6.2.10 released! xrl.us/hxnb | geoffb's column (/. ed): xrl.us/hxhk | pugscode.org | pugs.kwiki.org | paste: sial.org/pbot/perl6 | www.geeksunite.net
Set by autrijus on 10 October 2005.
stevan_ this is borrow right from CLOS, Ruby, etc 00:00
Class must be something right?
unless you assume that Class springs forth from the runtime fully formed
However that means that Class is a mystery to the other elements of your system (unless you explicity expose its internals) 00:01
if it helps,.. rename Class to Klass
Klass is just a class which describes classes (hash of methods, hash of attributes, array of superclasses) 00:02
if you create Klass, then you can bootstrap the rest of the system in 00:03
its actually very simple once your remove the nomenclature baggage
00:07 stevan_ is now known as stevan
stevan Juerd: as for how a subclass can also be an instance,.. it is an instance of Class, which has the ability to have subclasses 00:09
think of the basis of the system as being an instance type, instead of a class 00:10
Juerd: the MiniMetaModel.pl in the docs/ dir of the Perl6-MetaModel might be helpful in explain this 00:11
Juerd: this might help too en.wikipedia.org/wiki/Metaobject 00:22
00:28 mugwump is now known as samv, samv is now known as mugwump
Aankhen`` Um... question: what is this "Pugs.cabal.in" file that util/build_pugs.pl uses, and where/how do I get it? 07:14
obra you know what cabal is, yet? 07:18
Aankhen`` Not really.
obra It's the Haskell makemaker equivalent
misc something that doesn't exist ?
Aankhen`` Ah.
Does this mean I'm going to have to download something?
integral Aankhen``: Cabal comes with GHC. Pugs.cabal.in is just a source file, it's in the pugs svn, and should be in pugs distributions 07:21
Aankhen`` Oh.
In the root directory of the pugs/ tree? 07:22
integral it's actually not the final file Cabal uses, that's Pugs.cabal
Aankhen``: yep
Aankhen`` Strange, SVK doesn't seem to sync it for some reason...
integral weird 07:23
Aankhen`` Hmm.
Maybe I should just delete the entire local branch and create a new one. :-\ 07:24
Wha? It's in the local branch, but it's not being pulled down. 07:26
Neither is Pugs.sln 07:27
Bleh.
Aankhen`` deletes the checkout dir and tries again.
Cool, that worked. 07:32
OK, Pugs still won't build. 07:40
I am cursed. 07:41
pasteling "Aankhen``" at 59.144.181.34 pasted "Build fails" (14 lines, 633B) at sial.org/pbot/13804 07:42
Aankhen`` (in case anyone's interested :-)
rafl Aankhen``: What ghc version are you using? 07:49
Aankhen`` 6.4.1 07:50
rafl No idea. Seems to be ghc-related, though. 07:51
Aankhen`` :-\
I seem to be the only one with this problem.
Anyway.
BBL.
09:09 |cnhackT| is now known as cnhackTNT
wolverian is autrijus's Pugs slide 6/55 missing a [+] or so? 10:39
masak wolverian: i count only 46 slides in "Introduction to Pugs" 10:42
wolverian masak, pugs2.xul, that is 10:47
autrijus yeah, it is. 10:48
wolverian autrijus, in any case, nice presentation!
autrijus thanks!
wolverian autrijus, do you know if that @fib syntax is actually official yet? do we need to mark it lazy somehow?
autrijus (@fib Y @fib[1...]).map(&infix:<+>) 10:49
I think range is lazy by default, that's official
and binding respects that too
wolverian oh, right
and [+] is syntax level so that's too eager here?
ods15 so... whats pugs? perl6's bug tracking system?... 10:50
autrijus wolverian: no, [+] will yield one single value
wolverian autrijus, oh, duh. sorry. 10:51
ods15, are you asking seriously? :)
ods15 wolverian: not atm, but i did when i first saw it
wolverian ods15, ah. was it easy to learn what it really means? 10:52
ods15 uhhh...
what does it really mean
autrijus Perl6 User's Golfing System 10:53
wolverian ods15, it's the plural of a breed of dogs, obviously
autrijus but that's a backronym.
rafl autrijus: Did someone do recordings of your talks?
