Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2011-09-06

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:02 frhodes joined #perl6
00:13 jevin joined #perl6
00:32 Psyche^ joined #perl6
00:32 uasi joined #perl6
00:47 colomon rakudo: my $start = times[0];my $i = 0; for 1..1000 { $i = $i + $_; }; say $i; say times[0] - $start;
00:47 p6eval rakudo 493e23: OUTPUT«Could not find sub &times␤  in <anon> at /tmp/kzquodDmMe:1␤  in <anon> at /tmp/kzquodDmMe:1␤␤»
01:04 woosley joined #perl6
01:14 uasi joined #perl6
01:20 REPLeffect joined #perl6
01:46 sorear &times should be LHF
02:01 colomon A shame quickly loop 100,000 times is not.
02:02 colomon *looping
02:09 lue Could I get an account on feather? I've been using a different remote server, and ssh is now telling me of possible DNS spoofing [I've been thinking of moving to feather for a while now anyway]
02:26 daniel-s joined #perl6
02:41 daniel-s joined #perl6
02:45 abercrombie Hi #perl6, is it possible to re-visit $*ARGFILES?
02:47 abercrombie I used $*ARGFILES to read the command line file once, now I want to read the file again. Is there any way other than open() ?
02:53 sorear seen Juerd
02:53 aloha Juerd was last seen in #perl6 9 days 2 hours ago leaving the channel.
02:57 sorear lue: you could get an account easily if Juerd were here.  I'm sending em a short mail now.
02:58 lue Alright, thanks sorear
02:59 sorear lue: btw, I'm 95% sure that Juerd will ask for fairly detailed personal information - Juerd's company insists on having records of exactly who is using the machine they donate
03:00 TimToady rakudo: say $*ARGFILES.perl
03:00 p6eval rakudo 493e23: OUTPUT«ArgFiles<37112368>␤»
03:00 TimToady b: say $*ARGFILES.perl
03:00 p6eval b 1b7dd1: OUTPUT«IO;ArgFiles.new(args => [], filename => Any, PIO => Any, ins => Any, autoflush => Any, path => Any, stat => IO;Stat.new(path => Any))␤»
03:03 sorear yucky
03:03 sorear I still think that Perl 6 shouldn't require paths and filehandles to use the same class
03:03 samlh joined #perl6
03:04 sorear they might both do the Stattable role
03:04 sorear but I'll admit not grokking IO::Any
03:05 TimToady rakudo: $*ARGS = '/etc/passwd'; say lines[0]; $*ARGS = '/etc/group'; say lines[0]
03:05 p6eval rakudo 493e23: OUTPUT«Cannot assign to a non-container␤  in <anon> at /tmp/t2d7EkAA_n:1␤  in <anon> at /tmp/t2d7EkAA_n:1␤␤»
03:05 TimToady rakudo: @*ARGS = '/etc/passwd'; say lines[0]; @*ARGS = '/etc/group'; say lines[0]
03:05 p6eval rakudo 493e23: OUTPUT«root:x:0:0:root:/root:/bin/bash␤daemon:x:1:1:daemon:/usr/sbin:/bin/sh␤»
03:05 jevin joined #perl6
03:06 TimToady it does not appear that assigning @*ARGS resets $*ARGFILES currently; arguably a bug
03:09 abercrombie Do you mean after @*ARGS = '..', lines should restart from the first line of the new @*ARGS?
03:10 TimToady that is how P5 handles @ARGV and <>, I think, though I could be misremembering
03:11 * sorear looks for sensitive data in the p6eval chroot
03:11 agentzh joined #perl6
03:11 sorear it looks like moritz has been pretty good about this, I can't find anything *really* important
03:12 TimToady sorear: I think some of what is going on with IO is trying to minimize race conditions between filenames and inode identities; the only safe thing to do with a filename is to immediately open it, and do everything else with the file descriptor
03:13 TimToady otherwise there is no guarantee whatsoever that /foo/bar at time N is the same file at time N+1
03:14 TimToady many security breaks come from this
03:14 dalek rakudo/nom: c773a04 | Coke++ | t/spectest.data:
03:14 dalek rakudo/nom: track failure modes
03:14 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/c773a04d75
03:14 TimToady a normal stat should generally only be done for informative purposes like ls
03:15 TimToady so it comes down to naming, caching, *and* off-by-one errors :)
03:15 abercrombie TimToady: You are right. In perl5 it will reset to the new one
03:16 TimToady so whoever implemented ARGFILES was making some unwarranted assumptions, from the standpoint of "default to Perl 5 semantics"
03:23 Su-Shee_ joined #perl6
03:30 abercrombie Is b: for the 2011.07 version?
