pugscode.org/ | nopaste: sial.org/pbot/perl6 | pugs: [~] <m oo se> (or rakudo:, kp6:, smop: etc.) || We do Haskell, too | > reverse . show $ foldl1 (*) [1..4] | irclog: irc.pugscode.org/
Set by TimToady on 25 January 2008.
meppl good night 02:40
rindolf Hi all. 10:27
Which language or technology inspired Perl 6's junctions ? 10:28
b_jonas rindolf: a perl5 module
Quantum::Superpositions
rindolf b_jonas: ah. 10:29
b_jonas by Damian
rindolf b_jonas: was it original to Q::S, or was Q::S inspired by something else?
b_jonas: hmmm... en.wikipedia.org/wiki/Perl_6#Junctions 10:32
Seems like an original concept to Perl 6.
b_jonas I don't know 10:33
I think they're a bad idea the way they are
rindolf b_jonas: why?
Juerd dalek: The memory leak(s) in botnix must be fixed or worked around 12:25
dalek: At this rate, in a few days it'll crash feather again.
Grrrr b_jonas: I believe that Icon (en.wikipedia.org/wiki/Icon_programming_language) has constructs similar to junctions 12:29
moritz_ re 12:46
pugs_svnbot r20243 | moritz++ | [STD.pm] more perlhints 13:13
diff: dev.pugscode.org/changeset/20243
diakopter Juerd: yeah 14:26
Juerd PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
7185 dalek 15 0 366m 346m 3772 S 0 13.9 0:59.00 botnix
diakopter Juerd: a scheduled restart every night would be a decent workaround. Just have to figure out the right signal.
Juerd 16:24:52 up 18:34, 5 users, load average: 0.00, 0.01, 0.00
Well, if it uses 346 MB ram in less than a day, I'd prefer a real solution or a restart every 12 hours. 14:27
diakopter Juerd: yeah. 14:28
Juerd What is botnix anyway? :) 14:29
allbery_b <Grrrr> b_jonas: I believe that Icon (en.wikipedia.org/wiki/Icon_programming_language) has constructs similar to junctions 14:35
hm. more that the evaluation model makes any "or" a junction (and any list a nondeterminacy monad a la Haskell) 14:36
diakopter it's this perl ircbot framework 14:40
it replaced svnbotl on #parrot
because it polls/parses parrotvm.org/svn/parrot/rss instead of hitting the subversion repository (as well) 14:42
TimToady allbery_b: it's more like icon relies on success/failure as indicated by unthrown exceptions to implement a kind of short-circuit or 16:20
but the order of "or" is still strict, unlike junctions 16:21
from my vague recollections...
and it parses very differently from junctions 16:22
my impression is that $a == 1 | 2 | 3 parses == tighter, but I could be misremembering that... 16:23
it's been a *long* time since I looked at Icon... 16:24
[particle1 anybody here want to mentor a google summer of code student who wants to extend the perl 6 test suite?
moritz_ what does "mentoring" involve? 16:25
[particle] i was hoping you'd speak up :)
moritz_ [particle]: but you hoped I'd consent? ;-9
TimToady I know what mentoring involves, but I'm no good at it. :)
[particle] i expect that will follow shortly :) 16:26
i'll give you a link to read, moritz_
moritz_ [particle]: ok. in principle[tm] I'm available, and I know the infrastructure, but I'm not very good in the technical details
[particle] moritz_: basically keeping the student on track with his plan. getting weekly status. answering questions. 16:27
moritz_ sounds possible 16:28
[particle] moritz_: groups.google.com/group/google-summ...nistrators 16:29
every mentor will have a backup 16:30
[particle] and yes, pmurias can be your backup :) 16:30
[particle] what did i say? 16:30
moritz_: good luck to your SO 16:31
moritz_ [particle]: thanks.
[particle]: ok, I'm your man
[particle] thanks! sign up there, and join irc.perl.org#soc, and join the perl soc mailing list. i'll grab you a link in a sec 16:32
[particle] hexten.net/mailman/listinfo/psoc 16:33
TimToady moritz_ += (1..100).pick
and I'll be glad to provide some technical backup, of course 16:34
moritz_ pugs: say (1..100).pick
exp_evalbot OUTPUT[21␤]
[particle] blackjack! 16:35
moritz_ [particle]: half the truth!
TimToady hmm, if 42 the meaning of everything, and whether something is true/false is 50/50, then maybe 21 is the whole truth... 16:37
