6.2.7 released! | pugscode.org/ <Overview Journal Logs> | pugs.kwiki.org | paste: sial.org/pbot/perl6
Set by autrijus on 13 June 2005.
autrijus $ ./pugs -BPIR -e '1,2,3; say "riverrun!"' 00:00
riverrun!
that's right. I'll add that sentence.
putter Hmm...
autrijus added. 00:01
putter is puzzled. my @a = (1,2); gives set_pmc error... where did I get a Syn "," error...
autrijus maybe because I implemented a infix:, stub when you're not looking :) 00:02
putter ah! :) 00:03
svnbot6 r4830, autrijus++ | * oops, bad copy/paste. fix the Syn "," stub. 00:04
autrijus now with 4830 00:05
I don't get why I can't assign a ResizablePMCArray into a PerlArray.
so sadly your example is still broken -- you may wish to take a look
putter hmm... 00:08
autrijus ok, I'd call that we're 17% there toward converting the Eval runloop toward the PIR one. that doesn't seem to be too unrealisitic :) 00:11
and now I sleep. see you tomorrow! & 00:12
putter good night! fun day.
ingy hola 00:28
nothingmuch hola
ingy nothingmuch: you know basic physics? 00:29
nothingmuch uh, probably not 00:30
but try me
ingy say I want to weigh a big couch...
me and a friend grab either end
and step onto two bathroom scales
would the couch weight be the total minus our total weights? 00:31
it seems reasonable
this is a real use case btw
nothingmuch yes
ingy my movers charge by the lb 00:32
nothingmuch a couch isn't that heavy though
ingy is that your final answer? ;)
nothingmuch i would worry more about appliacnes
well, think of it this way
ingy I don't have any
nothingmuch if you step on a scale with one foot
and then with two feet
does it show the same weight?
ingy just a bed and a couch
sure
but if I step one two scales 00:33
s/one/on/
nothingmuch okay, so in that case when you put your weight down in two places then it's the same amount of load
ingy I guess that is a simple enough empirical test :)
now I need a 2nd scale
nothingmuch just weigh yourself on one scale and both to be sure
ingy of course 00:34
nothingmuch but unless there are funny angles involved
that is, the couch is pushing you guys outward
ingy the movers estimate my possessions at 2200 lbs
I think they are way off
nothingmuch and the scales experience mechanical stress, and thus become inaccurate 00:35
for a bed and a couch?!
ingy so I'm going to call them on it
no there is more stuff, but not very much
anyway, there is no need to guess 00:36
nothingmuch anybody got a link to the pair programming chair picture 00:38
oh wait, first google hit when not using image search 00:39
putter ingy: you dont need the second person/scale. both ends of level couch take same load. multiply by 2. turn around to make sure couch isnt odd. 00:40
ingy putter: I can't parse that 00:41
I can *lex* it but... :\
nothingmuch putter: it's probably not going to be easy to get a couch balanced on a scale 00:42
and 4 legs + tilt will make it not very accurate anymore
putter ingy: one end of the couch goes on a scale, one on a prop. you can hold the scale side, for instance, standing on the scale, and subtract your weight later. 00:48
The prop and the scale are both carrying the same weight. 1/2 each. if the couch is level and symetric.
level you can eyeball (small variation doesnt matter). symetric, one arm is heavier than the other :), you can check by repeating the exercise by switching which end is held and which propped. 00:49
ingy putter: neat! 00:50
putter clear?
ingy thanks
putter :)
ingy Like Mom told me, "Always ask a Man" 00:51
(from MIT)
putter Good night all. 01:07
clkao ingy!
Wheet sial.org/pbot/11201 01:58
can i get some help please?
QtPlatypus What is the problem your having? 01:59
Wheet its in the listbox
at the bottom
the url i mentioned above 02:00
kungfuftr Wheet: #perl-help 02:01
QtPlatypus I think you have the wrong channel. That is perl5 code, try looking for a IRC channel thats just #perl
Wheet whats the difference between perl 5 code and perl 6 code 02:02
kungfuftr Wheet: perl6 doesn't exist yet
Wheet so none of you know perl 02:03
i see.
kungfuftr lol
svnbot6 r4831, autrijus++ | * add hw2005.txt back since people linked to it... 02:15
r4831, autrijus++ | ...and good URIs never break.
kungfuftr QtPlatypus: hehe, gave the guy a hand... clueless goit... used some code from random documentation and expected it to work 02:26
putter nopaste.snit.ch:8001/2956 is my attept of this evening to get array assignment working on parrot. fyi. 02:32
_meppl_ gute nacht - good night 03:40
mugwump pugs++ ~ # tab-completing :l 04:17
nothingmuch autrijus: when you guys are ready I can hack my smoke loop to start running parrot 05:19
svnbot6 r4832, chromatic++ | r4832@windwheel: chromatic | 2005-06-18 18:24:08 -0700 05:45
r4832, chromatic++ | attribute fixing
r4832, chromatic++ | r4835@windwheel: chromatic | 2005-06-18 22:12:48 -0700
r4832, chromatic++ | Fix argument passing to object attributes in methods by attempting to bind
r4832, chromatic++ | named values regardless of sigil and twigil.
r4832, chromatic++ | r4836@windwheel: chromatic | 2005-06-18 22:14:04 -0700
r4832, chromatic++ | Untodo some tests that now work.
r4832, chromatic++ | Unfortunately, test 40 now fails (and I'm not what should go in :todo<???>).