03:31 imarcusthis joined #perl6
03:33 TimToady aye
03:33 abercrombie b: @*ARGV="a"
03:33 p6eval b 1b7dd1: OUTPUT«Cannot modify readonly value␤  in '&infix:<=>' at line 1:src/metamodel/RoleToInstanceApplier.nqp␤  in main program body at line 22:/tmp/yNuE0EDNMF␤»
03:33 abercrombie b: @*ARGS="a"
03:33 p6eval b 1b7dd1:  ( no output )
03:35 abercrombie lines[0] will only read the first line, right? Not read the whole file and retrieve the first line
03:35 * sorear needs to come up with something to say on the Parrot lists to offset the noise generators
03:36 sorear abercrombie: lines[0] will open the file, read the first line, and leave the file open indefinitely
03:36 sorear abercrombie: hopefully you're on an OS that closes files on exit
03:37 abercrombie So is this the lazy evaluation? Coz from the grammer I think it equals to lines()[0]
03:37 sorear it exactly equals that
03:38 abercrombie Since a mere lines() will give us the whole file, I guess it utilize the lazy eval tech, am I right?
03:40 sorear righht
03:41 abercrombie Thanks. Does niecza reset the line number when @*ARGS get assigned to a new value?
03:43 sorear does it make me a bad person to have an extremely low opinion of KatrinaTheLamia
03:43 sorear abercrombie: @*ARGS isn't magical.  No-one can tell you've changed it.
03:44 PerlJam sorear: no.
03:44 sorear (why did I even bother reading her last mail...)
03:44 abercrombie Coz in Perl5 the ARGV thing is "sensitive"
03:46 PerlJam sorear: I wonder if KatrinaTheLamia is really just a trick to derail interesting/productive conversation
03:48 sorear PerlJam: we have so many people on the perl 6 lists who have no firm grounding in reality
03:50 sorear I especially have to wonder what to think about Damian Conway.  I'm told he's a brilliant hacker but I find myself thinking "that's a horrible idea" every time he posts
03:50 sorear *on implementation matters
03:51 jamtech joined #perl6
03:53 abercrombie #1   @*ARGS="a.txt"; say lines[0]; lines; @*ARGS="a.txt"; say lines[0];
03:53 abercrombie #2   @*ARGS="a.txt"; say lines[0]; say lines; @*ARGS="a.txt"; say lines[0];
03:53 abercrombie the first one won't get reset while the second one does.
03:54 abercrombie So it seems the problem is due to optimization?
03:54 PerlJam sorear: Damian pushes the envelope of reality a little bit.  In a way, he's like a Perl 6 venture capitalist but with ideas instead of money.
03:58 envi_laptop joined #perl6
04:01 uasi joined #perl6
04:01 birdwindupbird joined #perl6
04:05 satyavvd joined #perl6
04:05 sorear abercrombie: laziness!
04:05 sorear abercrombie: lines; # this does nothing
04:06 sorear abercrombie: +lines; # tells Perl to simulate wc -l, and exhausts all the lists
04:12 abercrombie sorear: I see. Thank you.
04:30 molaf joined #perl6
04:37 kaare_ joined #perl6
04:52 kfo_ joined #perl6
05:00 uasi joined #perl6
05:08 zby_home_ joined #perl6
05:11 sftp_ joined #perl6
05:12 saaki joined #perl6
05:12 PZt joined #perl6
05:13 pmichaud joined #perl6
05:13 sjn joined #perl6
05:13 yves joined #perl6
05:13 cotto_work joined #perl6
05:13 BinGOs joined #perl6
05:13 ascent_ joined #perl6
05:14 tadzik joined #perl6
05:15 dju joined #perl6
05:28 ZapZ_ joined #perl6
05:31 ZapZ_ joined #perl6
05:35 ZapZ_ joined #perl6
05:36 moritz good morning
05:47 lue hello moritz o/
05:51 sorear hi moritz.
05:51 sorear TimToady++ discovered a new hole in our sandbox
05:51 sorear hmm, I wonder
05:51 sorear niecza: my @*ARGS = '/etc/passwd'; say lines;
05:51 p6eval niecza v9-10-g678102e: OUTPUT«Land der Berge, Land am Strome, Land der Äcker, Land der Dome, Land der Hämmer, zukunftsreich! Heimat bist du großer Söhne, Volk, begnadet für das Schöne, vielgerühmtes Österreich, vielgerühmtes Österreich!  Heiß umfehdet, wild umstritten liegst dem Erdteil…
05:52 sorear niecza: @*ARGS = '/etc/passwd'; say lines;
05:52 p6eval niecza v9-10-g678102e: OUTPUT«Land der Berge, Land am Strome, Land der Äcker, Land der Dome, Land der Hämmer, zukunftsreich! Heimat bist du großer Söhne, Volk, begnadet für das Schöne, vielgerühmtes Österreich, vielgerühmtes Österreich!  Heiß umfehdet, wild umstritten liegst dem Erdteil…
05:52 sorear niecza: @*ARGS = '/etc/passwd'; say $*ARGFILES.lines;
05:52 p6eval niecza v9-10-g678102e: OUTPUT«Unhandled exception: CORE open may not be used in safe mode␤  at /home/p6eval/niecza/lib/CORE.setting line 0 (CORE open @ 0) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2008 (CORE ArgFiles.get @ 16) ␤  at /home/p6eval/niecza/lib/CORE.setting line 1999 (…
05:59 * sorear out
06:03 wtw joined #perl6
06:09 orafu joined #perl6
06:15 SHODAN joined #perl6
06:24 REPLeffect joined #perl6
06:42 amil joined #perl6
06:42 amil left #perl6
06:47 koban` joined #perl6
06:47 wamba joined #perl6
06:47 koban` left #perl6
07:13 mj41 joined #perl6
07:15 PZt joined #perl6
07:32 cotto when is mls usually online?
