»ö« | perl6.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot usage: 'perl6: say 3;' or rakudo: / pugs: / std: , or /msg p6eval perl6: ... | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by moderator on 20 October 2009.
moritz_ I thought more about *why* it's not faster (and not that easy, for that matter) 00:01
00:03 IllvilJa joined
moritz_ nqp: try { say 1 } 00:03
p6eval nqp: Confused at line 1, near "try { say "␤current instr.: 'parrot;HLL;Grammar;panic' pc 519 (src/cheats/hll-grammar.pir:197)␤
jnthn ENOLISTOPS 00:04
ng_feed rakudo-ng: jnthn++
rakudo-ng: First cut at getting placeholder parameters in place. Seems to handle various cases I've tried.
jnthn (but sadly, don't know that we immediately win any more tests...) 00:05
chromatic I know I've written about that too, but mostly it comes down to marshalling and demarshalling. 00:06
00:24 charsbar joined 00:26 johnjohn101 joined
johnjohn101 so i hear that fedora 12 will have perl 6. 00:27
Tene Fedora 11 did too. 00:28
yum install rakudo 00:29
Fedora 12 *currently* has rakudo, btw.
F12 has been released. 00:30
00:31 ihrd left
johnjohn101 that's pretty easy. d 00:32
do you think it's that easy for ubuntu?
i guess I could try it out
Tene It will be, once ubuntu gets rakudo packages. 00:33
afaik, ubuntu currently does not ship Rakudo.
johnjohn101 when might that be?
Tene That's up to ubuntu, last I heard.
johnjohn101 is rakudo stable enough for real world? 00:35
Tene Depends on what exactly you're asking. For many uses, no. For some uses, yes.
I use it regularly. 00:36
The current plan is to release a Recommended For Use version in, I think March? Maybe it was May? Spring sometime.
That's likely to be much more suitable.
johnjohn101 so early to mid next year. 00:37
chromatic April
Tene So I was right on average. ;)
johnjohn101 will that be the official perl 6?
Tene johnjohn101: any implementation that passes the test suite is Perl 6. Perl 6 is defined by its test suite. 00:38
Rakudo is the only real candidate these days, though, so the answer to what you probably mean is probably "yes". 00:39
johnjohn101 not that i'm having any problems with perl 5. I love it.
Tene So do most of the people working on Perl 6.
TimToady I think Perl 5 was designed by an idiot, but I'm in the minority. :) 00:40
johnjohn101 it has some limitations. I'm getting a lot out of it for work. 00:43
perigrin TimToady: we could probably get a movement started behind that if you really *want* ... but I personally really like it. 00:46
TimToady it's not that bad, really, but still much more like the Wild West with free-range ranchers shooting at each other and at the farmers 00:48
japhb FWIW, I've heard that Tchaikovsky hated the Nutcracker Suite, which just goes to show you can never trust an artist's view of the work ... 00:49
TimToady or maybe it's a more modern era, where people think they ought to keep shooting wolves, and I'm trying to restore the ecosystem. :) 00:51
ah well, metaphors are cheap, and the really good ones are often spectactularly wrong. 00:52
japhb A human restoring an ecosystem is a dubious proposition at best. And I speak as the son of a man who spent his career restoring ecosystems.
TimToady Sure, but I've found that when I go a direction, I'm not usually all by myself. :) 00:53
japhb But clearly, a quality metaphor trumps any underlying facts. Truthiness is all we're really looking for.
A pack of hungry ecosystem restorers?
quietfanatic Perl 5: "I am your father!" Perl 6: "Noooooo!" 00:54
TimToady sounds like you've got the idea; carry on... :)
japhb quietfanatic, quite.
diakopter "that's not true! that's impossible!!!! [and you just severed my hand!]"
TimToady quitefanatic, quiet. 00:55
japhb Do we get to have a light saber battle now? Because here I find myself without a good blaster to play the part of the roguish ... er ... rogue. 00:56
01:07 quietfanatic joined
sjohnson heh 01:15
jnthn Ooh...I have a flight tomorrow. I should probably go to bed at a sensible-ish time. Night all! 01:17
01:18 agentzh joined
sjohnson cya 01:18
01:20 frew joined, Wolfman2000 joined 01:21 freenose joined 01:22 freenose left
slavik define "sensible" 01:23
we don't have "sensible"
01:32 awwaiid joined 01:38 pointme joined, zaslon joined 01:57 rfordinal joined 01:58 ihrd1 joined, mikehh joined 02:00 nihiliad joined
lisppaste3 mikehh pasted "rakudo fails to build with g++" at paste.lisp.org/display/90706 02:03
arnsholt mikehh: Why are you trying to build Rakudo with g++? 02:06
mikehh arnshoit: better diagnostics 02:07
arnshoit: stricter
arnsholt Well, g++ is stricter because of C++ 02:08
The conversion that makes g++ quit is perfectly legal C
mikehh arnshoit: and if I build parrot with g++ it automatically uses the same compiler top build rakudo
arnsholt Although gcc does emit a warning 02:09
Hang on, you're building Rakudo against a Parrot you've built yourself? 02:10
mikehh arnshoit: I mainly work on building and testing parrot - I've done at least four builds and tests - up to fulltest with gcc and g++ I also test rakudo and partcl and others onagainst the build 02:18
arnsholt Fair 'nuff 02:19
I think the easiest way to fix it is just to set CC by hand in the Makefile and live with it for now 02:20
mikehh It builds ok with parrot built with gcc but as I pasted it fails with parrot build with g++ - I could probably fix it 02:23
arnsholt Yeah, it's just a matter of adding explicit casts 02:24
mikehh right - although #include parrot/extend.h is needed in perl6.ops 02:25
02:26 ihrd1 left 02:36 lestrrat joined 02:45 kst joined 03:02 colomon joined
colomon phenny: tell sjohnson TCC LE used to be a commercial product, and I was a paying customer for 15 years or so. (First it was called 4DOS, then 4NT, now TCC LE.) Now they are focusing on fancier windowing stuff and giving away the command shell for free. 03:06
phenny colomon: I'll pass that on when sjohnson is around.
sjohnson thanks phenny 03:21
phenny sjohnson: 03:06Z <colomon> tell sjohnson TCC LE used to be a commercial product, and I was a paying customer for 15 years or so. (First it was called 4DOS, then 4NT, now TCC LE.) Now they are focusing on fancier windowing stuff and giving away the command shell for free.
sjohnson ;)
colomon: is there something i am missing? a "by hand" resizing thing for the black console window would be like... the best thing ever
maybe i'm just not doing something smart... i know cygwin's use of rxvt avoids this problem which is kinda cool
03:25 JimmyZ_ joined
sjohnson perhaps it's just unpossible in windows? 03:28
03:41 vamped joined 03:45 colomon_ joined 03:47 envi^office joined 03:53 colomon_ joined 03:58 colomon_ joined 03:59 nihiliad joined
diakopter sjohnson: Console2 04:07
sjohnson: sourceforge.net/projects/console/ 04:10
04:11 colomon joined
diakopter hmm, I seem to recall it having more features than this 04:12
04:12 colomon_ joined 04:14 colomon_ joined 04:30 dduncan joined, meppl joined
pugs_svn r29129 | lwall++ | [S04] as several folks have suggested, rename "blorst" to "blast" 04:34
04:37 dduncan joined
pugs_svn r29130 | lwall++ | [gimme5] force $<sym> to always be string for now 04:40
r29130 | [STD] rename 'blorst' to 'blast' so we can have phaser blasts
r29130 | parse +/- as part of complex numbers
r29130 | refactor infixish to accept [op]= and <<op>>=
r29130 | rename some terms to values
r29130 | don't parse colon as rad_number unless followed by \\d
TimToady std: :
p6eval std 29128: [31m===[0mSORRY![31m===[0m␤Illegal use of colon as invocant marker at /tmp/9Es3kMoqBT line 1:␤------> [32m:[33m⏏[31m<EOL>[0m␤FAILED 00:01 101m␤
TimToady std: say :
p6eval std 29128: [31m===[0mSORRY![31m===[0m␤Preceding context expects a term, but found infix : instead at /tmp/NamLvcywQh line 1:␤------> [32msay :[33m⏏[31m<EOL>[0m␤FAILED 00:01 103m␤
04:49 dduncan joined
diakopter std: say: 04:49
p6eval std 29128: ok 00:01 95m␤
diakopter std: :say
p6eval std 29128: ok 00:01 99m␤
04:51 phio joined
diakopter std: http//puredanger.com/tech/2009/11/18/closures-after-all 04:51
p6eval std 29130: Undeclared routines:␤ 'closures-after-all' used at line 1␤ 'http' used at line 1␤ 'puredanger' used at line 1␤ 'tech' used at line 1␤ok 00:01 100m␤
diakopter I mean.
puredanger.com/tech/2009/11/18/clos...after-all/
04:53 kst joined, Khisanth left 04:54 Khisanth joined, phio left 04:57 frew_ joined 04:58 meppel joined
pmichaud phenny: tell masak I plan to be updating spectest-results and announcements around 13h00 utc 05:12
phenny pmichaud: I'll pass that on when masak is around.
05:14 carltonf joined, justatheory joined
diakopter all: I *highly* recommend watching this. ecn.channel9.msdn.com/o9/ch9/2/7/8/...ok_ch9.mp4 *very* interesting/exciting. 05:49
several mentions of Perl. even.