Preferable video recordings. :-)
ods15 you should've called it ACME PUGS
autrijus I'm going to present the haskell talk in... 30 mins
theorbtwo ods15, I think Acme::'s goal is just to be fun. 10:54
pugs is supposed to be fun and useful... with the emphasis on the fun.
ods15 wtf is Acme:: ?
rafl autrijus: Turn on your laptops microphone during the talk! :-)
theorbtwo Anyway, while pugs' primary author was previously well known for his work on a BTS, pugs isn't a BTS, and doesn't even use a formal BTS. 10:55
wolverian ods15, CPAN's "fun" namespace
ods15 heh didnt know that 10:56
rafl What's a BTS? I only know it as Bug Tracking System.
ods15 i was refferring to the cartoon mnopoly
rafl: thats exactly what he meant...
theorbtwo Yeah, Bug Tracking System. I was refering to RT.
If you didn't know what Acme was, at least in that sense, then why did you think it should be ACME PUGS? 10:57
rafl Ah, OK. Thought of the Debian BTS which is simply called 'BTS'.
ods15 theorbtwo: i was refferring to the cartoon mnopoly 10:58
anyway, nm, i'm just brainfarting.. bbl
theorbtwo rafl: I think it's been named debbugs, but everybody in the debian world just assumes it's the one you mean when you say "the BTS". 11:00
rafl Ah, right. The package for it is called debbugs.
pdcawley Dunno. 11:22
nothingmuch I agree! 11:29
wolverian tentatively. 11:30
gaal rafl, hi. 'pbuilder build' doesn't do anything on my box :/ 11:39
rafl gaal: Please show me its output. 11:54
gaal rafl: no output at all. it returns to the prompt pretty quick 12:08
rafl gaal: May I see the output of strace -f ?
gaal sh -x show that it sourced some extra files, but then nada
sec
feather:~gaal/public_html/pbuilder.out 12:11
(i need the cvs version of ghc, not 6.4*) 12:13
rafl Hm, maybe your pbuilderrc is broken? That's the last file that is read before it exits.. 12:14
gaal doesn't look broken... 12:15
i don't have to run this as root, do i?
either way it doesn't work
rafl you do. 12:16
it uses chroot which normally gives you EPERM for non-root users.
gaal same behavior when run as root. 12:18
rafl Well, no idea. pure+reinstall?
What's yor commandline?
gaal where does the local mirror get created?
pbuilder install ghc-cvs*dsc 12:19
er
s/install/build
rafl Created a chroot before?
gaal pbuilder create --mirror my_mirror
yes
rafl pbuilder create --distribution=sid --mirror ... etc.
Hm, no idea then.
gaal though that didn't say much either :)
rafl The chroot is put into /var/cache/pbuilder/base.tgz.
Well, reportbug pbuilder 12:20
gaal hmm, that doesn't exist. okay, purging as you suggest.
autrijus yay
talk over
gaal whee
autrijus well received
rafl Recordings? :-P 12:21
autrijus pugscode.org/euroscon/haskell.xul
clkao autrijus++ 12:22
gugod slide-making technology improved every year
gaal wow, many changes since 24 hours ago :)
clkao gugod: so uyou need to think about next year now 12:23
autrijus gugod: Spork::Formatter::Takahashi bah
gugod: just take the resize algorithm in takahashi.js
gaal pugscode.org/euroscon/haskell3.xul doesn't load? 12:27
hmm does now. 12:28
rafl: apt-get dist-upgrade seems to have helped: pbuilder create now does something :) 12:41
xinming autrijus: 12:51
pugscode.org/euroscon/haskell.xul
How do you made this? :-S
impressive...
gaal xinming: view source and scroll down 12:52
xinming gaal: hmm, I read that. But... I can't find the tags I know... - -b 12:53
<html> <head> </head> <body> </body> </html> that what I thought a web page must have... :-S 12:54
can't open in IE. :-P 12:55
brother xinming: but it isn't html
xinming can be rendered in Firefox. But not in IE. for IE, it will ask you the location to save... :-( 12:56
gaal xinming: it's XUL, you have to use mozilla or ff
xinming gaal: Oh, Ok, I think I understand.
kgftr|konobi gaal: sounds like there isn't a <? XML declaration at the top
oh... there is.... hhhmmm... 12:57
ah... server needs to tell IE what filetype the document is
xinming M$ sucks. :-P 12:58
cm there should be a doctype as well 13:01
rafl gaal: Great. 13:19
svnbot6 r7691 | iblech++ | * Usual svn props. 13:24
r7691 | iblech++ | * t/builtins/strings/trans.t: unTODO
r7691 | iblech++ | * t/unspecced/evaluation_order.t: (die "first").method(die "second")
r7691 | iblech++ | * t/oo/delegation.t: Minor fix (s/handles join/handles "join"/).
r7692 | iblech++ | PIL2JS: Prelude::JS::Str: Copied src/perl6/Prelude.pm's implementation of
r7692 | iblech++ | $str.trans, but doesn't work 100% yet (dunno why; will investigate).
xerox Do you remember where are the pugs videos? 13:41
autrijus it's pugscode.org 13:44
"OSCON movies"
xerox Right! Thanks.
autrijus :) 13:45
stevan_ our pasting service (sial.org/pbot/perl6) is gone 14:48
I think we should switch to a LISP based one (paste.lisp.org/) :) 14:49
xerox Lisppaste il great.