07:34 moritz cotto: in on or two hours maybe
07:34 moritz late CEST (UTC+2) uptime :-)
07:35 cotto that might explain why I can't seem to catch him
07:35 cotto stupid earth with its wildly differing timezones
07:38 lue .oO(stupid Time with its wildly differing speeds and starting positions)
07:42 drbean joined #perl6
07:44 tadzik good morning #perl6
07:44 lue good evening tadzik o/
07:48 jnthn morning, #perl6
07:49 lue evening jnthn o/ [I'm getting the distinct feeling I'm in the wrong timezone]
07:49 tadzik timezones. They're all wrong
07:51 jnthn tadzik: Thanks for the benchmarking :)
07:51 lue .oO(There should either be one timezone (UTC), or ∞ timezones (setting your clock to the sun))
07:51 jnthn tadzik: Guessing that you'd --optimize'd the setting?
07:52 tadzik jnthn: yes
07:52 tadzik the size of the pbc didn't change at all
07:52 tadzik (as du CORE.setting.pbc says)
07:56 jnthn at all?
07:56 jnthn I'd not have expected much difference in either direction
07:56 im2ee joined #perl6
07:56 jnthn But something
07:57 im2ee Hi! :)
08:02 moritz jnthn: 'du' has file system block size resolution
08:02 moritz so probably 4kb
08:02 moritz so "no change at all" => "no change by more than 4kb"
08:03 lue .oO(does 'll' (or 'ls -l') do the same sort of block size resolution?)
08:04 moritz lue: no, as you can easily see from the numbers
08:05 moritz if you run it on a small file
08:06 lue Now that I think about it, it's pretty obvious. Maybe I'm just tired.
08:06 tadzik im2ee: hello
08:06 tadzik I think du -b could be more accurate, yes
08:06 moritz but 'du' is 'disc usage', and disc usage *is* measured in fs blocks
08:06 * moritz uses   wc -c  for byte counts
08:07 tadzik right
08:07 tadzik -b, --bytes
08:07 tadzik equivalent to `--apparent-size --block-size=1'
08:07 tadzik and --apparent-size "print  apparent  sizes,  rather  than  disk usage;"
08:32 jnthn moritz: ah, ok
08:32 jnthn I suspect for loops run faster because the optimizer improved the iterator code in the setting somewhat.
08:35 mls morning perl6!
08:35 mls so we're inlining now? jnthn++
08:35 moritz \o mls
08:36 mls (be careful to not to inline blocks that contain OUTER::)
08:36 mls s/to//
08:37 jnthn mls: Well, that's easy, OUTER:: ain't implemented yet :)
08:37 moritz we don't have OUTER:: yet :-)
08:37 jnthn mls: We're only inlining immediate blocks
08:37 mls yeah, I know. But we'll have it someday.
08:37 jnthn This isn't the really interesting inlining.
08:37 jnthn That's *much* harder.
08:38 mls moritz: what about callframe()? Do blocks appear in the callframe?
08:38 moritz mls: they do
08:38 * jnthn wonders how much space the spec gives us there.
08:39 mls hmm, I worry about callframe(4).leave(1) or such...
08:39 jnthn Yeah.
08:39 moritz mls: that would be very fragile code
08:39 moritz well, the spec talks about callframes, not about execpted callframes in the absence of optimizations :-)
08:39 mls Yes, I know. We should ask TimToady++ about his thoughts on inlining
08:40 jnthn Another interesting task will be an NQP optimizer. We can be a lot more aggressive there.
08:41 mls True.
08:41 jnthn oh heh
08:41 jnthn if I make an NQP optimizer
08:41 jnthn And then compile NQP under it
08:41 jnthn Then the optimizer will optimize itself
08:41 jnthn :)
08:42 mls Too bad it doesn't optimize even better then ;)
08:43 moritz maybe it does? it's hard to tell in advance what optimizations do to the code :-)
08:46 jamtech joined #perl6
08:47 wamba joined #perl6
09:41 envi_laptop joined #perl6
09:58 tadzik make spectest with --optimize is 5 seconds longer. Likely noise, or there's not much to inline in the roast
09:59 tadzik that's assuming https://gist.github.com/1197158 actually turns on the optimizer permanently :)
10:00 moritz tadzik: looks like it should :-)
10:01 jnthn tadzik: Noise
10:02 tadzik probably, yes
10:02 jnthn But tbh, it's walking the entire tree to do *one* kind of optimization.
