Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2010-01-14

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:03 cognominal joined #perl6
00:07 jan_ joined #perl6
00:16 ash_ colomon: well, i'll look into it to see if i can narrow it down
00:18 ash_ _bob_k master builds fine with the latest parrot trunk for me
00:18 _bob_k ash_: Interesting.  Which platform and such are you on?
00:19 ash_ os x 10.6.2, apple gcc 4.2, 64-bit if that makes a difference
00:20 _bob_k OK, so what can I do from here?
00:21 ash_ a bug report would be good
00:21 _bob_k Got an URL?
00:23 ash_ email perl6-bugs at perl.org the
00:23 ash_ the bug report
00:23 ash_ it automattically adds it in from there and notifies people
00:23 _bob_k Will do.
00:23 _bob_k Thanks for your help ash_ and colomon.
00:23 ash_ the bug tracker is at: http://rt.perl.org/rt3/Public if you want to take a look around and see if you can find any other bug reports like yours
00:43 dukeleto joined #perl6
01:05 ashizawa joined #perl6
01:15 simcop2387_ joined #perl6
01:16 justatheory joined #perl6
01:25 agentzh joined #perl6
01:32 Wolfman2000 joined #perl6
01:32 Wolfman2000 sorry for not being here in awhile. Finally got my school schedule figured out.
01:33 Wolfman2000 Did anything recent take place with Perl 6 development within the past few weeks?
01:34 Juerd I think that everything that took place within the past few weeks can be classified as recent.
01:35 Wolfman2000 I'm a bit behind on what took place.
01:35 Juerd Join the club :)
01:35 Wolfman2000 fair enough
01:35 Juerd I still feel I'm years behind
01:36 Juerd I wonder if I will ever catch up
01:36 Wolfman2000 and yet you develop it directly
01:46 xomas joined #perl6
01:53 Juerd http://juerd.nl/i/5e6f1b41f​b0beb7f878d01fe3b069776.png
01:53 Juerd Is googlebot killing feather, indirectly?
01:57 Juerd Oh my
01:57 Juerd I understand why generating the timeline is so consuming
01:57 Juerd http://juerd.nl/i/837c8ca33​8c036903129875e8147f0c0.png
01:59 Juerd svnweb re-enabled, trac disabled
02:00 diakopter I got chided the other week for disabling dev.pugscode.org
02:01 diakopter Juerd: ^^
02:02 Juerd I don't know what chided means but it's probably no good
02:03 Juerd Whoever finds and fixes the cause, can reenable it afterwards :)
02:03 diakopter amen
02:04 diakopter chide: To scold mildly so as to correct or improve; reprimand
02:04 Juerd Ah
02:18 orafu joined #perl6
02:26 japhb joined #perl6
02:27 colomon ng: my $a = list; say $a.WHAT
02:28 p6eval ng b65331: List()␤
02:28 colomon ng: my $a = list; say $a.elems
02:28 p6eval ng b65331: 0␤
02:31 JimmyZ joined #perl6
02:35 colomon defining my own List.new doesn't seem to help; it just doesn't get called when you say List .new.
02:35 colomon :(
02:51 synth joined #perl6
02:59 Exodist joined #perl6
03:00 agentzh joined #perl6
03:57 Baggiokwok joined #perl6
04:07 scp1 ng: sub postfix:<!>($n) { [*] 1..$n }; say [+] gather { for 3..50_00 -> $x { if $x == [+](map { $^a! }, $x.comb()) { take $x; } } }
04:07 p6eval ng b65331: Unable to parse blockoid, couldn't find final '}' at line 1␤current instr.: 'perl6;Regex;Cursor;FAILGOAL' pc 1664 (src/stage0/Regex-s0.pir:907)␤
04:11 solarion joined #perl6
04:14 astrojp left #perl6
04:32 pnate joined #perl6
04:47 Chazz joined #perl6
04:55 scp1 left #perl6
04:59 cognominal joined #perl6
05:45 Exodist joined #perl6
05:54 jaldhar joined #perl6
05:56 diakopter hum dee dum
05:58 c0tt0 joined #perl6
06:01 sjohnson rakudo: say 'hi'
06:02 p6eval rakudo 3867ff: hi␤
06:09 diakopter ng: {}()=0
06:09 p6eval ng b65331: Null PMC access in getprop()␤current instr.: '&infix:<=>' pc 14291 (src/builtins/Junction.pir:113)␤
06:09 diakopter ng: {}()()=0
06:09 p6eval ng b65331: Null PMC access in invoke()␤current instr.: '_block14' pc 29 (EVAL_1:0)␤
06:09 diakopter ng: {}()[0]=0
06:09 p6eval ng b65331: Null PMC access in can()␤current instr.: '!postcircumfix:<[ ]>' pc 278104 (src/gen/core.pir:33442)␤
06:11 diakopter ng: { { { { { Mu } } } } }()()
06:11 p6eval ng b65331: invoke() not implemented in class 'Mu'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤
06:21 diakopter ng: multi token() { }; token infix:<token> { };
06:21 p6eval ng b65331: Method 'isa' not found for invocant of class 'Undef'␤current instr.: 'perl6;Perl6;Actions;_block956' pc 145792 (src/gen/perl6-grammar.pir:29583)␤
06:25 JimmyZ joined #perl6
07:02 saschi good morning
07:09 Su-Shee joined #perl6
07:12 Su-Shee good morning
07:21 kaare joined #perl6
07:33 Baggiokwok joined #perl6
07:48 jaldhar Good morning.  Anyone know why when building the latest rakudo, I get "No ICU lib loaded" when trying to do unicode stuff?
07:49 jaldhar I do have ubuntus' libicu-dev package installed
07:50 jaldhar e.g. say "\c[GUJARATI LETTER JA]"; gives the error
07:53 Tene jaldhar: your parrot was probably compiled without ICU enabled.  Check the configure output.
07:55 jaldhar ah parrot.  Yes I noticed it didn't download parrot so it must be using the old ubuntu parrot or one of my previous experiments.
07:55 jaldhar Will try again...
07:55 iblechbot joined #perl6
07:55 Tene Yes, that would do it. :)
07:55 Tene Glad to help.
08:12 jaldhar say "\c[GUJARATI LETTER JA]\c[GUJARATI LETTER LA]\c[GUJARATI LETTER DHA]\c[GUJARATI LETTER RA]";
08:12 jaldhar જલધર
08:13 jaldhar \o/
08:13 cognominal joined #perl6
08:14 diakopter ja la dha ra
08:14 diakopter oh. heh
08:15 jaldhar yeah.  Gujaratis typically don't pronounce the a at the end of syllables which is why the English transliteration of my name is missing a couple
08:16 sjohnson snurghi
08:18 snearch joined #perl6
08:19 snearch_ joined #perl6
08:29 snearch_ joined #perl6
09:09 agentzh joined #perl6
09:11 dakkar joined #perl6
09:25 meneldor joined #perl6
09:28 jaldhar trans substitutes one character in an array with a character from another.  Which operator does one to many mapping?
09:28 moritz_ jaldhar: also trans
09:28 jaldhar hmm...
