6.2.11 released! | pugs.blogs.com | pugscode.org | pugs.kwiki.org | paste: sial.org/pbot/perl6 | < audreyt> my fingers juxpatosed
Set by nothingmuch on 24 February 2006.
00:09 michimueller joined 00:10 K joined
nothingmuch so 00:10
i have weird slides
that are 60% done 00:11
and I'm off to sleep
does anybody want to look at them and comment?
lumi: sorry for missing you
my mac::growl (and inherently irssi growl script) is broken
Khisanth waits for the release of perl 6.6.6 00:15
nothingmuch we saw weird license plates yesterday 00:17
one with 666 and one with 777
in the same traffic light
avar obra: Hurd is shipping right now, Perl 6 isn't 00:21
obra Hurd is shipping? 00:24
avar yes it is 00:26
until recently I had it running on one computer here
obra . It is not ready for production use, as there are still many bugs and missing features. However, it should be a good base for further development and non-critical application usage. 00:27
From their status page.
That sounds like Perl6.
avar yes, that doesn't mean that it isn't shipping
obra I took your question to be "Which will ship first?" 00:28
Pugs is shipping.
avar can I download the Perl6 compiler that's written in Perl6 and targets parrot and make it run even simple programs right now?
pugs doesn't really count, it's a seperate project
that's kind of like saying "Yes Linux is shipping" to "Is Hurd shipping" ;)
revdiablo Except pugs is implementing perl6, whereas linux is not implementing hurd 00:29
obra Linux and Hurd are basically unrelated
Pugs is part of the bootstrapping plan for Perl 6.0.0.
So, What was your actual question?
avar I didn't have a question
ayrnieu doesn't understand why people think that pugs doesn't count.
avar <obra> Hurd won't ever ship ;) 00:30
<avar> obra: Hurd is shipping right now, Perl 6 isn't
obra avar: I apologize. It was elmex' question.
avar I'll grant you that Perl 6 is kind-of shipping, but you're wrong about Hurd not shipping...
obra Their website says it's incomplete and not production ready. 00:31
Is their website out of date?
www.gnu.org/software/hurd/hurd.html#status
revdiablo obra: Maybe it's just that Hurd has been shipping incomplete and not production ready for a longer period of time?
=)
avar obra: no, but it's shipping and you can use it for various things 00:32
revdiablo: maybe;)
obra avar: and I think that's a fine description of the state of perl6. 00:33
It is not the state of "An implementation of perl6 in perl6 targetting parrot"
avar I'd say Hurd was more advanced atm, although Perl 6 will probably be "finished" sooner, for some definition of finished 00:34
they've been writing it since 1991 or something;)
obra But that wasn't even the question!
avar I.e. it generally works like it's supposed to works
s/s$//
I'm expanding on it!
obra *snicker* 00:35
avar The reference implementation of Perl 6 is supposed to be written in Perl 6 and target parrot, it's nowhere near that goal last I checked
obra Last I looked, Larry had described Pugs as the current reference implementation. Which is different than the "Production" implementation. But I'm splitting hairs because my blood-sugar is crashing 00:37
I'm going to wander off and fix that.
avar been smoking pot?;) 00:38
obra Not my thing, sadly 00:39
Been working all day
00:57 joepurl joined 01:09 lao_ joined 01:10 Nouk joined 01:25 dduncan joined 03:18 Ijon joined 03:20 Ijon left 04:36 mako132_ joined 04:49 amnesiac joined 04:54 stevan joined 06:27 spo0nman joined 06:50 scook0 joined 06:51 hoowa joined 06:55 drrho joined 06:57 GeJ joined
ayrnieu mmm, OpenBSD on PPC is wonderful, but I need another machine to run GHC on, now. 07:03
07:15 marmic joined 07:19 gaal joined
svnbot6 r9189 | lwall++ | Some modernizations to P5-to-P6 porting guide. 07:19
07:20 TimToady joined 07:21 shmarya joined 07:27 nothingmuch joined
nothingmuch moose 07:28
shmarya lol 07:29
07:33 ayrnieu joined, kakos joined
gaal elk 07:42
arcady caribou 07:56
nothingmuch now that audrey is done, we can finish writing my slides 07:57
gaal wapiti!
08:01 iblechbot joined
audreyt ...I did plan to go overtime 08:01
but not by 19 minutes
sorry :/
08:02 shmarya is now known as shmar
audreyt slides: xrl.us/j72t 08:03
I finished it practically on the stage, so didn't get to upload it before the talk
obra hah
TimToady audreyt: you had one "subtype" that s/b "subset".
shmar audrey: are they available in a downloadable form?
audreyt yeah, also :p5 08:04
both fixed
shmar: they are... just grab the .xul, and takahashi.js and euroscon.css in the same dir
shmar kewl... will do...
audreyt a simple wget should suffice iirc
wget -m -np perlcabal.org/~autrijus/osdc/ 08:05
shmar awesome...
got it... 08:09
audreyt cool
08:15 shmar left, shmar joined
gaal is -np the special audrey switch? :) 08:17
08:18 nothingmuch joined
nothingmuch ingy: ping 08:19
TimToady audreyt: yaml of all .t and .pm files in Perl 5 subs:
audreyt: scp 172.19.2.36:/tmp/p55yaml.tgz .
16 megs or so
audreyt TimToady: k 08:20
nothingmuch TimToady: access is blocked between us
push it to feather if they haven't fixed it so far
audreyt confirmed, ssh doesn't work intranetwise.
TimToady hmm.
audreyt we can revert to usbdisks
or crossover cable
TimToady I don't have a feather account 08:21
audreyt but after my lighting talks :)
oh wow
nothingmuch can we add a feather account?
audreyt Juerd: fix it :)
TimToady I could put it on my home public_html
nothingmuch there was some legal issue, right?
(if Juerd doesn't answer)
audreyt otoh, I _can_ adduser TimToady easily, as I can verify his rela name
nothingmuch TimToady: if it's too slow you can also use my home box
TimToady but that'd be slow.
audreyt and nonhologramness
nothingmuch but that's not too fast either
in that case, i think we can add an account
TimToady I have 256kb uplink
nothingmuch that's not that long 08:22
kbit or kbyte?
azuroth what's up, dude?
audreyt gaal: have crossover with you?
azuroth: the sky (which is also Blue)
nothingmuch i can autocross with my laptop, if we only have patch
or!
i can be an adhoc router
08:22 mj41_ joined
nothingmuch much simpler 08:22
audreyt ...to you perhaps.
TimToady I could put it on my camera card. :)
audreyt TimToady: now that's an excellent idea
who would've thunk of that! 08:23
azuroth audreyt: *cries*
nothingmuch TimToady: anyway, if you'd like you have an account, just /msg me ssh key
so that in the event of things breaking we already have at least something partially copied
(multi threading your fallbacks)
08:24 fandango joined
audreyt bbiab, finalizing binding for my two talks coming up in 10min (or was that 5) 08:24
azuroth are there web cams? 08:26
gaal oops darkning talks
audreyt: aw, sorry, no crossover - only saw Q now 08:30
nothingmuch $ingy.steal( :for<kwid>, URL.new("www.osdc.org.il/talk.html?id=52") ) 08:33
gaal yes, the ascii art -> SVN graphics is shiny
08:34 K joined
shmar If you're looking for a External HD, I got one here... plenty space.... 08:40
TimToady shmar: already made the tranfer using my 512MB camera memory, but thx. 08:43
shmar TimToady: np...
audreyt: is that link (to the slides) safe for blogging? 08:45
TimToady shmar: audreyt is currently lightening-talking. 08:49
08:49 elmex joined
shmar TimToady: Yah.. I noticed just after I hit <ENTER> 08:49
08:57 lypie left 08:59 avarab joined 09:08 xinming joined 09:11 xinming joined
xinming what does the word 'juxpatosed' mean? :-/ 09:11
shmar do you mean juxtaposed?
www.thefreedictionary.com/juxtaposed 09:13
09:13 Aragone is now known as Arathorn
xinming shmar: hmm, what does topic mean? 09:14
09:14 kane_ joined
xinming < audreyt> my fingers juxpatosed 09:14
shmar juztaposed is also sometimes used to say 'each placed in the place of the other' IE: audrey's fingers are all confused... 09:15
09:19 gaal joined
xinming I know it's a pain to illustrate idea of the word who's mother language isn't English... But... I still don't understand. :-( 09:19
ayrnieu xinming - if I type your name with my left hand shifted to the left (in dvorak), your name is 'kungmung' I might type that, pause, and blame my fingers for the mistake. 09:22
xinming gives up for trying to understand... :-) 09:23
thanks anyway...