10:02 tadzik which probably doesn't occur too often in the chain of is() is() is() :)
10:02 jnthn If we're breaking even already that's kinda nice...
10:03 jnthn Yeah, tests don't contain loopy hot pants
10:03 jnthn er
10:03 tadzik we are breaking, but only 2 tests :)
10:03 jnthn hot paths
10:03 jnthn wtf :)
10:03 jnthn tadzik: Two tests break?
10:03 tadzik t/spec/S04-exception-handlers/catch.rakudo and t/spec/S04-statements/given.rakudo
10:03 jnthn ah, good to know.
10:03 tadzik 2 tests files. Will paste
10:03 jnthn Thanks.
10:03 jnthn For a first cut, that's not much breakage.
10:03 tadzik https://gist.github.com/1197173
10:23 espadrine joined #perl6
11:06 pochi joined #perl6
11:10 cosimo joined #perl6
11:12 JimmyZ_ joined #perl6
11:13 daxim joined #perl6
11:16 wamba joined #perl6
11:34 Psyche^ joined #perl6
11:37 Chillance joined #perl6
11:39 espadrine joined #perl6
11:41 odoacre joined #perl6
11:42 plobsing joined #perl6
11:49 pernatiy joined #perl6
11:54 tlocalhos joined #perl6
12:00 bluescreen10 joined #perl6
12:08 mtk joined #perl6
12:17 benabik joined #perl6
12:17 hanekomu joined #perl6
12:18 bluescreen10 joined #perl6
12:36 sahadev joined #perl6
12:39 dalek nqp: 8ad13af | moritz++ | tools/build/PARROT_REVISION:
12:39 dalek nqp: bump PARROT_REVISION to after the pmc_is_ptr merge
12:39 dalek nqp: review: https://github.com/perl6/nqp/commit/8ad13af138
12:40 jnthn moritz++
12:40 jnthn The nommap entry about the while 1 leak can also now go \o/
12:40 moritz will make it go
12:41 * [Coke] yawns
12:42 * moritz commits
12:42 dalek rakudo/nom: 860fe17 | moritz++ | / (2 files):
12:42 dalek rakudo/nom: bump nqp revision, remove memory leak from NOMMAP
12:42 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/860fe17159
12:42 jnthn I have Perl 6 day on Thursday. I plan to deal with the parametric roles issues so they can be removed from nommap
13:01 sftp joined #perl6
13:01 envi_laptop joined #perl6
13:04 [Coke] one of the tests started hanging again, btw. (and by hanging, I could just mean "takes forever to complete and I didn't want to wait")
13:18 agentzh joined #perl6
13:22 araujo joined #perl6
13:28 stephanmg joined #perl6
13:30 stephanmg joined #perl6
13:36 Holy_Cow joined #perl6
13:37 stephanmg joined #perl6
13:41 plobsing_ joined #perl6
13:49 IRCReaderBOT joined #perl6
13:55 stephanmg joined #perl6
14:02 [Coke] anyone remember the "show links between github committers" groovy bouncy diagram?
14:03 tadzik gource
14:04 felher if i wan't a unflatted(is that even a word?) array of <a b c>, <d e f>, <and so on>. Whats the preffered way to write that? [ [<a b c>], [<d e f>], [<and so on>] ]?
14:05 tadzik <a bc c>.item should work
14:05 tadzik and I think you mean "flatten"
14:06 tadzik rakudo: say [ <a b c>, <d e f> ].perl
14:06 p6eval rakudo 860fe1: OUTPUT«["a", "b", "c", "d", "e", "f"]␤»
14:06 moritz [<d e f>] will return an Array, <d e f>.item a Parcel
14:06 tadzik rakudo: say [ <a b c>.item, <d e f>.item ].perl
14:06 moritz rakudo: say [ <a b c>; <d e f> ].perl
14:06 p6eval rakudo 860fe1: OUTPUT«[("a", "b", "c"), ("d", "e", "f")]␤»
14:06 p6eval rakudo 860fe1: OUTPUT«["d", "e", "f"]␤»
14:07 tadzik moritz: comma
14:07 moritz the one with ; didn't work
14:07 tadzik oh, it's meant to?
14:08 tadzik as a super-comma
14:08 moritz dunno if it should work
14:08 moritz frankly I don't know when ; is statement separator, and when it's super-comma
14:08 [Coke] tadzik: that wasn't what I was thinking, but is awesome.
14:09 moritz which is slightly worrying, but only slightly
14:09 PerlJam [Coke]: code-swarm?
14:09 tadzik I want to see the other one then
14:09 felher tadzik, moritz: thnx :)
14:09 tadzik felher: I hope that didn't confuse you even more :)
14:09 tadzik pmichaud++ gave an excellent talk on this on the yapc, but I feel I should watch it once or twice again :)
14:11 felher tadzik: a talk about arrays, parcels and such?