06:08 masak joined
masak good morning, #perl6. 06:08
phenny masak: 05:12Z <pmichaud> tell masak I plan to be updating spectest-results and announcements around 13h00 utc
masak pmichaud: ok, that means I should probably release in the evening after all. 06:09
06:12 agentzh joined
Tene hi masak 06:15
masak o/
06:18 pnate joined
masak phenny: tell moritz_ that I believe Mark Dominus wrote a "(implementing Perl in) C is not the answer" article, and that I've once read it online. with a bit of luck, I might find it, too. 06:19
phenny masak: I'll pass that on when moritz_ is around.
06:23 kst joined, colomon joined 06:25 colomon_ joined
masak phenny: tell moritz_ www.perl.com/pub/a/2001/06/27/ctoperl.html 06:26
phenny masak: I'll pass that on when moritz_ is around.
06:27 JimmyZ_ joined 06:31 gfx joined 06:33 colomon_ joined 06:50 JimmyZ__ joined 06:55 colomon_ joined 07:04 quietfanatic joined, dduncan left 07:05 gabiruh joined 07:07 am0c joined 07:10 JimmyZ_ joined 07:11 kaare joined 07:16 kaare joined, zloyrusskiy joined 07:17 kaare_ joined 07:19 Su-Shee joined
Su-Shee good morning 07:19
masak o/
in searching for Mark Dominus's Perl-to-C article, I stumbled over this one: www.perl.com/pub/a/2000/11/perl6rfc.html 07:20
now, I wasn't involved in Perl 6 at the time, but I found it interesting for historical reasons.
it's interesting, for example, to feel the vibes in that article that Perl 5 is translatable, and should be translated to Perl 6. 07:21
even in 2004/2005 when I jumped on the ship, I kept hearing fragments of this assertion.
today, I don't find it likely that people will make such an automatic migration on a large scale. 07:22
perigrin masak: you didn't have a talk opposite sky's PONIE talk at YAPC::EU 2003 :)
masak no, I most certainly did not. :)
what was that about? could you fill me in? 07:23
"I always said (with considerable regret) that I did not think it was useful, but that Larry might yet prove me wrong and salvage something worthwhile from the whole mess."
that sentence certainly came true.
TimToady++
perigrin I was giving a talk opposite it ... I didn't see it :) 07:24
however PONIE was the/an attempt to run Perl5 on Parrot
masak oh, ah.
I believe we'll make another attempt at that soon enough.
and by 'we', I mean 'someone else'. :)
perigrin (my first talk ever actually) 07:25
I wish them luck with that.
masak would be fun to find this sky and extract PONIE experiences from him. 07:26
perigrin sky == Artur Bergman
also Nicholas was involved at one point 07:27
masak ok.
perigrin if you google I'm sure you'll find documentation
masak oh, I recognize sky. he looks familiar.
moritz_ good morning 07:29
phenny moritz_: 06:19Z <masak> tell moritz_ that I believe Mark Dominus wrote a "(implementing Perl in) C is not the answer" article, and that I've once read it online. with a bit of luck, I might find it, too.
moritz_: 06:26Z <masak> tell moritz_ www.perl.com/pub/a/2001/06/27/ctoperl.html
masak mornin'! :)
07:35 TiMBuS joined 07:42 colomon_ joined
JimmyZ_ Good afternoon :) 07:44
07:44 colomon joined
moritz_ masak: thanks for the link 07:45
masak np.
JimmyZ_: 下午好! 07:46
JimmyZ_ masak: 你也好 07:47
masak ^^
07:47 ejs joined
masak 我真不好. I had to go look up 午 on Google Translate because I had forgotten it. :/ 07:48
07:50 Su-Shee left
JimmyZ_ masak: 临时性健忘，没有关系的 07:50
masak you're too kind. :) 07:51
no, the truth is that my 中文 is bit-rotting. 07:52
cls_bsd amazing I can read it w/o Google Translate :D 07:53
masak cls_bsd++
I'll try to remember that 性健忘 means 'amnesia'. I might need that a lot when talking to Chinese. :) 07:54
JimmyZ_ masak: 临时性-健忘 07:55
masak JimmyZ_: only if I re-learn what I've forgotten. :)
07:56 chromatic joined, ejs joined, kst joined
masak '对不起... 我有一个临时性健忘...' 07:57
07:59 barney joined
JimmyZ_ masak: learning, like eating Sugarcane, that will increase your memory 07:59
masak: I don't know whether I describe it correctly. ;) 08:00
masak aye. some of the people I admire most have kept learning all through their lives.
ng_feed rakudo-ng: moritz++
rakudo-ng: one more passing test file
08:00 iblechbot joined 08:01 robert___ joined
JimmyZ_ masak: review. ;) 08:01
masak I will, I will. thanks for being the cane and the carrot. 08:02
08:02 vamped joined
masak 复习了复习... :) 08:03
moritz_ cane and carrot? :-) 08:04
JimmyZ_ masak: 我们把复习比喻为吃甘蔗学习法
masak moritz_: JimmyZ_ mentioned 'sugar canes'. I believe the original expression is 'the stick and the carrot' or equiv.
JimmyZ_: oh. 08:05
moritz_ masak: ah. In German that's (translated back) "Sugar-bread and whip" :-)
masak JimmyZ_: 'reviewing is like eating sugar cane'? 08:06
moritz_: we say 'carrot and whip' in Sweden. :)
JimmyZ_ masak: sugarcne is a kind of Plant. 08:07
masak nod.
JimmyZ_ masak: it is images.google.com/images?q=%E7%94%9...amp;tab=wi
masak nodnod. 08:08
JimmyZ_ masak: not like, liken 08:09
masak JimmyZ_: 'we liken learning to eating sugar cane'? 08:10
JimmyZ_ masak: something like that, just say metaphor, analogy or figure of speech in english, I think. 08:12
masak I don't think I know what it's like to eat a sugar cane.
I imagine it's very sweet.
JimmyZ_ masak: we study a chapter of the article and after somedays we review it and then study the next chapter. 08:14
08:14 colomon_ joined
beggars only that no on can really eat sugar cane 08:14
masak JimmyZ_: sounds pretty standard.
08:15 kst joined
masak JimmyZ_: is the point that you have to eat the sugar cane in small pieces? 08:15
JimmyZ_: or that you eat the same cane twice? :)
JimmyZ_ masak: we also eat the sugar cane and enjoy its sweet and then eat the next piece. 08:16
masak 'its sweetness'. I see.
JimmyZ_ yes
masak I will re-learn my missing 汉字, and I will savour their sweetness! :) 08:17
08:17 rfordinal joined
JimmyZ_ masak++ 08:17
and question time. github.com/rakudo/rakudo/blob/ng/sr...num.pm#L12
masak :) 08:18
JimmyZ_ pmichaud, it needs no box__PN, I think.
masak ah, an ng question.
JimmyZ_: that's not a question. :P
JimmyZ_ just a advice. 08:19
08:19 zamolxes joined
moritz_ why not? returning a parrot int is likely to blow up all kind of things 08:19
JimmyZ_ or a suggestion?
08:19 kst joined
masak JimmyZ_: 'just advice', or 'just a suggestion'. 08:19
JimmyZ_: what moritz_ said. it seems reasonable to box things.
JimmyZ_: all HLLs will need to do that with Parrot-native values. 08:20
08:20 mariuz joined
JimmyZ_ pir::ceil__IN does autoboxing, IIRC. 08:20
masak wow, Any-num.pm is 太美了!
masak looks forward to ng 08:21
moritz_ JimmyZ_: if it would, you couldn't call box__PN on it
JimmyZ_ moritz_: why github.com/rakudo/rakudo/blob/ng/sr...-num.pm#L4 didn't use box?
moritz_ JimmyZ_: probably because the prefix:<~> does that 08:22
masak sorry, the what? 08:23
Any.abs doesn't use prefix:<~>...
08:24 colomon joined
JimmyZ_ moritz_, pir::op does the box op, use pir:box means that it does the box twice. 08:25
IIRC, it somebody make the misstake. and pmichaud pointed out it. 08:26
s/make/makes/
oh
maked
masak made. :) 08:27
JimmyZ_ oh :( why I had been using it for ten years along and it's still wrong.
masak English is a bitch. 08:28
doesn't have thousands of individual symbols that you have to learn, though.
JimmyZ_ ;) 08:32
08:46 rgrau joined
mathw Morning 08:46
masak o/ 08:47
chris.prather.org/what-stops-me-fro...today.html # a nice read
my favourite sentence: "While Perl6 may have more polish than I expected, and is a project that is *well* on it's [sic] way to being complete, it lacks some fundamental things I love about Perl5."
spinclad like all of CPAN... 08:48
masak that's his next sentence. :) 08:49
mathw o/ masak
the interesting thing with these people is seeing if there's anything they miss that we don't know about 08:50
08:50 kangas_ joined
mathw And I'm delighted to read that particular article. It's also making me think I really need to learn Moose... 08:51
masak me too. 08:54
and steal all the introspection for Perl 6 :)
mathw hee hee 08:55
just wish I could use Moose at work
there are people there who could really benefit from it
but our servers don't have Moose on them
and getting a module from CPAN approved and installed is worth more effort than I want to expend on a job I'm hoping to leave soon 08:56
08:56 ejs1 joined 08:57 Su-Shee joined 09:00 scottp joined 09:01 gfx joined
scottp My google has failed me... I have tried fresh Rakudo #22 on Mac OSX And get: src/string/api.c:2187: error: 'LONG_MAX 09:01
mathw Hello scottp 09:02
Is there any more of that error message?
moritz_ scottp: while running perl Configure.pl --gen-parrot?
