10 aborazmeh left 03:12 mcmillhj joined 03:13 vendethiel- joined, vendethiel left 03:17 mcmillhj left 03:21 mcmillhj joined 03:23 ChoHag left 03:26 mcmillhj left 03:27 drrho_ joined 03:30 pelev joined 03:31 pelev left, mcmillhj joined, pelev joined 03:32 pelev left 03:36 pedh joined, mcmillhj left 03:39 pelev joined, pelev left
AlexDaniel ===SORRY!=== Representation must be composed before it can be serialized 03:40
what does it mean?
03:41 pedh left
AlexDaniel ah-ha, it means I cannot ｢use …:from<Python>;｣ in a module 03:41
BenGoldberg was going to say, you need to do Classname.^compose, but that could be it instead. 03:42
AlexDaniel uh-oh, so how do I use something with Inline::Python if I'm doing it in a module?
BenGoldberg doesn't know. 03:43
In any case, g'night.
03:44 BenGoldberg left 03:46 darutoko joined 03:48 mcmillhj joined 03:51 Cabanossi left 03:53 mcmillhj left 03:54 Cabanossi joined
AlexDaniel .tell nine any ideas? irclog.perlgeek.de/perl6/2017-03-30#i_14348415 03:59
yoleaux AlexDaniel: I'll pass your message to nine.
04:00 mcmillhj joined 04:05 wamba joined 04:07 mcmillhj left
samcv looks rike that coverage i generated is incorrect because i think it copied over the old results every .t file was ran 04:08
so i guess i'll make a nqp shell script that sets a random one, then when it runs them all combine all the randomly named datafiles together
or maybe it combines them. i'm not really sure the files are the same size 04:10
04:11 mcmillhj joined 04:12 Resol joined 04:15 mcmillhj left 04:16 Resol left
u-ou I want to write a bot that connects to IRC via SSL. how do I do that? 04:19
yoleaux 29 Mar 2017 10:19Z <Zoffix> u-ou: Don't know if anyone mentioned, but if you're writing an IRC bot, we already have an IRC module you can use and not mess with sockets: modules.perl6.org/repo/IRC::Client
04:20 mcmillhj joined, cyphase left
u-ou I guess that's my answer in a way, but I'm not sure if I want to use a library yet. 04:22
04:24 mcmillhj left
AlexDaniel u-ou: so you want something to handle SSL stuff but don't want to use a module? 04:26
u-ou I didn't say "don't want to". just wondered if perl6 provided something for me. 04:28
04:34 mcmillhj joined 04:38 Cabanossi left 04:39 Cabanossi joined 04:41 dustinm` joined 04:43 cyphase joined 04:52 cyphase left 04:57 cyphase joined 05:01 curan joined 05:10 cyphase left 05:12 noganex joined 05:15 Exodist left, cyphase joined, noganex_ left, vytas left 05:16 vytas-local_ joined, AlexDaniel left 05:17 Exodist joined 05:20 gdonald left 05:21 gdonald joined 05:29 cyphase left, AmazingPudding is now known as DrEeevil
u-ou pony 05:29
05:30 mcmillhj left 05:32 Tonik joined 05:33 mcmillhj joined, cyphase joined 05:36 cyphase left
samcv u-ou, there's an IRC bot thing 05:37
05:37 cyphase joined
samcv i've built an irc bot using it 05:37
05:38 mcmillhj left
samcv buggable, eco IRC::Client 05:38
buggable samcv, Found 4 results: IRC::Client, IRC::Client::Plugin::HNY, IRC::Client::Plugin::Factoid, IRC::Client::Plugin::UserPoints. See modules.perl6.org/#q=IRC%3A%3AClient
u-ou ta, will look into it 05:39
05:39 wamba left 05:42 mcmillhj joined
samcv the source for my bot is here github.com/samcv/keira-perl6-ircbot in case you're interested. it all uses plugins 05:44
well the whole thing is a plugin. in the lib folder the Keira.pm6 file has the bulk of the code, that get simpored into the executable
05:46 domidumont joined
u-ou thanks :D 05:46
05:46 mcmillhj left 05:47 noganex left 05:48 noganex joined 05:49 xtreak left, mcmillhj joined 05:50 RabidGravy joined, xtreak joined 05:51 domidumont left 05:52 domidumont joined, Cabanossi left
u-ou gtg 05:52
05:53 xtreak left, xtreak joined 05:54 Cabanossi joined, mcmillhj left 05:55 Tonik left 05:58 cyphase left 06:02 cyphase joined 06:03 mcmillhj joined 06:08 mcmillhj left, labster left 06:09 lowbro joined, lowbro left, lowbro joined 06:10 cyphase left 06:13 labster joined 06:15 cyphase joined 06:18 domidumont left 06:19 mcmillhj joined 06:20 cyphase left 06:22 vendethiel joined, vendethiel- left, wamba joined 06:24 xtreak left, mcmillhj left 06:25 cyphase joined 06:27 xtreak joined 06:28 mcmillhj joined 06:32 mcmillhj left 06:33 cyphase left 06:36 mcmillhj joined 06:37 obfusk_ left 06:38 obfusk joined, cyphase joined 06:41 mcmillhj left 06:46 mcmillhj joined 06:49 cyphase left 06:51 mcmillhj left, Cabanossi left 06:54 mcmillhj joined, Cabanossi joined, cyphase joined, xtreak left 06:55 xtreak joined 06:59 pmurias joined, mcmillhj left
pmurias jnthn: re 'make spectest' on rakudo-debugger, does 'make spectest' test the rakudo debugger? 06:59
07:00 ufobat joined
samcv pmurias, i don't think it does 07:01
07:02 mcmillhj joined 07:07 mcmillhj left 07:09 cyphase left 07:11 dakkar joined 07:14 cyphase joined, mcmillhj joined 07:15 domi91c joined
domi91c hello 07:15
07:16 domi91c left 07:19 domidumont joined 07:20 mcmillhj left 07:24 cyphase left 07:27 kurahaupo__ left, skids left 07:29 cyphase joined, mcmillhj joined 07:32 xtreak left 07:33 xtreak joined, zakharyas joined, robertle left 07:37 mcmillhj left 07:38 Cabanossi left 07:39 Cabanossi joined 07:41 mcmillhj joined 07:45 cyphase left 07:46 mcmillhj left 07:50 mcmillhj joined, cyphase joined 07:54 mcmillhj left 07:56 wamba left, ChoHag joined 08:00 mcmillhj joined
pmurias why is nqp::if faster then if ... {}? 08:04
samcv do you think it is faster?