14:11 tadzik felher: yes
14:11 felher tadzik: is there a video?
14:11 tadzik felher: not yet
14:11 tadzik I think there are no videos from yapc::eu yet, but there are ought to be
14:11 tadzik I'm looking forward to them, I missed too much talks due to schedule clashes
14:12 PerlJam [Coke]: oh ... I know what you're talking about ... There's a URL where you can enter a github repo and it gives a wire-frame tree that bounces around a little bit that connects all of the committers.
14:12 felher tadzik: yeah, i'd really like to watch them, too. :)
14:12 PerlJam [Coke]: sorry, I don't remember the service
14:14 [Coke] perljam;aye, that's the one. ;)
14:14 [Coke] (the one you don't remember, not code-swarm ;)
14:14 moritz there are publich IRC logs available of this channel, you know...
14:14 moritz *public
14:14 moritz maybe I should implement an option to show only links from a page
14:15 PerlJam moritz: it's hard to search them when you don't know what you're searching for  :)
14:15 stephanmg joined #perl6
14:16 [Coke] nor am I sure that it was even mentioned here. I just know it was mentioned in a perl context.
14:16 PerlJam it was mentioned either here or on #parrot I think.
14:17 PerlJam I remember playing with it for a good 5 minutes or so :)
14:35 PerlJam [Coke]: great!  now it's going to bug me all day what that site is since I can't remember it
14:35 bluescreen10 joined #perl6
14:36 sorear good * #perl6
14:40 moritz PerlJam: maybe looking through http://moritz.faui2k3.org/tmp/perl6-links.txt helps you to find it
14:41 moritz PerlJam: it's the last 1000 lines in #perl6 which contained http links
14:42 PerlJam moritz++  !
14:42 moritz http://www.stargit.net/#rakudo maybe?
14:42 PerlJam http://www.stargit.net/#rakudo
14:42 PerlJam yep
14:42 tadzik oh, there's my dot there
14:43 tadzik the fisheye is funny
14:45 moritz SQL is sometimes very practical
14:45 moritz though that query took a minute or two to run
14:48 uasi joined #perl6
14:52 abercrombie joined #perl6
15:00 [Coke] PerlJam++ moritz++
15:01 thou joined #perl6
15:07 samlh joined #perl6
15:08 hanekomu joined #perl6
15:25 Sarten-X joined #perl6
15:43 REPLeffect joined #perl6
15:50 hanekomu joined #perl6
15:53 samlh_ joined #perl6
15:54 flussence .oO( argh... no wonder my laptop was running slow, that `now` bug was eating all my RAM in the background :( )
16:18 MayDaniel joined #perl6
16:26 molaf joined #perl6
16:28 abercrombie Hi, #perl6. One of my friends encountered a strange problem. His Rakudo version is 2011.07, his system locale is en_US.UTF-8. All those settings are same as mine. But for a same file which contains a Unicode var name, I can run smoothly while he cannot. His rakudo complained: Confused at line 1, near "my $\u5b87\u822a = \""    I've checked that \u5b87 is the UTF-16 encoding for that unicode
16:28 abercrombie character. But we are sure the file is encoded in UTF-8 according to the result of "file" command and "od -x" command. Any suggestion?
16:31 jevin joined #perl6
16:32 pmichaud good morning, #perl6
16:33 pmichaud abercrombie: I suspect that you have ICU loaded on your system and your friend does not.
16:33 jnthn hi, pmichaud
16:34 pmichaud (ICU is the library that Parrot uses for unicode handling)
16:34 pmichaud jnthn: o/
16:35 abercrombie How can I check whether I have it loaded?
16:35 pmichaud ng: print %*VM<config><has_icu>;  # maybe?
16:35 pmichaud b: print %*VM<config><has_icu>;  # maybe?
16:35 p6eval b 1b7dd1: OUTPUT«Any()»
16:35 pmichaud nope
16:35 pmichaud hmmm
16:35 Holy_Cow joined #perl6
16:35 jnthn pmichaud: What's to focus on in nom? Besides the parametrics stuff that I know I should do...
16:36 pmichaud jnthn: I need to re-orient myself with nom and its recent changes.  I'll likely do that a bit later today.
16:36 jnthn pmichaud: OK
16:36 jaldhar joined #perl6
16:36 jnthn pmichaud: Pretty much all my work at the weekend was performance oriented rather than feature/bug oriented.
16:36 hanekomu joined #perl6
16:37 pmichaud jnthn: good, that's where our biggest pain point was
16:37 jnthn Aye, but we probably have others now ;)
16:37 pmichaud agreed
16:37 jnthn Not that we don't need a lot more performance work, but... :)
16:39 abercrombie pmichaud: I just don't know why his Rakudo tried to explain the character in UTF-16? His file is encoded in UTF-8
16:40 mls pmichaud: hi! I found the cause of the high lineof() percentage
16:40 mls the one that made you suspicious of the profiler
16:42 pmichaud mls: ooooh, do tell!