09:29 moritz_ rakudo: say 'abc'.trans([<a b c>] => [12, 34, 56])
09:29 p6eval rakudo 3867ff: 123456␤
09:29 moritz_ rakudo: say 'aba'.trans([<a b c>] => [12, 34, 56])
09:29 p6eval rakudo 3867ff: 123412␤
09:30 moritz_ hugme: list projects
09:30 hugme moritz_: I know about book, gge, hugme, ilbot, json, november, nqp-rx, nqpbook, perl6-examples, proto, svg-matchdumper, svg-plot, temporal-flux-perl6syn, tufte, web
09:33 jaldhar moritz:
09:33 jaldhar my @tr = ( "\C[DEVANAGARI LETTER JA]" => 'ja' );
09:33 jaldhar my $s = "\C[DEVANAGARI LETTER JA]";
09:33 jaldhar $s.trans(@tr).say;
09:34 jaldhar doesn't give 'ja'.
09:35 moritz_ that's because you're passing an array, not a pair
09:36 moritz_ rakudo: my @tr = ( "\C[DEVANAGARI LETTER JA]" => 'ja' ); my $s = "\C[DEVANAGARI LETTER JA]"; $s.trans(|@tr).say;
09:36 p6eval rakudo 3867ff: jaaaaaaaaaaaaaaaaaaaaaa␤
09:36 moritz_ wtf?
09:36 jaldhar yup
09:37 moritz_ rakudo: say "\c[DEVANAGARI LETTER JA]"
09:37 p6eval rakudo 3867ff: ज␤
09:37 moritz_ rakudo: say "\C[DEVANAGARI LETTER JA]"
09:37 p6eval rakudo 3867ff: C[DEVANAGARI LETTER JA]␤
09:37 moritz_ jaldhar: that's the problem
09:37 moritz_ jaldhar: you need to use a lower-case \c
09:37 moritz_ \C only makes sense in regexes, iirc
09:38 jaldhar but I am using \c
09:39 moritz_ 10:33 < jaldhar> my @tr = ( "\C[DEVANAGARI LETTER JA]" => 'ja' );
09:39 moritz_ that's a capital C
09:39 moritz_ not c
09:39 moritz_ rakudo: my @tr = ( "\c[DEVANAGARI LETTER JA]" => 'ja' ); my $s = "\c[DEVANAGARI LETTER JA]"; $s.trans(|@tr).say;
09:39 p6eval rakudo 3867ff: j␤
09:39 jaldhar hmm my IRC client capitalized it?  It is definitely \c in my code
09:40 moritz_ then it's time to get another IRC client :-)
09:40 jaldhar and it should give ja not j
09:40 moritz_ that I agree with
09:41 moritz_ rakudo: my @tr = ( "\c[DEVANAGARI LETTER JA]" => 'ja' ); my $s = "\c[DEVANAGARI LETTER JA]"; $s.trans(@tr[0]).say;
09:41 p6eval rakudo 3867ff: j␤
09:41 moritz_ rakudo: my @tr = ( "\c[DEVANAGARI LETTER JA]" => 'ja' ); my $s = "\c[DEVANAGARI LETTER JA]"; $s.trans(|@tr[0]).say;
09:41 p6eval rakudo 3867ff: flattened parameters must be a hash or array␤in Main (file <unknown>, line <unknown>)␤
09:42 moritz_ rakudo: "\c[DEVANAGARI LETTER JA]".trans(["\c[DEVANAGARI LETTER JA]"] => "ja").say
09:42 p6eval rakudo 3867ff: j␤
09:42 moritz_ rakudo: "\c[DEVANAGARI LETTER JA]".trans(["\c[DEVANAGARI LETTER JA]"] => ["ja"]).say
09:42 p6eval rakudo 3867ff: ja␤
09:42 moritz_ jaldhar: ah
09:42 moritz_ jaldhar: if the RHS of the pair is an array, it uses the array elements
09:42 moritz_ jaldhar: if it's a string, it uses the first character of the string
09:43 moritz_ that's... confusing and not very DWIMmy
09:44 moritz_ but it conforms to the spec
09:44 moritz_ http://perlcabal.org/syn/S05.html#Transliteration
09:44 krakan joined #perl6
09:46 jaldhar even with an array I still get only j
09:46 jaldhar oh wait
09:54 jaldhar one last question.  In the line
09:54 jaldhar $line.comb.trans(@tr).join('').say
09:55 jaldhar The output still gives a space between each character group.  i.e. ja la dha ra
09:55 jaldhar how do I supress that?
09:55 moritz_ leave out the .comb
09:55 moritz_ it splits it into characters, and returns a list
09:55 moritz_ and then calling .trans on it treats it as a string again
09:56 moritz_ and the default joining uses a space as the separator
09:56 jaldhar ok got it.  The last time I tried it, rakudo needed the comb to split the line into characters.
09:59 payload joined #perl6
10:22 avar 6/w 4
10:24 riffraff joined #perl6
10:28 GeJ joined #perl6
10:32 jnthn morning #perl6
10:32 Juerd Good morning!
10:33 frettled Good #perl6!
10:33 Juerd frettled: It appears not to have been svnweb, but trac instead
10:33 frettled aha
10:34 frettled So no more bugtracking.  ;)
10:34 moritz_ I think taking down trac wouldn't hurt too much
10:34 Juerd http://juerd.nl/i/5e6f1b41f​b0beb7f878d01fe3b069776.png
10:34 moritz_ basically it's just spam
10:34 Juerd http://juerd.nl/i/837c8ca33​8c036903129875e8147f0c0.png
10:34 frettled eh, urh, yes
10:34 Juerd Digging up that timeline takes a long time, and when Googlebot does that for many dates at once, boom.
10:35 frettled Juerd: so it was just a red herring that the webserver was busy in the /svnweb/ directory when using lots of memory, I see.
10:35 Juerd I think so
10:35 Juerd I had disabled svnweb and then the gigantic load increase happened again
10:36 frettled Elimination works :)
10:37 Juerd Now let's hope that either nobody really needs trac, or someone can fix it.
10:37 Juerd 02:59 <@diakopter> I got chided the other week for disabling dev.pugscode.org
10:37 Juerd 03:01 <@diakopter> Juerd: ^^
10:37 Juerd 03:02 <@Juerd> I don't know what chided means but it's probably no good
10:37 Juerd 03:02 <@Juerd> Whoever finds and fixes the cause, can reenable it afterwards :)
10:37 Juerd (dev.pugscode.org is trac)
10:38 frettled aha
10:39 frettled If someone feels like implementing captcha plus a hidden input field that must be empty, then that could help a lot.
10:39 jaldhar I can't resist...one more question.  If I wanted to make an array of pairs out of two arrays.  Is this the right way to go about it? @c = @a >>=><< @b;
10:39 jaldhar ?
10:39 frettled The trick is to hide the input field with CSS; web scrapers don't know what to make of that.
10:40 Juerd Ehm
10:40 Juerd Most web scrapers don't even see css, so they don't see it's hidden either.
10:40 Juerd And it's likely that they have a prefab spam template for trac anyway
10:40 moritz_ rakudo: say (<a b> »=>« <c d>).perl
10:40 p6eval rakudo 3867ff: Confused at line 2, near "\x{bb}=>\x{ab} <c d>"␤in Main (file <unknown>, line <unknown>)␤
10:40 huf so add some js trickery to it
10:41 frettled Juerd: that's the point
10:41 Juerd huf: Do you feel like implementing this? :)
10:41 moritz_ jaldhar: I think it should work
10:41 Juerd frettled: Oh. Then I don't understand how it would work ;)
10:41 frettled Juerd: if the field has values in it, it means it's spam
10:41 Juerd Ah!