08:05 mcmillhj left
samcv i mean sometimes it's not any faster. do you have evidence showing it's slower or just assuming it's slower? 08:05
08:06 cyphase left
pmurias it's used in the CORE a lot, I'm just wondering if the regular statement could be optimized so it's no longer needed 08:06
samcv but yeah nqp::if can be faster depending on what is in the if condition. if it's a nqp type thing nqp::if can be faster somewhat if it's a hot codepath
ah
it would go here github.com/rakudo/rakudo/blob/nom/...imizer.nqp 08:07
there probably already is something there, but i'm not sure what it covers 08:08
but if you are finding nqp::if is being faster in one situation i would try and improve the optimizer
08:08 mcmillhj joined, Cabanossi left 08:09 Cabanossi joined 08:10 cyphase joined 08:13 mcmillhj left 08:15 cyphase left 08:16 tojo joined 08:18 xtreak left
lizmat has tried to get into the optimizer a few times 08:20
but I have not been able to wrap my head around it just yet :-(
this has probably to do with age 08:21
samcv i think i understand it. and then i keep going and then realize i know nothing at all
08:21 xtreak joined
samcv when that file starts to make sense and then suddenly you realize you have no clue what it's doing 08:21
08:25 wamba joined 08:26 dominix joined 08:29 dominix left 08:33 mcmillhj joined 08:38 mcmillhj left 08:43 xtreak left 08:44 xtreak joined 08:46 mcmillhj joined
El_Che samcv: that's the Socratic paradox :) 08:47
08:51 mcmillhj left 08:53 rindolf joined 08:54 mcmillhj joined 08:56 espadrine_ joined, pytuger left
pmurias do we want to implement count-only for stuff such as (-1.5.Num..^3).iterator? 08:58
we have a skipped test, but I'm not sure how useful the count-only method for such cases would actually be 08:59
08:59 mcmillhj left 09:08 xtreak left 09:13 mcmillhj joined 09:14 TEttinger left 09:16 pytuger joined, pytuger left, pytuger joined 09:18 mcmillhj left 09:21 gdonald left 09:22 gdonald joined 09:23 mcmillhj joined 09:24 M-Illandan left, Matthew[m] left, ilmari[m] left, unclechu left, dp[m] left, xui_nya[m] left, CIAvash[m] left, mienaikage left, tadzik left, wictory[m] left 09:26 wictory[m] joined, espadrine_ left 09:28 ufobat left, mcmillhj left 09:29 jonas1 joined 09:35 mcmillhj joined
timotimo There's no short-hand way to get a Promise that's already kept with a value you give, right? 09:36
like, you can't "Hello".Promise, or Promise.kept-value("Hello")
09:40 mcmillhj left 09:43 tadzik joined, dp[m] joined, M-Illandan joined, Matthew[m] joined 09:44 unclechu joined, ilmari[m] joined, mienaikage joined, xui_nya[m] joined, CIAvash[m] joined 09:46 jferrero left 09:47 xtreak joined, jferrero joined 09:48 labster left 09:51 ChoHag left 09:53 ChoHag joined 09:54 andrzejku joined
andrzejku hey where is Perl goo.gl/IxNzo8 09:54
how is with bug density in perl6?
10:00 mcmillhj joined 10:05 mcmillhj left 10:08 Actualeyes joined 10:09 mcmillhj joined 10:12 ufobat joined 10:13 mcmillhj left 10:18 mcmillhj joined 10:22 mcmillhj left 10:27 mcmillhj joined 10:30 astj left, astj joined 10:31 donaldh joined 10:32 mcmillhj left 10:35 astj left, astj joined 10:36 Cabanossi left, mcmillhj joined 10:39 Cabanossi joined 10:41 mcmillhj left 10:47 vendethiel left 10:49 tojo left 10:50 mcmillhj joined 10:55 mcmillhj left 10:57 wamba left 11:03 tojo joined 11:04 bjz joined 11:05 Actualeyes left, mcmillhj joined 11:10 mcmillhj left
SHODAN andrzejku, odd article 11:13
11:14 maasta left
andrzejku maybe 11:14
but how is with bug density in perl6?