pmurias [particle]: you meant pmichaud ? (i don't know the test suit well) 16:59
[particle] no, i meant you, because you entered the channel at just the right moment.
just teasing you.
cognominal_ TimToady: what is the convention to say the compiler, there is no inner eval, I won't use OUTER::, I will not keep reference around ...., so the compiler can use non PMC registers when appropriate instead of shoving every lexical in a PMC 17:36
TimToady there is no such convention. that's why it's called *auto*-boxing... 17:39
cognominal_ but for performance sake, we may want to avoid always dealing with the lexpad. 17:40
say we want to calculate fibonacci numbers... 17:41
currently in rakudo, everything is boxed
cognominal_ well, everything but params 17:43
pugs_svnbot r20244 | mj41++ | Renamed to TapTinder and moved. 18:06
r20244 | mj41++ | * mj41.cz/wiki/TapTinder
r20244 | mj41++ | * pr.perl6.cz/svn/taptinder/trunk/
diff: dev.pugscode.org/changeset/20244
allbery_b TimToady: it does, yes. but the eval model isn't just about unthrown exceptions; *every* expression node is a restartable generator. most of course trivially produce only one value then fail, but that every expression node i a generator makes unthrown exceptions and junction-like behavior possible 18:17
pmurias @tell mncharity could use make a list of things you want to use Moose for (i'll try to figure out how to do them in pure p5) 20:13
lambdabot Consider it noted.
stevan_ pmurias: I already talked to mncharity about that,.. i think Class::Accessor would fit all his Moose uses 20:20
it doesnt make sense to use Moose if you are not taking advantage of the features
[particle] moose! 20:23
PerlJam It's too bad Joose wasn't named Goose as I had some good suessian stuff to use on it (from The Sleep Book) 20:24
pmurias stevan_: emitting plain perl5 accesors is not a problem
jrockway if someone writes a new object system for Groovy it could be called Goose 20:25
pmurias sub foo {if (defined $_[1]) {$_[0]{'foo'} = $_[1] } else {$_[0]{'foo'} = $_[1]}} 20:26
PerlJam that would be perfect in conjunction with Joose as I could talk about Mooses drinking moose joose and gooses drinking goose joose :)
stevan_ pmurias: yes, and thats pretty much all you guys are using Moose for. its an awful lot of overhead for something as simple as that 20:31
pmurias: oh, and also the constructors too
jrockway there is that new XS accessor module now also
stevan_ pmurias: also, ditching autobox will help too
yeah, what jrockway said ^^ 20:32
Object::Tiny!
pugs_svnbot r20245 | buchetc++ | [t/spec] STM should be safe for arrays and hashes 20:40
diff: dev.pugscode.org/changeset/20245
lambdabot Title: Changeset 20245 - Pugs - Trac
pmurias jrockway: XS accessor module = Object::Tiny ? 20:47
jrockway no 20:48
Object::Tiny is a joke module
AutoXS::Accessor 20:49
that is the xs-based accessor one
also Class::XSAccessor
pmurias stevan_: ditching autobox require a lot of static analysis or a boxing everythin 20:59
pmurias q:code :lang(Perl5) {...} seems a challenging problem 21:10
pmurias working on elf and kp6 is a bit like xkcd.com/303/ 21:50
lambdabot Title: xkcd - A webcomic of romance, sarcasm, math, and language - By Randall Munroe
meppl good night 22:01
pmurias meppl: g'night
meppl ;) 22:02
deneme_ hi 22:17
my problem is extending Perl i am writing a library 22:18
where can I consult?
there is an error 22:19
that I can't recover
deneme_ Can Anybody help me? 22:19
allbery_b not here, this is perl6 development, not perl5 22:22
deneme_ It is a simple question about pointers 22:23
where is perl5 channel?
[particle] #perl
pugs_svnbot r20246 | pmurias++ | elf - begining of a non-moose backend 22:28
r20246 | pmurias++ | copied version of the p5 one
r20246 | pmurias++ | constructors are missing
diff: dev.pugscode.org/changeset/20246
lambdabot Title: Changeset 20246 - Pugs - Trac

Powered by Cro and the Raku® Programming Language.

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

Thank you!