r4832, chromatic++ | r4837@windwheel: chromatic | 2005-06-18 22:15:47 -0700
r4832, chromatic++ | Untodo several attribute-passing tests that now pass. 05:46
r4832, chromatic++ | r4838@windwheel: chromatic | 2005-06-18 22:16:53 -0700
r4832, chromatic++ | Fix a handful of typos and thinkos, and now this test completely passes!
r4832, chromatic++ | r4839@windwheel: chromatic | 2005-06-18 22:17:40 -0700
r4832, chromatic++ | Unskip the whole test file, fix the test number, untodo some passes, and clean
r4832, chromatic++ | up the rest and all of this test file passes too.
r4833, mugwump++ | Add test for chained accessors, too 06:16
r4834, mugwump++ | Add (failing) test for polymorphic chaining 06:28
r4835, mugwump++ | Add (some failing due to inheritance problems, some passing) tests for $object.isa() 07:12
r4836, mugwump++ | Add test for alternate style of using a Tree - subclassing it
r4837, mugwump++ | Add tests for chaining off return value from subs
Alias_ seen autrijus? 08:26
jabbot Alias_: autrijus was seen 8 hours 14 minutes 16 seconds ago
Alias_ hrm
svnbot6 r4838, iblech++ | Usual svn properties and unTODOed one test of class/basic.t. 09:33
r4839, iblech++ | Pugs.Compile.PIR -- Made the END blocks run in the correct order (last-defined 10:23
r4839, iblech++ | to first-defined).
r4840, iblech++ | Emit.PIR -- Context forcers (&prefix:<+>, &prefix:<~>, &prefix:<?>, and &true). 10:52
clkao sees chromatic using svk 12:14
svnbot6 r4841, mugwump++ | Some tests for first section of S10 (Packages)
r4842, iblech++ | Perl 6 Prelude for -CPIR, putter++ for suggesting this. 12:20
r4842, iblech++ | * Added src/perl6/Prelude/PIR.pm with definitions of &prefix:<?>, &chomp, and
r4842, iblech++ | &chop, and removed these definitions from Emit.PIR.
r4842, iblech++ | * Modified src/gen_prelude.hs to use @*ARGS[0] as module name to output.
r4842, iblech++ | * Modified Pugs.Compile.PIR to load the Prelude and added the PIR/ directory to
r4842, iblech++ | the svn:ignore property of src/Pugs/Compile (because the Prelude.hs is saved
r4842, iblech++ | in src/Pugs/Compile/PIR/Prelude.hs).
r4842, iblech++ | * Modified Makefile.PL to generate the Prelude.
r4843, iblech++ | Added the usual svn properties to new t/packages/*.
mugwump hmm, is there a way to set the default svn props for "svn add", I wonder 12:22
iblech mugwump: There is: ./util/add-svn-props file
mugwump: There is: ./util/add-svn-props.sh file
mugwump ah, thanks 12:23
meppl guten morgen - good morning 12:34
svnbot6 r4844, mugwump++ | (failing) tests for %Package:: hash 12:45
mugwump ?eval sub x { [email@hidden.address] } x<1 2 3> 13:11
?eval sub x { [email@hidden.address] ~ $_ } x<1 2 3> 13:12
mugwump looks around for evalbot6
latter yields: *** Undeclared variable: "@_" 13:14
stevan mugwump: ping 13:57
mugwump hi 14:03
how are you this weekend, stevan ? 14:04
stevan mugwump: I am good, and you? 14:10
svnbot6 r4845, mugwump++ | Various (failing) tests for AUTOLOAD and related package subs
mugwump Excellent! A very fulfilling weekend for me. saw two live gigs :) 14:11
stevan nice 14:12
who did you see?
mugwump The Nomad (thenomad.co.nz) and Shapeshifter (another NZ artist)
stevan cool 14:13
hey, this is a paper you might be interested in
research.sun.com/projects/plrg/core-calculus.pdf
a friend of mine found it recently 14:14
mugwump looks interesting ... I'll have to pull it up at work ... I'm on lower order cycles atm 14:16
stevan cool
mugwump I was quite happy that ext/Tree/t/subclass.t mostly works...
Alias_ maptastic?
stevan I didnt see that
stevan has been out of the pugs loop for a while now 14:17
mugwump Alias_: maptastic ... a bundle of map-like things that I started a while ago
It's a maintenance release, really... 14:18
Just as a pre-req for Perldoc-0.10-Perl5
stevan mugwump: so are you thinking about perl6 perldoc? 14:19
mugwump sure, I think it should be quite natural to write
stevan: check out svn://utsl.gen.nz/trunk/Perldoc if you want to see what I'm porting to p6... same as Perldoc-0.10-Perl5, just u/l'ed to CPAN 14:24
silly cpan-upload script didn't work for me :(
otherwise it would have been up for hours 14:25
gaal|work mugwump, hey 14:27
please check out t/builtins/undef.t
it has tests for AUTOLOAD and friends...