shmar audreyt: is that link (to the slides) safe for blogging? 09:24
09:24 bsb joined
ayrnieu thinks of 'aiya, wo do shou huai le', but gives up trying to explain... :-) 09:24
also, wo de.
xinming :-) 09:26
09:27 nothingmuch joined
nothingmuch lumi: nothingmuch.woobling.org/perl6_comp...start.html 09:27
audreyt shmar: sure yes 09:28
shmar audreyt: thanks... Linking to that URL. 09:29
audreyt: idp.shmarya.net/?p=57 09:33
audreyt k 09:34
also perlcabal.org/~autrijus/osdc/haskell.xul and vb.xul and mariner.xul in the same dir
shmar yah... I'll go back and edit the entries from yesterday... 09:36
I really liked your VB/perl6 parallel
rather: VB/pugs
audreyt thanks! :) 09:38
shmar: docs/Perl6/ can use some help if you are a English hacker -- cherrypicking from docs/* into the new hierarchy makes things much more accessible 09:39
audreyt ponders xul2pod
09:39 szabgab joined
shmar audrey: I am (engHack) and will take a glance now... busy compiling tho.... 09:40
audreyt :D 09:41
shmar++
shmar lol!
audreyt shmar: what's your email so I can send you a commit bit? 09:43
shmar audreyt: [email@hidden.address] 09:44
audreyt commit bit sent. welcome aboard! 09:45
you can add yourself to AUTHORS -- and help filling in the Hebrew name for Mikhael Goikhman in that file if you happen to know it 09:46
-- as your first commit to test that commit works
gaal OSDC takes its first vic^H^H^Hnewcomer!
shmar HEHE!
audreyt also, you can invite more people into committership by logging in to rt.openfoundry.org and go to the rt.openfoundry.org/Foundry/Project/...?Queue=270 page
szabgab I heard that D
gaal welcome, shmar 09:48
TimToady shalom, shmar 09:49
shmar ta... hope I'll be up to it!
commited... 09:53
svnbot6 r9190 | shmarya++ | Adding self to authors list
audreyt shmar++ 10:02
clkao !audone keynotes? 10:03
audreyt !cl yeah, can look at js now (and p5->p6) 10:07
clkao !au whoot
!au you might want to change the jenga picture to flickr.com/photos/clkao/100733111/ or flickr.com/photos/clkao/100732917/i...otostream/ 10:09
lypanov umm 10:10
lypanov wonders if nothingmuch's darcs talk slides are online?
lypanov sips his tea, wakes up slightly, and checks osdc.org.il
shmar lypanov: post a link if you find them... 10:12
szabgab we don't have any of the slides up on osdc.org.il yet
lypanov ah :) 10:13
szabgab but we have Larrys talk in mp3 a 52Mb file, anyone knows how to make it smaller ?
lypanov szabgab: you're expecting to place them online eventually? :)
szabgab: maybe try the ogg speex thing
lypanov isn't sure really
xinming szabgab: please.... don't shrink it... :-) 10:14
bsb Should some of the following do something useful: 10:15
./pugs -e '$_ = q:code{ "hi"+2 }; say; say $_.perl; say $_.yaml'
szabgab here is the full version: www.osdc.org.il/html/larry-wall-2006-02-26.mp3
audreyt bsb: sure... also a Perl deparser would be nice
./pugs -CParse-Perl6
clkao: fixed 10:16
I mean, the jenga
clkao LOL
and you see what the pumpkin is doing - polaying jenga while having wine! 10:18
lypanov hehe 10:19
audreyt come'on, you need alcohol when hacking p5 internals
xinming what's the ip of osdc.org please? DNS is broken here.
audreyt you of all people should know
xinming: www.osdc.org.il
xinming audreyt: yes, the ip of the site.... Dns is broken 10:20
audreyt 192.117.122.130
xinming thanks
???
shmar beat me to it...
audreyt er, nvm
xinming It's not wan ip. ;-)
audreyt indeed 10:21
that's totally broken
shmar yup just noticed...
audreyt szabgab: E-PUBLIC-SITE-BROKEN
shmar lol
szabgab audreyt: E what ?
lumi perlbot nopaste
perlbot Paste your code here and #<channel> will be able to view it: sial.org/pbot/<channel> 10:22
audreyt szabgab: see scrollback
szabgab: www.osdc.org.il resolves to 192.117.122.130
which is an intranet addr
and hence inaccessible to people not in this building
10:22 iblechbot joined
szabgab audreyt: why would that be an intranet IP ? 10:24
it is 192.168.* AFAIK
can you access www.perl.org.il ? 10:25
arcady it's a perfectly valid IP
and the same as www.perl.org.il
audreyt oh hm, sorry.
mea culpa.
so it's just that the .cn firewall is blocking dangerous israeli site
because you talk about sensitive subjects or something. 10:26
sorry for the false alarm.
pasteling "nothingmuch" at 192.118.100.254 pasted "pbotutil config" (29 lines, 509B) at sial.org/pbot/16047
10:26 nnunley joined
arcady clearly open source is a threat to communism 10:26
audreyt just like wikipedia is.
szabgab xinming: can you access the perl.org.il site ?
xinming audreyt: maybe, firewall even blocks freebsd...
clkao i think it's the biohazard sign
xinming szabgab: I can.
audreyt ahh right, which signifies biochemical weapons
but then, it's not on osdc.org.il
szabgab xinming: it is the same server so I wonder what could be the different between perl.org.il and osdc.org.il 10:27
clkao a friend told me they found security hole in the great wall and will publish a paper this week
szabgab the picture of Audrey ?
audreyt lol, right, it matches porn profile or something
;)
szabgab :) 10:28
clkao basically, if you just ignore the RST packet, it doesn't really block things
audreyt clkao: oooh
xinming clkao: No, There might be 3 levels of blocking in GFW 10:29
shmar audreyt: Where should I be cherry picking docs from/to? 10:33
audreyt shmar: pugs.blogs.com/pugs/2006/02/hackathon_3_p6d.html
actually, add the map to README or somewhere 10:34
docs/quickref/ goes to Overview/
shmar okey...
audreyt except for quickref/fear/ which probably belongs in FAQ
shmar getting into it...
audreyt improvise :)
xinming in early Feb this year, why freebsd.org is blocked, I think because the soome reason which about the Kylin OS.(hmm, stop mentioning it), and the freebsd.org is block through DNS, Not the content.... And some site are blocked through the whole connection. eg... type some sensive words in google. It will block you for 5 minutes, and it will drop all package send to google.
audreyt yup, I tried it in Beijing 10:35
szabgab xinming: www.perl.org.il/osdc/larry-wall-2006-02-26.mp3 10:37
xinming szabgab: thanks 10:38
audreyt just spent 10min reading up on the K OK 10:44
s/OK/OS/
not terribly interesting :/
TimToady audreyt: what is "EVA" in takahashi? 10:46
bsb This asterix guy is totally bagging php 10:47
Asked if anyone knows a good php coder then harrassed the audience member who said yes 10:48
shmar lol!
The question is what is he recommending instead?1? 10:49
TimToady AGI, I guess...
shmar apples and oranges?
szabgab Java 10:51
shmar you must be kidding!
szabgab see what I mean :)
bsb Now he wants more audience participation, no one is game
webmind java with apples ?
shmar java + apples = exotic coffee... 10:52
lypanov um... perl6 is looking cute.... (the quasiquoting/macro stuff in the blog) 10:53
audreyt: couldn't you better use an email 2 blog gateway of sorts to prevent so much lost journalling? 10:54
10:57 KingDiamond joined
azuroth does cvs keep track of the octal permissions of a file? 11:02
11:03 lazy1 joined 11:04 lazy1 left 11:11 Aankhen`` joined
audreyt TimToady: it toggles white/black I think 11:25
TimToady: unless I broke it
I don't know what it is called Eva... perhaps obscure reference to Neon Genesis Evangelion 11:26
lypanov: I could :/
TimToady Seems to be toggling some TT font metrics or something. 11:30
11:39 shmar joined
shmar is back from lunch.... 11:39
11:57 rantanplan_ joined 12:12 KingDiamond joined 12:19 azuroth joined, Southen joined 12:20 chris2 joined, nothingmuch joined
azuroth shmar's lunch seemed to cause a tidal wave of people 12:23
12:24 levengli joined
azuroth to bed 12:24
nothingmuch ciao 12:25
azuroth :-)
levengli i am new here (fresh from OSDC::2006::Israel). where can i find beginnner in 12:26
formation on perl6?
nothingmuch levengli: hi =)
levengli the equivalant to the camel book?
nothingmuch are you familiar with Perl ?
ah
dev.perl.org/perl6
levengli very familiar with perl.
nothingmuch this has documents called Synopses
which are the closest thing to a spec
levengli i say the appocalypsis as well as the exegesis as well. 12:27
i meeant "saw"
nothingmuch apocalypses are more of a food-for-thought kind of thing
they were the fusion of accepted RFCs
szabgab nothingmuch: Levenglick: www.osdc.org.il/person.html?id=44
nothingmuch and more
aha 12:28
I am www.osdc.org.il/person.html?id=10
the exegesis docs are outdated 12:29
they are basically language demonstrations
levengli i plan to hear your talk about perl 6 compilation
nothingmuch but since they are explicitly designed to show off everything they are actually pretty yucky
(code wise)
it's going to be very handwavy =)
levengli great. the synopsis it is
nothingmuch the synopses are a delta to Perl 5 12:30
levengli dont worry about the handwaving, it is what makes conferences go around :)
nothingmuch =)
i'm not worried about the handwaving
i'm worried about the slides
which are not 100% ready.... *aaaah*
levengli :) 12:31
shmar audreyt: Want to do the quickref/ops file... Am creating Reference subdir in docs/Perl6... Do I need to create a Reference.pod in the docs/ root as well? 12:36
audreyt er no, we renamed Refrence to Overview
so it's Overview::Operator I believe
Reference is too overloaded a term 12:37
shmar ok... good thing to note somewhere!
audreyt indeed! like, in Perl6/Doc.pod or the README
I mean, docs/Perl6/README
shmar I agree... ok... making it Overview::Operator
audreyt cool
nothingmuch ?eval class Int; method times (&block) { for 1 .. self { code($^i) } }; 5.times { say "foo" } 12:38
evalbot_9190 Error: unexpected "f" expecting ";", statements or "}" not a class name or reserved word
nothingmuch ?eval class Int; method times (&block) { for 1 .. self { code($^i) } }; 5.times { say "foo" };
evalbot_9190 Error: unexpected "f" expecting ";", statements or "}" not a class name or reserved word
nothingmuch ?eval class Int; method times (&block) { for 1 .. $?SELF { code($^i) } }; 5.times { say "foo" };
evalbot_9190 Error: unexpected "{" expecting operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
nothingmuch ?eval class Int; method times (&block) { for 1 .. $?SELF -> { code($^i) } }; 5.times { say "foo" };
evalbot_9190 Error: unexpected "{" expecting operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
nothingmuch =( 12:39
shmar actually according to Doc.pod, Operator is at Spec::
audreyt shmar: nono, the idea is that we have the same file
one in Spec and one in Overview
nothingmuch ?eval class Int; method times (&block) { for 1 .. $?SELF &code }; 5.times { say "foo" };
audreyt or even one in Tutorial
evalbot_9190 Error: unexpected "f" expecting ";", statements or "}" not a class name or reserved word
audreyt that serves different purposes
it's like perlre, perlretut, perlrequick 12:40
shmar aha. I get it... sorry am a bit thick... must be the atmosphere in here.
audreyt respectively
shmar got it...
nothingmuch ?eval class Int { method moosen (&block) { for 1 .. $?SELF &code }; 5.moosen { say "foo" }; }
evalbot_9190 Error: unexpected "f" expecting ";", statements or "}" not a class name or reserved word
nothingmuch URHGGJN#!T!#%
audreyt nothingmuch: you are not making sense to me or to the compiler
nothingmuch why not to you?
levengli \whois audreyt 12:42
sorry :(
audreyt np :)
lypanov windows user? ;)
levengli alas. please dont kill me 12:43
lypanov hehe
lypanov refrains
audreyt ?eval class Int { method moosen (&code) { (1..self).map(&code) } }; 5.moosen:{ say "foo" };
evalbot_9190 OUTPUT[foo foo foo foo foo ] (bool::true, bool::true, bool::true, bool::true, bool::true)
12:43 gaal joined
lypanov attempts to grok 12:43
gaal "the code just flows" doesn't sound that attractive to me...
audreyt "the point just floats" 12:44
szabgab ?eval say "hello world"
lypanov "the code is pretty and goopy"
evalbot_9190 OUTPUT[hello world ] bool::true
audreyt ?eval class Int { method moosen (&code) { for 1..$?SELF {code} } }; 5.moosen:{ say "foo" };
evalbot_9190 OUTPUT[foo foo foo foo foo ] undef
levengli i'll be beack in a year or two, after i ahve finished reading all the synopsis's 12:45
lypanov audreyt: i'm impressed. i can grok this
lypanov debates reading them also
audreyt lypanov: Zeno's Paradox detected. :)
shmar :)
audreyt lypanov: er, well, it's just ruby :) 12:46
lypanov audreyt: exactly ;)
lypanov is distracted by wikipedia's page on zeno's paradox now :P
hehehehehehe
audreyt: nice ref :) 12:47
audreyt lypanov: perhaps more interestingly: 12:48
?eval multi neg_repeat (Int $x, Str $s) { [~] $s xx -$x }; neg_repeat(-4, "foo")
evalbot_9190 "foofoofoofoo"
audreyt now that's not quite ruby :)
lypanov can i talk to evalbot_9190 in private or? 12:50
?eval 5
evalbot_9190 5
blm is self a keyword in perl6? ie no need to do my ($self) = @_;
lypanov evidently not
audreyt blm: yes it is
blm audreyt: Thanks
lypanov eval "h" xx 3
?eval "h" xx 3 12:51
evalbot_9190 ("h", "h", "h")
lypanov everyone just ignore the bot ;)
audreyt ?eval class Int { method ego { self } }; say 1.ego
evalbot_9190 OUTPUT[1 ] bool::true
audreyt lypanov: it might pay off in the long term to 1)apt-get install pugs and 2)use the interaction shell
(or check it out from svn, as building pugs is now much faster) 12:52
shmar :)
lypanov Get:4 ftp.nl.debian.org unstable/main pugs 6.2.10-4 [1662kB]
new enough?
shmar audreyt: testing is taking hella long though!
blm audreyt: You are adding a method to Int or creating a new Int class?
audreyt shmar: indeed
blm: adding a method to Int, as everything is by default open
xinming audreyt: hmm, will there be collision if get a method and sub with same name?
gaal ?eval sub hw($n) { $n ?? "~V" ~ hw($n-1) !! "~V~" }; hw 3
xinming in a class.
evalbot_9190 "~V~V~V~V~"
blm audreyt: Thanks :-)
audreyt xinming: yes there will, just like two subs with the same name 12:53
xinming eg: class A { method t { }; sub t { } };
lypanov makes funny faces at all the double symbols :P
gaal lypanov: s/hw/handwave/ and all will become obfuscated
xinming audreyt: hmm, then, How to make perl 6 act as perl 5 does? eg, make a sub both method and sub.
lypanov ummm
audreyt xinming: you say "submethod"! 12:54
lypanov is pugs shell really meant to bail on bad code?
audreyt (really, seriously, that's the way.)
xinming in perl 5, we can in fact call a sub directly, and also act as a method...
audreyt lypanov: well... 6.2.10 has a bad (non-ruby-influenced) shell
lypanov audreyt: want me to be a total git and report lots of things i dislike in perl6?
audreyt xinming: yup, and even if you declared something as a method, you can still call it directly 12:55
lypanov (/me already knows the response is yes hehe)
audreyt lypanov: it'd be better if you 1)write it down in a buffer and 2) nopaste/link to it
blm ?eval say "Hello, world!"
lypanov audreyt: would always do so
audreyt IRC is too noisy for this :)
evalbot_9190 OUTPUT[Hello, world! ] bool::true
audreyt lypanov: woot, then sure, definitely!
lypanov audreyt: it'd be "grar. this sucked" sessions while in the train
audreyt lypanov: sure, but in that case please pull the latest svn code
lypanov audreyt: shall do 12:56
audreyt to save us both times over already-fixed bugs
woot
lypanov++
xinming hmm, class A { method t is export { "test".say } }; t(); <=== will this be valid example?
audreyt my gripe with ruby can be summarized to: "I can't tell it to pretend to be a static language"
lypanov audreyt: "Currently, the P6Doc files are installed along with Pugs, into Perl5's sitelib path and formatted as manpages. "
shmar running make test now for at least an hour! 12:57
lypanov audreyt: there are html versions of this anywhere?
audreyt xinming: you are outside class A... so you need to call it A::t()
lypanov: will do so once I get sufficient sleep
which means e.g. tomorrow
xinming audreyt: Ok
shmar or is that another Xeno?
;)
audreyt lypanov: (packaging it to Perl6::Doc and uploading it to CPAN gets us HTML for free)
lypanov audreyt: okay. i'll work on compiling. whenever you managed to get this up ping me if you could then i'll read :)
audreyt: *nod*
audreyt lypanov: cool -- meanwhile there's always the docs/quickref/ tree 12:58
lypanov audreyt: has difficulty reading text without differentiating font styles :) 12:59
audreyt I see :)
lypanov gets distracted far too quickly alas without interaction
audreyt understands it, perhaps too acutely
lypanov heh
shmar ALL: are the Perl6 pods supposed to be differential (ie to perl5) or absolute? 13:00
IE: should I be writing as though the reader knows perl5 or not?
lypanov would say the latter, but has no real say ;) 13:01
shmar I agree...
In which case I will *cough* plagiarise *cough* liberally from the existing perldoc.... 13:02
lypanov maybe just provide references?
shmar hmmm.... 13:03
lypanov (assuming they are large sections)
shmar can we get a ruling
?
audreyt shmar: plagiarise is the one true way to go -- but in Overview, just provide enough info to get people started 13:04
think perlrequick and perlintro
shmar ok...
audreyt but just commit ahead -- we can always refactor :) 13:05
(and I agree, shouldn't assume too much p5 knowledge)
chris2 hey lypanov, audreyt et al.
audreyt we have the Perl6::Perl5::Porting and other ::Perl5::* for that
lypanov yoyo chris2
lypanov should stop reading up on the zeno's paradoxes :P
shmar ok.. gotcha...
gaal I am never forget the day I first meet great perldoc
lypanov goes back to $job
audreyt should really start writing to sasada/takahashi/matz/other_ruby_people for the Tokyo hackathon 13:06
13:06 kisu joined
audreyt I want to target YARV... or at least try :) 13:06
chris2 whew ;)
audreyt lypanov: after 2 hours or so (which is longer than our attention spans, combined), search.cpan.org/dist/Perl6-Bible/ should show a new batch of HTMLified Synopses 13:07
nothingmuch audreyt: please reload slides 13:10
i've clarified and cleaned up, i think
or at least hope
will continue to garbage collect
audreyt nothingmuch: url? 13:11
shmar argh.... no 'up' history in interactive mode!
audreyt shmar: can't be... if you have readline we use it
(for GHC 6.4.1)
alternatively, use rlwrap
which is cheap and cheerful
lypanov audreyt: thx. bookmarked 13:12
chris2 lypanov: rotfl, andy hunt asks me where he can download his *own* software :P 13:13
lypanov chris2: hehe
brb
13:13 iblechbot joined
shmar audreyt: you mean GNU readline? 13:18
13:19 kolibrie joined
audreyt shmar: yes, not the OSX one 13:19
13:26 lypanov joined 13:27 mj41_ joined
lypanov returns 13:29
audreyt finds this rails tutorial more amusing than the euro oscon one so far 13:30
lypanov wonders if he one day should learn rails 13:31
audreyt parsec++ 13:32
lumi: ok, some unrecoverable parse failures should now really terminate on the spot
committing 13:34
shmar szabgab: is there any particular reason that sourceforge.net is inaccessible? 13:36
svnbot6 r9191 | audreyt++ | * Parser: Refactor
r9191 | audreyt++ | between (symbol "{") (char '}')
r9191 | audreyt++ | into verbatimBraces
r9191 | audreyt++ | * Also, parse for q:code(:COMPILING)
r9190 | shmarya++ | Adding self to authors list
r9192 | audreyt++ | * Pugs.hs: In the interactive shell, lumi suggested we simply fail on 13:42
r9192 | audreyt++ | unrecoverable failures, instead of prompting "...>" to the user.
r9192 | audreyt++ | This is a first attempt at it -- anything that fails at the beginning
r9192 | audreyt++ | of a program (say a single ">") fails immediately.
r9193 | bsb++ | r9199@rex: bsb | 2006-02-27 15:33:09 +0200
r9193 | bsb++ | s/CODE/q:code/g + some waffle
r9194 | bsb++ | r9200@rex: bsb | 2006-02-27 15:39:26 +0200
r9194 | bsb++ | Haddock fixes, update docs and add docs/haddock_tips
lypanov chris2: yet again, my ibook keyboard is falling apart while it feels my typing wrath. muwhahahaha 13:44
13:44 scw joined
bsb gaal: how is src/Pugs/PIL1/Instances.hpp moosed? 13:44
chris2 lypanov: my f4 key feels weird, somehow
lypanov chris2: *wonders what f4 does* 13:45
chris2: you mean the volume decrease button? :P
chris2 volume down, yes
i dont think i use it otherwise :P
lypanov :P phew. i thought you emacs users used the f- keys for a moment then
chris2 f3 and f5 are like "click". f4 doesnt
lypanov: i have bound it to next-error :P 13:46
shmar adds my keyboard woes...
lypanov has had this keyboard for ~1 month and the letters on the n a and s are already half chipped away
shmar a spill killed my "O" and "P"
audreyt bsb: "make haddock" mooses it via GHC and then util/munge_haddock.pl
shmar very sticky.
chris2 i use f7 (gsub) and f12 (goto line) heavily
and the expose buttons, of course :)
lypanov oh yeah. you have panther at least, i forgot ;) 13:47
chris2 :P
shmar Tiger! :P
lypanov leopard! 13:48
chris2 etch! *ducks*
shmar lypanov: no way!
lypanov chris2: ugh
chris2 when there are no binaries for 10.3 anymore, i wont update, i gues 13:49
but never say never :P
lypanov chris2: you'll buy 10.5!
chris2 yeah. i skipped 2.4, can skip 10.4 as well :P
shmar why stay at 10.3?
lypanov chris2: lol. eww
bsb I meant "how is Instances.hs" moosed into existence by DrIFT?
chris2 shmar: $$ for stuff i dont need?
shmar good point... 13:50
lypanov lol
chris2 lypanov: hey, my debian survived 2.2 -> 2.6 while swapping the mainboard at the same time
lypanov you're nuts
my debian didn't even survive the last upgrade
audreyt bsb: by util/drift.pl -- see Makefile
shmar absoultely!
chris2 lypanov: i think it was slink when i installed it :P
lypanov ?eval "*shame* i started the ot" 13:51
evalbot_9194 "*shame* i started the ot"
audreyt nothingmuch: you never gave me the slides uri
lumi bsb: The makefile makes it happen.. Is this what you mean?
gaal fork-art.com/moose.htm
shmar gaal: LOL!
lypanov gaal: hehehehe
13:52 xinming joined 13:53 nothingmuch_ joined
audreyt nothingmuch: reading your slides now 13:54
audreyt thinks nothingmuch has the potential to become Whyish
gaal you mean stiff?
audreyt you just need to practice on foxdrawing
szabgab shmar: I guess because of the local sysadmin firewalled sourceforge 13:55
lumi Fox? I thought it was a cat
gaal definitely foxes.
audreyt Why's channel is Fox channel, not Cat channel
gaal devoid of legs, but foxes still.
gaal requires C8H10N4O2 13:57
audreyt only has C14H19NO2
lumi I can generate CO2 and N is very common around here 13:58
chris2 is embarrassed he forgot the formula of ethanol
nothingmuch_ the former is caffeine and the latter is amphetamines? 13:59
H2SO4 is quite cinematic
there is a geek rhyme someone told me once
"Little timmy had a drink but soon he'll drink no more
Odin- Johnny was a chemist's son,
audreyt nothingmuch: well, Methylphenidate
nothingmuch_ for what he thought was H2O was H2SO4"
Odin- but johnny is no more,
what johnny thought was H2O
nothingmuch_ Odin-: =)
Odin- was H2SO4 14:00
:p
There's numerous versions.
nothingmuch_ ofcourse
like the "great green globs of greasy grimey gopher guts" song
gaal lithium is quite cinematic when coming into contact with water
nothingmuch_ Methylphenidate - designer amphetamines?
Odin- audreyt: Methylphenidate is an amphetamine. :p
RitalinĀ®
nothingmuch_ aye
but Ritalin isa Amphetamine
not the other way 14:01
14:01 lisppaste3 joined
Odin- nothingmuch_: That's what I said. Methylphenidate is an amphetamine. :) 14:01
audreyt Odin-: right, hence "well"
Odin- And RitalinĀ® is someone or other's drug based on methylphenidate. :p
lypanov errr 14:03
profile.myspace.com/index.cfm?fusea...id=1609812
lypanov makes a face
remind me to not "feel lucky" again :P
audreyt make: don't know how to make a face. Stop
lypanov hehe
nothingmuch_ Odin-: aye, i meant "no sense in correcting audreyt" 14:04
Odin- Well, that's alright. There's not the least shred of sense in me either. ;) 14:05
audreyt nothingmuch: I start to think that the "Balance" in your slide may be replaced by "Eclecticism" 14:06
if not "Dialectics"
bsb Hey drifters, can I ci an attempt to make the drift output compatible with haddock 14:07
?
audreyt bsb: you asked "can I ci", the answer is always "yes" 14:08
bsb I just like being able to blame the person who said "yes" :)
shmar lol
audreyt bsb: you asked "blame the person", the answer is always "sure, you can blame audrey"
14:09 Qiang joined
lypanov hehe 14:09
audreyt nothingmuch: more seriously: s:g/Balance/Choice/
gaal I thought nothingmuch_ was our scapemoose 14:10
audreyt but I'm our moosescape
gaal bsb: if you do ci, redrift also.
LeTo audreyt: I've hacked feather ~lt/svnbot.p6 - it's using 'svn st -u README' to retrieve latest version (w/o log cluttering due to error message), but that needs that it's running in the local repo
bsb gaal: how? 14:11
svnbot6 r9195 | bsb++ | r9209@rex: bsb | 2006-02-27 16:10:58 +0200
r9195 | bsb++ | Try to make DrIFTs output compatible with make haddock
r9195 | bsb++ | munge_haddock.pl strips sections in #ifndef HADDOCK
audreyt LeTo: gotcha
gaal bsb, just run "make" after you modify the drift file; that should modify the Instnces.hs files
if it doesn't, find -name Insances.hs | xargs rm and make again 14:12
audreyt but you do need a copy of DrIFT in pugs/../
bsb I don't have drift installed so it don't do nothing
gaal that way people who don't have DrIFT see your changes
okies then let's get together and merge stuff
or just ci :)
wheee coffeeeeeee
audreyt bsb: perlcabal.org/~autrijus/tmp/DrIFT.tgz 14:13
bsb: you broke the build! blame audrey!
14:14 nothingmuch_ is now known as nothingmuch, kane__ joined, KingDiamond joined, nothingm2ch joined
szabgab anyone the list of our acutions www.osdc.org.il/auction.html 14:19
svnbot6 r9196 | audreyt++ | * add -cpp to make haddock happy(ish)
r9197 | bsb++ | r9212@rex: bsb | 2006-02-27 16:18:08 +0200
r9197 | bsb++ | removed cpp style munging from munge_haddock
audreyt TimToady++ # adopting the Takahashi style 14:21
nothingmuch wish me luck ;-) 14:28
shmar good luck\
bsb good luck, comrade Yuval 14:29
clkao audreyt: ping me if you happen to get js fixed 14:34
audreyt k 14:35
14:36 vel joined
audreyt working on it now 14:42
gaal I wonder if there's a trick to make pattern matching more succinct in Prim.hs 14:44
where we perform fromVal on almost all params
audreyt I've written some helpers
for numeric stuff mostly
but yeah, a better notation'd be a win
gaal you're our agent on the haskell' board ;) 14:45
audreyt clkao: somehow it seens to workforme
try again?
14:46 Ara4n joined
gaal nothingmuch is teh lord of teh photoshop!!!1one 14:47
okay okay google
@google lord of the pugs
audreyt www.eleanore.net/pug/pugs.sized.jpg
xinming lol 14:50
14:56 stevan joined
audreyt mm, future-resistant 15:00
it's like future-proof
(referring to nothingmuch's ongoing talk)
lumi Feature resistant
audreyt a feature-proof language wouldn't be fun
a feature-proved language otoh is nice 15:01
lumi no fun;
audreyt goto shop;
gaal feature hooved?
xerox feat ure
lumi De-moosified
audreyt xerox++ # oblique
xerox Yay. 15:02
15:08 hexmode joined
audreyt nothingmuch: bad example... it's an iterate loop you're shoing for the locality analysis in 1..$large example 15:12
nothingmuch: sorry for not catching it sooner
...but then, I doubt there's many people who are still following closely at this point anyway :)
bsb what? 15:13
audreyt bsb: you need a -fvia-C in FPS.hs
but I thought it's already there
shmar audreyt: I want to commit a mod of the Overview.pod file with basic operators... 15:17
audreyt shmar: go ahead :) 15:18
shmar requests validation 15:19
svnbot6 r9198 | shmarya++ | Added operators to the Overview doc
shmar audreyt: is that what you had in mind? 15:23
gaal shmar: <audrety> yes! 15:27
shmar ? 15:30
15:30 Amnesiac joined
shmar gaal: ?? 15:31
15:31 anatoly joined
lumi Gaal is channeling Audrey 15:32
Hi anatoly
shmar ahah!
anatoly hi lumi, hi gaal
shmar There is no associativity listed in the docs/quickref/op file... I am assuming (+5 second test...) 15:46
mj41_ hello, is win32 build ok? seems like nmake don't create src\gen_prelude.exe .... but 'C:/ghc\ghc-6.4.1\bin\ghc.exe -O0 -o src\gen_prelude.exe src\gen_prelude.hs ' from command line does 15:50
svnbot6 r9199 | audreyt++ | * Makefile.PL: "make ghci" should now work again, by compiling 15:52
r9199 | audreyt++ | the -fvia-C parts in a Prereq.hs (needs poetry for that)
15:55 nothingmuch joined 15:56 mako132 joined
nothingmuch moose ness 15:56
mooseness
svnbot6 r9200 | audreyt++ | * regen Instances.hs to repair "make haddock"
15:56 lypanov joined
nothingmuch m ooseness 15:56
oosem 15:57
15:57 bsb joined
nothingmuch osemo 15:57
YAY
shmar lol
anatoly moose, nothingmuch 16:03
bbl
bsb nothingmoose: sent my key 16:04
16:05 pmurias joined
svnbot6 r9201 | pmurias++ | Ported the Regular Expressions section, renamed as Rules, but 16:08
r9201 | pmurias++ | named rules and grammars are yet to be done...
nothingmuch bsb: yeah, i got it 16:09
thanks =)
16:11 rgs joined
pmurias Overview.pod referes people to perldoc -f should it be p6doc -f instead? 16:13
cognominal bon Larry est alle en Israel mais pas en France... 16:15
il faut dire qu'on fait pas des hackatons pugs en France.
shmar merci
cognominal Man get rape'd by two women zeugmatically 16:16
# je trouve que ca a une certaine poesie ce mot ajoute pour tromper les canner de spam
rgs I didn't knew this was a french-speaking channel. 16:17
integral o_O
rgs: pas normalement
shmar ;) 16:18
nothingmuch welcome, rgs
rgs integral: mais il est difficile de rabattre le caquet ōæ½xE0 ce bon vieux cognominal
re-hi nothingmuch
cognominal oops, I saw rgs, I thought I was on #perlfr
meaning, that's all rgs fault 16:19
rgs sure
shmar nothingmuch: there is not listing of precedence for MetaOperators and QuoteOperators..... I know that QuoteOperators are hectically difficult to document, but where do Metas fit into the hierarchy?
16:20 xinming joined
lumi Keysigning-p? 16:24
audreyt rgs: hey! 16:27
rgs hi audreyt ! 16:28
shmar there is not listing of precedence for MetaOperators and QuoteOperators..... I know that QuoteOperators are hectically difficult to document, but where do Metas fit into the hierarchy? 16:30
audreyt shmar: metaoperators are always nesting brackets 16:31
so they don't get prec issues
and they just convert infix to prefix
listop
and assumes normal listop prec
shmar ok.... I am building a skeleton for the core Operators.pod file... I am attempting to follow the existing style as much as possible. 16:32
16:32 Sandrina joined
anatoly hi audreyt! 16:34
audreyt anatoly: hey!
gaal yo anatoly 16:36
16:37 kane__ is now known as kane-xs
kane-xs waves at audreyt 16:37
shmar I assume that you'll be wanting sections on Meta and Quote Operators... 16:40
16:55 justatheory joined
integral anyone know how long a smoke takes on feather atm? 17:11
audreyt no idea
integral I guess I'll find out! 17:12
audreyt how exciting! :)
integral smoke doesn't upload by default? 17:13
clkao unles you have a very old smoke conf 17:14
17:15 orafu joined 17:17 nothingm1ch joined
clkao audreyt: what's the story with perlcabal.org/~autrijus/osdc/svk.xul 17:20
it looks so funny :p
this problem is solved in 1982, and then "Perl6: Parser!"
obra hah 17:24
integral has anyone else noticed that Data.Yaml.Syck is getting recompiled on every build? 17:29
PerlJam I haven't noticed that, but such behavior is usually indicative of someone using keyword expansion. 17:31
integral I'm not using svk in between. 17:32
I'm doing incremental builds where I'm only touching Pugs.Parser
PerlJam weird
anatoly integral: I noticed that, too, just a few minutes ago. 17:33
integral ah, there is a rm -rf dist/build/src/Data/Yaml being run before the util/build_pugs
anatoly integral: yeah, I guess it's generated on each run from the other file in that directory. But if it stays the same all the time... 17:34
PerlJam just looked at the imaginary timeline again 17:37
It's almost like it's not imaginary :-)
17:38 Amnesiac is now known as amnesiac
integral "Prisoners are always force fed when they decline to eat themselves" 17:40
*blink* t/01-sanity/08-test.t is failing currently 17:44
anatoly integral: svk log -r17227
integral: not about that, about that other thing
integral ugh, kludge eh? *sigh* 17:45
oh, I see why that's failing for me, I think
oh, maybe not :-/ 17:46
hmm, maybe.
ah, yes, missing blib6
anatoly whew :)
integral my brain finally spotted the /usr/share path for Test.pm 17:47
anatoly although Test.pm failing basic sanity for Test.pm seemed kinda nice
"I always lie"
17:48 ghenry joined 18:03 putter joined
putter tewk: ping? 18:03
anyone around who could give a brief summary of the op table parser algorithm? (I find myself rewriting assembly code trying to figure out what all the control flows are, but some fall throughs look a bit odd, and... blech). 18:07
18:17 Bit-Man joined
clkao audreyt:prelude for js can be generated, but it seems test.pm is failing 18:27
pugs: src/Pugs/AST/Internals/Instances.hs:(684,4)-(691,63): Non-exhaustive patterns in function showPerl5
tewk putter: pong? 18:29
putter hi tewk. just now discovered parrot's compilers/pge/PGE/OPTable.pir has nice algo documentation. yay. :) 18:35
was going to ask you for the 1k-foot level overview of parser. but will look at docs first.
tewk Yeah, it is nasty. It is essentially modeling a state machine. The pretty solution would be to put all the local variables of the parse function into a hash and turn the labels into functions that take the hash as a parameter. 18:36
18:37 rantanplan_ joined 18:38 Daveman joined
tewk Unfortunately it is just a transpose of OPTable.pir 18:38
$THESIS has stolen all my tuits lately:( I haven't gotten back to OpTable. 18:39
18:41 xinming joined
putter I'm thinking of doing a p5 reimplementation - clean, simple, and slow. Context is I'm going to try doing the p6-regex -> regex-parse-tree transition with an opparser, since it's the "right thing", and we'll need the oop anyway shortly thereafter for the full p6 parser. 18:41
18:41 dduncan joined
tewk putter: Cool I'd love to see how it comes out. 18:47
putter basically Regexp::Parser+continuation-backtracker (cleaned up version of spike) + opparser +p5/6-regexp-op-grammar provides p5/6-regexp -> match/parse-tree, tweak to regexp-ast, then regexp-ast -> rules-using-continuation-backtracker. currently have or know how to easily do all the bits but the ooparser. 18:48
audreyt putter: lwall and nothingmuch was syncing how to do that today
we just went back from dinner after conference, where I just suggested (5min ago) that we start with p5 18:49
putter oh, objective is to bootstrap a self-hosting "correct"/dynamic p5+p6 regexp engine.
audreyt and ignore the nonregular parts
eg => autoquoting the left
and just focus on proving that the regular grammatical-category, longest-token-match, model works
for a sufficient set of Perl 6 programs
putter tewk: yeah. we'll see. "copious free time". but the OpTable.pir docs look nice. 18:50
audreyt i.e. just parsing t/*, not thinking about evaluation at this moment
putter audreyt: putter thinks... 18:51
audreyt in other news, Larry serialized both OpTree and syntactic parse tree as YAML::Syck today, which handles 95%+ of perl5 core tests
putter oh, wow
audreyt I may do a spike tomorrow morning on compiling that to p6 -- but we can't emit p6 yet, though anatoly asked for a task and I suggested that
(anatoly is this wonderful lambdaperson I met in Arbel hackathon who grokked monads just from the Wadler papers in 4 hours) 18:52
putter ;)
audreyt: re regular being sufficient... I've had my doubts about that, but didnt want to make a fuss until things were working. but fine for bootstrap/spike. 18:53
audreyt putter: I have my doubts over the parsing model works or not
even just for regular stuff 18:54
because it's never proven for anything
putter Full backtrack works :) I've been thinking of the expr/statement extension stuff as simply a nice api, which may or may not be connected to how parser guts work, or the ast gets optimized. 18:57
svnbot6 r9202 | gaal++ | r9187@sike: roo | 2006-02-27 20:52:28 +0200
r9202 | gaal++ | * Adding "A Peek into Pugs Internals", OSDC.il talk
18:57 gaal joined
putter s/ast/parser ast/ 18:58
18:59 netanya_goy joined
putter admits never really liked monads. likes arrows _much_ better. 19:00
audreyt agrees completely
xerox doesn't really know 19:03
I mean, some Monad rocks. 19:04
..Monads rock, even
And if Monads are Arrows.. we agree ;-)
19:05 Arathorn is now known as Aragone
putter lol. putter goes to find arrows paper for xerox etal. googles for "arrows". not quite the right thing... ;) 19:06
anatoly lambda arrows works though! :) 19:07
19:08 nothingmuch joined
nothingmuch mooseness 19:08
xerox putter: ?! 19:09
19:09 calanya joined
xerox Say MonadRandom, that was a cool one. 19:09
putter runs away from the web. so many papers, so little time. in so many fields ("The Life Cycle Of Bacteriophage Lambda";). 19:12
anatoly: monad arrows too
re arrows, www.haskell.org/arrows/index.html citeseer.ist.psu.edu/hughes98generalising.html 19:14
svnbot6 r9203 | rootmj++ | * fix MSWin32 build with GHC default install path
r9203 | rootmj++ | * some tabs to spaces conversion
putter haskell.readscheme.org/ exists, but isnt getting updated. 19:15
haskell.org/hawiki/UnderstandingArrows 19:16
why does fp chose such unfortunate names. "attribute grammar"-- 19:18
nothingmuch putter: ping 19:20
putter actually, no. I'm confused. Hs attribute grammars do in fact look like what I think of the name meaning in a parsing context. So... err... why is everyone so excited?
nm: pong 19:21
audreyt putter: I'm not excited
putter lol
nothingmuch audreyt told me to bug you about parsing =>
audreyt I personally think generic programming, syb style, subsumes much of AG ability
nothingmuch syb?
putter "syb style"?
nothingmuch oh
scrap your boilerplate
i'll read that tonight
putter ah
audreyt er, no, not "bug" putter, but sync your visions of a p6-parsing-p6-grammar using p5 bootstrap
putter lol 19:22
19:22 Shachaf joined, SolidState joined
putter vision syncing++. putter sooo misses design conversations by a whiteboard. :( 19:23
nothingmuch come to hackathons?
audreyt where are you based? boston? 19:24
putter good suggestion. about time to really get a job/funds I think.
yes
nothingmuch nothingmuch.woobling.org/perl6_comp
audreyt I think I'm going to visit NYC for stevan, and hence Boston for obra, on the coming July or so
nothingmuch meet up with stevan
audreyt not 100% sure
nothingmuch 2-3 hrs drive, right?
stevan about that 19:25
audreyt: I am actually right in between NYC and Boston
well actually 3 hours from boston and 2 from NYC
nothingmuch isn't connecticut to the east?
stevan of NYC yes
nothingmuch and boston to the north?
stevan but west of boston
west and a little south 19:26
putter nifty. can come down to nyc.
nothingmuch ah
putter s/can/I can/
stevan I have already told my wife we might have a basement full of perl hackers this summer
nothingmuch =D
putter waits for the key bit, her response... :)
stevan putter: she sighed and said "ok" 19:27
putter :)
stevan audreyt: if we coordinate well, maybe we can have an east coast hackathon of sorts before/after YAPC::NA
gaal stevan: but did you tell her people who know falafel will be coming?
putter lol
stevan perl hackers who cook good falafel are always welcome :) 19:28
gaal actually i don't know how to make it
stevan however the pizza oven will be in the backyard,.. far away from the house
nothingmuch making falafel is nontrivial, i think
gaal just to eat it :)
nothingmuch fuck off
stevan LOL
nothingmuch all of you
gaal but i can make you pizza
stevan gaal: there is a great falafel place in New Haven
good Pizza too
gaal stevan: nothingmuch burned no pizza this hackathon! 19:29
stevan nothingmuch: congrats :)
nothingmuch nor did i evere burn pizza ever!
i burned *WAX PAPER*
stevan I suppose lables in hebrew are easier for him to read
nothingmuch which i thought was a synonym for *BAKING PAPER*
due to our host john misleading me!
anyway
shutup
stevan hands nothingmuch a thesaurus
fucking canadians :P 19:30
putter so, the first NewEngland P6 Mongers meeting is July at the latest :)
stevan putter: yes :) 19:31
nothingmuch pings stevan in private message for the FOURTIETH TIME!
stevan there is actually a nice (cheap) conference center down here,.. on the beach too
stevan points out to nothingmuch that he has long since forgotten his freenode password
nothingmuch: ping me on irc.perl.org
if we coordinate well enough I might be able to get us some cheap rooms ($29 USD a night) and a small conf hall 19:32
putter that could be quite neat. wonder who all could make it... 19:33
nothingmuch: re perl6_comp, empty, yes? 19:34
nothingmuch putter: darcs pushing
it's done now
putter looking...
kolibrie might be able to make it to an east-coast something-or-other 19:37
stevan kolibrie: nice 19:38
19:41 Ovid joined
putter nm: the talk slides only, yes? read them a few hours ago when backlogging :) let's see, thoughts... 19:42
mostly agreement, so I'll focus on the things which seemed odd... 19:43
szbalint Rules is the new name for regular expressions, right? 19:44
19:46 samuel joined
putter I think we're going to eventually need better vocabulary for talking ab out "parsing", etc. p6 doesn't combine parsing and execution only from the perspective of the extreme dynamic syntax corner of the universe. everyone else could consider it a combo. so need ways to talk about more fine grain distinctions. 19:47
biggest think is I don't think I really by the ast node replacement story. I think p6 is itself powerful enough to do it at the code level. 19:48
s/think/thing
szbalint: yes
in order of depreciation: rules regex regexp regular-expression
;)
nothingmuch: ? 19:49
19:49 nothingmuch joined
putter 3- i'm not sure I really understood the ffi as emitting argument. 19:50
szbalint ty.
audreyt needs to 1)shower 2)sleep
ttyl :)
lypanov night audreyt
PerlJam rules need a better name. 19:51
lypanov svn co's
PerlJam something more jargony
szbalint irregular expressions :)
audreyt expressive regulations 19:52
lypanov irregexps hehe
putter actually, that's probably not quite right. I kind of expect rules and regex to diverge and specialize. eg, regex ends up meaning the pattern itself, and "rules" the bigger-picture grammar/Rule context/object. maybe.
lypanov nice :)
PerlJam "regular expression" has a very specific, computing sciency meaning. It even sounds esoteric. "rules" just sounds plain by comparison
lypanov putter: thats what i'd always understood actually
putter lol # re iexps
integral like LALR(\infty), but not?
hmm, recursive descent is LA(\infty)?
PerlJam Anything eulectic and euphonic is good. (I wish I could think of such a name) 19:53
putter PerlJam: very specific, long ignored ;)
Kattana irregex? thats not very specific? are they context free or context sensitive or unrestricted? 19:54
PerlJam putter: nah, we're always telling people to use HTML::Parser rather than REs because of the limitations of REs (even though they really don't apply to perl, except when we try to limit ourselves to the tame)
putter Kattana: unrestricted i believe, though particular engines may have other ideas. 19:55
Kattana mhm I think so too, I have been reading wikipedia to catch up on all thise stuff 19:56
PerlJam heh ... I just thought "well, I guess we'll have to just keep calling them rules because we already have the "rule" keyword" hahaha; am I insane? ;-)
audreyt Grammar Rules
lypanov naah, grammar sucks
oh!
PerlJam GRs
audreyt P6GRs
lypanov grar's
audreyt Grammar Rule Expressions 19:57
PerlJam not to be confused with the GRT
audreyt => GRE
lypanov gre's
i could dig gre
it has another meaning, but different domain
PerlJam Grammars Ready to Eat!
lypanov lol
audreyt everybody knows GRE signifies intelligence and dedication and silliness
PerlJam That works for me. Now we just need to start using that term always so that the meme will propigate :) 19:58
putter oh, that's the other thing. every time I start talking about grammars, I want to distinguish grammar, the concept, from Grammar, the p6 construct. (also from grammar, as in p6, but not tied to some particular Grammar object)
what should the grammar(1) parse, and how should we architect the grammar(3) from the available grammars(2). 19:59
audreyt putter: p6 jargon so far is optimized for pun
which has been my #1 roadblock as implementor
putter oo, #1-ish, interesting... I wouldn't have expected that. 20:00
audreyt like, in day 5 or so I was confused about what Array means
and it took me 10 months to figure out the punny parts and what each means 20:01
putter is still confused about what Array means... is it role? a class? (is a class now a role?) a role/class parameterized on argument type? on underlying implementation (how?)? an abstract class? I'm very fuzzy on it all. 20:03
lypanov its a rolling class
putter lol
stevan rolls up another Array and passes it to putter 20:04
lypanov honestly has no clue
Kattana is furry with confusions, way beyond fuzzy.
Although I now see how all the puns crept in.
Daveman :o 20:05
putter :)
putter ponders puns, prototyping prophalactic precautions 20:07
gaal a lolling crass 20:10
putter nothingmuch: my biggest question for you is do you think a ast node modification scheme is still needed, or can normal p6 be the ast and modification mechanism. 20:11
nothingmuch sorry 20:12
my brain is overflowing
i have to do 3 things at once
putter oh, no problem 20:13
szbalint plenty powerful projects progress Perl: [paste preferred, people] 20:15
putter notes steven keeps handing me things I don't quite know what to do with. ;)
spinclad TWIMC: the new Perl6-Bible seems to be missing S03.
stevan wonders what he gave to putter recently ? 20:16
spinclad could make a New England hackathon, would love to come
putter stevan: was trying to remember. dont quite.
stevan spinclad: nice, where are you located?
putter szbalint: lol
stevan takes away the Array he rolled for putter, youv'e had enough I think ;)
spinclad brookline, MA (almost surrounded by boston 20:17
)
putter oh, awesome.
lypanov spinclad: its right at the end
spinclad: missing a title maybe?
lypanov has no idea how this stuff works
putter thinks it's time for a NE P6 Mongers wiki or some such. If only to keep track of all the people.
spinclad likes nep6wiki 20:18
stevan a mailing list might be easier to get started
audreyt spinclad: fixed, cpan-uploaded, thanks 20:19
(and yes I'm in the middle of shower)
spinclad audreyt: thx and wow
lypanov lol
spinclad xinming: juxpatose: to juxtopose adjacent characters, eg ':$foo' -> '$:foo'. (almost an autonym) 20:20
s/juxto/juxta/ #oop
similar to 'spinneroosm' 20:22
gaal actually, "juxtapose" just means putting one thing next to another
putter stevan (CT), obra (Bos), spinclad (Bos), putter (Bos), anyone else NewEngland-ish?
gaal not placing them in the wrong order
but the two are often mixed up :-) 20:23
kolibrie DC
putter lol # gaal
spinclad juxpatose: to interchange juxtaposed characters, eg ':$foo' -> '$:foo'. (almost an autonym)
gaal lol # spinclad 20:24
stevan juxtaposes gaal's definition with spinclad's
spinclad i think i had that in mind even, but couldn't squeeze it out my fingers
putter stevan (CT), obra (Bos), spinclad (Bos), putter (Bos), kolibrie (DC); anyone else NewEngland-ish?