If people agree I could try getting it. 14:50
tehmaze perl6.pastebin.com/
;)
stevan_ *cough* PHP *cough* 14:51
xerox Vote?
stevan_ (+ 1 (+ 1 LISP))
The lisp paste allows you to annotate too, which is very nice 14:52
14:52 stevan_ is now known as stevan
xerox ...or, given the anarchistic plan, it's ok to get it now :-) 14:52
stevan xerox: our old one doesnt work,.. so lets get it
xerox The administrator is currently away, I'll leave him a message.
stevan besides,.. we have invaded the haskell community,.. Lisp was bound to be next :)
xerox++
Khisanth pastebin.com is quite possibly the the crappiest paste site around >_> 14:53
stevan xerox: if the Lispaste admin adds us, just change the channel topic to reflect please 14:55
Juerd tehmaze: How dare you.
xerox: Just get pastebot.
xerox: pastebot.sf.net
You can run it on feather if you like. 14:56
tehmaze Juerd: it works
Juerd tehmaze: It's PHP.
tehmaze so what
xerox stevan: will do.
Khisanth it's not even the PHP that makes it crappy :)
Juerd tehmaze: And in that case, whether it works is highly irrelevant.
stevan tehmaze: if we cared about something "just working" we wouldn't be here would be ?
tehmaze Juerd: so you won't buy anything of off an auction site too, if it's in php? 14:57
Juerd tehmaze: That's another thing. This is a Perl 6 channel.
stevan xerox: excellent, thanks
tehmaze ok, what *ever* ;)
Juerd tehmaze: Things we use must be written in Perl 6 or equally ranked language.
tehmaze Juerd: lots of my collegues will be attending amsterdam.pm btw
Juerd tehmaze: Who will?
tehmaze Juerd: Liz, Jouke, John, Bert
Juerd You work for Bookings? 14:58
tehmaze right
Juerd Ah, hm
Which John is this?
I don't think I know a Bert
tehmaze Loembang
Juerd I don't think I know him
gaal Um, new sigil!
tehmaze ok
Juerd gaal: Wha?
tehmaze Juerd: oh and Giel probably too
Juerd gaal: We have glyphs left?
tehmaze Juerd: Bert is my sysadmin-collegue 14:59
gaal Sounds like April Fools, right? How do you type the cent sign?
cm Ā¢T xD
Khisanth gaal: there are still symbols left on the keyboard?
Juerd gaal: wtf?
gaal p6-l
cm www.nntp.perl.org/group/perl.perl6....uage/23630 15:00
tehmaze it's not the 1st of april yet, Larry 15:01
stevan eak!
gaal hey! what happened to sial.org!? 15:02
stevan gaal: it's nothingmuch's fault I think
xerox The lispb0rgs are assimilating them.
gaal stevan: you like to blame him for everything huh!
stevan gaal: sure, why not :) 15:03
Khisanth gaal: ctrl+shift+a2 if you are using a gtk app :)
gaal Khisanth: thanks :)
(but i'm not :) 15:04
Juerd replied with a few questions
gaal what other pastebots are there around?
Juerd Why are dollars instances of cents? :)
gaal: pastebot.org
gaal: Eh, pastebot.sf.net
stevan gaal: alt-4 on OS X
Juerd gaal: You may run it on feather if you wish. It's a Perl thing. 15:05
Khisanth anyone know why the cent sign is not on a qwerty keyword?
Juerd Khisanth: Which QWERTY is that? There are several dozen. 15:06
US QWERTY? Because this keyboard has only ASCII characters.