16:43 Holy_Cow left #perl6
16:43 mls the profiler overhead for each op was too big
16:44 mls as lineof() consisted of an insane amount of very fast ops, the result was skewed
16:44 mls (the profiler timed each op, which didn't make much sense)
16:45 mls I changed the code so that it only does the tick calculation when the context changes, and old ng's lineof went from 20% to 7%
16:45 mls 5% is correct
16:47 mls in other words, functions with many ops had a too high tick value
16:47 pmichaud makes good sense
16:48 mls I'm currently changing the code so that the profiler can also do "annotation segment" profiling
16:48 mls (aka HLL profiling)
16:49 mls then we'll have a real perl6/nqp profiler
16:50 pmichaud mmmm, annotation segment profiling
16:50 pmichaud that will be... awesome
16:50 pmichaud mls++  # fantastic patches
16:50 jnthn mls++ indeed
16:50 mls thanks!
16:51 mls (I like doing low level stuff ;) )
16:51 jnthn Yes, it's fun :)
16:51 pmichaud your low-level stuff is helping our high-level stuff immensely.
16:51 jnthn Very much so.
16:56 [Coke] mls++
16:58 masak joined #perl6
16:58 masak good evening, lambdaparrots.
16:59 jnthn o/ masak
16:59 pmichaud o/ masak
17:00 benabik (\x -> "Good evening, " ++ x) "masak"
17:00 masak ooh, Haskell!
17:01 benabik lunch &
17:08 mls afk -> home...
17:11 donri joined #perl6
17:14 am0c joined #perl6
17:16 17SABAQJF joined #perl6
17:18 jevin joined #perl6
17:19 hanekomu joined #perl6
17:21 MayDaniel joined #perl6
17:24 mj41 joined #perl6
17:33 wamba joined #perl6
17:49 pyrimidine joined #perl6
17:55 mberends joined #perl6
18:06 masak #phasers in 55?
18:07 colomon Bool::True
18:07 mberends ooh! thanks for the reminder
18:07 jnthn oops, I didn't dinner yet
18:09 masak rakudo: if 21 % 4 -> $r { say "Thanks for the remainder: $r ;)" }
18:09 p6eval rakudo 860fe1: OUTPUT«Thanks for the remainder: 1 ;)␤»
18:10 jevin joined #perl6
18:17 baest_ joined #perl6
18:21 jevin joined #perl6
18:22 samlh joined #perl6
18:33 pmurias joined #perl6
18:37 moritz masak++ # Thanks for the remainder
18:39 diakopter %
18:40 Su-Shee "uhm?" ;) src/core/YOU_ARE_HERE.pm  > src/gen/core.pm
18:45 wamba joined #perl6
18:48 jnthn ;)
18:55 Su-Shee that's what you see if you watch rakudo building while knitting..
18:55 takadonet joined #perl6
18:55 takadonet left #perl6
18:59 birdwindupbird joined #perl6
19:22 Holy_Cow joined #perl6
19:23 pmurias joined #perl6
19:26 Holy_Cow left #perl6
19:31 pmurias joined #perl6
19:38 tyatpi_ joined #perl6
19:44 plobsing joined #perl6
19:44 masak I finally understand why we have both the names "CORE" and "setting" in the terminology.
19:44 masak "CORE" is just a setting among settings.
19:44 PerlJam masak: so that we can pun CORN and CORP like niecza does?  ;)
19:44 masak sort of the "official", recommended one.
19:45 masak PerlJam: I just assumed that pun was from STD/TimToady. :)
19:45 masak maybe I underestimated sorear.
19:45 tadzik CORP is a setting for CORPorations? :P
19:45 PerlJam I don't know the provenance, so I can't say for sure.
19:45 PerlJam I just noticed it with niecza.
19:46 * masak .oO( dark wings, dark puns )
20:01 molaf joined #perl6
20:06 pernatiy joined #perl6
20:08 moritz darkpan?
20:09 masak I rest my dark case. :)
20:09 ab5tract joined #perl6
20:15 supernovus joined #perl6
20:18 jevin joined #perl6
20:19 pmurias masak: re recommended one, it's the default one
20:20 masak aye, that's probably a better way to phrase it. :)
20:21 masak but, just like with many other things in Perl 6, we do recommend our defaults ;)
20:32 diakopter watching http://irclog.perlgeek.de/parrotsketch/2011-09-06 live is interesting
20:33 masak aye.
20:36 masak rakudo: lines("README".IO)
20:36 p6eval rakudo 860fe1:  ( no output )
20:36 im2ee joined #perl6
20:36 masak rakudo: say lines("README".IO)
20:36 p6eval rakudo 860fe1: OUTPUT«Method 'readline' not found for invocant of class 'Mu'␤  in method get at src/gen/CORE.setting:4681␤  in sub coro at src/gen/CORE.setting:3732␤  in method reify at src/gen/CORE.setting:3702␤  in method reify at src/gen/CORE.setting:3506␤  in method reify at src/gen…
20:36 masak hm.