has had them commented out
mugwump dang, I was grepping for (?+i:AUTOLOAD) :) 14:30
gaal|work :/
mugwump gaal|work: my tests test a bit more, I think. In particular, that the AUTOFOO method is only called once 14:31
gaal|work does it parse now? we can uncomment stuff now
mugwump however I didn't write anything for the AUTOSCALARDEF etc
its failure gets caught by try{ } now
gaal|work oh, goodie, the undef tests weren't meant to be comprehensive
(i mean the AUTO* tests in undef.t of course) 14:32
ah, i see why your parses 14:33
you say 'sub AUTOx'
mugwump oh. yes :)
gaal|work my tests say 'AUTOx { ... }'
i'm actually not sure it's valid
mugwump I did notice that the specification of AUTOLOAD as in S10 can't be specified via a normal signature, because it relies on the passed-in $_ being out-of-band to @_ 14:35
gaal|work ok, i think the thing to do is to add mutual textual links (i.e., comments :) between the tests in hte two files; and i *think* your tests should remove the sub and therefore need to be commented out for now 14:37
like you don't say try { ... sub CATCH }
mugwump that's because CATCH is a closure trait, not a sub 14:38
gaal|work brb - phone 14:39
mugwump dev.perl.org/perl6/doc/design/syn/S...ure_traits
gaal|work interesting - can't the various autothings be thought of as traits to the containing class? (i haven't thought much about this) 14:42
svnbot6 r4846, iblech++ | Prelude::PIR, Emit.PIR, Compile.PIR -- Implemented &sleep and made &exit call 14:44
r4846, iblech++ | all END blocks.
gaal|work oh -
t/syntax/symbol_table.t 14:45
also :)
mugwump aha. that's the problem with this test suite, I haven't been through it well enough, and there's ("not enough"&"too many") top level directories
gaal|work unless i'm mistaken, there wasn't a package/ when i added these tests 14:46
sorry :/
also the kwid mentions namespaces, symbol tables and symbolic refrerences, but not packages :-( 14:47
mugwump more tests ā‹™ less tests :) 14:48
^-- U+22d9 ASCII equiv: >>> 14:49
gaal|work mugwump++ # saved me the lookup :)
Khisanth it doesn't really look as good as the >>
gaal|work is reminded of the arrow paper with its weird typography
mugwump Unicode's great for arrows: āžµ āž­ āžø :) 14:51
Juerd Let's give every right pointing arrow a different meaning in Perl 6
We already have ->, =>, and ==> 14:52
gaal|work >
Juerd True
But there are many unicode arrows that are just waiting to be used 14:53
Hmmm
$value <= key?
gaal|work luckily we do not have -}
Juerd Oh, that's num le
gaal|work -] ?
Juerd |} # bow
------> # arrow
gaal|work -) ?
mugwump lots in page 21: ā†’ ā†  ā† ā‡‰ ā‡’ ā‡Ø ā‡¢ ā†¦ 14:56
Juerd ōæ½xE2FR ōæ½xE2Fōæ½xA0 ōæ½xE2F] ōæ½xE2GI ōæ½xE2GR ōæ½xE2Gōæ½xA8 ōæ½xE2Gōæ½xA2 ōæ½xE2Fōæ½xA6
Right, but of course. 14:57
gaal|work my client just renders lots of ōæ½xE2's and surrounding noise for that
mugwump mine too... encodings suck
Juerd Terminals suck.
gaal|work ōæ½xE2 - looks like a lambda, actually a hebrew letter
we had this conversation before :) 14:58
mugwump Well, with irssi-head, screen running in utf-8, the fairly high coverage 9x15 font, a nice XKB map, and SCIM for good luck, I can type these chars in easy ... ā˜® 14:59
Juerd I can't get the combination of ssh and screen working with utf-8 15:00
Otherwise I'd be over already.
ssh without screen works well
screen without ssh works well
But the combination doesn't want utf-8 :(
mugwump I found going ^A :encoding utf-8 utf-8 worked... 15:01
gaal|work is ....UTF-8 in the environment on both ends of the connection?
say in LANG
mugwump but when you reconnect to the screen it resets its idea of what your terminal (the second setting) is
so, I run uxterm with a LANG=en_NZ.UTF-8 on my laptop ... 15:04
ssh to my online machine, run screen with LANG=en_NZ.UTF-8, and irssi-head with the same
that seems to work really well - I can actually type in double width characters (etc) without weird display bugs 15:05
autrijus yo 15:19
Juerd Hi
autrijus <- just woke up
Juerd gaal|work: Yes 15:20
gaal|work: utf-8 works locally on both boxes, and through ssh
gaal|work: And it works through screen, locally, but screen in ssh insists on outputting crap for utf-8
I last tried months ago, though
Perhaps it's been fixed already. 15:21
autrijus mmm on O'Reilly's book sales graph, Perl is for the first time eclipsed by Javascript.
# www.flickr.com/photos/36521959321@N01/19788490/
(and looks like will soon be eclipsed by .net, and then by ruby)
gaal|work Juerd, ack
autrijus-san, good morning
i think you stashdotted flickr :) 15:22
Juerd autrijus: Yes. I've had a talk with a book salesman, and he said Perl sells less every year. 15:23
autrijus: They had had two copies of Perl 6 Now in stock for half a year without selling any of the two.
autrijus gaal|work: radar.oreilly.com/archives/2005/06/...f_ope.html too
Juerd: *nod* I think your p6l post is largely ccorrect. 15:24
Juerd Thanks
I couldn't resist any longer
autrijus resist? 15:25
Juerd Resist commenting further on the colon business 15:26
Or are you referring to another post?
gaal|work i think autrijus means the '15 years' post :)
Juerd Oh, that post. Forgot about that one. 15:27
It's a pessimistic, but unfortunately probably realistic view
gaal|work re colon - sorry about the necromancy, i'll try to keep it well-controlled with garlic - i think ./ is *wonderful*, $: is just fine, and $. and @. moreso are finger breaking :) 15:28
i say this after writing a bit of actual code
Juerd $: is fine typing-wise, and visually 15:29
But it completely destroys the last hopes for consistency
putter autrijus: any thoughts on which path to pursue...?
or something else, entirely on pugs-side...?
Juerd It would be fine if $:foo meant $.:foo, even if nobody actually typed $.:foo in full.
svnbot6 r4847, iblech++ | Prelude::PIR, Emit.PIR -- Refactored the context forcers (&prefix:<+>, 15:30
r4847, iblech++ | &prefix:<~>, etc.) and added &int, &lc, &lcfirst, &uc, &ucfirst, &infix:<x>,
r4847, iblech++ | and &pi.
gaal|work iblech the mad *.
iblech Thanks :D
Juerd mad asterisk?
gaal|work mad any(...)
iblech autrijus: You know the synchronisation of svn.perl.org <-> svn.openfoundry.org is still down? 15:31
Juerd Not all?
gaal|work porter, tester, etc. etc.er.
Juerd Or, better: mad none()
autrijus iblech: fixing 15:35
iblech autrijus: Thanks
mugwump āˆ—! now *that*'s a mad asterisk! 15:41
(U+2217 "ASTERIX OPERATOR" :))
er, ASTERISK ;-)
are we in preflight? or even taxi'ing? 15:43
svnbot6 r4848, mugwump++ | Quick port of the first 'layer' of Perldoc - the DOM tree, based this time on stevan++'s Tree module 15:44
wolverian mugwump: taxiing 15:45
(yes, it looks weird, but that's just how it is :)
mugwump fixes up previous test 15:46
s/test/commit/ # 3:46am localtime ;)
wolverian sometimes I wish cdparanoia wasn't quite so slow by default 15:47
but then again, I've many times been grateful that it really was.
mugwump jitter correction needs to be done...
wolverian yes.
gaal|work abcde++
someone should port that to p6 :) 15:48
wolverian gaal|work: I use Sound Juicer, but abcde looks nice if you prefer a commandline app.
mugwump gaal|work: system("abcde", @*ARGS)
wolverian (SJ is great because it just works - without messing with any settings.)
gaal|work i was happy with Grip when I ripped most of my collection, that and CDex on windows. 15:49
none of those required much messing around.
wolverian Grip is nice, but the interface is pretty crummy.
gaal|work actually a good getopt library is what we really need
wolverian (you can tell I'm a Gnome guy - keep it simple, stupid!)
gaal|work for massive portage
mugwump nothing wrong with grip, but lame-- bladeenc++ 15:50
gaal|work oggenc++ :)
wolverian oggenc++ # yes!
autrijus iblech: should be fixed now 15:53
iblech autrijus++ 15:55
svnbot6 r4849, iblech++ | * Testified more examples/s.
r4849, iblech++ | * Added t/packages to pugs::hack.
r4849, iblech++ | * Modified util/add-svn-props.sh to add the usual svn props to all files given
r4849, iblech++ | in @*ARGS, not only to first one.
mugwump *** No compatible subroutine found: "&add_daughter" 15:56
autrijus putter: ok, let me look at the @a=@b closer
mugwump heh, love the way that Tree::DAG_Node used feminine terms :)
putter the parrot clone fix, or at least one version thereof, turned out to be trivial. patch in progress. then I'll look again at assign. 15:57
autrijus oh cool. danke
mm optional args work now
mugwump & # time for a few hour's sleep methinks 16:06
Juerd autrijus: Why does -CPIR output to a dump.ast rather than stdout? 16:08
autrijus Juerd: there's absolutely no good reason for that.
Juerd Can I request that it uses stdout, for purity reasons? 16:09
svnbot6 r4850, mugwump++ | Make tests fail gracefully
autrijus sure can.
Juerd (And for scriptability :P) 16:10
autrijus ...done.
Juerd Thanks much
autrijus committing
no prob.
svnbot6 r4851, autrijus++ | * no longer use dump.ast -- -C now prints directly to stdout. 16:19
r4851, autrijus++ | Suggested by Juerd.
Alias_ autrijus: Question? 16:26
autrijus Alias_: Answer!
Alias_ Is there any reason whatsoever that we can't solve Module::Install's ExtUtils::AutoInstall problem by just always irc'ing it?
autrijus inc you mean? sure, please do so
just put a include in the M::I::AU 16:27
whenever you use it
Alias_ It's illogical that any single M:I user should be expected to add it as a dep
I mean just add it automatically for EVERY package
autrijus I think just use it when you use auto_install()
is enough -- otherwise probably is of no use 16:28
Alias_ blinks
one sec
let me check my notes
right, that
I don't even know what it does :/
heh
I just cut and paste
what does it actually do
autrijus it is a thin proxy to EUAI. 16:29
as I said, we should probably take that block away (in M::I::AutoInstall line 20)
and replace it with a simple include('ExtUtils/AutoInstall.pm') and require
can you try hacking it in?
I'll roll a release when you tell me so :) 16:30
Alias_ I'll take a look
autrijus danke!
Alias_ I've just had enough with people blaming me :)
"You didn't add a dep for ..."
<me> But it's not!
"We don't care. Make it work so we can CONSUME!"
autrijus *nod* please fix away. I can find historical circumstances as excuses
but it's simply Wrong at this point now
Alias_ encounters line 20 and throws new Exception::WTF() 16:33
svnbot6 r4852, autrijus++ | * oops, unbreak Main.hs
autrijus Alias_: Just Say Delete 16:34
fagzal g'day
Alias_ places one hand over his eyes and holds down the delete key
fagzal autrijus: I've seen this ==> www.flickr.com/photos/36521959321@N01/19788490/ Do you by any chance have a link to this graph with higher res? :) 16:36
kungfuftr fagzal: um... have you tried asking o'reilly? 16:38
Alias_ That looks like something from gnat's talk
He has all sorts of book trend stuff
fagzal kungfuftr: not yet... 16:39
Alias_ not good to see Perl on the decendancy :( 16:40
I mean, to be overtaken by JavaScript...
fagzal Alias_: Yep...
I
I'm OK with anything else but JavaScript :)
autrijus Alias_: it's updated
kungfuftr javascript++
autrijus javascript++ # sane, minimal, nice, evolving, active 16:41
Alias_ btw
fagzal Incompatible? ;-)
Alias_ TorgoX, Theory and myself are starting up a JavaScript CPAN if anyone wants to help
autrijus er, javascript for browsers is another matter altogether :)
kungfuftr fagzal: that's an implementation problem, not a language problem
autrijus it's largely a DOM problem.
I'd love javascript if my brain has been formatted to OO...
fagzal autrijus you're right...
kungfuftr ecmascript 262
Alias_ ali.as/CESAN/ for overview and current status 16:42
autrijus ...however I've tasted something stronger, so I merely think javascript is nice now :)
fagzal Alias_: JavaScript CPAN? Woa...
nothingmuch evening
where is it in use other than that?
Alias_ It's as much to embarrase python as it is to get a nick JS repository :)
nothingmuch i think most people hate javascript because of browser incompatibility in the end of the 90s, right?
kungfuftr can't believe he burnt the bottom of his feet... OW!
fagzal nothingmuch: yep 16:43
Alias_ fagzal: Anyways, check out that URL. Volunteers welcome
kungfuftr nothingmuch: that and most people who've done javascript are worse than php developers
nothingmuch yup
fagzal Worse than php developers? Strong words, those are :))
nothingmuch this one time I saw some clean javascript 16:44
kungfuftr fagzal: totally true though
Alias_ At least JavaScript people tend to only cut and paste
nothingmuch i was like "wow, cool, javascript I can read"
Alias_ Which keeps the diversity of crap to a minimum
I caught my minion doing it yesterday
He actually nicked two .js files off of Apple's website, so he could do a popup 16:45
fagzal Typical.
Alias_ I was shocked, SHOCKED I tell you
He's already gotten a Dreamweaver ban, this only compounded it
kungfuftr copies and pastes some perl6
autrijus hm, on that graph, I don't see ruby anywhere
Alias_ it's the bottom one
trending upwards 16:46
autrijus oh? where is python then.
fagzal Alias_: Say no more :-) I could not explain the difference between $var{key} and $var->{key} to one of my developers for like half a year... I mean I could explain, he just did not get it :-)
autrijus I was sure that was python
Alias_ autrijus, oh wait... see those pixels scraping along the bottom line
it ducks above ever so briefly
autrijus ah. right.
aw.
fagzal Alias_: I mean he once got it... then forget :-) 16:47
autrijus that's like, one book at that month
Alias_ autrijus, well there's only one book. Maybe now there's another book it will help
All that RAILS RAILS RAILS AJAX SPOOOOOOOGE will have to do something...
autrijus oy... :)
Alias_ After Maypole and Catalyst and (my stuff) Rails is... well... quaint 16:48
fagzal, he couldn't get pointers?
Juerd re js: I write my javascript (for browsers) from scratch!
:)
Alias_ Juerd, as do all good developers :)
Juerd, wanna help out a little from time to time on CESAN? 16:49
Juerd Well, I copied that keyboard mapping tool for dvorak.nl
fagzal Alias_ Unfortunately. I told him "like in C, you now, *ptr". he-he... not helping.
Juerd But it was GPL and good style :)
Alias_: What is CESAN?
Alias_ CPAN for ECMAScript
Me, Torgox and Theory
Juerd What help is needed?
I have very little time to spend programming|writing
Alias_ TorgoX is bringing POD, Theory has got Test::Simple and friends ported...
ali.as/CESAN/ has status details 16:50
fagzal Anyways... I am into doing some marketing form Perl. Both 5 and 6. I need a bigger image of this book-selling stuff to scare people :)
s/form/for/
Alias_ Juerd: We could use some admin'ed server space... 16:51
Or even just a recruiterererer
Juerd I'm currently severely out of budget to sponsor server space 16:52
Alias_ Well, I has hoping company or something
Juerd And I'm afraid I can't be of much help by doing things either
Alias_ but that's ok
Juerd I == company
fagzal Alias_: what kind of server space you need?
Alias_ As I said, if we get this rolling, there's one fairly obioooooous JavaScript user that might help 16:53
autrijus fagzal: I'm sure you can simply zoom it a lot, use it as a background, and trace the lines using gimp or somethiing :)
Juerd Alias_: A machine of feather's proportions is a huge investment, worth it only because I expect to make money using Perl 6 in the long term
Alias_ fagzal: It would be nice to have a properly adminned master FTP site
feather?
fagzal autrijus: arrgh :-)
Juerd Alias_: Do you not know about feather?
autrijus there may already be filters that does this :)
fagzal Alias_: I have a few servers... can't give you a dedicated host, though.
Alias_ Juerd, I'm not a big Perl 6 person... 16:54
fagzal: I don't expect to need a lot of either space or bandwidth for a while
Juerd Alias_: Feather is the ongoing Perl 6 BOF session, in the form of a sponsored linux box
autrijus nice explanation, Juerd.
Alias_ We won't let people point their web pages at CESAN
autrijus "it's the party that just go on... and on"
Juerd autrijus: :)
Alias_ so it's what?
A box that does...
Juerd Alias_: If CESAN has to do with Perl 6, it can run on feather 16:55
Alias_ it doesn't :(
Juerd But only if -- and I'm not sure Perl 6 is that ready for production yet :)
autrijus goes back to hacking parrot, as if on cue
Juerd Alias_: It's a box that does... absolutely nothing if the users don't give it things to do
kungfuftr heh
Alias_ of course! 16:56
Juerd Alias_: It's used for communication (irc and email), compilation, testing, hacking, ...
Alias_: By people who lack bandwidth or linux box
Or who wish to use feather for other reasons
Any Perl 6 developer can get an account
if you're not a Perl 6 developer, you can get an account too, as long as you'll be contributing (bug reports, for example) 16:57
The box is also great for one-on-one chats with talk :)
nothingmuch why is israeli public transportation so useless? 16:58
Alias_ Well I have some code in Perl 6
nothingmuch my flight is at 06:00, and the last way out of where I live (1.5 hour drive) is 20:00
Alias_ Mostly the comments left over after Config::Tiny and Alg::Dep were ported :)
I'm sure I show up in the blame log somewhere 16:59
Juerd Well, if you need access to do Perl 6 things, just let me know by email and I'll set up your account 17:00
autrijus hm, Nemerle looks nice. 17:02
even nicer than Boo, in fact, but not as ducky 17:03
Juerd autrijus: Can you inspect feather:~juerd/pugsmake.output? 17:04
autrijus Juerd: weird. it's recent? 17:05
Juerd Current
autrijus I'll try building with p5embed here
and see if I can duplicate. meanwhile turn p5embed off?
Juerd Will do
Do you want a snapshot of my build dir?
Just in case?
feather:~juerd/pugs_p5embedbug.tgz 17:06
autrijus 'k 17:07
fagzal K, by all, have fun. 17:11
Alias_ autrijus, so, do I have this correct. All I'll need it 17:15
$self->include('ExtUtils::AutoInstall');
require ExtUtils::AutoInstall;
The first instructs it to be bundled at packaging time, the second assumes it got packaged
autrijus yes. 17:16
Alias_ and I presume you have to add it as an explicit dep for M:I itself..
autrijus yes. 17:17
in fact I'm thinking about bunding the two into one single dist.
Alias_ It's yours as well?
autrijus eh, sure.
all this toolchain is from a single crazy person :)
brb... 17:18
putter autrijus: fyi, update: have set_pmc working. now dealing with implementation. first cut didnt like the flavor of arrays being compiled. puttering... 17:21
Alias_ BTW, it looks like I just snagged cesan.org 17:30
:)
autrijus putter: cool! meanwhile I found "down with lambda-lifting" at research.microsoft.com/~emeijer/ 17:37
which contains a nice algorithm for compiling to register machines that will totally eliminate useless stack pushing and popping
autrijus prases the research community
praises, even
Juerd microsoft?
autrijus yeah, MSR 17:38
...and I just mailed emeijer to see if there's a free software implementation of Emit.CLR that we can use.
(in haskell, that is -- meijer is the Mondrian/Haskell.net guy)
putter Ok, this works ./pugs -B PIR -we 'my @a = (10,11);say(join("x",@a));'. But (10) ends up a PerlInt, and () a null something. How much of that should array.pmc or perlarray.pmc deal with? 17:39
ok, btw, woot!
autrijus putter: the PerlInt is fine 17:40
() is a null?
putter err...
autrijus also, please check in your work :D
putter "Null PMC access in elements()" I havent looked at it closely. 17:41
re check in, its a parrot patch... so finish, test, nopaste, someone else commits... 17:43
autrijus mumbles something about asking for committership
dinner, brb 17:45
putter re lambda lifting, havent read it, but sounds nifty. re research community, isn't civilization great :) 17:50
gaal feather users: cd pugs ; ~gaal/ctags -R # haskell tags too
nothingmuch gaal: now patch cscope 18:03
gaal shto eta cscope
nothingmuch it's like ctags, only nicer 18:04
in vim+ctags, go on a symbol
and hit ^\+g
you jump to it's def 18:05
^t goes back up the stack
gaal that sounds like tags?
nothingmuch this way it's very very easy to trace calls
dudley \
gaal how's it different
?
nothingmuch it's got a pretty curses ui
gaal ^] and ^O for taags.
nothingmuch and support for more types of searches
you can search for use of a symbol 18:06
and you can grep an indexed DB
Khisanth cntrl+\ + g? that looks like emacs! :p
nothingmuch it does =/ 18:07
but it works in insert mode, which makes it sort of useful
gaal what languages does it support other than c and fuzzily java?
nothingmuch c++
and, err, that's basically it
which is why you need to patch it 18:08
gaal did you see the haskell patch for ctags? :)
nothingmuch ;-)
not the source
gaal it was *very* short. 2 1/2 regexps short.
nothingmuch okay okay ;-) 18:09
gaal heh, weird p5ity. did you know what $SIG{foo} = "bar" does? (note the string) 18:22
it sets the hander for foo to the _string_ "main::bar", and will look for that at SIGfoo time.
autrijus right. 18:23
unless it's IGNORE or DEFAULT.
which assume the magical behaviour.
gaal however! you can give a fully qualified name, in which case no defaulting is done
(yes)
autrijus dwim, eh
gaal no dwid woulda been to use the caller's package!