09:04 rlarson87 joined 09:31 rfordinal left
scottp moritz_: yes - running "perl Configure.pl --gen-parrot" after a fresh checkout to be sure. 09:33
Ahhh... here is something that may help: /usr/include/secure/_stdio.h:53:1: warning: this is the location of the previous definition 09:34
moritz_ scottp: that's an error I've never seen before, so I'm afraid I can't help. You could try asking in #parrot on irc.perl.org though 09:36
scottp I had one other idea. The version of perl may be causing an issue. I seem to have two.
Ta will do
Compiling now with the other perl - will let you know. Thanks. 09:38
Su-Shee moritz_: did you by any chance read the recent iX hype about "functional and oo hybrid languages" and how suddenly language which can do both become fashionable again? ;) you should consider beating the same bush in your p6 article. 09:40
moritz_ Su-Shee: too late, it's already in print
Su-Shee: but I'll remember it for my next one, thanks :-)
scottp YES IT WORKED.... OK One we need to add to the FAQ. Lots of Mac apps install other versions of perl, and that was my prob 09:42
Su-Shee and don't forget the comparison how "go" has only interface and mention roles. ;)
moritz_ scottp: so which one was successful? 09:43
scottp the standard mac one
The /opt - which for some reason was default in my path version was wrong. My guess is it is due to compile time options, which Parrot uses to work out which C and libraries to use.
i.e. the /opt perl might have libraries/headers etc that are no longer available 09:44
perigrin is there a way to test after teh fact which perl you built against?
moritz_ which is wrong, and being worked on
parrot_config perl 09:45
rakudo: say %*VM<config><perl>
p6eval rakudo 7347ec: /usr/bin/perl␤
perigrin heh
'perl'
but I suspect that's my 5.10 09:46
scottp horrah all tests passed
moritz_ YaY
scottp where is the right place to document the problem/solution for others to find/
moritz_ probably in the README 09:47
scottp: you can submit a README patch to rakudobug@perl.org 09:50
masak I'll go very soon (lunch+packing+flight), but before that, does anyone have any suggestion for the name of today's release? 09:53
out of the suggestions in the release guide, the only one I have a relation to is Lisbon.pm
moritz_ well, that's a start, isn't it? 09:54
and...
don't you go to some kind of perl meeting?
is that organized by a .pm group?
masak yes, but Stockholm.pm has already been a release.
moritz_ then take Lisbon 09:56
masak goodie. 09:57
will switch countries, then do a release.
see you.
moritz_ bye, good flight
jnthn o/ 10:01
jnthn here for a little bit before going to the airport.
ooh...I just missed masak!
vamped how does one view .pod documents? gedit doesn't provide any formatting. 10:03
moritz_ vamped: perldoc $file.pod
vamped moritz: thx 10:04
10:05 colomon_ joined 10:06 Arf joined 10:14 colomon_ joined
jnthn blast! We need to update our grammar! 10:16
10:16 colomon joined 10:28 ejs2 joined 10:30 sundar joined, colomon joined 10:32 dakkar joined 10:35 colomon_ joined
jnthn airport time! \\o/ 10:42
zaslon loljnthnhazblogged! jnthn++ 'Progress and minor frustration': use.perl.org/~JonathanWorthington/j...7?from=rss
10:49 pmurias joined 10:53 colomon joined 10:59 colomon_ joined
pugs_svn r29131 | pmurias++ | [mildew-j] improve error messages for a missing method, 11:01
r29131 | CORE/{Signature.pm,Multi.pm} are loaded my mildew-js
11:10 vamped left 11:18 payload joined 11:21 colomon_ joined 11:23 colomon_ joined 11:36 justatheory joined 11:48 cognominal joined
pmichaud good morning, #perl6 11:54
colomon \\o 11:57
11:59 SmokeMachine joined 12:08 payload joined 12:09 envi^home joined
carlin "I had to find out what bloody Twitter was. I thought it was a new guy playing five-eighth for England" - All Blacks Coach Graham Henry is bewildered by modern life. 12:16
dalek kudo: b5d7e51 | pmichaud++ | docs/ChangeLog:
Update ChangeLog for release.
 12:26
kudo: 6e6a5ac | pmichaud++ | docs/announce/2009-11:
Add 2009-11 announcement draft.
kudo: 40b555d | pmichaud++ | docs/ChangeLog:
Oops, named argument binding to positional parameters was already done.
kudo: b1a2db7 | pmichaud++ | docs/spectest-progress.csv:
spectest-progress.csv update: 453 files, 32753 (85.5% of 38318) pass, 5 fail

S02-lexical-conventions/unicode.rakudo aborted 5 test(s) S06-signature/named-parameters.rakudo passed 81 unplanned test(s)		 12:32
12:37 pmurias joined 12:39 payload1 joined 12:45 xinming joined
pmichaud phenny: tell masak I've added a draft announcement for the release -- update it as you see best! thanks! 12:51
phenny pmichaud: I'll pass that on when masak is around.
moritz_ phenny++
pmichaud++
12:53 ejs joined 12:54 ruoso joined
dalek kudo: c5c2aae | pmichaud++ | docs/announce/2009-11:
Update 2009-11 announcement with latest spectest numbers.
 12:56
kudo: 0fd222b | pmichaud++ | build/PARROT_REVISION:
Bump PARROT_REVISION to Parrot's 1.8.0 release.
ruoso bom dia...
13:06 takadonet joined
takadonet greeting all 13:06
moritz_ oh hai
colomon o/ 13:07
pmichaud: do roles work in ngyet? 13:09
pmichaud colomon: there's some support for roles, but I don't know how far along it is yet. 13:11
colomon I've been thinking of Iterator again, and how roles might interact with it. 13:12
pmichaud I'm thinking Iterator is likely a role. 13:14
colomon like having an Iterator role with both get and generate methods, and a default implementation of generate($n) which just calls get repeatedly
pmichaud could be. There's something to be said for allowing .get to take an argument, or have some equivalent of .lines that allows a number 13:15
colomon I like the idea that implementing a really simple get function should be enough to your iterator get up and running. 13:21
pmichaud sure, makes sense
colomon and still have a system which allows smarter behaviors. 13:22
13:25 ejs1 joined
colomon btw, one thing that came up that had me confused. The idea that map is lazy, so it doesn't actually do anything if you don't do anything with the results of the map. 13:26
pmichaud yes, that's correct. (remember that void context is eager, though)
colomon I don't mind map working that way (though it is troublesome for a lot of baby idioms with side effects)...
is eager void context not currently working? 13:27
pmichaud correct, it's not.
colomon ah!
okay, I'm cool with the language design, then.
just need to get the implementation to match. :)
pmichaud I'm trying to decide how to get map to detect that it's in void context 13:28
13:28 Lorn joined, takadonet left
moritz_ pmichaud: that's not map's job 13:28
pmichaud the alternative is to generate an "eager(...)" call around every statement... and I'm not sure I want to do that.
moritz_ that's basically what TimToady suggested
except he called it .Void or so
pmichaud oh, .Void might not be so bad. 13:29
13:33 payload joined
pmichaud afk for a bit 13:35
moritz_ OT question: in bash, what's the wildcard for "everything but A"? 13:36
ah, [^A]* 13:38
13:43 payload1 joined 13:48 jferrero joined
PerlJam greets 13:55
13:55 arthur-_ joined 13:57 mikehh joined
moritz_ hello PerlJam 13:57
carlin thedailywtf.com/Articles/Introducin...fsets.aspx # Brilliant 14:06
14:07 kst joined 14:08 mikehh left
PerlJam masak++ and mst++ posts triggered lots of good responses in the blogosphere 14:09
moritz_ and KyleHa++ has brought it to perlmonks, but the monks don't seem to be interested in a serious discussion 14:10
Wolfman2000 *yawn* morning 14:23
PerlJam: what's the latest discussion on? I've mainly been idling here
14:26 iblechbot joined 14:29 KyleHa joined
KyleHa I tried to run update_test_whatever_thats_called with ng, and it took over my machine for a while--load average spike, all memory consumed. Has this been identified already, or should I try to figure out what happened? 14:30
moritz_ KyleHa: I've experience the same thing... 14:32
KyleHa: I suspect it's a test in an infinite loop 14:33
which makes parrot cosume all memory
so nothing to worry about for now :-)
just start with resource limits set
KyleHa Well, we'd like to know WHICH test, yes?
moritz_ you can easily find out by letting `top' or `htop' running 14:34
and then you can see which parrot process consumes all that memory, and you can also see its command line arguments
KyleHa Yes, I was going to do something like that.