20:37 diakopter rakudo: say "README".IO.readline
20:37 p6eval rakudo 860fe1: OUTPUT«Method 'readline' not found for invocant of class 'IO'␤  in <anon> at /tmp/eOdfB3CxJy:1␤  in <anon> at /tmp/eOdfB3CxJy:1␤␤»
20:37 masak rakudo: lines("words")
20:37 p6eval rakudo 860fe1: OUTPUT«Too many positional parameters passed; got 2 but expected 1␤  in method lines at src/gen/CORE.setting:2906␤  in sub lines at src/gen/CORE.setting:4777␤  in <anon> at /tmp/DJP5vqcJu3:1␤  in <anon> at /tmp/DJP5vqcJu3:1␤␤»
20:37 masak that's the error I was looking for.
20:37 * masak submits rakudobug
20:38 masak breaks hangman in the adventure game book.
20:38 masak b: lines("README"); say "alive"
20:38 p6eval b 1b7dd1: OUTPUT«Operation not permitted in safe mode␤  in 'Safe::forbidden' at line 2:/tmp/KUoemhjfvP␤  in 'lines' at line 5932:src/gen/core.pm␤  in main program body at line 22:/tmp/KUoemhjfvP␤»
20:38 masak b++
20:38 moritz b--
20:38 moritz it tried to open file README
20:38 moritz but iirc it shouldn't, should it?
20:38 masak well, that was right at some point ;)
20:39 moritz nom: say "a\nb\nc".lines.perl
20:39 diakopter what's b
20:39 masak diakopter: welcome back :P
20:39 moritz rakudo: say "a\nb\nc".lines.perl
20:39 p6eval rakudo 860fe1: OUTPUT«("a", "b", "c").list␤»
20:39 moritz diakopter: it's the beijing release
20:39 masak diakopter: we decided to confuse everyone with our branch names.
20:39 moritz diakopter: last release of the old rakudo/master
20:40 diakopter k thx
20:40 masak from what I can see in the setting source, the sub form &lines tries to call the method form .lines with an argument that doesn't bind to a parameter.
20:40 masak $limit
20:40 zby_home_ joined #perl6
20:41 moritz where is $limit not bound?
20:42 masak in Str.lines
20:42 * moritz doesn't understand
20:42 moritz oh
20:43 moritz it's used in IO.lines
20:43 masak doesn't help us much :)
20:45 moritz so what should "a\nb\nc\n".lines(1) do?
20:45 moritz return "a"? or the whole string?
20:45 TimToady rakudo: sub MAIN { say "hi" }
20:45 p6eval rakudo 860fe1: OUTPUT«hi␤»
20:45 TimToady innersting
20:45 moritz TimToady: nom has a *very* basic MAIN implementation
20:45 jnthn rakudo: sub MAIN ($outfile = 'sphere-perl6.pgm') { say $outfile }
20:45 p6eval rakudo 860fe1: OUTPUT«sphere-perl6.pgm␤»
20:46 Tene rakudo: sub MAIN($a,$b) { ... }
20:46 p6eval rakudo 860fe1: OUTPUT«Not enough positional parameters passed; got 0 but expected 2␤  in sub MAIN at /tmp/uGrlaFDuR9:1␤  in sub MAIN_HELPER at src/gen/CORE.setting:5956␤  in <anon> at /tmp/uGrlaFDuR9:1␤  in <anon> at /tmp/uGrlaFDuR9:1␤␤»
20:46 Tene :(
20:47 diakopter rakudo: my sub MAIN { say "hi" }
20:47 p6eval rakudo 860fe1: OUTPUT«hi␤»
20:47 TimToady it's actually been some time since I "ported" sphere to nom, so MAIN showing up is not too surprising
20:50 diakopter rakudo: our sub MAIN { say "hi" }
20:50 p6eval rakudo 860fe1: OUTPUT«hi␤»
20:54 masak rakudo: .say for "README".IO.lines
20:54 p6eval rakudo 860fe1: OUTPUT«Method 'readline' not found for invocant of class 'Mu'␤  in method get at src/gen/CORE.setting:4681␤  in sub coro at src/gen/CORE.setting:3732␤  in method reify at src/gen/CORE.setting:3702␤  in method reify at src/gen/CORE.setting:3506␤  in method reify at src/gen…
20:54 masak I think this should work, too.
20:54 jnthn b: .say for "README".IO.lines
20:54 p6eval b 1b7dd1: OUTPUT«Method 'eof' not found for invocant of class ''␤  in 'IO::eof' at line 5675:src/gen/core.pm␤  in <anon> at line 5691:src/gen/core.pm␤  in main program body at line 1:src/metamodel/RoleToInstanceApplier.nqp␤»
20:55 masak /o\
20:55 diakopter wow
20:55 masak screw this, I'm learning Python instead.