gaal putter: I plan to be New English in the temporal vicinity of YAPC 20:25
SolidState hi 20:26
stevan looks into falafel scholarships in the Chicago area for gaal 20:27
spinclad tuxjapose: to interchange adjacent concepts, eg 'interchange' and 'adjacent' 20:28
20:28 calanya left
putter NewEngland P6 Mongering: in vicinity: stevan (CT), obra (Bos), spinclad (Bos), putter (Bos), kolibrie (DC). plus, for a possible July(YAPCish) hackathon: audreyt, gaal 20:28
stevan click 20:29
SolidState I'm not sure I'm reading Synopsys 12 correctly, should classes have a default constructor called "CREATE"? Or do I have to define my own (e.g. "sub new{bless{}}") the same way I did in Perl5?
stevan s putter's edit button
spinclad too
stevan SolidState: most of the time the construction will be done for you, you just augment it with BUILD
SolidState: Perl 6 will build your instance structure for you (if you want it to that is) 20:30
SolidState Is that currently implemented in Pugs?
stevan SolidState: most of it yes, but it is temporary until the real metamodel gets done
20:31 justatheory joined
SolidState That's okay, but I have a file with the contents "class Foo; 20:31
\n
1;
stevan SolidState: class Foo { has $bar; }; my $foo = Foo.new() will create a Foo instance with a slot for $bar
SolidState and in Pugs I load Foo 20:32
and then I try to create a new object:
my $obj=Foo.CREATE();
and it doesn't word :(
work*
*** No compatible subrountine found: "&CREATE"
stevan SolidState: use Foo.new()
CREATE will almost never be called outside of the inner workings of the class 20:33
putter can attributes be applied remotely? eg, instead of sub foo() is frabjous {...}, it's just sub foo() {...} and then later &foo.is<frabjous> or somesuch. 20:34
hmm, if signature, return type, body(?), are all attributes, that would allow complete separation of definition and assembly. :) 20:36
pity they can't be lexically scoped... :) 20:37
20:37 samuel joined
spinclad lypanov: re S03, i see what you mean now 20:37
PerlJam putter: what do you mean? 20:38
20:38 GeJ joined
PerlJam putter: you can make a lexical alias that hides the sub and has whatever attributes you want on it. 20:39
putter PerlJam: right, but say... 20:40
SolidState stevan: How come your suggetion (the one-line class definition+instance) works but when I put the class definition in a file and load it using :l in Pugs and then try the second part I get '*** No compatible subrountine found: "&new'"? Any ideas?
putter { infix:<+>.is<somethingorother>; 3+2+4 } I just want the somethingorother property for... oh, you could 20:41
20:41 rantanplan_ joined
stevan SolidState: no idea, sounds like it is not loading your file right 20:41
putter { my infix:<+> := $OUTER::infix<+>.clone.is<somethingorother>; ... }
stevan SolidState: try
say ::Foo.defined ?? "true" !! "false" 20:42
see if the Foo namespace is loaded basically
putter though that changes infix:<+>'s identity, which was not a desired sideeffect.
spinclad { &fib.is<cached>; fib(17).is<fast> }; fib(17).is<slow again> 20:43
putter I just remembered how is() is implemented. so I guess separate identity actually is the right thing, rather than a sideeffect. 20:45
spinclad my foo is Sub; foo.signature(Int -> Int); foo.body:{...} ??
putter is() is (optional) rule does()ing plus (optional, general) object mutation. 20:46
SolidState stevan: I got "true" 20:47
putter my foo is Sub; ... putter goes to look at Snn...
spinclad defining a sub should desugar to a series of steps something like this, I guess... 20:48
SolidState stevan: can u reproduce my problem? 20:50
stevan SolidState: sorry no, I havent compiled pugs in ages :) 20:52
putter sub foo is Sub; foo has signature; signature.trait_auxiliary:<has>(&foo,\(Int $x -> Int)); ... or something like that
SolidState stevan: I didn't compile pugs, I downloaded the binary zip file
stevan: 6.2.11
putter a runtime mixin role, plus a hand call to <is>, which doesn't get done for you with runtime mixins. 20:53
stevan SolidState: I would avoid the :l part and just "use Foo" in a file
SolidState stevan: www.jwcs.net/~jonathan/perl6/ is where I downloaded from
stevan: okay, I'll try that out 20:54
stevan: From a file, the code works okay. So thanks :) 20:59
stevan SolidState: cool :)
putter nothingmuch: multi emit(Any $x) {die "dont know how to emit $x"} multi emit(MultiSub $f where $f.object_id == Perl6::Prelude::infix:<+>.object_id) {...} emit(parse($p6code)); 21:00
or something like that, maybe?
multi emit(MultiSub $f where $f ~~ Perl6::Prelude::infix:<+>) {...} 21:01
21:01 ananiev joined
ananiev ummm 21:02
integral hello?
ananiev hi 21:03
just wondered if perl6 is going to have a nice threads model, because itreads sucks for me and I have to use java to implement my server 21:05
integral definately! 21:06
azuroth what makes nice, for you?
integral There's going to be lots of stuff on the lock-free programming/transactional memory front for a start
putter nothingmuch: ok, I'm simply confused. of course you have the ast anyway for compiling. and now I understand you ffi slides. looks good. leaving the question, how do we bootstrap?
integral also I hope for stuff like: @a = map -> $x { async { is atomic; f.($x) } }, @b; 21:07
ananiev lightweight, everything shared
integral hmm, there's a draft spec for this somewhere! 21:08
putter ananiev: en.wikipedia.org/wiki/Software_tran...nal_memory 21:09
azuroth is atomic? I would've thought use would make more sense there
integral azuroth: it's not a pragma but a adverbial modifier thingie to the block
ananiev: try svn.openfoundry.org/pugs/docs/Perl6...rrency.pod 21:10
ananiev thanks
azuroth hmm, okay 21:11
21:14 rantanplan_ is now known as Snoop|Work, Snoop|Work is now known as rantanplan__
lypanov wonders how long this compile is gonna take 21:16
(ibook, 1.2ghz, -O0, no prelude build)
putter memory may also be significant. but things should be faster than a week ago. 21:17
let us know how long it takes! :) 21:18
integral the default 256MB of ram is not good for pugs builds :) 21:19
svnbot6 r9204 | gaal++ | r9222@sike: roo | 2006-02-27 23:12:35 +0200
r9204 | gaal++ | * Updates recovered from dead laptop
21:20 rantanplan_ joined
putter notes dev.perl.org/cvs/ has a broken link cvs.perl.org/viewcvs/project , and even when you work around it (cvs.perl.org/viewcvs/ ), parrot is still unavailable. 21:24
is there a working parrot repository viewer anywhere?
got it, svn.perl.org/viewcvs/parrot/ 21:26
sigh, though it just broke. 21:27
21:28 xinming joined
putter back. here is OPTable.pir which was mentioned earlier (docs at the bottom) svn.perl.org/viewcvs/parrot/trunk/c...iew=markup 21:29
lypanov integral: i'd 3x that luckily :) 21:30
errr 21:31
its finished...?
s/d/ve/
very neat. more than
integral it'll be the gain from the prelude change :) 21:32
lypanov okay 21:33
first bug found
interactive shell, imo ctrl-d after a first line shouldn't cause a bail but rather "end the entered script"
how would i report such an issue?
putter lypanov: so how long did the build take? 21:34
lypanov putter: 10:12 -> 10:30 ish maybe
max
putter good to know. thanks. 21:35
re shell suggestions... might want to batch observations into a paste... or there is an interactive shell written in p6 around someplace which could be dusted off ;) 21:36
lypanov hehe 21:38
wheres the code for the shell?
justatheory Someone should send Orwant this mug: www.usatoday.com/tech/news/techinno...htm?csp=15 21:39
lypanov all source files are .hs right?
integral lypanov: Pugs.Shell?
lypanov or does haskell use anything other than this?
integral yep, .hs or .lhs, but we don't use .lhs for pugs-original things 21:40
21:40 r0nny joined
lypanov k 21:40
21:41 salty-horse joined
putter lypanov: the current pugs shell, yes. there is also a examples/eval.p6 21:43
21:44 salty-horse left
putter bbl & 21:44
21:44 nnunley joined 22:03 SamB joined 22:36 justatheory joined 22:55 pdcawley joined 23:03 jp-autar1 joined 23:26 epsalon joined 23:53 theorbtwo joined