Many other QWERTYs => because the keyboard's country doesn't use cent signs 15:07
Yet others => because in practice, 'c' works equally well
Khisanth ok I guess the better question would be why ascii doesn't include the cent sign :)
kgftr|konobi autrijus: do you still have your webl10n presentation anywhere available? 15:08
Khisanth never noticed that
gaal Juerd: pastebot.sourceforge.net/ is a rather uninteresting page with a link to the parent directory. :/
Juerd Khisanth: It's not an official symbol, iirc
sourceforge.net/projects/pastebot then
And blame dngor for the stupid page on pastebot.sf.net
gaal Can I blame nothingmuch instead? Lower overhead for me. 15:09
Juerd Heh 15:10
Hmmm
I wonder if I can have a t-shirt made before tonight with a cent symbol and a big red cross
Oh, it's past 17:00
Then probably not
Damn :)
tehmaze Juerd: make it a `50 c' with a cross then :p 15:11
Juerd I suggest 1c as the ascii equivalent :)
tehmaze oh, there is a printshop still open on the Vijzelstraat
Juerd I'm in Dordrecht. 15:12
tehmaze ah, too bad
Juerd tehmaze: We could get annoying and demand that the euro sign be used 15:15
tehmaze: It also has a C in its glyph
tehmaze: There's now a very americanized imbalance with both $ and (cent thingy)
tehmaze no, I agree that ^ is the next logical sigil to use :)
Khisanth hmm the sideways Quake symbol
Juerd tehmaze: Only if we call # the comment *sigil* 15:18
tehmaze and ! the not *sigil* ? :)
Juerd After all, #foo and #foo are practically the same sigil.
tehmaze: No, that's an operator.
tehmaze yes, ! is not a sigil, so that will fit
Juerd Hehe 15:19
tehmaze easy to remember too
Juerd not a sigil, not the not sigil.
right.
tehmaze or ōæ½xAC 15:20
gaal Juerd: pastebot wants some modules from cpan (eg. POE). ok to install directly or do you want me to shoot for debs?
Khisanth that would make ??!! unpleasant to look at .. $foo ?? ... !! !!!bar; 15:21
tehmaze (shift-` on a UK QWERTY keyboard ;)
Juerd gaal: Install directly only if it's not installed already.
gaal: All available debian perl module packages ARE installed.
gaal OK. Some things have newer versions on CPAN but hopefully this'll still work. 15:23
Juerd It will 15:24
If not, please try to use "dh-make-perl --build --install --cpan Module::Name" before installing without debian packages
And when you do need to not-use a debian package, remove the installed one first
I want to keep the machine upgradable :)
gaal wtf are Server::Web and Client::IRC that it wants?
Juerd gaal: Those are its own modules. 15:26
gaal ah, those are parts of pastebot itself. urrr, it's not runnable from the local dir?
Juerd gaal: pwd?
gaal (want to test this before make installing)
Juerd Skip the test. Be a man! :)
gaal ~gaal/src/pastebot-20040908
if you say so :) 15:27
Juerd It can *find* the modules
But they don't compile
Because of runtime errors
(Yay for mixed compile and run times)
gaal yay for silent evals :( 15:28
Juerd Silent?
It's an mkdir error
pastebot.perl Error while loading Server::Web: mkdir pastestore failed No such file or directory at Util/Data.pm line 268.
Compilation failed in require at Server/Web.pm line 20. 15:29
gaal s/silent eval/blind users/
:(
Juerd :P
gaal they probably wanted mkpath instead of mkdir
fixing
Juerd Is that an assumption? 15:30
gaal a conjecture. 15:31
Juerd That's the same as an "educated guess" :)
gaal no, a conjecture doesn't have to be educated. :) 15:32
hareldvd__ Any idea which channel for perl-Tk issues? 15:36
Juerd #perl
gaal weird! the cwd is the current one (added a debug print), so it isn't a mkpath error, but strace doesn't show a mkdir call at all.
Juerd gaal: See Ulit/Data.pm line 268 then 15:37
Util even
gaal that's where i'm looking.
Juerd Print the argument to mkdir|mkpath. Is it what you want it to be? 15:38
gaal oh, it's just a perl bug:
mkpath($dir) unless -d $dir or die "mkdir $dir failed $!"; 15:39
should be
Juerd ()
gaal do { mkpath($dir) or die "mkdir $dir failed $!" } unless -d $dir;
Juerd mk or die unless -d
Cight
Right
No need for do, though
gaal s/path/dir, but mkpath is better anyway.
Juerd It's a rather silly bug. 15:40
gaal $SILLY_BUGS++
counterwise, not karmawise.
hello featherpaste. 15:41
Juerd pastebot!
:)
Oh, this one doesn't respond
autrijus rehi \camels 15:42
Juerd Hi autrijus
autrijus: Had you seen juerd.nl/takahashi/icons.html?
featherpaste "gaal" at 192.115.25.249 pasted "pbuilder error" (20 lines) at feather.perl6.nl:8888/1
gaal rafl: any ideas?
Juerd 6.2.10 released! xrl.us/hxnb | geoffb's column (/. ed): xrl.us/hxhk | pugscode.org | pugs.kwiki.org | paste: feather.perl6.nl:8888/ | www.geeksunite.net 15:42
gaal Juerd: the README explains how to set up apache to forward requests to the pastebot (that's its own server), so that people behind firewalls can still access it. 15:43
autrijus Juerd: no, checking 15:44
gaal Juerd: anymoose, now that it's tested to work *g*, i'll make install.
autrijus Juerd: what is that? 15:45
xerox www.sooff.com/database/perform_imag...500022.wmv 15:46
Juerd gaal: feather.perl6.nl/paste/ 15:47
autrijus: See juerd.nl/, most recent item
gaal Juerd: k, made install. Now to figure out how to have it start automatically. 15:48
autrijus Juerd: you coming to the 8pm meeting?
Juerd autrijus: Yep
autrijus: Have to find a way to park my car, but will be there.
gaal: crontab -e
gaal I heard parking is terrible in Amsterdam 15:49
Juerd gaal: @reboot /full/path/to/whatever/starts/the/daemon
autrijus Juerd: cool! the icons looks crazy enough
Juerd autrijus: This won't be an actual talk, by the way
gaal: It is.
gaal: Especially in the city centre, where this is. 15:50
gaal Juerd: Tel-Aviv is terrible that way too. Especially in the city centre, where I live! 15:51
Juerd gaal: I've never been there, or even near there :)
wb featherpaste
gaal one more quit 15:52
wolverian Juerd, wow, SVG icons would work a bit better, methinks :)
Juerd wolverian: I can't find them :) 15:53
wolverian: This icon theme is called krystalsvg, even! 15:54
Doesn't matter for the main idea, though
autrijus kgftr|konobi: try... web.archive.org
I've put what I have at autrijus.org/webl10n/
Juerd I will not complain any more that Larry is sometimes inactive in p6l ;) 15:58
autrijus roflmao
he was sitting opposite the table of me when he posted this c| thing. 15:59
Juerd (Not that I have)
xerox Un progretto con molti collaboratori internazionali si evolverĆ” piĆŗ in fretta e meglio 16:00
hah!
(no 'si'!) 16:01
Juerd in fretta e meglio => ?
xerox faster and better
(autrijus, I meant to say that 'si' shouldn't be there) 16:02
autrijus ok :)
bbiab, dinner... 16:03
gaal Juerd: *wanna* be there for YAPC on Feb? :)
Juerd gaal: Yeah, but by then I won't have enough cash yet.
wolverian oh, haha. larry is on a rampage
Juerd afk # ..amsterdam! 16:05
gaal heh, best away message ever :) 16:06
Juerd: re: comment spam. If you don't mind coding some Javascript, you could have the posting form mint a hashcash stamp on every comment. That is quite effective. 16:07
theorbtwo I'm not sure I understand why this needs a new sigil -- what was wrong with ::, or with making it $type does Class? 16:19
geoffb Khisanth, two things: 1) Thanks for the info about ctrl-shift-<hex> in GTK apps. That makes my life MUCH easier. 2) You wondered why ASCII didn't include the cent sign. Oddly enough I came across an old document that indicated you probably shouldn't use '$' because it was the only printable character in the ASCII range that was not standard on all systems of the time . . . . 18:43
Anyone happen to know what the emacs equivalent for the GTK ctrl-shift-<hex> trick for inputting arbitrary Unicode characters would be? 18:44
jp-autark geoffb: M-x insert-ucs-character 18:54
geoffb jp-autark, does that have a default binding?
jp-autark don't know. 18:55
geoffb (Just curious if I'll have to add a binding everywhere)
OK, thanks.
ingy seen whiteg
jabbot ingy: whiteg was seen 1 days 1 hours 35 minutes 28 seconds ago
geoffb Well that's interesting. My emacs has no insert-ucs-character.
jp-autark, which emacs are you using? 18:56
jp-autark geoffb: 21.4.1
geoffb Ditto, from Debian testing.
OK, that's just odd.
Add-on emacs package, perhaps? 18:57
jp-autark maybe
geoffb apt-cache search emacs |wc
332 2762 18095
ouch
jp-autark geoffb: maybe you need the mule-ucs package? 18:58
geoffb Giving it a try 19:00
jp-autark geoffb: there also seems to be a ucs-insert function. Dunno if that's part of mule-ucs as well though.
geoffb With mule-ucs I now have insert-ucs-character 19:02
jp-autark where you able to do utf stuff with emacs without it? 19:03
geoffb Interesting, it seems to expect decimal codepoints
jp-autark geoffb: #xHEX
geoffb jp-autark, I use(d) emacs purely as an ASCII editor
I don't think I've ever loaded any file into emacs that wasn't 7-bit clean, AAMOF
jp-autark or #oOCT, #bBIN. But maybe ucs-insert is easier to use? 19:04
geoffb ah, yes, does direct hex
Khisanth ucs-insert seems to default to hex here 19:05
geoffb Nope, no *ucs* in the default keybindings
Khisanth, yeah, that's what I just found.