OK, thanks! 14:35
moritz_ you can also merge or rebase (locally) the autounfudge-with-limits branch
which starts each parrot process with a resource limit
I haven't merged it yet because it's an ugly hack that only works on *nix 14:36
14:40 kst joined 14:47 pmurias joined, colomon joined 14:49 rfordinal joined
pugs_svn r29132 | pmurias++ | [mildew-js] t/signature.t,t/signature-named.t 14:49
14:50 perlygatekeeper joined, ejs1 joined
pmurias considers blogging about mildew-js but is concered mildew-js is not blog worthy yet 14:53
moritz_ pmurias: why shouldn't it be? 14:54
PerlJam What is "blog worthy"? What does that mean? 14:55
pmurias PerlJam: worthy of a blog post?
moritz_ I think the question boils down to "do I waste other people's time when I blog about it?" 14:56
pmurias yes
moritz_ usually the answer is "no", because people stop reading very quickly when they are not interested
so you only waste 5 seconds of their time, that's fine IMHO
PerlJam never thinks of it that way. 14:57
If I write something, it's about a topic that I find interesting or useful. If someone else also finds it interesting, great; if not, oh well.
moritz_ the second question is "do I waste my time writing it?"
PerlJam moritz_: aye, that's the more important one IMHO 14:58
moritz_ and that depends on many factors
usually I blog things because otherwise I write the same thing a few times in different discussions
that's why I attempted to write the ulimative answer to "how can I handle Unicode in perl (5)", and "why are LTM and proto regexes so important" and such pieces 14:59
and then there are things I blog about because I want people to know, and by blogging I reach many more people than via IRC 15:00
and I think that's what pmurias is aiming at 15:02
15:02 cdarroch joined
pmurias moritz_: what i'm unsure of is it worth to blog about mildew-js till it can run enough stuff to be interesting 15:03
moritz_ pmurias: so what *can* it run today? 15:04
pmurias v6/re-mildew/TESTS-js - a subset of test mildew can run 15:05
moritz_ so it's so much that you can't summarize it in one line? :-) 15:06
then it's blog-worthy
pmurias it has roles,subs,control exceptions
moritz_ that's blog worthy 15:07
it also seems to have stuff that rakudo doesn't have yet
like t/caller_my_pseudo_packages.t 15:08
and t/throw_inside_catch.t
Wolfman2000 ...back from breakfast. pmurias: If I'm understanding you...there is a Javascript implementation of Perl 6 that is farther along than Rakudo/
pmurias t/throw_inside_catch.t is incorrect
Wolfman2000 ?*
lambdabot Maybe you meant: . ? @ v
moritz_ Wolfman2000: it's only further along in some specific areas
pmurias Wolfman2000: Rakudo is much farther
15:09 kst joined
Wolfman2000 ...right, forgot about this thought. 15:10
15:10 Psyche^ joined
Wolfman2000 rakudo: my $tmp = any(1, 3); say $tmp; 15:10
p6eval rakudo 7347ec: any(1, 3)␤
Wolfman2000 rakudo: my $tmp = any(1, 3); say $tmp[0]; 15:11
p6eval rakudo 7347ec: Method 'postcircumfix:[ ]' not found for invocant of class 'Int'␤in Main (file src/gen_setting.pm, line 324)␤
Wolfman2000 ...right. I need to figure out how to get something out of an any
pmurias rakudo: say any(1,3).pick
p6eval rakudo 7347ec: maximum recursion depth exceeded␤in Main (file <unknown>, line <unknown>)␤
Wolfman2000 rakudo: say any(1,3).pick(1) 15:12
p6eval rakudo 7347ec: maximum recursion depth exceeded␤in Main (file <unknown>, line <unknown>)␤
Wolfman2000 Alright, that didn't quite work
15:12 pnate joined
pmurias moritz_: i'll blog about mildew-js once i get multis working 15:13
Wolfman2000 multis... :)
moritz_, TimToady, et al: you really got me liking the multis
15:24 payload joined 15:26 ejs joined 15:27 phio joined, nihiliad joined 15:32 KyleHa joined 15:33 phio left 15:34 mberends joined
KyleHa That infinite loop we talked about earlier seems to be in sub-ref.t, the RT #63974 test, when Test.pm tries to @topic_array.perl 15:40
moritz_ ouch 15:44
15:45 phio_ joined 15:46 rfordinal joined 15:52 am0c joined 15:57 gfldex joined
pmichaud re-good morning, #perl6 16:00
KyleHa Howdy, pmichaud! 16:01
mathw You know you're in trouble in a job interview when the interviewer says 'I really don't know who comes up with these questions to ask. Have a look at this code...' 16:04
And then your eyes get all tangled up in it, because it turns out to be a semi-obfuscated way of making doubly linked lists out of hash references in Perl. 16:05
16:12 phio_ joined
moritz_ :/ 16:13
16:13 kst joined
mathw moritz_: It was a new experience for me. Perl data structures are usually based around lists and hashes mixed together, rather than inventing one's own list 16:18
pnate it sounds stupid to me
were they testing you to see your reaction to it or something? 16:19
mathw pnate: yes, it was, but it's a job interview question
they're often a bit stupid in an attempt to get you to demonstrate knowledge in a constrained environment
There was also a delightfully weird bit of C which ends up just telling you if your stack gets allocated top down or bottom up
moritz_ is that interesting? I mean unless you write malware, of course :-) 16:21
mathw No it's not interesting at all
I told him what it does
moritz_ for a VM maybe
mathw and then said I have no idea why you'd care
And he said he isn't sure either
Clearly not writing his own questions 16:22
pnate hahaha
sounds like it was equally awkward for both parties
moritz_ and he can't tell you "because we write and sell exploits", obviously :-)
mathw lol
pnate haha
mathw Well that's not the job I'm going for...
They probably do have people whose job it is to find exploits though 16:23
16:42 Coke joined
Coke is there a way with straight regex to get balanced {}'d ? (e.g. so {{}} matches but {{} doesn't?) 16:42
16:42 takadonet joined, cspencer joined
moritz_ Coke: with recursion that should work 16:44
cspencer hello, all 16:45
moritz_ rakudo: regex b { '{' <b>? '{' }; say "yes" if '{{}}' ~~ / ^ <b> $ /
Coke moritz_: just found that on perlmonks 179555. Good idea.
p6eval rakudo 7347ec: ( no output )
moritz_ hi cspencer, long time no see
rakudo: regex b { '{' <b>? '}' }; say "yes" if '{{}}' ~~ / ^ <b> $ /
16:45 am0c joined
cspencer moritz_: indeed, six day work weeks took me away from the rakudo world for a while 16:45
p6eval rakudo 7347ec: yes␤
moritz_ cspencer: ouch 16:46
rakudo: regex b { '{' <b>? '}' }; say "yes" if '{{}' ~~ / ^ <b> $ /
p6eval rakudo 7347ec: ( no output )
moritz_ six days is tough
cspencer moritz_: yeah, i'm making up for it now with a three day a week schedule
which leaves me time to look at rakudo again 16:47
i imagine a lot has changed in the last 8 months though, so i best start re-reading all the docs :)
16:47 hercynium joined
moritz_ cspencer: yes, there's currently a branch on the way where we basically rewrite rakudo from scratch 16:48
cspencer: using the new-and-shiny grammar engine which suports proto regexes
mathw it is a very, very cool branch though
cspencer i saw that on planetsix!
mathw everything's much nicer there
mathw out
cspencer it sounds pretty awesome, though i've not had a chance to check it out yet
is that what everyone is working from now?
PerlJam cspencer: mostly. 16:49
moritz_ mostly, yes 16:50
we've had a few small bug fixes in master
16:53 Fancycakes joined 16:54 Fancycakes left, iblechbot joined 16:57 ProperNoun joined
PerlJam There have been 27 commits to master since the creation of the ng branch. There have been 325 commits to the ng branch in the same time frame. 16:59
moritz_ wow, I wasn't aware it was that many
PerlJam (roughly) 17:00
I should say "since the common commit ancestor of the master and ng branches" since that's all I can really tell with git merge-base.
But I'm assuming there haven't been many (any?) merges between the branches. 17:01
17:01 rfordinal3643 joined
moritz_ afaict just one cherry-pick 17:01
PerlJam and there were about 30 commits since the last release before the ng branch was committed, so total commits between releases is 354 (so far) 17:04
wait ... wasn't masak going to do the release this morning his localtime? That should have already happened, no?
moritz_ PerlJam: nope, was re-scheduled for the evening 17:05
PerlJam smart move :) 17:06
cspencer git question: how do i pull a non-master branch from github? (in this case, ng) :)
PerlJam I bet if I look at the times that the tags were made, I bet they tend to always be in the evening :) 17:07
moritz_ cspencer: when you pull, you always obtain all branches
PerlJam cspencer: after you clone the repo, just do "git checkout ng"
oops, that's wrong.
cspencer ah, ok :) thanks
moritz_ and then git checkout -b ng orgin/ng
PerlJam yeah, that'll do
cspencer great, thanks :) 17:08
PerlJam is annoyed that he keeps forgetting the syntax for making a tracking branch in git
moritz_ --track 17:09
to the command above
PerlJam yeah, but is is "git checkout -t -b ng origin/ng" or just "git checkout -t ng" or what?
moritz_ whenever you create a new branch you need -b 17:10
and to track something you have to create a local branch
or with branch.autosetupmerge you can make --track the default
PerlJam "git checkout -t origin/ng" will make a local branche for you. 17:11
that's what it is.
moritz_ oh, I didn't know that
PerlJam I think it's only on newer gits.
moritz_ I was just stupidly typing git checkout -b ng --track origin/ng and felt like a monkey 17:12
PerlJam checkout is git's swiss army knife 17:14
17:27 masak joined
masak OH HAI 17:27
phenny masak: 12:51Z <pmichaud> tell masak I've added a draft announcement for the release -- update it as you see best! thanks!
masak I can haz Riga Hotel wifi connection!
pmichaud \\p/
jnthn lolitsmasak
masak \\o/
jnthn!
I presume you're somewhere nearby? :P
jnthn :P
masak ok. release. :) 17:28
jnthn \\o/
masak pulls
jnthn Yeah, it's easy in Latvia, huh? ;-)
spinclad Latvia must be a small country, huh? Hardly bigger than one hotel. 17:29
masak spinclad: actually, it's at least as big as three hotels.
spinclad (well, Riga, anyway)
just happening to bump into another just happening to be in town 17:30
jnthn Yeah, dunno how it happened.
masak complete happenstance. 17:31
it's not like we planned to meet at the airport or something like that. :)
pmichaud afk, lunch 17:36
17:38 mberends joined
masak mberends! \\o/ 17:38
mberends 'allo 'allo masak
jnthn oh hai mberends
mberends hai jnthn, are you in Riga now? 17:39
jnthn mberends: Oh yes.
mberends: And you know what?
I ran into masak at the airport too.
masak \\o/ 17:40
mberends -Ofun
17:40 stephenlb joined
jnthn lol...I just gave masak a merge conflict! 17:40
mberends is in Friesland with limited web access. Catching up backlog from last Monday... 17:41
next week will be better, hotel in Groningen promises free WiFi
masak: I like your blogs and historical references, but last year you said it was going to be less Western/Euro-centric 17:43
dalek kudo: d351043 | jnthn++ | docs/ (2 files):
Tweak the release announcement and change log to clarify what the nested siggies thing really is.
 17:44
jnthn mberends: The Dalai Llama ain't so western. ;-)
mberends troo
masak mberends: dang it, your memory is too good! :)
mberends so this month's Rakudo release comes from Latvia :-)
moritz_ so... Latvia.pm? 17:45
masak mberends: the trooth is that I meant to scour Wikipedia for lots of non-Western history, but I never got around to it.
mberends nvm
masak it does indeed :(
s/:\\(/:)/
mberends heh
jnthn Erm. I may have meant Lama, not Llama. 17:46
masak moritz_: actually, Lisbon.pm
I haven't met Latvia.pm yet, don't know if they're good guys. :)
...or if they HATE PERL 6.
jnthn
.oO( wish I had photoshop )
masak llol
masak has disturbing mental visions 17:47
mberends we all use Perl 5, even if it is to build Perl 6 ;-) like I use IE to download and install Firefox
masak "Oh no! Dalai Llama spit in my face!"
mberends: nice slogan. very... non-offensive. :P 17:48
mberends masak: you're too kind
masak mberends: we really should start thinking about that blog engine. 17:49
dalek kudo: b237f9c | masak++ | docs/announce/2009-11:
[docs/announce/2009-11] filled in the name "Lisbon"
masak mberends: do you think we could have a prototype for Christmas? :) 17:50
mberends masak: the problem is not only tuits, but also the numerous more shiny toys to play with. I haz a sekrit one too...
masak oh noes! 17:51
jnthn masak: Which Christmas?
cognominal sekrit?
masak mberends: is it Perl 6-related?
ssszekkritt! 17:52
mberends masak: you show me yours, and I'll show you mine, and to nobody else ?
TimToady
.oO(sounds like getting married...)
 17:53
mberends suspects masak has a very interesting sekrit
17:54 masak joined
jnthn oops, didn't mean to tread on ya cable... 17:54
spinclad onoezwelostmasak.. ohthereheis
masak kicks the wifi
mberends: you'd be getting the short end of such a deal -- I go public on Saturday :) 17:55
mberends masak: roll on Saturday then! 17:56
masak \\o/
jnthn Yay, I'll be in The Talk Where masak Reveals all
masak all spectests pass here.
masak goes to read the release guide rather than just winging it 17:57
jnthn I can't believe I'm watching someone reading something in the docs folder.
cognominal is there a doc folder? :)
masak I'm not reading it very carefully... :)
hm, should I update the ROADMAP? 17:58
masak czechs
Tene masak: I'm not exhausted so far today! If it holds out for another six hours or so, I might get something done tonight!
KyleHa jnthn: You were using 'docs' for your diary?
masak Tene: sounds promising. :)
moritz_ masak: that's not a typical release manager job
mberends on that topic, does anyone where the docs are to get mildew-js running? 17:59
Tene KyleHa: no, for his recipe book.
jnthn KyleHa: No, just my password list.
Tene docs/tasty_curry
jnthn KyleHa: Oh, and archiving all those love letters I wrote to...whatever her name was.
moritz_ mberends: presumably either in v6/re-mildew/ or in pmurias' head
masak apart from 'Completed ROADMAP items' at the bottom, the ROADMAP looks fine.
KyleHa She was very special. 18:00
Tene I'm getting nervous, not having written any code lately... I might have to start using caffeine again. 18:02
mberends moritz_: v6/re-mildew doesn't look very doc-y and I cannot ssh into pmurias++ head
moritz_ I think that's a feature :-) 18:03
mberends security feature, yeah :-)
KyleHa If my head had a ssh port, I'd plug it with my finger. 18:04
Tene If my head had ssh enabled, I'd take a week off of work and do some tidying and code cleanup.
The bug tracker for my brain hase more tickets than Parrot's. 18:05
mberends plugging with a finger is a good old Dutch tradition (dykes etc)
mberends tries 'make' in re-mildew # JFDI 18:08
cognominal ho, I did not know that meaning of dyke... 18:09
I should remember this is not #perl 18:10
mberends cognominal: :)
18:12 SmokeMachine joined
jnthn cognominal: I, erm, learned the meaning about the water control type later too. 18:15
dalek kudo: 1db377a | masak++ | docs/ChangeLog:
[docs/ChangeLog] added a few more changes
 18:18
masak people might want to help review the announcement. gist.github.com/rakudo/rakudo/blob/...ce/2009-11 18:20
I'll especially welcome more nice things to say about Lisbon.pm.
moritz_ reads 18:21
one and a half lines are enough IMHO :-)
masak very well. 18:22
mberends +1
moritz_ looks fine to me
18:23 payload joined
mberends drat. re-mildew needs re-smop, and that has some b0rken C atm 18:24
colomon masak: looks good to me.
dalek kudo: f5065b6 | masak++ | docs/release_guide.pod:
[docs/release_guide.pod] rolled releases forward
jnthn masak: Apparently they organized a YAPC. 18:26
colomon we have release! \\o/ 18:42
jnthn masak++
PerlJam masak++
(a rakudo release is a nice thing to come back from lunch to :-) 18:43
masak and by that, I think I'm done.
and my battery is running out.
and I've forgotten my charger :/
pmichaud masak++
mberends masak++
masak wishes karma charged his battery :P
moritz_ masak++
jnthn masak: We can has cheez burger now plz?
masak oh yes. 18:44
jnthn \\o/
masak I still think we prioritized things right. :P
PerlJam has some reservations about a certain university's ability to prepare grad students for graduating. 18:46
zaslon lolmasakhazblogged! masak++ 'Rakudo Perl 6 development release #23 ("Lisbon")': use.perl.org/~masak/journal/39918?from=rss
masak food &
PerlJam "When I give my project report, will they ask me questions?"
18:46 reid06 joined
pmichaud masak: what, no "On today's date in .... " quote? 18:46
(yes, I know he left)
PerlJam: answer is "If your report is good enough to merit questions, yes. If they don't ask questions, you've probably done something really wrong." 18:47
mberends offline_hotel & 18:48
PerlJam pmichaud: the follow up was funnier ... "Where can I find a list of these questions?" :)
pmichaud lol
KyleHa masak++ # release of Lisbon
pmichaud "I don't understand why I have to write a report. I already paid for the course... isn't that sufficient?" 18:49
18:51 astrojp left, kst joined
TimToady we're in the wrong business--we should sell fake Doctorates in Linguistics... 18:59
PerlJam TimToady: fake? Just found an educational institution and make them real! 19:01
TimToady I've already paid for an educational institution...isn't that sufficient? 19:05
PerlJam The Larry Wall Center for Advanced Linguistics :)
KyleHa Last time I found an education institution, all I learned was that I don't know what a linguistic is. I assume it's pasta. 19:06
19:08 chromatic joined
TimToady Used to be I couldn't spell Advacned Lingristics, and now I are one... 19:08
19:11 kst joined
TimToady maybe we should rename 'void' to 'suck', since it's eager... 19:12
Perl 6, the language that sucks by default 19:13
PerlJam goes to void his bladder
diakopter Perl 6, the language that makes [line] noise in a vacuum. 19:15
TimToady only when you vacuum up the cord, silly
actually, 'vacuum' wouldn't be a bad name for an eager void 19:16
KyleHa Vortex. 19:18
PerlJam vacwm (hehe :) 19:19
gyre
TimToady you don't cheat fair
Sucker
or we could call it BlackHole and guarantee it won't be used in Russia... 19:20
we could be the first language to implement the Singularity
pmichaud ...except that black holes do not suck. :) 19:22
PerlJam You could call it the Wall (at least 3 meanings there)
diakopter hole
(implied gravity well)
TimToady well, maybe we could make mst happier by changing Suck to F-ck
pmichaud heh 19:23
I think he already named it that.
at least, that was the title of a recent blog post :)
"We've decided to rename Perl 6 to F-ck, which means that mst is now our most ardent marketer for Perl 6"
diakopter I thought we were naming 'eager void' 19:24
19:24 astrojp joined
[particle] am i in #perl? 19:24
diakopter /me whois you 19:25
TimToady [particle]: we're not swearing, just meta-swearing 19:26
perigrin F-ck.HOW seems wrong 19:27
KyleHa rakudo: WHAT( $are_you_deaf ) 19:28
p6eval rakudo 7347ec: Symbol '$are_you_deaf' not predeclared in <anonymous> (/tmp/ei6cL3qH7s:2)␤in Main (file <unknown>, line <unknown>)␤
KyleHa That was funnier in my head. 19:29
19:30 am0c joined
diakopter rakudo: sonic 19:30
p6eval rakudo 7347ec: Could not find non-existent sub sonic␤in Main (file src/gen_setting.pm, line 324)␤ 19:31
diakopter rakudo: standard 19:33
p6eval rakudo 7347ec: Could not find non-existent sub standard␤in Main (file src/gen_setting.pm, line 324)␤
pmichaud "eager void" sounds like the topic of a FloMax commercial. 19:34
diakopter or Activia. 19:35
TimToady Ex Lax, for that get-up-and-go feeling... 19:36
TimToady channels Mad Magazine from his misspent youth 19:44
PerlJam as long as you don't channel Mad Max
TimToady more like Maxfield Smart 19:45
dukeleto imagines TimToady riding on a hog in a post-apocalyptic future
TimToady I think we haven't discovered anything with this discussion, other than that freud rhymes with void 19:47
KyleHa ...and that starts with V and that rhymes with T and that stands for Trouble!! 19:48
diakopter and with The Noid
TimToady and with Pink Floyd 19:50
PerlJam and now we're back to the Wall 19:52
pmichaud tosses a brick at PerlJam.
colomon wishes he were comfortably numb... 19:54
TimToady I think 'vacuum' is still the best, metaphorically speaking
PerlJam you could use "gap" and then we'd have fun with "mind the gap!" 19:56
TimToady colloquially, vacuum can be shortened to 'vac' 19:58
and that's even shorter than void
PerlJam and can be punned in a few more useful ways too 19:59
But "vac" makes me think of "car" and "cdr" and that reminds me of lisp and lisp gives me crazy surreal nightmares. 20:00
:-)
TimToady never heard of a CdrVac...
diakopter how about 'well'
TimToady wells aren't supposed to return Nil 20:01
PerlJam but you can get oil and/or water from them sometimes
TimToady there's the 'null' of /dev/null as well, which semantically is much like a vacuum 20:02
in the take but don't give back sense 20:03
dukeleto TimToady: what about 'sink' . mathematicians love talking about sources and sinks 20:05
TimToady that's...not bad 20:06
KyleHa Sounds like the mathematicians are really thinking of drains.
TimToady sink does imply some degree of eagerness
oh, dear... 20:07
my Sink sub {...}
TimToady is tempted to kick himself
all things considered, sink would be an improvement over void 20:08
20:08 kst joined
TimToady apart from prior art, of course 20:09
fo() ==> bar() ==> sink;
20:09 ShaneC joined
TimToady *foo # balky o key again... :( 20:10
moritz_ better than KitchenSink 20:11
TimToady KitchenSink is the role
pmichaud are we still thinking that expressions in void context end up with .Void ? 20:17
or would that now be .sink or .Sink or ...?
and why does .sink suddenly remind me of .sinh ? ;-)
KyleHa has a sinking feeling. 20:18
TimToady rakudo: my $_ = 'sinh'; .++; .++; .++; .say
p6eval rakudo 7347ec: Syntax error at line 2, near ".++; .++; "␤in Main (file <unknown>, line <unknown>)␤
pmichaud rakudo: my $_ = 'sinh'; .=succ; .=succ; .=succ; .say 20:19
p6eval rakudo 7347ec: sink␤
pmichaud there
sinks succ three times more than sinh's
TimToady whimpers
dukeleto my Sink sub :adiabatic { ... } 20:20
20:23 dukeleto left, dukeleto joined
sjohnson diakopter++ thanks for the console tip 20:24
will try it out
diakopter... you GENIUS 20:25
20:29 kst joined
diakopter i see 20:37
TimToady but vacuum is still better than sink, because vacuums are eager, while sinks are lazy
cognominal :) 20:38
diakopter unless you think in terms of gravitons
well, I guess I meant
moritz_ the existence of gravitons is not yet generally accepted
TimToady sinks don't do anything; it's the earth that sucks 20:39
perigrin unless you have a disposal
moritz_ so rename to .Earth? :-)
perigrin .gravity_well
diakopter .hempty_dempty
_sat_on_a_Wall 20:40
it's an odd day.
TimToady the question is, who's to be master, that's all...
diakopter .pit 20:41
.abyss
.abysm
.gehenna 20:42
moritz_ .bikeshed_void
diakopter that's a tad more destructive
heh
TimToady black holes are evaluated only for their side effects...
mathw Evening
TimToady btw $a++; return $b; 20:44
diakopter std: return return
p6eval std 29132: ok 00:01 99m␤
TimToady .shed 20:45
diakopter std: /me sheds a tear in the spime continuum 20:46
p6eval std 29132: [31m===[0mSORRY![31m===[0m␤Regex missing terminator at /tmp/GnzU8vHVbI line 1 (EOF):␤------> [32m/me sheds a tear in the spime continuum[33m⏏[31m<EOL>[0m␤FAILED 00:01 100m␤
diakopter I mean
yeah
TimToady .nsa 20:48
spinclad no such acronym? 20:49
TimToady now sink anything 20:50
moritz_ you think that's cute today :-)
spinclad ===SORRY!===\\nAcronym missing referent at 20:48:21 20:51
TimToady sync
diakopter TimToady: I thought you hid the deep recursion warning on std 20:52
.nom
TimToady it seems to keep cropping up in odd places, like it's not properly lexical scoped
diakopter .nom_nom_nom
use 4chan_memes; 20:53
TimToady funny think is, a sink really is a sync point
KitchenSync
dukeleto sink; sink; sink; #just to be sure 20:54
diakopter KvetchingSink
20:55 [particle]1 joined 20:59 pmichaud joined, zaslon joined, pointme joined, Woody2143 joined, hsb joined, christine joined, eiro joined, c9s joined, buu_ joined, mdxi joined, sjohnson joined, leedo joined, estrai joined
TimToady I really like the sink === sync correspondence, but maybe that's just my brainfever talking 20:59
20:59 christin1 joined, c9s_ joined, mdxi joined 21:00 buu joined, leedo joined
dukeleto TimToady: what exactly are you thinking about using 'sink' for? 21:01
TimToady replacement for statement_prefix:void mostly
sjohnson cleaning dishes
TimToady but something that implies eager context more than 'void' does 21:03
21:03 baest joined 21:04 justatheory joined
dukeleto TimToady: 'engulf' ? 21:07
moritz_ what about 'eager'?
it does comunicat eagerness :-) 21:08
dukeleto TimToady: should statement_prefix:void be an English verb or adjective or does it not matter?
moritz_: there you go being logical again ;)
spinclad as a verb it means more like 'source', though 21:10
TimToady eager returns its results
Tene 'ignore'?
'discard'? 21:11
spinclad 'ignore' could be lazy
spinclad would think 'slurp', but it's already taken
Tene I'd think slurp would return its results. 21:12
spinclad no, that would be 'void' ?8-[
moritz_ likes 'sink' 21:13
spinclad too 21:14
21:14 Woody2143 joined
pnate yeah, sink's good >_> 21:15
PerlJam sink +1
TimToady are those sink points?
dukeleto i have a sinking feeling that we are making progress 21:18
21:19 pmichaud joined, eiro joined
moritz_ if somebody finds an alternative name you can't easily make puns on -- that would get myvote 21:20
21:21 Su-Shee left, zaslon joined, pointme joined, hsb joined, sjohnson joined, estrai joined, hsb_ joined, estrai_ joined 21:22 justatheory_ joined
TimToady but can we bear to say 'sink context' everywhere we've been saying 'void context'? 21:25
dukeleto TimToady: so 'sink context' is just shorthand for saying 'eager void context' ? 21:26
TimToady no such thing as lazy void context...
[particle] why does it seem that when rakudo enters the "next 80%", it always means totally reimplementing the last 80% again? 21:27
moritz_ because it's software.
dukeleto [particle]: why do you say that?
[particle] rakudo passes 85+% of spectests
dukeleto [particle]: is this in reference to their nqp-rx changes?
[particle] what's the next step? reimplement! :) 21:28
moritz_ that's just a meaningless number
[particle] yes
TimToady well, if a dataset of one element can be called "always"
[particle] moritz_: i know.
rakudo has been totally reimplemented before
moritz_ if you substract the trig tests, rakudo is at 70%
[particle] i was there
21:28 payload joined
dukeleto TimToady: so there will be no such think as 'void context' in perl 6, only 'sink context' ? 21:28
moritz_ if you count all tests yet to be written, rakudo is 40% there
[particle] i'm not down on it, i'm just noticing a pattern
dukeleto s/think/thing/
moritz_ [particle]: it's called "agile", that pattern 21:29
dukeleto [particle]: it could be that they made implemenation decisions due to limitations in parrot that they are going back to correct
[particle] i agree with all of that 21:30
dukeleto [particle]: i agree with your observation, just wondering about it
TimToady but you're measuring angularly, when timewise the middle of the whirlpool turns faster (we hope)
jnthn [particle]: The reasons for the current set of refactors were clearly outlined in a blog post by pm. Plesae read it. 21:31
[particle] yes, we hope, or it's not a good sink
jnthn: i understand completely
perhaps i should have used a winking smiley in my statement
jnthn [particle]: :-)
[particle] it's zeno-esque
21:31 zamolxes joined
[particle] everytime we reimplement the 80%, we do it faster 21:32
PerlJam seems to recall that being paradoxical :)
[particle] eventually, we'll reimplement that 80% as fast as the new cray finishes an infinite loop!
dukeleto PerlJam: The Zeno Paradox is only a paradox if you don't take into account quantum effects. At some point, it is no longer possible to "half your distance" 21:34
dukeleto is pretty sure perl6 development harnesses quantum effects 21:35
PerlJam indeed
dukeleto when multiple people obvserve the spec, an implementation comes into being where one did not exist before 21:36
21:37 hsb joined
[particle] waves 21:39
moritz_ dukeleto: it uses transistors under the hood, so "yes" :-)
there you have your quantum effects
dukeleto Perl 6 : We Still Use Transistors 21:50
japhb I wonder if, as per quantum gravity, when the implementation gets dense enough, it bounces back violently (possibly creating a new universe just for the fun of it)
dukeleto japhb: i think you are talking about the 'big bounce' ? also, there is not such thing as 'quantum gravity', just a lot of out of work quantum gravity physicists :) 21:52
21:53 pmurias joined
japhb dukeleto, last I heard as of this year there was actually a quantum gravity theory that didn't produce infinities. And it predicts bouncing at maximum density, yes. 21:53
dukeleto japhb: the hard thing is justifying why, when the density reaches the critical value, the force is 'negative' instead of 'positive' feedback
japhb: yes, i read up on john baez's work. what i can understand, anyway 21:54
japhb: i think i can grasp categories, and maybe even 2-categories, but when he busts out n-categories, I reach for some good scotch
japhb When 'loop quantum gravity' was all the rage, I think it had to do with quantized spacetime, such that you got a spacetime equivalent of Pauli Exclusion going on. 21:55
I have no idea of the basis under the current theory, unfortunately. The above was my memory of reports in SciAm et al.
pmurias mildew: say "hi" 21:56
japhb hasn't taken a physics class in a really long time.
p6eval mildew: Can't locate MooseX/Declare.pm in @INC (@INC contains: src perl6-snapshot /etc/perl /usr/local/lib/perl/5.10.0 /usr/local/share/perl/5.10.0 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.10 /usr/share/perl/5.10 /usr/local/lib/site_perl .) at src/AST.pm line 2.␤BEGIN failed--compilation a…
dukeleto japhb: john baez is a loop quantum gravity/quantum foam guy
[particle] i read that as joan baez 21:57
japhb A relative of mine got a PhD in high-energy particle physics a few years back, and now works in a basically unrelated field. I wonder if he keeps up with current theory any more?
dukeleto japhb: math.ucr.edu/home/baez/ 21:58
TimToady I have a relative who keeps up :)
pmurias mildew: say "hi"
p6eval mildew: Can't locate MooseX/Declare.pm in @INC (@INC contains: src perl6-snapshot /etc/perl /usr/local/lib/perl/5.10.0 /usr/local/share/perl/5.10.0 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.10 /usr/share/perl/5.10 /usr/local/lib/site_perl .) at src/AST.pm line 2.␤BEGIN failed--compilation a…
pugs_svn r29133 | pmurias++ | [re-mildew] remove unused dependency on MooseX::Declare
pmurias mildew: say "hi"
p6eval mildew: Can't locate MooseX/Declare.pm in @INC (@INC contains: src perl6-snapshot /etc/perl /usr/local/lib/perl/5.10.0 /usr/local/share/perl/5.10.0 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.10 /usr/share/perl/5.10 /usr/local/lib/site_perl .) at src/AST.pm line 2.␤BEGIN failed--compilation a…
japhb TimToady, who would that be? 21:59
dukeleto p6eval is wreaking havok on my irc client
TimToady my son has just applied for a postdoc position at the Perimeter Institute
pmurias how can i leave a message for someone? 22:00
japhb dukeleto: ooh, that Mandelbulb fractal is very cool looking. Nice rendering!
TimToady, NICE. Good luck to him!
dukeleto TimToady: congrats!
22:00 quietfanaytic joined
diakopter pmurias: hi 22:01
pmurias: p6eval needs MooseX::Declare?
dukeleto japhb: yes, john's "this week in mathematical physics" is usually entertaining an mind-bending
quietfanatic Gotta watch out for those fanaytics.
dukeleto quietfanatic: hola
quietfanatic dukeleto: yo
dukeleto quietfanatic: how is Link::C doing these days? 22:02
pmurias diakopter: hi
diakopter: there was a leftover 'use MooseX::Declare' left in mildew 22:03
quietfanatic It's on the backburner
I need to rewrite the parser again
I've been working more on Gamebase, which depends now on Events, which I am still contemplating 22:04
dukeleto quietfanatic: are you hitting any Parrot road blocks lately, or is it staying out of your way?
quietfanatic dukeleto: Not immediately
dukeleto quietfanatic: good to hear
quietfanatic besides the occaisonal wild null PMC 22:05
*occasional
I expect to hit a couple blocks soon though
such as speed 22:06
or, the inability to introspect constants out of signatures
rakudo: say :(3).perl
p6eval rakudo 7347ec: :(Int ::TYPE_CAPTURE20 where all({ ... }))␤
quietfanatic There's no 3 in there :)
Oh, something I came across recently: 22:07
rakudo: my &routine; say (:&routine).WHAT
spinclad it's there in the ... ... you gotta introspect harder
p6eval rakudo 7347ec: Symbol '&routine' not predeclared in <anonymous>␤in Main (file <unknown>, line <unknown>)␤
quietfanatic versus:
rakudo: my $routine; say (:$routine).WHAT
p6eval rakudo 7347ec: Pair()␤
dukeleto quietfanatic: come talk to me if you hit speed issues, i am the parrot benchmark guy
quietfanatic Okay, I will when I get there 22:08
Tene rakudo: my $routine; say (:$routine).perl
p6eval rakudo 7347ec: "routine" => undef␤
Tene quietfanatic: what's wrong with that?
22:08 KyleHa joined
dukeleto quietfanatic: also, we are attempting to clean up our null pmc issues 22:08
quietfanatic "Symbol '&routine' not predeclared in <anonymous>"
when I clearly predeclared it.
dukeleto everytime someone uses p6eval, irssi craps itself. is p6eval sending out funny characters? 22:09
quietfanatic :&thing syntax doesn't work
Tene ah
dukeleto: WFM
dukeleto: it uses ␤
sjohnson weechat > irssi :)
dukeleto Tene: that is really pissing me off 22:10
Tene but weechat has an annoying name.
dukeleto: eh?
dukeleto: I was asking if that's the character that's causing problems.
sjohnson Tene: that is my one gripe about weechat... silly name
otherwise pretty rocksolid *nix cli client
dukeleto Tene: all i see is line noise. YES. whatever you attempting to type, messes up my terminal
spinclad (need font?) 22:11
dukeleto gotta go. keep it quantum, peeps
22:11 viklund joined
Tene I use irssi and I see it just fine. encoding issues? 22:11
pmurias shachaf: why is weechat better?
Tene .u ␤
phenny U+2424 SYMBOL FOR NEWLINE (␤)
pmurias diakopter: hows you past interpreter doing?
sjohnson hey thats great
shachaf pmurias: Hmm, I'd be curious as to the answer to that. I've considered switching to another client.
sjohnson unicode bot 22:12
Tene .u NEWLINE
phenny U+2424 SYMBOL FOR NEWLINE (␤)
Tene phenny is very useful.
moritz_ indeed
TimToady sounds like your irssi (or screen) isn't properly in UTF-8 mode
pmurias shachaf: sorry, that was meant to be aimed at sjohnson 22:13
quietfanatic TimToady: Are there any plans in the future to standardize Perl 6's syntax tree?
diakopter pmurias: dunno..
quietfanatic say, for macros and modules to dink around with?
diakopter haven't worked on it in a day or two
shachaf pmurias: Yep. But I'd like to see the answer too. :-)
diakopter so, naturally, I've forgotten entirely about it
sjohnson pmurias: i would be happy to answer that question 22:14
... by screenshot :)
TimToady quietfanatic: it would be nice if our ASTs were well-typed, so even if there's more than one, we know which one it is 22:15
quietfanatic TimToady: If I could mess with the syntax tree, then I could, for instance, figure out how to reverse a reversible sub. 22:16
TimToady reverse on which level? 22:17
moritz_ except that the question if a sub is reversible is not computable (if you mean the same as I am)
quietfanatic reverse meaning flip the input and output
sjohnson pmurias: members.shaw.ca/smujohnson/img/weec...enshot.png
Tene insufficient unicorns and ponies. 22:18
quietfanatic like e(x) reverses to log(x), etc
sjohnson pmurias: this is 95% out of the box behaviour and look
Tene what's all that stuff on the sides of the chat?
sjohnson pmurias: scripts can be written in 5 scripting languages + C if you decide to
Tene: channel list 22:19
pmurias sjohnson: colorfull nicks are nice
Tene ew
I don't want that.
and a user list too?
sjohnson Tene: colorful nicks make it a lot easier to read a multiway convo
Tene: these are all easily disableable :)
ctrl-N toggles my user list
Tene pmurias: scripts.irssi.org/scripts/nickcolor.pl 22:20
quietfanatic s/e(x)/e^x/
pmurias quietfanatic: that's a very hard problem 22:21
unless you write your program using bijections only
TimToady such a program could run on a quantum computer with arbitrarily low energy requirements :) 22:22
since it's the forgetting that takes energy 22:23
moritz_ so we just need to simulate a quantum computer in polynomial time :-)
TimToady I saw a polynomial dance once... 22:24
mathw Quantum Perl
that could be interesting
TimToady something that might interest both my sons :) 22:25
pmichaud "We've decided to change the name of Perl 6 to 'Future Perl'."
:-)
TimToady Perl 5 is the "Once Perl"?
pmichaud "Alternate Reality Perl"
moritz_ or qperl for quantum perl
pmichaud or just qerl
PerlJam ARPerl? ARP?
pmichaud although we did once consider "perm", since that's "perl".succ 22:26
quietfanatic pmurias: You could at least figure it out for compositions of unary functions known to be reversible
22:26 Whiteknight joined
PerlJam Quantum Perl wouldn't be a bad alternative to Perl 6 if ever we contemplated a name change :) 22:26
TimToady the Perly Exclusion principle says you can't have Perl 5 and Perl 6 in the same spot
huf it's quite quantum as it is, with so many name-ideas floating around
diakopter mind your p'5 & q'5
TimToady ow, I can dish it out today, but I can't take it... 22:27
quietfanatic pmurias: er, right, bijections, that's what those are...
TimToady no sqrt for you, young man! 22:28
quietfanatic TimToady: Okay I'll just use ±sqrt instead. 22:29
sjohnson diakopter: :)
moritz_ you can reverse injective functions; they don't need to be bijective
TimToady Good News: I have solved your problem! Bad News: The answer is a junction.
moritz_ :-)
sjohnson worse news: won't be available till christmas 22:30
;)
quietfanatic
.oO(or an autothreading set that you're *allowed* to pull values out of)
TimToady there's some kind of buzzing in my ear... :P 22:31
slavik will we have "for all (@list) { }" ?
moritz_ rakudo: my @a = 1..5; .say for all(@list) 22:32
p6eval rakudo 7347ec: Symbol '@list' not predeclared in <anonymous> (/tmp/FMY7JkXig0:2)␤in Main (file <unknown>, line <unknown>)␤
TimToady slavik: hyper for @list { }
moritz_ rakudo: my @a = 1..5; .say for all(@a)
p6eval rakudo 7347ec: all(1, 2, 3, 4, 5)␤
quietfanatic rakudo: my @a = 1..5; say all(@a)
p6eval rakudo 7347ec: all(1, 2, 3, 4, 5)␤
slavik I mean the automatic multithreading type of for all 22:33
I believe it was changed to something else
KyleHa rakudo: my $parse = Perl6::Grammar.parse( 'say "Hello world"' ); $parse.^methods().join('|').say
p6eval rakudo 7347ec: ( no output )
moritz_ slavik: all() autothreads already
slavik who implemented it? 22:34
moritz_ rakudo: say 1+all(3,4)
p6eval rakudo 7347ec: all(4, 5)␤
moritz_ slavik: jnthn++ did
slavik jnthn: I <3 you!!!
TimToady since all() are really the only ones that are true sets, maybe we could let you get the values from an all 22:35
slavik *cough* I can haz assertions in grammer? *cough*
moritz_ slavik: in nqp-rx yes, in rakudo not yet
slavik nqp-rx ???
moritz_ the thing that rakudo's grammar and action methods are written in 22:36
slavik how does that get put into rakudo?
moritz_ ng: grammar A { token TOP { (\\d+) <?{ +$0 < 256 }> } }; say A.parse('127') 22:37
p6eval ng 2a3149: Unable to parse blockoid, couldn't find final '}' at line 1␤current instr.: 'perl6;Regex;Cursor;FAILGOAL' pc 1630 (src/stage0/Regex-s0.pir:888)␤
moritz_ ng: grammar A { token TOP { (\\d+) <?{ +$0 < 256 }> } }; say(A.parse('127'))
p6eval ng 2a3149: Unable to parse blockoid, couldn't find final '}' at line 1␤current instr.: 'perl6;Regex;Cursor;FAILGOAL' pc 1630 (src/stage0/Regex-s0.pir:888)␤
quietfanatic TimToady: Something to be said for that, namely this.
Of course, then people will want to say all(0..*) to get the set of natural numbers... 22:39
moritz_ std: grammar A { token TOP { (\\d+) <?{ +$0 < 256 }> } }; say(A.parse('127'))
p6eval std 29132: ok 00:01 113m␤
moritz_ ng: grammar A { token TOP { (\\d+) <?{ say($/) }> } }; say(A.parse('127'))
p6eval ng 2a3149: Method '!cursor_start' not found for invocant of class 'Perl6Str'␤current instr.: 'perl6;A;TOP' pc 194 (EVAL_1:91)␤
TimToady and note that none(0..*) is also an infinite set 22:40
moritz_ ng: grammar A { token TOP { (\\d+) <?{ say('alive') }> } }; say(A.parse('127'))
p6eval ng 2a3149: Method '!cursor_start' not found for invocant of class 'Perl6Str'␤current instr.: 'perl6;A;TOP' pc 194 (EVAL_1:91)␤
TimToady but then, so is none(0)
moritz_ ng: grammar A { token TOP { (\\d+) } }; say(A.parse('127'))
p6eval ng 2a3149: Method '!cursor_start' not found for invocant of class 'Perl6Str'␤current instr.: 'perl6;A;TOP' pc 194 (EVAL_1:91)␤
quietfanatic TimToady: I'm wondering if a true mathematical set type wants to be implemented in terms of matchers, so that you can have things like that. 22:42
TimToady it's very easy to turn a set into a list, and a list into a junction
22:43 gbarr joined
TimToady especially if the set already has a % or @ sigil, then it's just any(%set) == $whatever 22:43
quietfanatic So you get Int for the set of integers, and Int&none(0) to get non-zero integers
TimToady assuming set types listify to only keys
quietfanatic but then you're back to junctions 22:44
Does any(%hash) only get the keys?
or is that special for a Keyset? 22:45
oh,you just answered that
TimToady Set types return only the keys, since the value is known to be 1
quietfanatic is slow today
rakudo: class X {has &.routine}; X.new(:routine({say "x"})).routine.() 22:47
22:47 IllvilJa joined
TimToady and there's a reason subsets are called subsets 22:47
p6eval rakudo 7347ec: ( no output )
quietfanatic ah, I suppose that's right 22:48
though you still can't call methods on them :(
moritz_ currently p6eval times out for everything that contains a class definition
TimToady you can, but you'll just get the method on the underlying object type
quietfanatic moritz_: is that what it is? Okay
yeah and that's rarely useful
moritz_ quietfanatic: I have no idea what it is or why, I just observe the symptoms 22:49
quietfanatic moritz_: Is it true for the machine p6eval is running on, or only for p6eval? 22:50
Anyway, the above code snippet prints nothing. 22:51
it doesn't call the routine
why doesn't it?
(I mean prints nothing locally as well) 22:52
ah, but: 22:54
saying .routine.().() does run it.
I think that first .() is interpreted as an argument list to the method call.
This is not what I expected. 22:55
22:55 jan_ joined
moritz_ quietfanatic: only in conjunction with evalbot - for example the spectests run fine on it 22:55
quietfanatic moritz_: huh. 22:56
s/is not what I expected/seems wrong to me/ 22:57
TimToady wrong according to S12:441 too 22:58
moritz_ agreed on both
22:59 constant_ joined 23:02 BinGOs joined
quietfanatic rakudo: say.(3) 23:02
p6eval rakudo 7347ec: 3␤
quietfanatic I think this is the same issue. 23:03
pmurias did anyone have any succes with integrating vim with a debugger? 23:04
i'm thinking of adding debugging/tracing to mildew-js
23:06 PacoLinux joined
colomon TimToady: I need to head off to get dressed for a concert, but before I go -- do you think it's right for Range.Str to list all the elements in the Range? That's Rakudo's old definition, but it seems odd to me for any Range with "more than a few elements" in it. 23:09
23:17 explorer joined
quietfanatic re eager void: for statement prefix, how about 'just'? :) 23:25
just do {...}
23:28 ihrd joined 23:30 mfollett joined
mfollett I have a question regarding Rakudo. My understanding is that Parrot is moving to LLVM underpinnings. If Parrot does that will Rakudo be able to JIT to GPUs? 23:36
diakopter mfollett: afaik, parrot is still only considering building llvm underpinnings. However, I don't think a move to LLVM necessitates enabling "JIT to GPUs". 23:39
23:39 beggars joined 23:40 cotto_work joined
mfollett diakopter: Ok, thanks, I've seen some hand waving from other groups talking about using LLVM and I've been curious what it would really mean in the context of Rakudo, if anything. 23:41
diakopter: It was making a lot of assumptions that didn't really seem reasonable, but I figured I'd ask around. 23:42
23:46 shachaf joined 23:47 mfollett left

Powered by Cro and the Raku® Programming Language.

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

Thank you!