*dwim 18:24
autrijus true.
gaal *now* all we need is to find an obfuscation for which the string "main" is useful, and we have a way to generate it! :-p
autrijus but this being jenga, it's too late to fix it :)
gaal i know a language that should fix it :) 18:25
autrijus and incidentally you have control over that language's prelude too
:)
gaal however in HashClass instances I'm not sure I can find the env, so i don't know :(
don't know how 18:26
autrijus huh?
env <- ask
just ask :)
curpkg <- asks envPackage
gaal but i'm not in the Env
oh!
autrijus you're in Eval
gaal *Eval* monad, right
silly me
autrijus eval always keeps track of env :)
gaal of course
i am dropping packets :(
autrijus :( 18:27
gaal heh heh, p5 bug: (well, footshoot opportunity): $SIG{foo} = "have:colon" 18:35
so, to see if a string contains a colon, elem ':' str is okay, right?
stevan autrijus: I will start the changelog now :) 18:46
gaal figuring out which signals the system supports at runtime is kinda hard. i'm going to to punt and exploit perl5 at build time for this (you don't wanna know how *it* got that information :) 18:52
gaal reads the "poem" for PugsConfig.pm -> Pugs.Config and giggles uncontrollably 18:59
svnbot6 r4853, iblech++ | Emit.PIR, Pugs.Compile.PIR -- Large Haddocking update.
stevan gaal: where is your l33t interpreter? I can't find it 19:18
gaal examples/obfu/l33t.p6 19:19
stevan ok
thanks
gaal my pleasure ;)
stevan any idea on the current test count? 19:24
iblech stevan: 7562 19:25
gaal btw, changeloggin tip: svn log -v
stevan iblech: muchas gracias
what is -v
gaal gives also files changed
stevan ah
very nice
gaal and also, use -r4500:HEAD to make the log quicker to fetch 19:26
for some value of 4500
stevan yeah that is what I do 19:27
(and why i started including the r numbers in the changelog :)
gaal :))
stevan dates are far too wide a scope for this project
gaal i'm getting careless =<< tired. i'm afraid i won't be flying with you guys tonight... 19:28
"you guys".. any female pugs hackers? 19:30
19:31 crysflame is now known as a[] 19:32 a[] is now known as crysflame
gaal anyway, i'm off to bed, see 'ya! 19:33
stevan thinks that is a "no" 19:34
nite gaal :)
gaal night :)
stevan autrijus: changelog is in 19:37
its a little sad
I am a little out of the loop these days 19:38
svnbot6 r4854, Stevan++ | Changelog for 6.2.8 (this needs MUCH work as I am not very clear on this PIR stuff) ... sorry, I have been outta the loop for too long, damn that $work :P 19:46
autrijus stevan++ 19:55
autrijus has burned too much time reading papers
but it's compensating my lack of formal education, and hence mandatory ;) 19:56
stevan lack of formal education ++ :) 20:05
I have been reading too many meta-class papers myself 20:06
I am seeing everything as an abstraction of an abstraction of an abstraction .....
jql once you get "educated" in CS, everything becomes so abstract you don't need to implement it anymore 20:08
you can just write papers. :)
stevan why write code, when you can write proofs :) 20:09
jql esoteric proofs are always the best. Implementing a turing machine with compile-time C++ templates sounds fun. 20:10
svnbot6 r4855, autrijus++ | * get HTTP::Server::Simple compiled to PIR. 20:37
r4855, autrijus++ | * initial work at replacing Pugs.Compile.Parrot altogether.
r4855, autrijus++ | * generalise metacompuatational compilation using FunctorM
r4855, autrijus++ | so one can compile [x] and (Maybe x) easily (and maybe more
r4855, autrijus++ | in the future).
r4855, autrijus++ | * handle `$x += 3` style post modifiers.
autrijus why not just prove your papers can exist, then. :) 21:04
21:07 knewt___ is now known as knewt_
mugwump āˆƒ paper ā†’ āˆƒ academics 21:21
putter :) 21:24
hmm... unicode text messaging.... eep.
jql too meta, even for CS. leave that to the philosophy Ph.D's 21:27
although philosophy leaks into CS via AI and the semantic web... ontology my ass. </rant> 21:29
autrijus actually ontology makes lots of sense in closed settings... 21:30
...but I'm not sure of how well OWL-style vocabularies adapt
putter flora2 is kind of nice. pity the "can mixin nondeclarative stuff" hasnt seemed to quite work. 21:33
putter dreams: prolog on p6, tabled prolog on p6, (working)flora2 derivative on p6, ahhhhhhhh.... 21:34
mugwump yay! C-q 2 6 2 f RET will be the way to enter a ā˜Æ in upcoming emacs 21:43
anyone have feet in the vim camp and want to push a similar feature there? 21:44
crysflame a ? ?
mugwump \x{262f} - YIN YANG symbol :)
putter peers closely at a fuzzy round smudge... ahhhh 21:45
mugwump I'm getting very used to the 9x15 font which has a huge coverage of unicode... that and good ol
good ol' fixed 21:46
so, it's better than the current sequence to enter it: C-q 1 1 1 0 6 5 6
(yes, that octal number is 0x491ae in hex) 21:47
crysflame octal unicode.. owowow 21:48
autrijus ow indeed
putter hmm.. I get the peace sigh for 1 1 1 0 6 5 6
autrijus *Main> runPIR "my @x = 100..200; say @x[1]" 21:49
101
putter++ # for making it possible
(needs trunk parrot)
crysflame neat
autrijus ponders how to deal with "1..."
mugwump that's not unicode btw, that's an emacs codepoint :) 21:50
crysflame whee, another encoding system to convert to/from 21:51
putter FC3 GNU Emacs 21.3.1. perhaps I broke my fonts at some point.
mugwump putter: try 9x15 from the shift-leftclick menu... 21:54
svnbot6 r4856, autrijus++ | * support for range and array assignment.
putter mugwump: following your suggestion (thanks:) that's what I tried it in... 21:55
Juerd justatheory: You know Perl is a rather practical language, right? 22:01
mugwump putter: my 9x15 (-misc-fixed-medium-r-normal--15-140-75-75-c-90-iso10646-1) is about 88k gzip'ed, from xfree86 4.3.0 fwiw
justatheory Juerd: In theory, yes.
kungfuftr Juerd: nope, it's just pathologically eclectic 22:02
Juerd kungfuftr: Perl is nothing "just".
kungfuftr Juerd: LIES!
autrijus er, I always thought it stands for Polymorphic Existential Recursive Lambdas
putter mugwump: thanks :) 22:03
Juerd Pizza Eating Rocks Lovely. 22:04
autrijus PHP Eiffel Ruby Lisp.
putter ;) 22:05
Juerd pugs: Pizza sind Ueberhaupt Geil und Super
s/Pizza/Pizza's/
putter Eiffel we could do, Erlang... hmm... 22:06
autrijus Prolog Erlang Rebol Lua
svnbot6 r4857, autrijus++ | * Pair and => support
autrijus (much farther)
crysflame autrijus++ # ha 22:07
kungfuftr autrijus: you should at least prefer Python over PHP... tho both are evil 22:08
autrijus kungfuftr: I'm trying to be orthogonal... python is too close to ruby and eiffel
mugwump prolog!
kungfuftr =0) 22:09
mugwump sees autrijus is 2 minutes ahead of him :)
svnbot6 r4858, putter++ | Added modules/Grammars/MatchX.pm - anyone want to help write a Match class? 22:45
putter It's just a shell. Then again, the class is largely a shell. But it is supposed to have a boolean value, an int, a string, an array, and a hash value. I'm not sure how much is currently possible with pugs, or even if the spec is consistent. 22:47
autrijus woot. 22:48
*Main> runPIR "my $x = [1..10]; say $x[3]"
4
putter (pointer to spec included in file) But if someone wanted to fiddle...
putter is trying to load shed.
!:) 22:49
(ninja smiley?)
autrijus XD 22:50
ingy XB
svnbot6 r4859, autrijus++ | * anonymous arrays; references; etc 22:51
autrijus torked tongue?
er, forked
ingy bucktoothed
putter lol
# torqued
obra re 22:52
ingy Squinting Beaver
putter packman ghost
ingy autrijus: I don't think 'D' is a tongue 22:53
autrijus and you'd be correct 22:54
ingy it's more like that smile with a big orange wedge in your mouth
putter autrijus: a while back you said "grammars are simply _" packages? modules? "and rules are simply _"?? 22:57
autrijus classes methods 23:02
so we have .goto now for PIR. yay ;) 23:16
svnbot6 r4860, autrijus++ | * PIR - `&foo.goto` support. trivial, via tail calls.
putter bye all. 23:20
autrijus bye putter!
chromatic Hello. 23:28
autrijus greetings chromatic-san. I see you have finally conquered the binding monster. 23:29
chromatic Yes, thanks. 23:30
Something's weird still in that area, but I think the named bindings finally work correctly.
autrijus yup.
I, on the other hand, is still struggling with segfaults.
chromatic Parrot? 23:31
I'm looking at DESTROYALL now in Pugs. You have my sympathies.
obra reads the amazon summary of his book
"It's primary purp0se is to maintain a history of events and the current status."
autrijus yeah, parrot.
segfaulting is simply parrot's way of showing affection.
obra They've translated it into l33t!
autrijus wouldn't that be purps0e?
obra :) 23:32
chromatic autrijus, do you know any reason why Pugs doesn't seem to call DESTROYALL for objects? 23:34
autrijus chromatic: I do not know. try scattering random 23:36
liftIO performGC
or even export it as a Pugs::Interansl::performGC
to see if it helps
chromatic I'm seeing GC on Hash::Const objects, but not objects I create. 23:38
autrijus maybe they are somehow visible. 23:39
is this a new bug?
autrijus ponders how to do the "eval" primitive.
chromatic I haven't checked in several days, so I don't know how new it is.
autrijus but it was once working 23:40
iirc.
chromatic Yeah, my destructor tests used to pass.
ingy autrijus: soon I will have some sporky new goodness for you 23:43
autrijus chromatic: I wonder if you can triage revisions a bit... I didn't recall anything breaking it
chromatic I'll take a look.
autrijus but I had not kept close track over the recent prelude work. 23:44
ingy: ooh.
ingy Imagine a single slide of Perl code, that changes hilights as you hit enter
without needing to duplicate the code in your slides file 23:45
autrijus o/~ Imagine there's no dupe code / it's easy if you try o/~ 23:47
ingy o/~ papa don't take no dupe o/~
ingy needs to switch autrijus from John Lennon to James Brown 23:48
chromatic: Test::Chunks is now Test::Base. I'll write the article after YAPC, because I'm giving a talk on it now :) 23:50
autrijus $ ./pugs -BPIR -e 'my @a; @a[0] = 3; say @a' 23:56
3
Odin-LAP The 'Imagine' lyrics are quite interesting. Anarchist, even. 23:58
svnbot6 r4861, autrijus++ | * assignment into arrays.
autrijus the melody hides the thoroughly revolutionary lyrics :) 23:59
autrijus ponders Test::Base::All::Yours

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!