20:56 masak :P
20:56 tadzik NO CHOICES MEAN YOU CAN NEVER BE WRONG
20:56 tadzik ha! That's like straight from my fortunes file!
20:56 masak AT LEAST WE'LL BE WRONG TOGETHER
20:57 tadzik masak: https://gist.github.com/1198940
20:57 masak I swear I didn't know about that quote
20:57 masak at least not consciously ;)
20:58 tadzik (:
20:58 masak anyway, I hear Python 3k is great, and only was 8 years in the making.
20:59 flussence those print() parentheses were crafted out of pure unobtanium!
20:59 tadzik Python 3 to Python 2 is like Perl 5.12 to Perl 5.10, no?
20:59 Woodi hi #perl6
20:59 dalek roast: 0ae420b | moritz++ | S32-str/lines.t:
20:59 dalek roast: test Str.lines(limit), lines(Str)
20:59 dalek roast: review: https://github.com/perl6/roast/commit/0ae420be90
20:59 tadzik hi Woodi
20:59 PerlJam hello Woodi
20:59 dalek rakudo/nom: fe590ef | moritz++ | src/core/Str.pm:
20:59 dalek rakudo/nom: implement $limit in Str.lines
20:59 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/fe590ef357
21:00 Woodi ii cannot sleep tonight and decided to put online "collection" of "text"s :)
21:00 flussence tadzik: I guess so. The major version increment is there because they changed the syntax of some fairly fundamental things, and There's Only One Way To Do It
21:00 tadzik yeah
21:01 Woodi i think such colection can be idea for blogs to get more new users...
21:01 PacoLinux_ joined #perl6
21:01 Woodi i mean, especially part new-features-in-p6...
21:01 masak yo Woodi
21:02 Woodi its at https://github.com/slunski/CrazyP6/ , pls not ban me....
21:03 pernatiy joined #perl6
21:03 Woodi just idea, becouse, it's possible, you look on this from close distance...
21:03 masak we don't usually ban people.
21:04 Woodi i strongly hope...
21:04 tadzik we should have a wiki page with those
21:05 masak it's a bit hard to read in its current form because some of the lines are very long.
21:05 flussence masak: click the "raw" link, it wraps better there :)
21:05 Woodi i rather think about that all as sensation not something faq :)
21:06 im2ee Ehh.. Im ill. Good night! Dobranoc :)
21:06 tadzik dobranoc im2ee
21:06 Woodi it's just copy-paste...
21:06 Woodi czesc :)(
21:07 Woodi ok, i will go now too. i will try to repair redability. gn
21:07 tadzik gn Woodi
21:15 diakopter moritz: your irclog is hanging for me
21:16 diakopter as of 4 minutes ago
21:16 flussence hm, same
21:17 diakopter oo now it's back
21:21 flussence .oO( I get the somewhat dubious honour of having the last word in a file named "CrazyP6"... )
21:22 tadzik (:
21:22 natureboy joined #perl6
21:24 moritz diakopter: seems to work again here... for you too?
21:28 flussence moritz: works alright now. Seems like the HTTP server went to sleep, it was getting to the "waiting for response" part
21:29 bluescreen10 joined #perl6
21:29 * moritz doesn't like that, but finds it much less disturbing than hanging logging bots
21:34 diakopter moritz: which web server is it
21:34 masak 'night, #perl6
21:34 diakopter n
21:34 moritz diakopter: mine
21:34 * moritz -> sleep
21:34 diakopter moritz: I mean, which software
21:34 moritz diakopter: apache2
21:35 pmurias joined #perl6
21:36 jdhore joined #perl6
21:43 pmurias joined #perl6
21:47 y3llow joined #perl6
21:48 pothos_ joined #perl6
21:49 pmurias joined #perl6
21:49 y3llow joined #perl6
21:51 pothos_ joined #perl6
21:54 lue joined #perl6
21:57 pmurias joined #perl6
21:58 molaf joined #perl6
22:01 samlh joined #perl6
22:17 pmurias joined #perl6
22:37 araujo joined #perl6
22:37 araujo joined #perl6
22:40 japhb joined #perl6
22:46 plobsing_ joined #perl6
22:48 pmurias joined #perl6
22:54 whiteknight joined #perl6
22:54 pmurias joined #perl6
22:55 [Coke] joined #perl6
23:00 felher joined #perl6
23:03 pmurias joined #perl6
23:29 benabik joined #perl6
23:31 pmichaud comments on my performance today in #parrotsketch and #parrot extremely welcome.  email or /msg if you don't want your comments to be public
23:36 cognominal joined #perl6
23:45 felher pmichaud: i know very little about parrot so i guess i won't be of any help on that matter. I would like to read what you've written (for the sake of learning), though, if there is a public log?
23:46 felher pmichaud: oh, seems like there is the same log available as there is for #perl6. So never mind my question. :)
23:48 cotto_work yup.  moritz++ does a nice job with that
23:51 felher yeah, moritz++ indeed.
23:52 jerome_ joined #perl6
23:54 molaf_ joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo