»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 Zoffix on 25 May 2018.
00:02 mcmillhj left 00:06 pecastro left 00:07 markoong left 00:13 Kaiepi left 00:14 Kaiepi joined 00:24 comborico1611 left 00:26 comborico1611 joined 00:28 comborico1611 left, mcmillhj joined 00:33 mcmillhj left 00:39 mcmillhj joined 00:43 mcmillhj left 01:05 mcmillhj joined 01:11 mcmillhj left 01:15 MasterDuke left 01:19 mcmillhj joined 01:20 wamba left, wamba joined 01:23 mcmillhj left 01:44 cgfbee left 01:50 mcmillhj joined 01:54 mcmillhj left 02:02 araraloren left 02:03 nnunley left, molaf left 02:05 mcmillhj joined 02:08 Xliff joined
Xliff \o 02:09
02:09 mcmillhj left 02:10 Xliff left, lainon joined 02:11 Xliff joined 02:12 MasterDuke joined 02:16 molaf joined 02:29 lainon left 02:40 mcmillhj joined 02:41 wamba left 02:45 mcmillhj left 03:05 mcmillhj joined 03:10 mcmillhj left 03:18 mcmillhj joined 03:22 mcmillhj left 03:52 mcmillhj joined 03:57 nnunley joined, mcmillhj left 03:58 rindolf joined 04:08 mcmillhj joined
samcv is it normal for it to take 5gb of memory to try and install Inline::Perl5 04:13
04:13 mcmillhj left
samcv getting that on a clean build of master 04:13
04:21 mcmillhj joined
geekosaur it's building a custom perl 5, isn;t it? 04:21
samcv no clue but it shouldn't lock up my computer
04:24 sacomo left 04:25 mcmillhj left 04:28 sacomo joined
samcv must have used at least 10GB 04:38
ok probably more. just got control of my system again and ctrl + c'd it
this is all zef had to say so far cry.nu/p/cxsi
04:42 mcmillhj joined
geekosaur that seems odd. Something must have gone wrong with it trying to do the dep search 04:44
wonder if it got garbage JSON somehow 04:45
04:47 mcmillhj left 04:56 eliasr left 05:16 kurahaupo left, kurahaupo joined 05:32 kurahaupo_ joined 05:33 kurahaupo left 05:34 psychoslave joined 06:00 Xliff left
ingy how do I convert a string to a regex? 06:09
samcv ingy: my $str = '^test'; 'test ' ~~ /<$str>/ 06:18
06:31 darutoko joined
u-ou I get a segfault trying to compile rakudo on the rpi3 06:33
I'd cross compile but my PC died :I 06:34
06:37 molaf left 06:38 sacomo left 06:40 sacomo joined
ingy e: my $s = '^test'; my $r = /<$s>/; say $r.gist 06:40
evalable6 /<$s>/
ingy is it lazy expanded? 06:41
samcv: ^
I know it works, because I already tested it
can't you do something like: Regex.new($s) 06:42
?