10:41 frettled usually, they fill it with junk values
10:41 moritz_ rakudo: say %(<a b> Z <c d>).pairs.perl
10:41 Juerd That's a smart trick
10:41 p6eval rakudo 3867ff:  ( no output )
10:42 frettled huf: javascript is bad because it means that paranoid people can't use your website.
10:42 moritz_ jaldhar: the trick with the zip operator works locally, I have no idea why p6eval doesn't like it
10:42 huf frettled: yeah, i knew that was going to come up ;)
10:42 huf w3m users ftw
10:42 frettled I just use NoScript when using Firefox.
10:42 moritz_ jaldhar: (and need to apologize for asking, these are interesting questions)
10:43 frettled In Opera, I disable JS by default, and then edit site preferences for trusted websites.
10:44 jaldhar moritz: I hope my questions help in some way.  I'll brb and try your suggestions.
10:49 jnthn std: <a b c> >>=><< (1,2,3)
10:49 p6eval std 29531: ok 00:01 108m␤
10:49 jnthn Probably should work, then...
10:52 frettled std: <a b c> »=>« (1,2,3)
10:52 p6eval std 29531: ok 00:01 106m␤
10:52 frettled just checking ;)
10:53 moritz_ rakudo: say Pair.new('a', 'b').perl
10:53 p6eval rakudo 3867ff: undef => undef␤
10:53 frettled Is there an appropriate unicode symbol for =>, which is supported by Perl 6?
10:53 moritz_ nope
10:53 frettled ng: say Pair.new('a', 'b').perl
10:53 p6eval ng b65331: Pair.new(:key(Mu), :value(Mu))␤
10:54 Juerd Hm. Can't we just go one step further and use µ instead of Mu? :)
10:54 frettled But that's an important symbol!  Quick, halt all development until we can fix that!  ;)
10:54 frettled Juerd: \o/
10:54 Juerd It's in latin1 even.
10:54 Juerd The meaning of Mu would get SO obscure :)
10:54 frettled Juerd: except that µ isn't «mu» :(
10:54 frettled as if it wasn't obscure already :D
10:54 frettled it could be a symbolic cow.
10:54 Juerd Good point there :)
10:55 Juerd No, that would be Moo
11:04 clintongormley joined #perl6
11:10 meppl joined #perl6
11:18 frettled Yes, but the pronunciation is very nearly the same.  :D
11:18 mathw Is there a cow in unicode?
11:19 moritz_ copy on write? :-)
11:19 mathw no, a cow that goes moo
11:20 moritz_ COW implemented in Moose? *g*
11:20 mathw argh
11:21 clintongormley in Perl6, cows go Mu
11:21 mathw Kopfschmerzen
11:25 frettled besser als Herzschmerzen
11:26 moritz_ besser als Herzversagen :-)
11:32 mathw probably
11:40 Su-Shee in Perl 6, there is no nothing, there is mu.
11:45 baz joined #perl6
11:47 baz left #perl6
11:47 baz joined #perl6
11:49 Su-Shee isn't this something where a rakudo example blogged would nicely fit in? http://transfixedbutnotdead.com/201​0/01/13/anyone_for_metaprogramming/
11:51 draegtun joined #perl6
11:52 mathw Su-Shee: yes
11:53 sjohnson mu == nothingness in Japanese
11:53 huf but in english it means somethig different
11:53 bluescreen joined #perl6
11:54 Su-Shee sjohnson: I know. :)
11:54 Su-Shee german cows say muh.
11:54 mathw english cows say moo
11:54 mathw or sometimes mooooooo
11:54 mathw if they're trying to get an urgent point across
11:55 Su-Shee yes, german cows can extend the u to muuuuuuuuh.
11:55 huf ours say "mú", unless they're fighting for the right of bovine freedom
11:55 Su-Shee huf: as long as they're not suddenly saying mü ;)
11:56 huf no, that's μ
11:56 saschi well
11:56 saschi that's true
11:57 draegtun Su-Shee: Like this - http://gist.github.com/276591
11:57 Su-Shee draegtun: ah, yours anyway. :)
11:57 draegtun :-)
11:58 * Su-Shee is looking forward to the book "metaprogramming in perl 6" and "higer order six" ;)
11:59 draegtun I have a question on my gist.   How do you define dynamic methods in perl6?
11:59 draegtun ie. in Perl5:  *{'subname'} = sub { ... };
12:00 draegtun lots of trawling didn't turn up anything :(
12:00 sjohnson well bed time!!! cya
12:00 moritz_ draegtun: it probably will work through the metaclass
12:01 jnthn draegtun: That'd be something done through the metamodel; Rakudo master doesn't provide a good way to do it yet. We've got a branch that would let you do it though, or almost would.
12:01 moritz_ draegtun: so you can say things like YourClass.^add_method('name', sub { ... })
12:01 draegtun mortiz_: can u give me some pointers to docs.  I trawled for metaclass but couldn't find anything specific
12:02 draegtun moritz_: ah... tried something close to that.. let me just try it again :)
12:02 moritz_ as jnthn said, it probably doesn't quite work yet
12:02 moritz_ draegtun: http://github.com/rakudo/raku​do/blob/ng/docs/metamodel.pod
12:05 draegtun just tested it and it does work!
12:05 moritz_ cool
12:05 moritz_ draegtun: then you're encouraged to reply to http://transfixedbutnotdead.com/201​0/01/13/anyone_for_metaprogramming/ :-)
12:06 draegtun I'm doing a followup post to that with Perl6 (re: gist)
12:07 moritz_ cool; please post the URL of your post here once you're done
12:07 draegtun btw:  draegtun == transfixedbutnotdead :)
12:07 moritz_ oh
12:08 Su-Shee though I really liked the blogname fingernailsinoatmeal
12:10 Baggiokwok joined #perl6
12:11 draegtun I've just updated the gist with the ^add_method change
12:11 draegtun I'll do a blog on it later on or tomorrow morning
12:12 draegtun many thanks for you help moritz, jnthn
12:16 Su-Shee hm, the perl 6 example looks very nice, visually compared to the ruby and javascript code.
12:18 draegtun Su-Shee: Methinks so to
12:20 Su-Shee and I really mean it, it just hit me how clean perl 6 looks.
12:22 draegtun Su-Shee: yep... I found it very intuitive.  Everything worked as I expected. Only bumps in road was "is also" to reopen class (trawling web seemed to bring up "is augment*" most which didn't work) and the dynamic method (^add_method)
12:23 draegtun I love the:  "$drew does ThrowStar;" bit
12:23 cognominal joined #perl6
12:24 moritz_ ng: class A { }; augment class A { method b { say "hi" } }; A.b()
12:24 p6eval ng b65331: hi␤
12:24 draegtun did try:  $drew does role { ... };
12:24 draegtun but that didn't work
12:25 moritz_ it's just rakudo (master branch) that's behind
12:25 moritz_ std: my $drew; $drew does role { ... };
12:25 jnthn draegtun: It will in the future; anonymous roles are just not yet implemented.
12:25 p6eval std 29531: ok 00:01 109m␤
12:25 draegtun moritz_:  yes came across that but I preferred   class A is also {}    Is there a difference?