11:15 mcmillhj joined 11:17 mcmillhj left
SHODAN i think you'll have to perform that study yourself if you're interested. i haven't heard of any such numbers :\ 11:17
timotimo there's not enough user code publically available on github for a sensible study, i'd say 11:18
SHODAN that might be a problem too
11:19 ufobat left
andrzejku okay but what do you think about it? 11:25
Perl6 vs C++ bug density 11:26
11:29 xtreak left
pmurias the way the measure "bug density" doesn't seem very reliable 11:29
11:31 ufobat joined
timotimo opened github issues per line of code? 11:32
pmurias commits with a message that looks like it fixes a bug
timotimo ah
andrzejku timotimo: I think it should be developer feelings statistics 11:33
11:33 donaldh left
timotimo no, definitely not 11:33
"how buggy does your language make your code?" "my language is perfect, go away" 11:34
that's got to be helpful!
"php provokes bugs" - "no php is flawless, you dumb bully!"
11:35 Actualeyes joined, mcmillhj joined
pmurias disgruntlement seems like a very useful thing to measure 11:35
andrzejku I don't think so 11:39
many developers which program for many many years can told you many stories with different projects and languages
11:40 mcmillhj left, ChoHag left
andrzejku I heard one a story about guy who need to fix some bugs in C/C++/scripting mixed code 11:40
and he was crying
after couple days
nebuchadnezzar interesting the GraphQL and Perl6 slides 11:41
11:42 ChoHag joined
pmurias andrzejku: most (all) technologies suck and annoy people in different ways, what would be interesting to measure is to what degree they do that 11:44
11:44 Actualeyes left 11:45 mcmillhj joined 11:50 mr-foobar left, mcmillhj left
profan pmurias: i think it's very difficult to come to any objective measure about it 11:56
while bug density may be hard to measure, "disgruntlement" is.. also quite up there in terms of difficulty i'd say :P 11:57
11:57 mcmillhj joined 12:02 mcmillhj left
DrForr timotimo: Were you the one playing with Synthesia? 12:02
timotimo ayup 12:03
DrForr Are you using your laptop keyboard, or an actual MIDI controller?
timotimo i'm the proud owner of an alesis vi61 12:05
12:07 geekosaur left
DrForr Shiny. I'm looking for something portable, having had to deal with selling at least one 61-key unit. 12:07
12:07 Cabanossi left 12:08 geekosaur joined 12:09 Cabanossi joined, Actualeyes joined
timotimo yeah, the vi61 is definitely not portable :) 12:09
my dad started my interest when he got some pretty tiny thingie for almost-free one time and showed it to me 12:10
it had pretty cheap keys, and not many of 'em 12:11
like one and a half octave maybe?
DrForr Indeed. I bought an 88-key Yamaha MIDI controller (weighted, velocity) and had a fun (for geeky "Okay, how the hell do I set this up..." values of fun) in order to practice as close as I could get to a real piano, but it's nothing really useful w/o GarageBand... 12:12
timotimo pff, garage band. there's so many cool tools out there to try 12:13
12:13 mcmillhj joined
timotimo if you'd enjoy fiddling with synth parameters and filters and such, have a look at sunvox 12:13
DrForr I know; it took me quite a while to figure out basic VS..R? sound stuff, I played w/ cubase.
timotimo oh, you mean the attack/decay/hold/fourth-thing 12:14
Woodi andrzejku: if you check some link in this article and link there, you will get that Perl5 is almoust as good as TypeScript :) but this article is just completly waste of time with terrible name with terrible arguments and conclusions...
timotimo ah, attack, decay, sustain, release
DrForr Yeah, ASDR envelopes.
It was just the keys, full 88-key multitouch velocity sensitive, and just figuring out the options to play back live sound was a challenge. 12:15
timotimo hehe 12:16
DrForr Hence garageband.
12:18 mcmillhj left
DrForr Synthesia looks fun enough (and I've watched enough video using it) to get me back into it, but I'd either need something like a Yamaha PSR taking up a huge amount of space (but with the fire-and-forget advantage) or take a more multimedia approach with a simpler keyboard. I can span more than an octave (I know, quit bragging) so at least a 49-key would work for me. 12:18
timotimo what, you can reach further than the same note on two octaves with just one hand? 12:20
DrForr Just barely.
12:21 drrho_ left
DrForr Double-jointed and all that. 12:21
12:21 drrho_ joined
andrzejku Woodi: hey 12:23
Woodi: priv?
timotimo FWIW, i just hook synthesia up to a qsynth aka fluidsynth 12:24
not much setup required
timotimo heads out now
12:24 tojo left 12:28 bjz left 12:42 donaldh joined 12:43 mcmillhj joined 12:47 raschipi joined 12:49 mcmillhj left, perlpilot left 12:51 perlpilot joined 12:56 curan left 13:00 ChoHag left 13:02 ChoHag joined 13:07 Cabanossi left 13:09 Cabanossi joined 13:10 mcmillhj joined 13:13 sufrostico joined
[Coke] www.platohistory.org/blog/2012/09/p...sited.html 13:14
raschipi Someone has to write a proposal so that Unicode can represent that. 13:16
DrForr youtu.be/AcS3NOQnsQM?t=115 13:19
[Coke] actually ponders making a module that, given a char or int, will print the str/codepoint/uniname 13:21
I've written so many one liners ending with a very long "say" at the end to dump that out…
raschipi Make it also able to dump property tables. 13:22
Like the bot.
masak [Coke]: may I point you to the excellent App::Uni? metacpan.org/pod/App::Uni 13:23
[Coke]: these days, I never leave home without it
13:24 xtreak joined
raschipi Looks like a good template, but it would be useful to have a p6 version so we can query what Perl6 thinks of them. 13:24
13:25 wcummings left
masak I won't stop you from porting App::Uni to Perl 6, of course. my guess is it ends up being much smaller. 13:26
[Coke] masak++ nifty 13:28
I could see adding a -v to get rashipi's desired output also. 13:29
raschipi There is Perl6 code for all of this already, just factor it out of the bots. 13:30
jnthn react/whenever example of the day: quick and dirty async SSH port forwarder github.com/jnthn/p6-ssh-libssh/blo...forward.p6 13:31
13:32 skids joined
raschipi "dirty" as in: cleaner than something that would be written in any other language except Erlang. 13:33
jnthn ;) 13:34
Well, more things like "it just goes and tears down the SSH session without caring about ongoing connections at shutdown" :) 13:35
What I actually needed to implement was reverse forwarding, alas... 13:36
jnthn goes to try and figure that out
raschipi "tears down the SSH session without caring about ongoing connections": Just let TCP deal with it. That's it's job anyway. 13:37
jnthn Well, there is that :)
nebuchadnezzar is enjoying the Damian's talk “On the shoulder of Giants” and wonders where Quantum::Computation module is :-D 13:45
13:51 Cabanossi left 13:54 Cabanossi joined
skids jnthn: had any time for any initial thoughts on my "rolevolution" branch? I'm thinking maybe people will lean towards reversing the sense of the directive so renaming it something like "insist role {} and insist method {}". Or maybe TimToady would have a better word. 14:02
14:04 jonas1 left
jnthn skids: Not yet, I'm afraid. :( Not really had time for Perl 6 anything this week (except $dayjob bits that happen to be done with Perl 6) 14:10
raschipi I admire Timtoady, he makes the most difficult task on Computer Science look so easy. 14:11
14:13 ChoHag left 14:15 wamba joined
llfourn_ skids: what's the rolevolution? 14:16
14:20 cdg joined 14:23 lowbro left 14:24 cdg left, cdg joined
jast raschipi: naming things? 14:26
raschipi Yeah.
The second one in the list being cache invalidation, of course. 14:31
14:33 ChoHag joined
Woodi why cache invalidation is hard ? 14:36
14:38 xtreak left
jnthn llfourn_: gist.github.com/skids/18fa6fb1de77...6e82e9fcc2 14:39
14:39 xtreak joined
Geth doc: d1f2c6ac97 | (Zoffix Znet)++ | doc/Type/Method.pod6
Correct and expand &lastcall docs
 14:39
raschipi It's a tradeoff between keeping everything consistent and actually getting better performance out of it. How do you know a cache is stale without checking the source or the peers (which is what a cache is meant to avoid, after all). 14:40
And then doing it without race-conditions.
Woodi raschipi: I think it's a bit like fighting with physics laws 14:42
14:44 xtreak left
Woodi ok, that's answer why it's hard ;) 14:46
DrForr Alternatively, thinking like a physicist, you have to observe the cache in order to attempt to invalidate it. 14:47
skids (This is going to evolve into a discussion on Lamport clocks and light cones, isn't it?) 14:50
14:53 donaldh left
DrForr Only relatively. 14:53
raschipi I just find it almost unbelivable CPU algorithms have to wait for clock propagation, that's all. People have been researching assynchronous CPUs to get around it. 14:54
DrForr Clock pulses take up a good chunk of the CPU. 14:55
timotimo yeah, it has to be a nice and even tree that gets the clock to all parts at the very same time 14:56
skids I actually think the basic problem might be more solvable by highly syncronous CPUs.
Like a lock location that CPU#1 will only write to on clock cycles that are mod 1, CPU mod 2, etc.
timotimo cache invalidation is actually really easy. as soon as you've written to the cache, flush the cache. 14:57
raschipi For CPUs, yeah. But CS wants to deal with Caches in general, not just CPU caches.
timotimo no, for all caches 14:58
raschipi That doesn't solve even a read-only cache, because clients don't subscribe to events necessarily. They just retrieve and keep it in a cache. 14:59
15:01 rindolf left, donaldh joined
raschipi The answer of course, is that they just use stale data anyway, it's an unsolvable problem in it's most general form. And the joke is that it's easier to solve than naming things. 15:02
15:02 cdg left 15:03 cdg joined 15:04 rindolf joined
timotimo you don't understand my solution 15:04
when you've written to the cache, immediately clear the cache. problem solved
[Coke] m: (0..0x1FFF).grep({$_.chr.uninames ~~ /GREEK/}).say
camelia (834 837 880 881 882 883 885 886 887 890 891 892 893 895 900 901 902 904 905 906 908 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 9…
timotimo you can never read stale data from the cache
because the cache never has data in it
[Coke] m: (0..0x1FFF).hyper.grep({$_.chr.uninames ~~ /GREEK/}).say
camelia ()
[Coke] shouldn't those be the same? 15:06
raschipi People, I'm going, we'll talk later.