06:45 epony left 06:46 kurahaupo_ left
samcv ingy: well you can do: my $rx = rx/<$str>/; and then 'blah' ~~ $rx 06:49
u-ou here's my compiler output pastebin.com/FgktGRuA 06:50
I think it is the latest Rakudo Star 06:51
2018.04
06:54 kurahaupo joined 06:57 jmerelo joined 07:00 psychoslave left 07:01 Tison joined 07:04 lizmat left 07:05 wamba joined 07:17 psychoslave joined
Tison r: use nqp; say nqp::index("xxy","y",-1) 07:23
camelia -1
2
07:26 psychoslave left
AlexDaniel m: my $s = '^test'; my $r = /<$s>/; say ‘test world’ ~~ $r 07:27
camelia ｢test｣
AlexDaniel m: my $s = '^test'; my $r = /<{$s}>/; say ‘test world’ ~~ $r
camelia ｢test｣
07:29 psychoslave joined
Tison r: use nqp; say nqp::index("xyy","y",-2) 07:29
camelia 1
-1
Tison r: use nqp; say nqp::index("xyy","y",1)
camelia 1 07:30
Tison r: use nqp; say nqp::rindex("xyy","y") 07:33
camelia 2
Tison r: use nqp; say nqp::rindex("xyy","y", -1) 07:35
07:35 psychoslave left
camelia -1 07:35
2
Tison jnthn: is it worth to keep consistance in MoarVM that index results are all non-negative integer? 07:36
oops 07:37
I misunderstand that 07:38
-1 represents not found, so it's another problem
the problem is should we regard start-pos `-1` as `.elems -1` 07:39
but you can see that there is even inconsistence within the same vm 07:40
github.com/MoarVM/MoarVM/blob/mast...ops.c#L659 07:42
github.com/MoarVM/MoarVM/blob/eed6...ops.c#L558
r: use nqp; say nqp::rindex("xxy","y", -1) 07:47
camelia -1
2
07:56 lizmat joined, cono joined
Geth doc: ba7cd4ea69 | (JJ Merelo)++ | doc/Language/traps.pod6
Adds set trap order, closes #1107
 07:58
synopsebot_ Link: doc.perl6.org/language/traps
lizmat u-ou: I can't make head or tails from that, I suggest you create an issue for it so it won't fall through the cracks 07:59
jmerelo Could this be a bug? stackoverflow.com/questions/507776...nativecall 08:00
u-ou ok. I will do it tomorrow because I'm super tired.
lizmat u-ou: thanks!
u-ou np
08:03 sena_kun joined
jmerelo BTW, see if you recognize what's on the screen on the right on this piece that was on national TV yesterday :-) www.flickr.com/photos/atalaya/4265...ateposted/ 08:04
lizmat :-) 08:05
I assume that's #perl6 ? I can't read it actually :-(
jmerelo lizmat: right :-) 08:06
lizmat Cool :-)
jmerelo lizmat: I actually warned, so it's GDPR-compliant colabti.org/irclogger/irclogger_log...06-08#l482 :-)
lizmat weekly: www.flickr.com/photos/atalaya/4265...ateposted/ 08:08
notable6 lizmat, Noted!
lizmat jmerelo: re jj.github.io/TPF-Grant/May.html , looks like something got left behind/unfinished: "If we look at the general" 08:16
jmerelo lizmat: right. I'll try and see what was it. Thanks!
lizmat weekly: jj.github.io/TPF-Grant/May.html 08:17
notable6 lizmat, Noted!
09:03 evalable6 left 09:34 pmmmm left 09:36 aindilis left 09:50 aindilis joined 10:07 pecastro joined 10:18 markoong joined 10:19 cpage_ joined 10:22 cpage left, cpage_ is now known as cpage
El_Che jmerelo: so, you're not only a professor in real life, but you play now one on tv? 10:23
10:24 eliasr joined
jmerelo El_Che: Tele5's offices are real close to the CS school. I don't know how they found me once in 2002 when looking for someone that could talk about steganography, and I've been handled from one generation to the next ever since... 10:25
El_Che lol
be honest, you're all illuminati! 10:26
did you do the secret handshake? :)
jmerelo El_Che: I offered them tea, but they were in a hurry :-) They didn't leave without me telling them about Perl 6, of course :-)
El_Che: unfortunately, they didn't found it noteworthy. 10:27
El_Che I bet you asked them to post question on SO
jmerelo I mean newsworthy.
El_Che: :-) that will have to wait until next time
El_Che did you tell them it's a new blockchain language?
jmerelo El_Che: Lol 10:29
El_Che I pretty sure that rakudo would be way faster if we removed the bitcoin mining 10:32
jmerelo El_Che: :-) That's probably NOT SPECCED, so I don't have to document it :-) 10:33
El_Che on the other hand, that specific grant would pay for itself, quite literally 10:34
jmerelo: good luck with the grant final evaluation, btw. Your work was very visible (and appreciated) 10:44
jmerelo El_Che: hope so. I don't intend to stop, anyway, you know. Only take it a little bit easier... 10:45
El_Che the fact that you can combine grantwork with staying a few days at home and cook for the family is a huge win 10:46
it keeps the batteries charged
jmerelo El_Che: it definitely is. Quality of life ++
El_Che: I've also had a lot of help (and lots of patience in the beginning) 10:48
11:09 ExtraCrispy joined 11:23 Kaiepi left 11:29 jmerelo left 11:47 lizmat left
rindolf hi all. this p6 program - github.com/shlomif/freecell-pro-0f...lve-log.p6 takes close to 2 minutes to run while the pypy equivalent finishes in under a second. with this as input - www.shlomifish.org/Files/files/text....lg.txt.xz 11:50
11:50 kiwi_51 joined, lizmat joined
rindolf python2 finishes in under 5 sec 11:51
El_Che rindolf: and p5? 11:52
rindolf El_Che: didnt try
11:58 kiwi_51 left 12:04 ChoHag left 12:08 APic left 12:13 robertle joined 12:28 camelia left 12:29 camelia joined 12:30 ChanServ sets mode: +v camelia 12:33 wamba left 12:34 wamba joined
masak rindolf: the script looks OK; I wouldn't be surprised if IO was the bottleneck and not (say) the regex. but I guess I shouldn't speculate too much without measuring. 12:41
12:43 lkjh joined 12:44 lkjh left
rindolf masak: ok 12:46
12:51 espadrine left
timotimo i'm not sure if P5 regexes get any optimization at all from the optimizer 12:54
jnthn masak: I'm somewhat doubtful; I/O these days is optimzied to the point that UTF-8 line reading in Perl 6 matches or slightly beats Perl 5. Perl 5 still wins on the "encodings? what is those?" case. :)
But only by a factor of ~3
MasterDuke rindolf: most of the time is spent in Match and Str methods. i'm not sure there's much to do until perl 6's regex related stuff gets some optimization thrown at it 12:55
rindolf MasterDuke: ah
12:56 Kaiepi joined
MasterDuke timotimo: the top entry in a profile is `<anon> <unknown>:<unknown>`. is that cause something got inlined and/or jitted? 12:56
masak would be fun to hand-code that regex in C or something, to see how fast it could run in theory
timotimo shouldn't be
MasterDuke also, a couple of the top things aren't jitted. the BAILs are coerce_Si (which i think samcv's recent PR adds) and ctxouter 12:59
timotimo the input file has not a single line that would match the regex 13:10
in fact, the input files look like they are actually the output files, and the actual input files aren't in the repo 13:11
dogbert17 timotimo: the latest nqp bump added jstuder's PR containing the chain ops fix. That should hopefully improve the performance of rindolf's previous example 13:14
AlexDaniel rewriting the regex in perl6 gives a slight improvement 13:16
m/^(<[0..9]>+)‘ = Verdict: ’(Intractable|Solved|Unsolved)‘ ; Iters: ’<[0..9]>+‘ ; Length: ’[<[0..9]>+|‘-’1]$/ 13:17
rindolf timotimo: which input file? 13:18
timotimo the file that you would run the .p6 script on 13:19
dogbert17 rindolf: probablu www.shlomifish.org/Files/files/text....lg.txt.xz does it contain the right data?
rindolf timotimo: it is dogbert17 's link
timotimo ah
i did not see that
m(
excuse me, i have not had any caffeine yet
AlexDaniel hm, or maybe not… 13:20
rindolf timotimo: heh, i avoid caffeinated drinks
El_Che rindolf: do you keep all these different python-pypy-perl6 comparisons in a central repo? 13:21
rindolf El_Che: not really 13:22
El_Che rindolf: it looks interesting to run them constantly or at least with new versions
AlexDaniel changing $0 to ~$0 shaves off some seconds
rindolf El_Che: some of them are in github.com/shlomif/project-euler
timotimo AlexDaniel: how does ~ instead of .join do? 13:23
AlexDaniel timotimo: same I think 13:24
El_Che rindolf: you added perl6 along the way it seems
AlexDaniel yeah, looks same to me
dogbert17 rindolf: did the solutuion to your previous code example end with ...496 ?
rindolf dogbert17: which one? 13:25
dogbert17 rindolf: my $N = 24; my $E = 1 +< ($N-1); ...
rindolf dogbert17: let me see 13:26
13:28 darutoko left, Schepeers left
rindolf dogbert17: yes, it does 13:28
13:29 Schepeers joined
dogbert17 rindolf: ok, then it takes 3m15s on my system. Don't know if that's any good though? 13:29
rindolf dogbert17: it is better than my p6 time, but much slower than pypy and c 13:30
AlexDaniel rindolf: eh, does changing .say to .put show any difference on your machine? 13:31
rindolf AlexDaniel: should i also add a "\n"? 13:32
AlexDaniel hm… false alarm, I guess there's no difference, and the stars simply aligned just right :)
rindolf AlexDaniel: ah
AlexDaniel rindolf: put does that by itself
rindolf AlexDaniel: ah
dogbert17 rindolf's example above takes ~1m to execute on my system, dealing with the regex takes 45s and writing data to the result file ~15s 13:35
13:38 evalable6 joined
Geth doc: 2a03fe22a5 | (Tom Browder)++ (committed using GitHub Web editor) | doc/Language/newline.pod6
remove period from title

for conistency with other docs		 13:39
synopsebot_ Link: doc.perl6.org/language/newline
AlexDaniel is there something wrong with doing regex matches in hyper for? 13:42
I have this: gist.github.com/AlexDaniel/8c3a9cf...6ed0be12eb 13:43
but the output is different if you remove hyper
the run time is half though
timotimo i've written a replacement that uses .words on the line and does some manual validation and it's no faster 13:45
AlexDaniel timotimo: can you take a look at my hyper version? 13:49
timotimo throwing out the verification of the numeric fields that we aren't using makes it a bunch faster 13:50
completely without verification it gets down to about 30 seconds, which still isn't good 13:52
jnthn AlexDaniel: You probably need to declare `my $/` inside of the loop 14:00
AlexDaniel: Arguably the `hyper` or `race` on a loop should do that automatically for you, as we do for `start` blocks 14:01
AlexDaniel oh, that helps
14:01 brrt joined
timotimo yeah, it does help 14:01
jnthn Anything other than that resulting in different behavior would be an unknown (to me) bug
14:01 wamba left
timotimo the only behavioural difference now is that this script doesn't write a final newline character 14:02
jnthn Feel free to ticket the idea to implicit decl $/ in a hyper/race loop body
AlexDaniel already doing that 14:03
jnthn $! should come along with it
14:05 wamba joined
AlexDaniel R#1912 14:07
synopsebot_ R#1912 [open]: github.com/rakudo/rakudo/issues/1912 [hyper/race] hyper and race should do `my $/`
jnthn Thanks. Will comment on it and/or deal with that when I'm a bit less hot/sleeply. 14:09
El_Che jnthn: take care of your self 14:10
jnthn :) 14:11
Yeah, after a couple of working weekends I'm enjoying having this one pretty much completely off :)
The hot and sleepy are mostly thanks to the weather 14:12
I hoped the rain would mean "nice and fresh air", which it did for a little time, but now it's already feeling humid again :/
jnthn is really not a fan of summer
El_Che jnthn: join the millions of complaining europeans now :) 14:13
jnthn wonders if any of the nearby pubs have air-con :) 14:14
14:19 zakharyas joined 14:24 zakharyas1 joined 14:26 zakharyas left
brrt jnthn: I feel you (re: summer) 14:38
I was actually thinking of replacing the central heating system in my appartment with a set of aircos 14:39
energetically, this is a win. home-value wise, probably not so much
El_Che brrt: is electricity that cheap there? 14:40
here gas is cheaper
brrt no, but realistically, i've never really used anywhere near the potential
to put it in other words; the last 5 years or so i've paid more on the gas pipes than on the gas itself 14:41
El_Che brrt: the sale is true here for electricy 14:42
distribution is expensive
brrt tax is expensive :-) 14:43
Tison has TimToady given a talk about slang? what does "slang" stands for in Perl 6 world? 14:48
brrt sub-language 14:49
14:51 damnlie left 14:52 damnlie joined
Tison P6Regex is a slang of Perl 6, and even Perl 6 is a slang of Perl 6? 14:54
15:00 Tison left 15:02 Tison joined
brrt i suppose that is a reasonable way to put it yes 15:05
15:17 dct joined 15:20 xtreak joined 15:21 n1ge left 15:30 xtreak left 15:32 xtreak joined 15:36 xtreak left 15:37 musca left 15:42 n1ge joined 15:43 espadrine joined 15:49 jmerelo joined 15:51 domidumont joined 15:52 skids joined, musca joined 15:55 domidumont left 15:56 domidumont joined 16:03 lookatme left 16:04 lookatme joined 16:22 zakharyas1 left 16:28 zakharyas joined 16:31 jmerelo left, zakharyas left 16:33 zakharyas joined 16:37 n1ge left 16:44 jmerelo joined 16:49 zakharyas left 17:03 rindolf left 17:04 brrt left
lizmat weekly: brrt-to-the-future.blogspot.com/201...m-jit.html 17:30
notable6 lizmat, Noted!
jmerelo lizmat: wow, cool :-)
buggable New CPAN upload: Desktop-Notify-0.3.3.tar.gz by FRITH modules.perl6.org/dist/Desktop::Not...cpan:FRITH 17:34
17:41 psychoslave joined
Geth ecosystem: ec78d3cc70 | (Fernando Correa de Oliveira)++ (committed using GitHub Web editor) | META.list
Add Trie to the ecosystem

  github.com/FCO/Trie		 18:13
18:15 psychoslave left 18:37 raschipi joined 18:50 rindolf joined 18:59 domidumont left 19:07 pmmmm joined 19:11 jmerelo left 19:16 Kaiepi left 19:17 Kaiepi joined 19:21 Kaiepi left, Kaiepi joined 19:22 aindilis left 19:23 aindilis joined 19:25 sena_kun left 19:26 sena_kun joined 19:28 sena_kun left 19:29 sena_kun joined 19:42 APic joined 19:49 dataangel joined 19:52 sena_kun left 19:53 sena_kun joined 19:55 comborico1611 joined 19:56 vike left 20:14 ChoHag joined
Geth perl6.org: e013c14c3d | (Tom Browder)++ (committed using GitHub Web editor) | source/resources/index.html
remove superfluous space
 20:21
doc: 0ecbd02a4e | (Tom Browder)++ (committed using GitHub Web editor) | doc/Language/system.pod6
add a simple example of sub “prompt”
 20:42
synopsebot_ Link: doc.perl6.org/language/system
21:24 vike joined 21:35 imcsk8 left 21:36 imcsk8 joined 21:50 sena_kun left 22:14 raschipi left 22:27 raynold joined 22:30 dct left 22:39 dogbert2_ joined 22:42 dogbert2 left 22:45 pecastro left
Kaiepi m: "\x[0]" ~~ / \x[0] /; say $0 22:52
camelia Nil
Kaiepi m: "\x[255]" ~~ / \x[255] /; say $0 22:53
camelia Nil
Kaiepi m: "\x[255]" ~~ / \x[255] /; say $/[0]
camelia Nil
Kaiepi m: "\x[255]" ~~ / \x[255] /; say $/
camelia ｢ɕ｣
Kaiepi m: "\x[0]" ~~ / \x[0] /; say $/
camelia ｢␀｣
Kaiepi ok weird
on my system "\x[0]" ~~ / \x[0] / just returns an empty match object 22:54
22:54 |oLa| joined
Kaiepi m: say $*PERL.compiler 22:56
camelia rakudo (2018.05.78.g.20.bf.96.f.9.e)
Kaiepi oh wait i was trying to get $/[0] when i wasn't capturing it 23:05
never mind lol
23:30 Tison left 23:44 w_richard_w joined 23:46 MasterDuke left 23:52 wamba left 23:54 Xliff joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!