12:25 cognominal joined #perl6
12:26 payload joined #perl6
12:26 moritz_ draegtun: 'is also' is removed from the spec
12:26 draegtun moritz_:  re: does role - Excellent... very intuitive
12:26 moritz_ draegtun: because it collides with a potential class named 'also'
12:27 draegtun moritz_:  OK... I'll change that to "augment class A {}" then
12:27 moritz_ draegtun++ #nice blog post
12:28 moritz_ draegtun++ # exploring Perl 6
12:30 draegtun moritz_:  augment doesn't work for me in my version of rakudo :(
12:30 moritz_ draegtun: yes, it only works in the "ng" branch :/
12:31 draegtun moritz_:  OK... I'll put that in the blog post about "augment"
12:31 meneldor guys is this http://cog.cognitivity.com/perl6/1_intro/ enough up to date for OO learning?
12:32 jnthn draegtun: The spec changed and Rakudo is still catching up. It's done in a branch, so it'll land in master in not too long, hopefully. :-)
12:33 moritz_ meneldor: I wouldn't recommend it, it's full of unclear terminology
12:33 ruoso joined #perl6
12:33 fridim_ joined #perl6
12:33 meneldor moritz_:  please tell me where to learn about p6 OO
12:33 moritz_ meneldor: the perl 6 book currently being written has one chapter on classes and one on roles
12:34 moritz_ http://github.com/perl6/book/
12:34 meneldor thank you
12:34 meneldor im starting to learn from wrong places for second time
12:34 meneldor there are too much obsolete tutorials
12:35 moritz_ aye
12:35 moritz_ that's why I started perl6.org, to link to those which are (mostly) up-to-date
12:37 baz joined #perl6
12:37 moritz_ (just an example for the inaccuracies: http://cog.cognitivity.com/perl6/1_intro/2.html the "properties" are actually called "traits", and you can't just make them up; either you have to use existing traits, or write roles which are specifically designed to work as traits)
12:38 moritz_ also it doesn't show what you can do with these "properties", so I guess the author hasn't really understood them
12:39 meneldor :) so if i learn p6 "that way" ill be completely confused later
12:39 meneldor those people have to remove these documents or keep them up to date
12:40 moritz_ on http://cog.cognitivity.com/perl6/ at least they write "IMPORTANT: This document is obsolete. The design of Perl6 has changed in important ways since this was written: refer to the Perl6 documentation efforts elsewhere. These pages will stay up for archival interest only: don't rely on them! -- Mike"
12:42 rodi joined #perl6
12:48 meneldor so this is my bad :) i have to read more carefuly
12:50 payload joined #perl6
12:50 moritz_ well, if google sends you straight to a sub page, you'll never see that warning
12:50 mikehh joined #perl6
12:52 takadonet morning all!
12:53 SmokeMachine joined #perl6
13:04 takadonet left #perl6
13:17 moritz_ btw I thought http://cog.cognitivity.com/perl6/  looked like a wiki, and registered, but I couldn't actually edit anything
13:18 zibri left #perl6
13:25 ignacio_ joined #perl6
13:26 huf hai guis, what does "ningen nante" mean in japanese?
13:26 huf i have a hunch someone here speak it
13:34 Su-Shee afaik it's some singer.
13:35 breinbaas joined #perl6
13:42 plainhao joined #perl6
13:42 bpphillips joined #perl6
13:53 Lorn joined #perl6
13:54 perlpilot joined #perl6
13:55 perlpilot feather is out of ptys again.
13:56 payload joined #perl6
13:56 frettled Hmm, that's a typical xen problem IIRC.  Is this a xen thingy?
13:57 perlpilot dunno if Juerd uses xen or one of the other virtualization thingys
13:57 frettled s/xen problem/xen configuration problem/
13:57 frettled 2.6.18-6-xen-686
13:58 frettled looks like it
13:59 frettled Juerd: the pty problem may have something to do with udev support
14:04 Juerd It is a xen thing, yes
14:05 Juerd I have a few dozen xen boxes, but only feather users a lot of pty's.
14:05 Juerd Do you happen to know how to fix it?
14:05 Juerd s/users/uses/
14:05 takadonet joined #perl6
14:06 * Juerd upgrades udev. Maybe that does something useful :)
14:06 frettled Juerd: http://www.xen-support.com/?p=16
14:07 Juerd | You are currently upgrading udev to a version that is not compatible with the currently running   |
14:07 Juerd | kernel.
14:07 Juerd Woops. Bailing out.
14:07 frettled :D
14:07 perlpilot Juerd++ frettled++
14:08 Juerd Oh my. Do an ls /dev
14:08 Juerd I think the current udev is also not quite compatible.
14:08 Juerd At least not terribly functional
14:08 frettled hmm
14:08 Juerd 1;0 root@feather:/dev# /etc/init.d/udev start
14:08 Juerd udev requires a kernel >= 2.6.26, not started. ... failed!
14:08 Juerd Hah.
14:08 Juerd And nobody noticed thus far? :)
14:09 frettled ah, and Debian doesn't package Xen kernels > 2.6.18
14:09 frettled (IIRC)
14:09 Juerd They do
14:09 Juerd Somewhere.
14:09 frettled oh
14:09 frettled perhaps in volatile or backports, then.
14:09 Juerd But I don't really have time for that right now :(
14:10 frettled and that requires a bit of attention to get right
14:10 PhilHibbs joined #perl6
14:10 frettled since you probably ought to do aptitude update && aptitude safe-upgrade
14:10 frettled Is there any particular reason why it's running a mix of unstable and experimental=
14:11 frettled s/=/?/
14:13 takadonet left #perl6
14:13 frettled perlpilot: I had two shells on feather, I logged out from one now.
14:13 frettled so that might help a tiny bit
14:14 perlpilot frettled: thanks.  :)
14:14 slavik joined #perl6
14:14 perlpilot Juerd: anyone else have root on the "real" feather that could do it?
14:14 moritz_ any ideas what I could blog about in the upcoming 5 hours of train ride?
14:15 slavik moritz_: how to install rakudo?
14:15 mathw moritz_: http://transfixedbutnotdead.com/201​0/01/13/anyone_for_metaprogramming/
14:15 moritz_ slavik: that's covered in three different places already
14:15 perlpilot slavik: that's been done a few times and hasn't changed since.
14:15 slavik oh, ok
14:16 moritz_ mathw: the author has already told us that he'll write a Perl 6 followup
14:16 mathw woo
14:16 perlpilot moritz_: blog about book material  :)
14:16 moritz_ perlpilot: :-)
14:17 frettled I'm also looking for something to blog about :)
14:19 iblechbot joined #perl6
14:20 Juerd frettled: Too many people have had root access on feather, and most choices were never documented
14:21 ignacio_ joined #perl6
14:22 frettled yay
14:23 Juerd It's unlikely that you'll encounter a bigger mess in debian context, except perhaps a box where even more things are custom compiled :)
14:24 Juerd I've seen the "Yes, do as I say" safety prompt a few times on feather, when trying to upgrade.
14:25 Juerd (For those who are unfamiliar with that: apt asks you to type this sentence when you're about to uninstall a vital package; in general that shouldn't happen when upgrading.)
14:26 Juerd I think that unless someone wants to find out what goes on on feather1, perhaps a new virtual machine should be erected and the old one deprecated
14:27 frettled I think that's a very sound idea.
14:28 frettled Also, it could pay to separate functionality a bit more.
14:28 frettled Perhaps use mpm-itk for different virtualhosts, that will make it easier to see which is the culprit/victim.
14:29 frettled Unfortunately, I haven't had a chance of learning how to use mpm-itk yet, too little time on my hands.  :(
14:29 Juerd These things are also from before the time feather was virtualised
14:29 frettled /o\
14:29 Juerd There's a different VM for production stuff, but some things were never migrated there.
14:29 frettled they were never production ready, then ;)
14:29 moritz_ it would be good to migrate svn there
14:29 Juerd frettled: Agreed.
14:30 Juerd moritz_: Now all we need is someone to do it :)
14:30 moritz_ Juerd: right
14:30 Juerd (and maintain it)
14:30 moritz_ and migrate commitbit along with it :/
14:30 Juerd Commitbit has been the cause of several problems in the past.
14:30 Juerd I forgot what kind of problems.
14:31 * moritz_ thinks about writing something more lightwight on top of an IRC bot
14:31 Juerd That would perhaps make a little bit too much sense.
14:32 PerlJam moritz_: you can blog about your experiences setting up the new feather  ;-)  When's that train ride again?  :)
14:32 moritz_ too soon :-)
14:32 moritz_ I'd much prefer to just convert it all to git, and put it up on github
14:33 PerlJam \o/
14:34 moritz_ but it requires some more thoughts and planning
14:34 moritz_ but that's something I could blog about, yes
14:34 frettled I still need a blogging topic! :D
14:34 moritz_ I think I've missed to 10 days deadline for the ironman anyway
14:35 moritz_ but since my score was never recorded properly... whatever
14:40 Juerd frettled: Blog about the mess you found when you first logged in on feather yesterday :)
14:41 mathw Blog about something utterly terrible but ultimately insignificant you heard on the news today, and how it made you think of a cool thing in Perl 6
14:45 frettled Juerd: heh
14:46 frettled mathw: also a nice idea
14:47 mathw of course, it only works if he did think of a cool thing in Perl 6
14:49 jaldhar joined #perl6
14:51 * Juerd closed a shell, so there's a pty available now :)
14:52 Juerd Be quick!
14:52 Juerd :P
14:54 colomon__ joined #perl6
14:58 takadonet joined #perl6
15:03 colomon_ joined #perl6
15:05 cdarroch joined #perl6
15:05 takadonet morning colomon
15:09 nihiliad joined #perl6
15:09 nihiliad left #perl6
15:13 perigrin_ joined #perl6
15:13 orevdiabl joined #perl6
15:14 Psyche^ joined #perl6
15:14 hatseflats joined #perl6
15:14 tomaw_ joined #perl6
15:23 KyleHa joined #perl6
15:23 stepnem joined #perl6
15:25 leedo joined #perl6
15:30 soupdragon joined #perl6
15:36 saschi noch wer da?
15:45 araujo joined #perl6
15:51 ash_ joined #perl6
15:52 ash_ jnthn: ping?
15:52 colomon takadonet: morning
15:53 * colomon has glasses.  what a strange thing!
15:53 mberends joined #perl6
15:54 mdxi glasses++
15:56 ash_ how do i leave a message for someone again? i forget
15:56 jnthn ash_: phenny tell ...
15:56 jnthn oh, wait...
15:56 jnthn ash_: pong! :-)
15:56 ash_ lol
15:56 colomon ash_: phenny tell masak ....
15:57 * jnthn was afk making a cuppa and missed the ping :-)
15:57 mberends phenny, tell ash_ something
15:57 phenny mberends: I'll pass that on when ash_ is around.
15:57 ash_ $P1= get_hll_global 'Role'
15:57 phenny ash_: 15:57Z <mberends> tell ash_ something
15:57 ash_ $P0 = '&infix:<~~>'(mixin_role, $P1)
15:57 ash_ say $P0  # inside add_meta_method returns 0
15:57 ash_ that's why the infix:<does> isn't being dispatched
15:58 jnthn ash_: Oh, that is...odd...
15:58 jnthn ash_: But good find.
15:58 pnate joined #perl6
15:59 jnthn That's an easier point to debug from.
15:59 ash_ so, P6role ~~ Role = 0
15:59 jnthn ash_: Oh, gah, yes. Hm.
15:59 jnthn ash_: Does liberalizing infix:does's signature a bit help?
15:59 jnthn e.g. if you s/Role/Mu/
16:00 ash_ probably, i'll try
16:00 jnthn It's in src/core/operators.pm iirc.
16:01 colomon jnthn: don't know if you saw this weirdness in the backlog:
16:01 colomon ng: my $a = List.new; say $a.WHAT; say $a.elems;
16:01 p6eval ng b65331: List()␤Method 'elems' not found for invocant of class ''␤current instr.: 'perl6;Mu;' pc -1 ((unknown file):-1)␤
16:01 ash_ jnthn: it gets further, but fails on 'Method '!select' not found for invocant of class 'P6role''
16:03 zaphar_ps joined #perl6
16:03 jnthn ash_: aha.
16:04 jnthn ash_: That probably means mixing in parametric roles is broken.
16:04 clintongormley joined #perl6
16:04 jnthn Which woudln't surprise me.
16:05 jnthn ash_: I can try and fix that.
16:05 jnthn ash_: In the meantime rather than passing a P6Role you could try doing:
16:06 jnthn oh, wait...
16:06 jnthn No, that's a messy fix. :/
16:06 jnthn I guess answer is to fix the real issue.
16:06 jnthn ash_: I can do that later today...
16:06 jnthn ash_: Part of it relates to an open question I postd here a day or two ago though.
16:08 ash_ ah, okay
16:08 ash_ well, i guess i'll go find something else to break and fix then, like Lists.elems :p
16:09 colomon List.new is the real issue, I'm pretty sure.  :)
16:10 ash_ ng: my $a = List.new; say $a.WHAT;
16:10 p6eval ng b65331: List()␤
16:10 ash_ hmm
16:10 colomon ng: my $a = list; say $a.WHAT; say $a.elems;
16:10 p6eval ng b65331: List()␤0␤
16:10 ash_ so is it probably not initializing the attribute?
16:10 ash_ o.0
16:11 jnthn ash_: I'll try and think of other OO-ish tasks you may be interested in too. :-)
16:11 ash_ sure, would be gald to help
16:11 ash_ AUTO*?
16:11 jnthn ash_: My cold has been getting worse rather than better, so my thinkiness today is not on great form... :-/
16:12 ash_ no good, its no biggy, i'll just try to find ng bugs if you don't think of anything, its not a big deal
16:12 * colomon went to the doctor his morning for his cold...
16:13 colomon *this
16:21 araujo joined #perl6
16:26 sjn jnthn: hey, I read that you're moving north? :)
16:27 jnthn sjn: Yes!
16:27 sjn jnthn: Malmö or Lund or something?
16:27 sjn or further north
16:27 jnthn sjn: Around there, yes.
16:27 jnthn Not arctic circle far north. ;-)
16:28 sjn Well, it'll be "close" to cph.pm at least :)
16:28 jnthn Yeah.
16:28 jnthn Closer to Oslo.pm :-)
16:28 sjn do tell if you take a trip to Oslo though
16:28 sjn really?
16:28 jnthn erm
16:28 jnthn closer than I am now ;-)
16:28 sjn ah :)
16:28 jnthn Not closer than to Copenhagen. :-)
16:28 ash_ i have class, i'll see if i can find the List.elems issue out, be back later
16:29 jnthn Maybe I cam make an Oslo.pm meeting some month though. :-)
16:29 jnthn It's a train journey rather than a flight now. :-)
16:29 sjn jnthn: well, I hope you'll enjoy staying in the land of the Mai-stang, Surströmning and Must :)
16:31 jnthn sjn: I may give the surströmning a miss... :-)
16:34 sjn booo! ;-)
16:34 sjn ypi
16:34 sjn you'll miss out on some of the best food in sweden!
16:34 sjn ("best" on some as-of-yet undetermined scale, of course)
16:34 jnthn There are many fermented substances I like, but I don't class herring as one of them. :-)
16:35 sjn I'm sure you can aquire that taste, with a little effort ;-)
16:35 sjn (and enough aquavit)
16:41 IllvilJa jnthn: kudos to you for regarding 'surstr�mming' as fermented (which it is) and not rotten (which it really isn't, despite some slander ignorant souls spreads)...
16:43 jnthn There's More Than One Way To Taste Bad. ;-)
16:43 jnthn .oO( oops, no more kudos for me )
16:45 IllvilJa jnthn: heh, 'surstr�mming' isn't the 'worst' Sweden has produced.  The candy known as 'Jungelvr�l' (English translation: "Djungle roars") at least makes US workmates of mine to retreat in horror when I offer it to them.
16:45 araujo joined #perl6
16:46 IllvilJa Maybe that says more about US candy taste than it says about 'Jungelvr�l'  though ;-).
16:46 Exodist joined #perl6
16:46 kaare_ joined #perl6
16:47 IllvilJa ('Jungelvr�l' is licorice candy covered with salmiak powder or something similar that have a darn intense, almost corroding, licorice taste...)
16:47 jnthn Oh, I like licorice.
16:48 nihiliad joined #perl6
16:48 payload joined #perl6
16:48 sjn IllvilJa: that's not so bad....
16:49 Su-Shee hm, my favorite licorice shop isn't available in english...
16:49 sjn the intensity doesn't last long, right?
16:49 uniejo joined #perl6
16:51 justatheory joined #perl6
16:52 IllvilJa Well, I like them.  Yes, the intensity lasts for a moment just when you put them in your mouth (it is some sort of salmiak salt powder that covers them).  After that, they are just like any tasty licorice candy you chew on.  A URL is here: http://sv.wikipedia.org/wiki​/Djungelvr%C3%A5l#Djungelvr.C3.A5l.  Swedish page, but check out the picture with 8 black candy pieces to the right, there you got them.
16:52 KyleHa joined #perl6
16:52 jnthn IllvilJa: Oh, I have had those!
16:52 jnthn IllvilJa: I actually like them.
16:52 jnthn Bit of a surprise the first time though. ;-)
16:55 jnthn IllvilJa: I prefer these though: http://my.opera.com/kirstycat/blog/billar
17:00 xinming joined #perl6
17:09 pnate2 joined #perl6
17:09 frettled Juerd: blogified, sortof :)
17:09 cotto_work joined #perl6
17:11 ash_ joined #perl6
17:13 hejki ng: sub postfix:<!>($n) { [*] (1..$n) }; say 5!;
17:13 p6eval ng b65331: Unable to parse blockoid, couldn't find final '}' at line 1␤current instr.: 'perl6;Regex;Cursor;FAILGOAL' pc 1664 (src/stage0/Regex-s0.pir:907)␤
17:14 hejki ng: sub postfix:<!>($n) { [*] (1..$n); }; say 5!;
17:14 p6eval ng b65331: Unable to parse blockoid, couldn't find final '}' at line 1␤current instr.: 'perl6;Regex;Cursor;FAILGOAL' pc 1664 (src/stage0/Regex-s0.pir:907)␤
17:15 colomon hejki: I'm pretty sure .. still doesn't work properly in ng.
17:15 colomon [*] probably doesn't either.
17:15 hejki ng: sub fac($n) { [*] (1..$n); }; say fac(5);
17:15 p6eval ng b65331: Unable to parse blockoid, couldn't find final '}' at line 1␤current instr.: 'perl6;Regex;Cursor;FAILGOAL' pc 1664 (src/stage0/Regex-s0.pir:907)␤
17:15 hejki mm..
17:16 ash_ ... works though
17:16 ash_ ng: say 1...3;
17:16 p6eval ng b65331: 123␤
17:16 colomon ng; sub postfix:<!>($n) { say $n }; 5!
17:16 ash_ but they kinda mean 2 different things, if only slightly different
17:16 hejki ng: sub fac($n) { reduce { $^a * $^b }, (1..$n); }; say fac(5);
17:16 p6eval ng b65331: Could not find non-existent sub &reduce␤current instr.: '&fac' pc 119 (EVAL_1:50)␤
17:16 hejki rakudo: sub fac($n) { reduce { $^a * $^b }, (1..$n); }; say fac(5);
17:16 p6eval rakudo 3867ff: 120␤
17:16 hejki =(
17:16 colomon ng: sub postfix:<!>($n) { say $n }; 5!
17:16 p6eval ng b65331: sh: ./perl6: No such file or directory␤
17:17 colomon I kind of think operator overloading doesn't work for new operators in ng yet either.
17:17 rodi joined #perl6
17:17 hejki ng: sub fac($n) { my $prod = 1; for 2..$n -> $x { $prod *= $x; }; $x }; say fac(5);
17:17 p6eval ng b65331: sh: ./perl6: No such file or directory␤
17:18 colomon yeah, it doesn't.
17:18 hejki dang. it seems a real long way to got
17:18 hejki s/t$$//;
17:19 colomon ng: sub fac($n) { my $prod = 1; for 2...$n -> $x { $prod *= $x; }; $x }; say fac(5);
17:19 p6eval ng b65331: Symbol '$x' not predeclared in &fac␤current instr.: 'perl6;PCT;HLLCompiler;panic' pc 137 (src/PCT/HLLCompiler.pir:101)␤
17:19 ash_ there is still always master rakudo
17:19 colomon :\
17:19 hejki y :<
17:19 hejki ng: sub fac($n) { my $prod = 1; my $x; for 2..$n -> $x { $prod *= $x; }; $x }; say fac(5);
17:19 p6eval ng b65331: Method 'Num' not found for invocant of class 'Range'␤current instr.: 'perl6;Mu;' pc -1 ((unknown file):-1)␤
17:21 ash_ wonder what it would take to get new operators working in ng... that seems difficult
17:22 hejki what's the current suggested parrot revision?
17:23 PerlJam hejki: for master or ng?
17:23 hejki master
17:25 ingy joined #perl6
17:30 ash_ can you do a backtrace in parrot to see the stack frame if your not sure where you are?
17:30 colomon ash_: I believe it would take modifying the grammar on the fly to add new operators in ng.
17:34 ash_ colomon: sounds like a reasonable assumption, how does master handler it? by augmenting the grammar on the fly?
17:36 jackyf joined #perl6
17:36 IllvilJa jnthn: Regarding those 'bilar' (Swedish word for 'cars' BTW), our Swedish astronaut Christer Fugelsang got a special edition of those manufactured for him.  Instead of having the shapes of cars, the candy had the shape of space shuttles.  Only 100 bags were made though and unfortunately it were just a one off :-(.
17:38 IllvilJa jnthn: but he took the opportunity to eat those in space in microgravity during a real time press conference, which were sort of fun :-).  Most of the bags were given to readers of his blog down here on earth.
17:45 Juerd "Juerd had already stopped Apache, but someone -- I don't know who -- started it again at 12:00, probably anxious that SVN and such didn't work."
17:45 Juerd frettled: Yea, that was probably /usr/local/bin/incredibly_u​gly_hack_to_restart_apache
17:45 frettled Juerd: oooh
17:46 frettled my goodness, that's actually what its name is!
17:46 Juerd That filename is one of the rare cases of good documentation on feather ;)
17:46 frettled hee-hee
17:46 frettled I don't understand what's happening, though.  I try to write shorter blog entries, but they just end up becoming longer and longer and longer.
17:47 Juerd So far I'm enjoying the read :)
17:47 frettled Soon I'll need to create a wiki page for each of my blog entries!
17:47 frettled Juerd: \o/
17:47 rodi joined #perl6
17:48 Juerd "It's just as if someone was waiting for me to go idle in order to produce the problem that I was looking for." -- So horribly familiar.
17:49 Juerd frettled: I don't think the spam itself is the DOS attack. I think Google's indexing of the spam might have been it :)
17:49 frettled mmm :)
17:49 frettled Juerd: good point.
17:49 Juerd The IPs in the server-status screenshot are googlebots
17:49 frettled aha!
17:50 Juerd s/plural/singular/
17:50 frettled excuse me while I edit reality.
17:50 Juerd Getting that trac timeline page took 50 seconds when I tried it myself
17:50 Juerd So that, times a few, lead to server death
17:51 Juerd Googlebot isn't always very friendly. It has been crawling vortaro.nl for more than a year, averaging at 1 GB per month
17:51 Juerd (More, if you count headers, but Apache doesn't)
17:51 ignacio_ joined #perl6
17:51 frettled Yahoo! is worse.
17:51 * Juerd likes to think that Googlebot is in love with vortaro.nl
17:51 Juerd It just can't get enough
17:51 frettled I've blocked access to our webservers (we're a web hosting provider) from Yahoo!'s /16.
17:52 frettled Juerd: hee-hee
17:52 frettled But Google's bots usually don't come in large hunting packs.
17:53 Juerd Well, it initiates a new request once every so many seconds, usually
17:53 frettled mm
17:53 frettled Yahoo! have no such petty concerns.
17:53 Juerd But apparently that interval is not increased when the responses take very long.
17:53 frettled okay, that's hardly ideal.
17:54 Juerd At least, that's what I think was happening to feather
17:54 Juerd In fact, perhaps the simple fix could be a robots.txt
17:54 Juerd (Although Googlebot would probably kill the server again before checking robots.txt)
17:55 frettled I think they check for it every time.
17:55 frettled So if robots.txt blocks for Trac, that might help a bit.
17:56 Juerd What makes a "time"?
17:56 frettled Every time a bot starts indexing.
17:57 Juerd When do you reckon that will be on dev.pugscode.org?
17:58 frettled I have no clue :D
17:59 Juerd :)
18:01 frettled Ooh, masakbot is submitting again!
18:04 bluescreen joined #perl6
18:16 |JackYF| joined #perl6
18:21 snearch_ joined #perl6
18:26 soupdragon joined #perl6
18:26 Chillance joined #perl6
18:26 ash_ joined #perl6
18:29 nihiliad joined #perl6
18:30 ash_ colomon: so... i think the issue is the List.new doesn't assign a value to $.values
18:31 ash_ ng: my $a = List.new(values => (1, 2, 3));
18:31 p6eval ng b65331:  ( no output )
18:31 ash_ ng: my $a = List.new(values => (1, 2, 3)); say $a.elems;
18:31 p6eval ng b65331: get_pmc_keyed() not implemented in class 'List'␤current instr.: 'perl6;List;!generate' pc 11209 (src/builtins/Str.pir:187)␤
18:31 ash_ gives you a whole new error
18:33 ash_ ng: my $a = List.new(values => 'a'); say $a.elems;
18:33 p6eval ng b65331: Method 'succ' not found for invocant of class ''␤current instr.: 'perl6;Mu;' pc -1 ((unknown file):-1)␤
18:34 ash_ so, List needs a new method to initialize all of its internal variables properly, what properly is, i am not sure yet
18:34 ash_ values, i think, needs to be a ResizablePMCArray but I am not sure
18:39 stephenlb joined #perl6
18:41 jnthn ash_: Sounds reasonable
18:41 jnthn ash_: Maybe try like
18:41 jnthn $P0 = new ['ResizablePMCArray']
18:42 jnthn setattribute self, '$!values', $P0
18:42 jnthn Somewhere
18:42 jnthn brb, store
18:42 ash_ in src/core/List.pm i tried adding a method new(*@args) there but it's not being called
18:46 Su-Shee_ joined #perl6
18:50 snearch_ joined #perl6
18:51 colomon joined #perl6
18:52 snearch_ joined #perl6
18:52 TimToady .oO(backlog, or fix my mailserver...choices...choices...)
18:53 mj41 joined #perl6
18:54 simcop2387_ joined #perl6
18:54 jnthn TimToady: Fixing your mail server gives you more things to read at the end of it. Backlogging gives you less things to read at the end of it. :-)
18:55 ash_ jnthn: where would i do that? Should i add a BUILD or a new to List? also, should I do it in the .pir or .pm?
18:56 jnthn ash_: BUILD may be a good place.
18:56 jnthn Provided list doesn't have a PIR new method that's getting called, anyways.
18:56 ash_ no, .new is falling back to Mu's
18:58 jnthn OK, that should wind up calling a BUILD you write then.
19:00 ash_ but List.new is supposed to be able to take a bunch of arguements to make the list, so you can do List.new(1, 2, 3); doesn't that mean i'd need to do a .new also?
19:01 jnthn BUILD is passed that argument.
19:01 riffraff joined #perl6
19:01 ive joined #perl6
19:02 jnthn ash_: Take a look at some of the tests for a better idea of BUILD.
19:02 ash_ kk, i still don't know when to use new vs BUILD
19:04 jnthn new = needs to actually manufacture an instance of the object
19:04 jnthn BUILD = does the init work for this particular class assuming we already have an object.
19:09 ash_ well that fixes 1 issue... sorta, now i get:  get_pmc_keyed() not implemented in class 'Array' in List;!generate
19:10 rgrau` joined #perl6
19:15 simcop2387 joined #perl6
19:17 ash_ oh, got it
19:19 supernovus joined #perl6
19:20 ash_ hm, no, it not raising an error but its not working right
19:29 KyleHa joined #perl6
19:31 lisppaste3 supernovus pasted "rakudo master build error" at http://paste.lisp.org/display/93395
19:32 ruoso joined #perl6
19:32 ash_ supernovus what version of parrot are you using?
19:32 supernovus 1.9.0-devel
19:32 supernovus I haven't been able to build rakudo in a while.
19:33 ash_ on ubuntu?
19:33 supernovus I tried the 'ng' branch and it died as well in a similar manner.
19:33 ash_ or linux?
19:33 supernovus Ubuntu 9.10 64bit
19:35 supernovus blizkost dies on me too, haven't tried any other parrot languages.
19:35 ash_ yeah, i am not sure exactly what the problem is, are you using trunk parrot? or PARROT_REVISION parrot?
19:36 supernovus trunk
19:36 ash_ can you try with PARROT_REVISION trunk?
19:36 ash_ its in build/PARROT_REVISION
19:37 ash_ or do perl Configure.pl --gen-parrot and it automatically does it
19:41 ash_ brb, supernovus if your still here when i get back i'll see if i can help you figure out whats wrong
19:54 dakkar joined #perl6
19:55 TimToady email back up, can backlog in peese
19:58 ms16004 joined #perl6
20:03 IllvilJa joined #perl6
20:14 Chillance joined #perl6
20:14 ash_ joined #perl6
20:15 supernovus Well, rakudo built fine with the PARROT_REVISION version of parrot. So, it's parrot 'trunk' that is currently broken. blizkost still bails horridly, but I think it's been broken for a while.
20:18 ash_ blizkost hasn't been updated in a while
20:19 ash_ thats good PARROT_REVISION built for you, i think masak has been notified of the trunk one
20:19 ignacio_ joined #perl6
20:22 ash_ well, i think there is a ticket in the bug tracker about it, i don't know if its all linux or just ubuntu but so far, someone else using ubuntu 8.10 had problems building it too, i am on os x and it builds fine for me with trunk parrot, gcc 4.2, might be gcc related, i am not sure
20:23 fglock joined #perl6
20:25 fglock hi
20:25 PerlJam greetings
20:26 jonasbn joined #perl6
20:26 supernovus ash_: using gcc 4.4.1 over here, so it may be a gcc related issue.
20:27 fglock some news in miniperl6,
20:27 fglock $ perl mp6.pl -v -Bgo util/mp6.pl
20:28 fglock $ ./6.out util/mp6.pl
20:28 colomon ash_: apologies for disappearing on you there.
20:28 ash_ no worries
20:28 fglock the Go backend now can compile mp6 source
20:29 ash_ wait, so you can have miniperl6 with a Go backend?
20:29 ash_ doesn't Go use the gcc as a backend?
20:29 fglock not bootstrapped yet, needs writing some tools
20:30 ash_ thats cool
20:30 fglock $ ./6.out -Clisp util/mp6.pl    # compile mp6 to lisp, using Go
20:31 fglock the speed is about the same as the perl5 backend
20:32 PerlJam fglock: um ... how fast is the perl5 backend?
20:33 fglock I can only compare between backends, Perl5 is 1x, Go is 1x, Javascript is 3x, Lisp is 5x
20:34 fglock Parrot only runs 'hello, World!' for now
20:35 simcop2387_ joined #perl6
20:35 fglock oh, and today I also tested with Java, through Rhino:
20:35 fglock $ perl mp6.pl -Brhino -e ' say "hello, World!" '
20:36 fglock checking Rhino speed now
20:36 fglock hmm - not so easy, needs some scripting
20:37 fglock will do later
20:38 PerlJam fglock: having a comparison between parrot and perl5 and others would be meaningful
20:39 PerlJam without the parrot metric, I can't get too excited about "as fast as the perl 5 backend"  :)
20:39 fglock actually, Rakudo should be able to run mp6 directly
20:42 fglock $ ./perl6 ~/projects/perlito/lib/Test.pm   # no errors!
20:46 fglock (updating rakudo)
20:52 ignacio_ joined #perl6
20:54 meppl joined #perl6
20:56 arthur-_ joined #perl6
20:56 mikehh joined #perl6
20:59 meppel joined #perl6
21:00 fglock how can I set @*INC for rakudo?
21:02 fglock PERL6INC and -I didn't work
21:02 ash_ fglock: are you in ng?
21:02 supernovus A rather obscure question... I can find the hex unicode representation of a character using "æ".ord.fmt("%04X");  Is there a way to return the character description that you can use in \c[]  (in this case "LATIN SMALL LETTER AE") ?
21:02 fglock no, plain rakudo
21:03 arthur-_ joined #perl6
21:04 cotto_w0rk joined #perl6
21:07 TimToady supernovus: not that I know; but check out qx/perl -e 'print do "unicore/Name.pl"'/ maybe
21:08 TimToady rakudo: say qx/perl -e 'print do "unicore/Name.pl"'/;  # probable fail
21:08 p6eval rakudo 3867ff: Confused at line 2, near "Name.pl\"'/"␤in Main (file <unknown>, line <unknown>)␤
21:09 TimToady eh?
21:09 TimToady rakudo: say qx[perl -e 'print do "unicore/Name.pl"'];  # probable fail
21:09 p6eval rakudo 3867ff: operation not permitted in safe mode␤in Main (file /home/p6eval//p1/lib/parrot/1.9.0-​devel/languages/perl6/lib/Safe.pm, line 24)␤
21:10 cotto_working joined #perl6
21:10 TimToady works locally though
21:11 riffraff joined #perl6
21:11 Tene fglock: PERL6LIB
21:11 Tene supernovus: you'll need to write some code that talks to libicu
21:11 ash_ fglock: in master rakudo it checks your ~/.perl6/lib folder and the $perl6install/lib folder by default i know, i think it checks an enviorment variable dont' know which
21:12 ash_ but Tene++ knows apparently
21:12 fglock Tene: thanks!
21:42 japhb joined #perl6
21:43 fglock I've fixed a couple of problems, but mp6 doesn't run in rakudo yet
21:43 fglock sleep &
21:43 fglock left #perl6
21:46 patspam joined #perl6
22:00 lisppaste3 joined #perl6
22:01 pnate joined #perl6
22:10 GeJ Good morning everyone.
22:12 tylerni8 joined #perl6
22:26 iblechbot joined #perl6
22:31 zamolxes so were those actual search engine crawlers, or were they spammers?
22:48 diakopter zamolxes: ?
22:49 diakopter oh, on feather/trac?
22:59 jferrero joined #perl6
22:59 like joined #perl6
23:06 zamolxes cause yep
23:06 zamolxes yep
23:09 Juerd joined #perl6
23:10 zamolxes so, one important thing is adding %D to the format log. Then you can do stuff like : what types of requests are most expensive
23:10 PerlPilot joined #perl6
23:10 zamolxes frettled: o/
23:12 zamolxes and if I can help with the varnish part, i'm here, just let me know
23:15 jferrero joined #perl6
23:29 hicx174 joined #perl6
23:29 ash_ joined #perl6
23:30 Psyche^ joined #perl6
23:39 ash_ isn't BUILD supposed to be passed all of the arguments from new?
23:45 rjh joined #perl6
23:47 ash_ ng's build isn't functioning properly
23:47 diakopter ng: ''
23:47 p6eval ng b65331: sh: ./perl6: No such file or directory␤
23:51 frettled zamolxes: Hey!  That's great!
23:51 diakopter ng: ''
23:51 p6eval ng b65331:  ( no output )
23:52 frettled zamolxes: if you're around here sometime during the day, I'm sure Juerd is here, too, and we can talk a bit more about it, okay?  (I'm just popping by because of sleeplessness now)

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

Perl 6 | Reference Documentation | Rakudo