15:07 raschipi left 15:08 xtreak joined
[Coke] masak: gist.github.com/coke/008e6e5e937a2...258701af7b 15:10
masak: uni -n /(random regex)/ doesn't quite work yet, but everything else seems fineish. 15:12
timotimo [Coke]: hyper's broken, though
[Coke] also need to make DWIM mode DWIMMIER. 15:13
timotimo: ok.
15:14 itaipu joined, Zoffix joined
[Coke] wonders if he should support "uni -n / 'ROMAN' .* 'HUNDRED' /" as an arbitrary command line. 15:16
as opposed to uni -n /'ROMAN'.*'HUNDRED'/ 15:17
15:24 Khisanth left 15:26 khw joined 15:27 pytuger left
ilmari [Coke]: note that there's already App::Uni on cpan, which ships a 'uni' command 15:27
15:29 gdonald left, gdonald joined
timotimo he was already told 15:30
15:33 tharkun left, tharkun joined
ilmari oh, indeed 15:34
moritz \o 15:35
moritz back home 15:36
ilmari boggles at Ↄ U+02183 - ROMAN NUMERAL REVERSED ONE HUNDRED
does that mean -100?
it's not a number as far as unicode is concerned 15:37
Category: Lu (Letter, Uppercase)
15:37 Khisanth joined 15:46 pytuger joined, pytuger left
Zoffix New blog post: "But Here's My Dispatch, So callwith Maybe": perl6.party/post/Perl6-But-Heres-M...with-Maybe 15:50
samcv unidump: Ↄ 15:51
unicodable6 samcv, gist.github.com/e2ce1728b26360660c...409f389cfe
15:55 Zoffix left 15:56 araraloren joined 16:00 dakkar left
araraloren m: enum A < C D E >; enum B < C D E>; 16:01
camelia Potential difficulties:
Redeclaration of symbol 'C, D and E'
at <tmp>:1
------> 3enum A < C D E >; enum B < C D E>7⏏5;
araraloren Anyone can help me for this compile error? 16:03
16:07 robertle joined
perlpilot araraloren: don't do that? 16:08
16:09 raschipi joined 16:11 mcmillhj left
jnthn It's a warning rather than an error 16:12
And it's because enums automatically export the symbols inside of them into the surroudning scope 16:13
16:15 xtreak left
araraloren jnthn, but rakudo complain this and gave up compile 16:21
16:22 zakharyas left, mcmillhj joined 16:24 zakharyas joined
samcv ===SORRY!=== Error while compiling: 16:28
Redeclaration of symbol 'A'
16:29 mcmillhj left
samcv m: my enum A < C D E >; { my enum B < C D E>; } 16:29
camelia ( no output )
samcv just put them in different scope if you have to have enums with the same names 16:30
araraloren, or you can store them into variables 16:31
araraloren That's a slightly inflexible rather than C enum :( 16:32
samcv m: my $var = my enum B < C D E>; say $var; my $var2 = my enum A < C D E>; say $var2
camelia Potential difficulties:
Redeclaration of symbol 'C, D and E'
at <tmp>:1
------> 3 say $var; my $var2 = my enum A < C D E>7⏏5; say $var2
Map.new((:C(0),:D(1),:E(2)))
Map.new((:C(0),:D(1),:E(2)))
araraloren No way, rakudo reject this, haha ...
samcv m: my $a; { $a = enum A < C D E >;}; my $b; { $b = enum B < C D E >; } 16:33
araraloren I put enum inside class now
camelia ( no output )
samcv yeah or do that
many ways you can do it but in a class souds nicest 16:34
and everybody likes nice programmers
araraloren ~~ OK thanks 16:35
16:38 Cabanossi left
TimToady if you export the enums, only the colliding ones should be "poisoned" and require qualification 16:39
16:39 Cabanossi joined, domidumont left
TimToady (one of the ideas we stole from Ada) 16:39
16:41 mcmillhj joined 16:46 mcmillhj left
b2gills m: constant A := do { enum A < C D E >}; constant B := do { enum B < C D E>; }; say A::C; say B::C 16:55
camelia C
C
b2gills m: constant A := do { enum A < C D E >}; constant B := do { enum B < C D E>; }; say A::C.perl; say B::C.perl
camelia B::C
B::C
b2gills m: constant A := do { my enum A < C D E >}; constant B := do { my enum B < C D E>; }; say A::C.perl; say B::C.perl
camelia Could not find symbol '&C'
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
b2gills m: constant A := do { my enum A < C D E >}; say A.perl 16:56
camelia Map.new((:C(0),:D(1),:E(2)))
b2gills m: constant A := do { my enum A < C D E >}; say A<C>
camelia 0
b2gills m: constant A := do { my enum A < C D E >}; say A<D> 16:57
camelia 1
16:57 araraloren left, mcmillhj joined 16:59 donaldh left 17:00 itaipu left 17:01 sufrostico left 17:03 mcmillhj left 17:08 AndyDee left, darutoko left, MasterDuke left
skids m: our class A { our enum A is export(:enums) < C D E >}; our class B { our enum B < C D E>; }; { import A; say A::C.perl; }; { import B; say B::C.perl }; { import A :enums; say C.perl } 17:10
camelia A::C
B::C
A::C
skids That's how I'm handling it in my xcb wrapper.