Now I just have to memorize some codepoints, and I'm good to go 19:06
jp-autark though insert-ucs-character may be useful if you only can remember the binary code for say 0x30a2 ;)
geoffb Ah, and ucs-insert is unique in the first two characters. Makes M-x uc<tab> work 19:07
or rather,
M-x uc<enter>
Thanks much, jp-autark 19:08
jp-autark np, good luck remembering
geoffb Actually, I'm choosing to just learn codepoint insertion and a list of codepoints, because that makes using unicode an N + M problem, rather than an N * M using magic combinations in each app. 19:10
jp-autark hopefully your list isn't too long :) 19:11
geoffb And frankly, GTK and Emacs cover most of the places I need unicode.
jp-autark, almost entirely limited to those to be used by Perl 6, actually. :-)
jp-autark grins 19:12
Khisanth AB and BB are useful :)
geoffb Now if we could just convince irclogger to set the charset for #perl6 to UTF-8
Khisanth, nod. Now remembered.
Khisanth irclogger?
geoffb goes to look up yen 19:13
colabti.de/irclogger/irclogger_log/...-10-20,Thu
Khisanth hrm yen is a5
just a few spots from Ā¢ :)
theorbtwo OS-level compose combinations are happy things. 19:14
Khisanth $Ā¢Ā£ā‚¬Ā„foo :D
geoffb Khisanth, heh 19:15
theorbtwo compose,c,/ ... yey, got it on the first guess.
Khisanth oh forgot %
theorbtwo Ā¢
Ā«Ā»Ā¢Ā„... what am I forgetting?
geoffb theorbtwo, OS?
theorbtwo xorg/linux 19:16
geoffb theorbtwo, what is the compose key sequence there? 19:17
theorbtwo Not sure I understand the question. 19:20
<< = Ā«, >> = Ā», c/ = Ā¢, y- = Ā„
imperator_ is perl6 doing selector namespaces? and if so, does one of the apocalypse/exegesis cover it? I just want to check out the proposed syntax 19:22
theorbtwo Selector namespaces?
imperator_ (no luck with google)
theorbtwo, encapsulation within scope 19:23
geoffb theorbtwo, I meant, what key(s) do you use to begin a composition
?
jp-autark geoffb: pugs/docs/quickref/unicode seems to be highly relevant. 19:28
stevan imperator_: do you mean private packages/classes?
imperator_: package Foo { my package Bar {}}
imperator_: Foo would be accessible from the outside, but Foo::Bar would not (IIRC) 19:29
geoffb jp-autark, thanks again 19:30
theorbtwo geoffb, you can set it to what you want; on most keymaps there isn't one by default. 19:31
xmodmap -e 'keycode 94 = Multi_key'
imperator_ stevan, that sounds like it :) 19:32
geoffb theorbtwo, ah, thanks
Hmmm, more is preserving unicode prettiness, but less is mangling.
theorbtwo Hold on a sec, that probably won't be useful for you. 19:33
94 is the key to the right of left shift on a 102 key keyboard.
I'm using a German keyboard with an American keymap, making that key extra. Most people either need it or don't have it. 19:34
To find the number for an arbitrary key, use xev, put your cursor in, hit the key, and look for a line stating "keycode NN". 19:35
geoffb Yeah, saw that at the bottom of the doc jp-autark pointed out
And FWIW, 'less -r' seems to be needed to see the unicode
(as other than hex crappage)
theorbtwo Oooh. 19:36
wolverian Juerd, would it be possible to get irssi 0.8.10rc5 or some other, newer than 2003, svn version on feather? 19:39
Juerd, ubuntu packages rc5, at least. let me check debian. (it's for the /recode support) 19:40
theorbtwo I suppose it's a bit early to add Ā¢ to the list at the top of pugs/docs/quickref/unicode, esp since the ascii fallback hasn't been defined yet.
wolverian Juerd, ah, no, feather's apt-cache doesn't see a newer irssi. :/ 19:41
ErRandir can anyone tell me what to put in the -time parameter in Tk's eventGenerate() ? 19:43
geoffb ErRandir, I'm assuming you want #perl 19:44
And BTW, why doesn't #perl appear to so many people?
Limbic_Region define doesn't appear
geoffb Limbic_Region, many people come to #perl6 when they want #perl, saying that #perl didn't appear in their channel list, but #perl6 does. 19:45
Limbic_Region ErRandir - this channel is primarily for the discussion of the development of perl6 - not to discuss issues with per5
geoffb correctly indicated you should probably be in #perl
ErRandir sorry. thx
Limbic_Region geoffb - oh, not knowing what a channel list is I hadn't ever seen that
geoffb Limbic_Region, ah 19:46
Limbic_Region my client allows me to type in the channel I want to join
integral #perl is +s. someone should fix that *hint*
Limbic_Region doesn't have ops in #perl
geoffb integral, what does +s mean? Invisible? 19:48
wolverian secret
integral "let's hide from all the spambots!" 19:49
kolibrie hmm, why does my terminal allow the Ctrl-Shift-<hex> normally, but in irssi I get weird symbols? 19:52
Khisanth irssi is probably intercepting those 19:53
kolibrie for shame!
workaround? 19:54
Limbic_Region don't use irssi 19:58
;-) 19:59
kolibrie :(
Limbic_Region kolibrie - I use CGI:IRC from work because I can't IRC directly - it does silly things like intercepts plus symbols (+) and doesn't display them 20:00
Juerd was kind enough to hack in a fix for that specific problem - but that's clients 20:01
err - that's the nature of clients
theorbtwo I could have sworn that there was a way to set the charset of a chan, but I may be thinking of gaim.
kolibrie I'm sure autrijus types unicode into his irssi 20:03
theorbtwo Oh, you should start your irssi with a unicode locale set, possibly? 20:04
Er, utf8, not just unicode.
Khisanth doesn't irssi have a /recode thing? 20:05
I don't use it but I have seen that mentioned quite a few times :)
theorbtwo "No help for recode."
kolibrie has never heard of recode
theorbtwo: LANG=en_US.UTF-8 ? 20:07
theorbtwo Looks OK to me. 20:08
I have LANG=en_US.utf-8.
jerakeen.org/blog/2005/06/23/screen-irssi-utf8 20:09
Juerd wolverian: .deb does not automatically mean it's compatible. In fact, it's likely to not be. 20:10
Hi from the amsterdam.pm post-oscon meeting!
wolverian: If you can get a feather-compatible .deb for irssi, go ahead. If you want to build your own irssi in your homedir, of cuorse, you're also free to do so. 20:11
kolibrie theorbtwo: sounds complicated
theorbtwo Most of that is the extra screen step.
kolibrie it worked!! 20:13
theorbtwo Cool. 20:14
xinming hmm, I saw the oscon05-autrijus.mp4 But I don't understand some idea on pugs. :-S
kolibrie actually, it looks like I only needed the irssi set command, everything else was already set that way
xinming hmm, Will pugs be the tool to bootstrap perl 6? hmm, If true. Then, after perl 6 is written, What will pugs be? 20:15
I mean where should pugs project go. :-S 20:16
wolverian Juerd, right. thanks :) 20:17
PerlJam xinming: pugs doesn't have to go anywhere 20:19
theorbtwo The Road goes ever on and on,
Down from the door where it began.
Now far ahead the road has gone,
And I must follow, if I can,
Pursuing it with eager feet,
Until it joins some larger way
Where many paths and errands meet.
And wither then? I cannot say. 20:20
We'll worry about that when we get there.
Anything else would be premature pessimization.
Limbic_Region well, unless someone gets the bright idea of parrot targeting haskell ....
xerox \o/ 20:21
theorbtwo You mean haskell targeting parrot, Limbic.
Limbic_Region no, but that too
;-)
I was just saying that pugs has multiple backends currently that work the opposite way parrot is designed 20:22
we turn p6 code into other languages
xinming a bit confused by the relation for pugs and perl 6 now. :-S 20:24
theorbtwo pugs is an implementation of perl 6. 20:25
It's up in the air as to if it will become /the/ implementation of perl 6. Sources point to no.
However, it will almost certianly be a very useful thing in writing the perl 6, if for no other reason that people have played with it some and figured out what works, what doesn't, what is underspecified... 20:26
Khisanth hrm pasteling is back :) 20:27
xinming hmm, what does the backends do then? :-S 20:28
PerlJam theorbtwo: it *is* a very useful thing.
xinming there are many backends. :-S
GHC, JS, JSON, PIL, PIR, Parrot, Perl5, Pugs
Limbic_Region xinming - forget pugs for a second
p6 is intended to run on Parrot 20:29
so p6 parses p6 and compiles to parrot asm
introduce pugs
haskell parses p6 and compiles to haskell
xinming Limbic_Region: yes, I know that, p6 bootstrap...
Limbic_Region haskell parses p6 and compiles to p5
haskell parses p6 and compiles to JavaScript
xinming ....
Limbic_Region haskell parses p6 and compiles to etc 20:30
xinming ...
Limbic_Region xinming - I didn't say anything about bootstrapping
xinming amazing...
Limbic_Region that is a bi-product of parsing/compiling p6 code
xinming Limbic_Region: hm, I know, I mean p6 is parses p6 circle...
Limbic_Region or rather - that is a bi-product ability of
so what I was saying is even if Pugs is used as the means for bootstrapping p6 20:31
it still does things p6/parrot weren't intended to do
or even ponie
svnbot6 r7693 | kolibrie++ | added some missing twigils
Limbic_Region it takes the p6 code and targets a completely different backend instead of making those languages target parrot
xinming Limbic_Region: hmm, So, what does these do in real life? Or It's just for fun? :-S 20:32
PerlJam xinming: for one, pugs allows us to hammer out design issues without having "the" perl6 compiler around. 20:33
xinming If so, the efficiency is too "low" IMHO...
integral -Ofun!
Limbic_Region pugs is optimized for fun 20:34
err integral beat me to the punch
theorbtwo xinming: Fun is the point; being useful is just something that happens by accident sometimes.
Limbic_Region PerlJam - right, but that's not xinming's question
the question was regarding post p6 1.0
theorbtwo The more interesting question is "how are we going to get to perl 6 1.0". 20:35
Sadly, I have no idea.
PerlJam Post p6, pugs can still fulfill that role.
xinming If So, we can compile the perl 6 code into haskell or perl 5 or other $language... and using the compiler which compiles other $language to get perl 6 done. right? 20:36
PerlJam pos p6, pugs will still have abilities that p6,ponie,parrot, etc. don't have. Having features in pugs might also encourage those features to be added to the "real" p6 compiler (or ponie or p5 or parrot or whatever) 20:37
s/^pos/post/
theorbtwo I think the most reasonable way to get to a self-bootstrapping perl6 is to make something that compiles haskell to perl6. 20:38
...then run it on pugs.
xinming hmm, If what I said is true. then, which will be chosen to compile the official perl 6? :-P 20:39
PerlJam theorbtwo: I don't know about that. It depends how quickly we can get a PGE-like thingy in haskell I think.
xinming Oh, lost again... :-S 20:43
ingy seen autrijus 20:44
jabbot ingy: autrijus was seen 4 hours 41 minutes 40 seconds ago
Juerd ingy: autrijus was seen less than an hour ago IRL 20:54
ingy: He left the meeting, probably to sleep
ingy Juerd: thanks...
maybe you can give me an opinion on a Perl module UI issue 20:55
I'm reworking YAML.pm so that it can support multiple backend parser/formatters
including YAML::Loader::Syck (which wraps libsyck) 20:56
so do I expose these choices to the user as:
use YAML -syck;
geoffb ingy++ on that
ingy or
use YAML::Syck;
I started on the former but am now leaning towards the latter 20:57
geoffb I would agree with that, ingy (using the latter form)
ingy the idea is that YAML::Syck would subclass YAML 20:58
theorbtwo I'd use the latter.
ingy and get all the basic UIness for free
ok thanks
theorbtwo If you make them different modules, it makes it easier to have them be different os-level packages. 20:59
ingy in either case they would be separate modules / distributions 21:00
theorbtwo libyaml-perl has no interesting dependencies, libyaml-syck-perl depends on libsyck and libyaml-perl.
If they'd be sepperate modules/distributions, you'd need extra code in the superclass to support the subclass being able to have -syck support in the import. 21:01
ingy theorbtwo: right, I already wrote that and it is ugly
fmadero is there a command one could use to list all perl modules 21:36
nothingmuch gladly announces that he managed to gently bend the enclosure of his laptop out 21:42
so that it's no longer scraping the hinges when the screen is opened
rafl gaal: Maybe try to use a version of sid that's not that recent. snapshot.debian.net - 5-days-ago or something like that. 21:57
ingy nothingmuch: you are always breaking things ;) 21:59
rafl gaal: Also do you have cdeboostrap installed? 22:00
mugwump stevan: I like your graphviz charts 22:16
gaal rafl: hmm, really i just need a ghc from cvs on linux. can't that bug in the binary version simply be fixed? it's "just" a wrong dependency in the package metadata apparently. #319812 23:52