17:10 jast left 17:11 espadrine_ joined 17:12 jast joined 17:15 AndyDee joined 17:16 mcmillhj joined 17:24 zakharyas left
raschipi huggable: hug me 17:33
huggable hugs raschipi
17:33 pistacchio joined 17:40 raschipi left, grondilu joined 17:44 khw left
[Coke] do we have a way to interrogate which version of unicode we're using yet? 17:47
17:47 sufrostico joined 17:48 AlexDaniel joined
[Coke] .seen rjbs 17:54
yoleaux I saw rjbs 10 Mar 2017 21:04Z in #perl6: <rjbs> /win 7
17:56 ilmari_ joined 17:57 z-b joined, kipd_ joined, hahainte1net joined, alnewkir1 joined 17:59 charsbar_ joined, charsbar left, ilmari left, kipd left, hahainternet left, alnewkirk left, charsbar_ is now known as charsbar, ilmari_ is now known as ilmari, hahainte1net is now known as hahainternet 18:00 kipd_ is now known as kipd 18:03 cdg left 18:08 Cabanossi left, Sound joined 18:09 Sound is now known as Guest72951, Guest72951 left 18:10 Sound_ joined, Sound_ left 18:12 Cabanossi joined
lucs Why does it seem to me that all English speakers pronounce "integral" as "intregal"? 18:20
geekosaur not all of us but it's pretty common 18:28
[ptc] as a native, I've always been amazed at the pronunciation of 'Wednesday'; where I come from it sounds more like "Wensday" 18:35
hobbs lucs: I usually hear "intergral", but the short answer is rs are slippery 18:37
nothing else explains Brett Favre 18:38
lucs hobbs: Hmm... Yeah, I think I've heard it like that too. 18:39
[ptc]: I've never heard Wednesday pronounced other that like you said, "Wensday" :) 18:40
hobbs the only people who consistently say it otherwise are Indian English speakers
who pronounce it exactly as written
geekosaur sometimes you hear the second d but in the wrong place ("wendsday")
18:41 AndyDee left, pistacchio left
hobbs (which is pretty amusing, in my book) 18:41
18:41 AndyDee joined
[ptc] lucs: ah well, where I'm from mumbling is part of the accent :-) 18:41
lucs :)
hobbs (that last bit was about Wednesday, not integral) 18:42
[ptc] fwiw, my version is "innegrl"
lucs Yikes :)
[ptc] :-) 18:43
hobbs sounds like Appalachia :)
18:46 Tonik joined 18:54 Cabanossi left 18:56 Cabanossi joined 18:58 labster joined, kyan joined 19:00 obfusk left, obfusk_ joined
skids "nucular" is a verbal tick you may easily acquire if you were exposed to too many evangelical talk radio shows during the 80s. I don't know where "intergal" would have come from though. 19:03
19:04 TEttinger joined 19:05 raschipi joined 19:13 LeCamarade left
[Coke] (uni) added a -w so you can easily search for things with "CAT" and not get MULTIPLICATION 19:14
19:16 sufrostico left 19:17 kyan left 19:34 ugjka is now known as xyk, xyk is now known as ugjka 19:40 kurahaupo__ joined 19:42 kurahaupo__ is now known as kurahaupo 19:52 bjz joined 20:02 mcmillhj left
gfldex i used `temp` in real code today the first time 20:07
20:07 mcmillhj joined
gfldex there are so many nice things in Perl 6 I didn't use yet :) 20:07
raschipi What did you use it for? 20:11
gfldex github.com/gfldex/perl6-meta6-bin/...meta6#L148 20:12
you can't init a container with itself, unless you use temp
raschipi I see, you're accessing a variable in the outer scope. I was thinking about using temp on parameters and not seeing how it was different from "is copy" 20:14
20:16 mcmillhj left
gfldex m: my $foo = 42; sub s { temp $foo //= $foo ~ ':'; say $foo }; s 20:16
camelia 42
gfldex this feels wrong
m: my $foo = 42; sub s { temp $foo; say $foo }; s
camelia 42
gfldex temp should first create a copy and then apply the //= 20:17
20:17 cdg joined
raschipi m: my $foo = 42; sub s { temp ($foo //= $foo ~ ':'); say $foo }; s 20:20
camelia 42
20:20 xfix joined
raschipi I think temp doesn't happen at runtime. 20:20
Like "my". 20:22
20:24 Peter_R left 20:26 mcmillhj joined, Peter_R joined 20:27 ChoHag left 20:31 itaipu joined, setty1 joined 20:32 mcmillhj left 20:33 setty1 left 20:34 bjz left, LeCamarade joined 20:35 setty1 joined 20:38 gdonald left, gdonald joined 20:41 mcmillhj joined 20:44 itaipu_ joined
Voldenet i can't express how much I like the usage of method-name instead of method_name or MethodName 20:45
I wish I could use perl6 syntax in .net's world, anyone working on perl6 for it? :P
20:46 mcmillhj left
raschipi What about $method's-name ? 20:46
geekosaur niecza hasn't been maaintained in a while
20:46 timotimo left, timo joined 20:47 timo is now known as Guest15399, Guest15399 left, itaipu left, timo1 joined
Voldenet raschipi: eh, is that even correct? :o 20:48
hmmm, that's kind of insane syntax :D
timo1 finally i'm on an up-to-date version of weechat again
Voldenet but I can see its uses
raschipi m: my $method's-name = "aaa"; say $method's-name
camelia aaa
AlexDaniel I've used ' a couple of times 20:49
It was something like $let's-save
Voldenet :)
i tend to use it to delimit strings only 20:50
raschipi What's the name of the Unicode property the codepoint needs for Perl6 to allow it in variable's names?
AlexDaniel raschipi: general category? :P
raschipi: docs.perl6.org/language/unicode_te...Characters 20:51
m: say ‘π’.uniprop
camelia Ll
AlexDaniel m: my $πππ = 42; say $πππ
camelia 42
AlexDaniel unidump: π 20:52
unicodable6 AlexDaniel, gist.github.com/759dd48da1132d5990...78310dad7e
raschipi m: say ‘'’.uniprop
camelia Po
AlexDaniel raschipi: ‘gc’ or ‘General_Category’, and uniprop defaults to it 20:53
I'm pretty sure ' and - are exceptions
raschipi They are treated differently by the parser, but they do have a property that's looked for everything else.
20:54 grondilu_ joined
Voldenet m: my $; = 42; say $; 20:54
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of $; variable; in Perl 6 please use real multidimensional hashes
at <tmp>:1
------> 3my $;7⏏5 = 42; say $;
Voldenet m: my $􏿽xCD􏿽xBE = 42; say $􏿽xCD􏿽xBE
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of $; variable; in Perl 6 please use real multidimensional hashes
at <tmp>:1
------> 3my $;7⏏5 = 42; say $;
20:54 mcmillhj joined
raschipi They are used as letters in English and other languages, therefore have the property set. 20:55
20:55 ggoebel joined
raschipi I just don't remember the name. 20:55
20:56 grondilu left
AlexDaniel u: 􏿽xCD􏿽xBE 20:56
unicodable6 AlexDaniel, U+003B SEMICOLON [Po] (;)
AlexDaniel … something is wrong
mc: ‘􏿽xCD􏿽xBE􏿽xE2􏿽x80􏿽x99.uniname.say 20:57
committable6 AlexDaniel, ¦2015.12: «SEMICOLON»
AlexDaniel or am I missing something
.u 􏿽xCD􏿽xBE
yoleaux U+037E GREEK QUESTION MARK [Po] (􏿽xCD􏿽xBE)
raschipi How do I ask the bot to produce the property tables?
AlexDaniel raschipi: uniprop: …
raschipi: you can include more than one character
raschipi uniprop: '
AlexDaniel oh sorry… unidump 20:58
raschipi unidump: '
unicodable6 raschipi, gist.github.com/4904e5860a97fe5782...e564a6c1e9
20:58 kurahaupo left, Tonik left
AlexDaniel raschipi: fwiw, apostrophe appears here: github.com/rakudo/rakudo/blob/nom/...r.nqp#L504 20:59
20:59 mcmillhj left
raschipi Like I said, I know they are treated exceptionally by the parser, because of it's other uses in the language. 21:00
AlexDaniel ah, <apostrophe> is also ' :)
raschipi Found it: Pattern_Syntax , www.unicode.org/reports/tr31/ 21:01
21:01 wamba left
raschipi Loves, it's time for me to leave. XOXO 21:02
Voldenet m: my Int $int:hungarian-notation'2 = 42; say $int:hungarian-notation'2
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3my Int $int:hungarian-notation7⏏5'2 = 42; say $int:hungarian-notation'2
expecting any of:
constraint
infix
infix stopper…
Voldenet uh, at least there are limits to insanity
m: my Int $int:hungarian-notation = 42; say $int:hungarian-notation 21:03
camelia 42
Voldenet ...not that shallow either :)
timo1 right, you're not allowed to start with a number after a hyphen (or apostrophe) 21:06
Voldenet but I bet colon does something special
timo1 yup
21:06 raschipi left
timo1 it's what we use for &infix:<+> and such 21:07
21:07 timo1 is now known as timotimo, nadim_ joined, mcmillhj joined 21:09 RabidGravy left, Cabanossi left
Voldenet hm, is it invalid to use $type:name as a variable name though? 21:10
timotimo m: my $type:name = 5; my $type:othername = 9; 21:11
camelia ( no output )
timotimo m: my $type:name = 5; my $type:othername = 9; dd $type:name, $type:othername
camelia Int $type:name = 5
Int $type:othername = 9
Voldenet It obviously works. ;)
timotimo seems to work just fine
21:11 skids left, Cabanossi joined 21:12 MasterDuke joined 21:14 mcmillhj left 21:28 itaipu_ left
gfldex m: my $name = 42; my $name:foo = 43; dd :: 21:29
camelia PseudoStash.new(("!UNIT_MARKER" => !UNIT_MARKER,"\$!" => Nil,"\$/" => Nil,"\$=finish" => Mu,"\$=pod" => [],"\$?PACKAGE" => GLOBAL,"\$_" => Any,"\$name" => 42,"\$name:foo" => 43,"\$¢" => Nil,"::?PACKAGE" => GLOBAL,:EXPORT(EXPORT),:GLOBALish(GLOBAL)))
gfldex what is $¢ ? 21:30
it's in roast, so it's a thing 21:31
mst dollarcent? 21:33
21:34 nadim_ left
MasterDuke the changelog for 2015.09 say: '$¢' in grammars now refers to the current cursor 21:35
perlpilot gfldex: I tend to think of it as the "in-flux" version of $/
timotimo yeah, it's the same as $/.CURSOR
21:38 bjz joined
Voldenet a propos, special characters in perl6 are widely used, but is there any good way to use them with regular keyboard on windows? 21:39
21:40 itaipu joined
timotimo there's a page about that on the docs site 21:41
Voldenet except not on windows ;(
timotimo oh? damn
21:42 grondilu_ is now known as grondilu, grondilu left, grondilu joined 21:45 cpage_ left 21:46 dct joined
gfldex Voldenet: most ppl on windows cheat and use a VM 21:47
21:47 mcmillhj joined
Voldenet using alt + 0171 for « and alt + 0187 for » is kind of painful 21:49
also, I use a VM too, but xorg sucks
so I chose not to use it :P
gfldex Voldenet: that's what I use gfldex.wordpress.com/2016/08/31/on...ingertips/ 21:50
21:52 mcmillhj left 21:53 skink joined 21:54 Cabanossi left
Voldenet atm I'm using plain << and >> instead, which looks worse but kind of works :) 21:54
too bad all methods are bound to using unix tools 21:55
21:55 sufrostico joined, rindolf left
geekosaur Voldenet, github.com/samhocevar/wincompose ? 21:55
21:56 pmurias left, Cabanossi joined
geekosaur huggable, wincompose :is: github.com/samhocevar/wincompose a compose key for Windows, allowing choice of compose key and user-defined compositions 21:58
huggable geekosaur, Added wincompose as github.com/samhocevar/wincompose a compose key for Windows, allowing choice of compose key and user-defined compositions
22:00 bjz left
geekosaur ahahah just discovered one of its predefined bindings uses the konami code for a table flip 22:01
Voldenet heh :D
Hm, I'm yet to learn how to use it though 22:02
22:02 ufobat left 22:03 rindolf joined
geekosaur "compose key" is a unix thing. so <compose> < < gets « 22:03
22:03 mcmillhj joined
Voldenet wincompose comes with too many sequences, and I just need a "perl6" sequence :) 22:04
...or i might just use google japanese ime ★ 22:05
skink Who has issues installing Crypt::Bcrypt? 22:08
grondilu off-topic: rocket launch tonight www.youtube.com/watch?v=xsZSXav4wI8
22:11 mcmillhj left
perigrin isn't this the space tested rocket? 22:13
(yes yes it is) 22:14
labster yup, I'm excited!
perigrin is on the wrong end of the country for a change 22:16
:/
22:18 setty1 left 22:22 mcmillhj joined
AlexDaniel grondilu: another one that is supposed to land back, right? 22:23
grondilu AlexDaniel: I think so 22:24
there is a "landing" mark on the timeline
22:27 mcmillhj left
Voldenet skink: I don't have issues, it just plainly does not work for me 22:28
> Testing [FAIL]: if:ver('0.1.0'):auth('github:FROGGS')
whatever that should mean
22:29 dct left
perigrin AlexDaniel: and the one that already did it once. 22:31
Voldenet oh
t/if.t (Wstat: 0 Tests: 1 Failed: 0)
perigrin basically "can we do it again"
Voldenet Parse errors: Bad plan. You planned 5 tests but ran 1.
skink Oh, that's roundabout 22:33
That's a dependency of Crypt::Random
Voldenet Yeah, it's /if/ 22:34
not sure if it's not my setup's problem, but "Cannot find method 'symtable' on object of type GLOBAL", got that too? 22:37
skink Ah, okay zoffix already put a ticket in for that module
Voldenet indeed 22:38
"let's just wait"
skink Everything actually builds fine for me
22:38 mcmillhj joined 22:39 Cabanossi left 22:41 obfusk joined, obfusk_ left, Cabanossi joined 22:43 mcmillhj left
labster If anyone's still in suspense, successful launch, and success retrieval of first stage. The second stage of SES-10 is moving from LEO to geostationary orbit. 22:54
22:54 mcmillhj joined 22:59 mcmillhj left 23:03 Zoffix joined
Zoffix skink: it probably works for you 'cause you're on 2017.01 or earlier rakudo. 2017.02+ had changes to some internal features that if.pm6 is actually using. 23:05
skink Yeah that's what I figured 23:07
23:09 anant joined 23:10 anant left, mcmillhj joined 23:14 cdg left 23:15 mcmillhj left 23:17 Zoffix left 23:24 AlexDaniel left 23:31 skink left 23:34 rindolf left 23:38 skids joined 23:39 itaipu left 23:45 espadrine_ 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!