Check your feather email | spec.pugscode.org | paste: sial.org/pbot/perl6 | pugs.blogs.com | www.treehugger.com/files/th_images/paradigm.jpg [set by audreyt on 2006-08-29 05:26:57 -0700]
Set by Akwa|user on 14 September 2006.
00:04 penk joined 00:22 nekokak joined 00:27 mako132_ joined
audreyt [particle]: hey, I made you metacommitter 00:40
[particle] audreyt, thanks, what does that mean? 00:41
audreyt meaning you can freely hand out committer bit the next time someone asks :)
most people regularly appearing here in #perl6 has metacommit bit
[particle] excellent, now i need to figure out how...
audreyt a sec 00:42
login to rt.openfoundry.org first and then rt.openfoundry.org/Foundry/Project/...?Queue=270 00:43
lambdabot Title: Pugs -- OSSF, tinyurl.com/fc5ql
audreyt which I've just added a DNS alias to commitbit.pugscode.org/ (takes a while to propagate)
[particle] del.icio.us, thanks! 00:46
audreyt cool!
[particle] so, i've got ~5000 static html pages, and i'm looking for a search tool. any leads? 00:47
audreyt uhm, htdig?
[particle] yeah, that's the first thing i thought of. you seem to know of everything bleeding edge, so i figured i'd ask ;) 00:48
TreyHarris audreyt: i'd be happy to help with that as well... i was here too when yts asked 00:49
audreyt TreyHarris: done 00:50
TreyHarris thanks :)
audreyt :) 00:51
zptao grep 00:53
or write a script for it 00:54
ajs_home I gave up. These days I use MediaWiki, which has decent (but not great) search built in.... I'm thinking of applying Perl 6 to improving its search as an early project once pugs is stable.... 01:03
audreyt pugs is by definition no more stable than the perl6 specs, though... :) 01:06
01:10 aufrank joined
[particle] pugs is stable, if you don't C<svn up> ;) 01:12
audreyt [particle]++ # jhi will approve of this definition 01:13
01:22 xinming_ joined 01:23 Aankhen`` joined
audreyt TreyHarris: the "your" declarator is already in perl5 01:26
via Acme::Your on CPAN
svnbot6 r13717 | audreyt++ | * Prefix +^ now takes arbitrary-precision integers; it's simply: 01:36
r13717 | audreyt++ | sub (Int $x) { -$x - 1 }
[particle] a 2's compliment op? 01:37
TimToady yes 01:38
[particle] hrmm, S03 says 1's compliment ...checking which is correct... 01:40
01:40 kanru joined
TimToady they're both correct 01:41
2's complement merely changes the numeric meaning of flipping all the bits.
[particle] okay, so this is pure bit twiddling, it's not really -$x - 1 01:42
TimToady it's assuming we're using arbitrarily large integers to store bits as 2's complement
[particle] ok. but the true internal representation is ieee, at least for parrot (i think that's specced, too) 01:43
audreyt perl6's Int is not parrot's I32...
[particle] (and ieee is sign-magnitude) 01:44
TimToady it doesn't actually have to keep track of all the bits, since there are an infinite number of either 1 or 0 bits there.
[particle] audreyt, yes i know.
does this extend to int, or just Int?
audreyt +^ on int probably flips bits using native repr
[particle] ok, great.
audreyt however, undecorated numeric literals assume Int
not int, behaviour 01:45
so +^0 is always -1, I think
[particle] yes, that's sane.
unless pragmata apply
TimToady we'd like to think so o_O
[particle] ack *thppt*
audreyt TimToady: S03:926 is misleading, then, as it talks about machine dependent behaviour
svnbot6 r13718 | audreyt++ | * Make interpolating <<$x>> behave a bit more like <<literal>>: 01:48
r13718 | audreyt++ | my $x = 'foo';
r13718 | audreyt++ | << $x >>; # Str('foo')
r13718 | audreyt++ | << $x $x >>; # Seq(Str('foo'), Str('foo'))
r13718 | audreyt++ | It used to be that it produces a list no matter what, but that
r13718 | audreyt++ | wasn't the behaviour for noninterpolating literals.
01:49 MacVince joined
audreyt TimToady: should +^1.5 coerce, or use native Num flipbit? 01:53
TimToady Wouldn't the standard implicit semantics coerce unless Num defines flipbit? 01:56
audreyt sure, but the opcode is spelled +^ not int^ 01:58
but otoh, bitwise on floating is something perl5 didn't do
maybe surprising if it starts doing that by default 01:59
TimToady but if there's prefix:<+^>:(Int) and not :(Num)...
audreyt nodnod.
ok, as-is then
TimToady I think it works out in floating anyway: int(-1.5 - 1) - 1 == 1 02:00
audreyt good (floating) point 02:01
TimToady real-ly 02:02
audreyt triages on 02:03
02:03 weinigBBL is now known as weinigZzZ
TimToady how difficult will it be to rewrite @.foo() to @(self.foo()) or self.foo().[] 02:03
lots of the ext tests you're working your way toward fail on that... 02:04
audreyt hmmm 02:05
TimToady (on account o', that's what I translated ./foo() to.)
well, sometimes.
hard part is picking up the .()
audreyt yeah. need case analysis
seen @.foo, don't parse it as Var, instead parse it as call 02:06
jabbot audreyt: I havn't seen @.foo,, audreyt
audreyt but "has @.foo()" invalid
right?
so this only makes sense in expr
not sig position
TimToady yeah, unless we want @.foo to have a sig for some reason. :/
actually, didn't S09 allow sigs on arrays and hashes? Lemme check... 02:07
yeah, it does currently. 02:08
audreyt hrmph. 02:09
what about 02:10
@::('.foo')
?
lambdabot Unknown command, try @list
audreyt is that even legal?
TimToady It could probably be made legal, but I'd have to see a decent use case for separate a twigil from its sigil. 02:11
audreyt ok, not considering it then 02:12
theorbtwo @.::('foo') is legal, though?
lambdabot Unknown command, try @list
TimToady in fact, one could argue that the reason for separating them is to denature the twigil.
audreyt TimToady: what's 02:13
TimToady much like $object.'*foo' calls the *foo method
audreyt $.foo.(1)
?
it's the same as $.foo(1) ?
TimToady I think .() and () are always the same as postfix. 02:14
audreyt okie.
zptao did they cut your balls off audreyt?
or do you just tuck it 02:15
audreyt TimToady: also, I wonder if @.+foo should be made legal
probably not.
(quantified call)
theorbtwo gets "I My Me Strawbery Eggs" flashbacks.
TimToady we can always add it later if it makes sense. 02:16
audreyt @.foo: 1,2,3
lambdabot Unknown command, try @list
audreyt sane?
TimToady I think it should be equiv to @.foo(1,2,3) 02:17
it really is a method call underneath, after all. 02:18
self.foo(1,2,3).[]
it's just an attribute with a fancier name.
ajs_home I have a question about my recent RFC on p6l if anyone has a second. It was supposed to be a quick thing, but at this point, I'm not sure. Is it worth pursuing, or should I just drop it? 02:19
Hmm... I'm sensing "just drop it." 02:20
TimToady I think it's just something to bear in mind as we go along. 02:21
We're all feeling our way into the roles thing, and how it relates to scopes.
my original proto notion was really directed more at lexical scopes, but if it can be extended to class scopes, that seems like a cool idea. 02:22
ajs_home Fair enough. I was just getting antsy, seeing all of the places that the same multi method would be exported from each class... with the hope that they would all play nicely, but I have to admit that my idea was just a stop-gap, and not a full solution 02:23
I'll let it go for now.
Thanks
TimToady at the moment, I think the overall sentiment is that we mostly just want to get the thing out the door without boxing ourselves in
ajs_home Out the door sounds like a nice plan
TimToady but there's plenty of wiggle room left
(fingers crossed)
02:24 mako132_ joined
audreyt TimToady: implemented... 02:26
resmoking to make sure
mugwump audreyt: do you know of any words with Ēš (or other tonal Ć¼) in the Hanyu Pinyin? 02:33
audreyt the umlaut U 02:34
mugwump or anyone with some zh knowledge? :)
audreyt is used not as a tone
but as alternate spelling of yu
mugwump oh, so it's not a distinct vowel? 02:35
so, HĆ nyĒš would be correct?
audreyt as in nyu (woman), lyu (travel)
actually "yu" doesn't need the umlaut 02:36
it's only needed after n/y
so yu3 is correct in itself, the umlaut is probably not canonical 02:37
zptao where do you keep your genitals audreyt 02:40
02:40 TimToady_ joined
zptao a special jar? 02:40
guess thats what happens when you spend so much time in a code window you lose your own personality 02:42
audreyt zptao: would you like a pugs commit bit? 02:43
zptao no
02:44 ChanServ sets mode: +o audreyt
audreyt too bad, then... 02:44
02:44 audreyt sets mode: +b *!*n=k0k0@*.dsl.pltn13.pacbell.net, zptao was kicked by audreyt (audreyt)), audreyt sets mode: -o audreyt, agentzh joined
[particle] audreyt++ # style points 02:44
audreyt *curtsies* 02:45
TimToady: this test:
class A { method a { $.a = 1 }}; A.new.a;
hangs after the commit. expected, right? 02:46
TimToady_ yes, should use $!a for that.
audreyt k 02:47
02:47 mauke_ joined
TimToady_ course, then it should fail, since there is no such attr... 02:47
pasteling "agentzh" at 210.22.200.67 pasted "Anyone can help me to translate the following poem to modern english?" (8 lines, 304B) at sial.org/pbot/20043 02:48
TimToady_ and in any case .a isn't an rw method...
02:48 pfenwick joined
TimToady_ that is modern english, according to the dictionaries. 02:49
agentzh TimToady_: i can't understand it...
TimToady_ 'course, that's basically anything after Chaucer... 02:50
agentzh wow
TimToady_ you don't need the poetry part of it preserved?
agentzh i prefer something after the world war II
TimToady_: no, i don't.
02:50 bcorn joined
agentzh i'm just trying to translate that to chinese. 02:50
i only want to fully understand it. 02:51
TimToady_ then they built him a new ship
agentzh okay
TimToady_ they made it out of elves' silver and elves' glass
agentzh yeah
TimToady_ the front of the boat was shiney, it didn't run on oars 02:52
or with sails (whether or not the boat actually had a silver mast is unclear)
agentzh how about "shaven oar"?
pfenwick only just arrived... Are we discussing tales of Gil-Galad?
TimToady_ shaving as in carving, so an oar (paddle) carved out of wood
agentzh ah 02:53
i see!
please go on...
TimToady++
TimToady_ the Silmaril (a big jewel) and a banner were put on the boat by Elbereth 02:54
the jewel was like a lantern
pfenwick finds the IRC logs. ;)
TimToady_ and the bannar was like a living flame.
anyway, this Elbereth chick shows up... 02:55
agentzh aye!
TimToady_ something like that...
agentzh how about "who thither came"? 02:56
she came from the thither?
or the thither itself came to her?
TimToady_ thither == "to there"
she showed up 02:57
agentzh oh!
got it!
Thank you very much, TimToady!
TimToady_ np
agentzh i'll tell the audience that larry helped me with the translation... :)
that'll simply rock! 02:58
:D
03:02 mauke_ is now known as mauke
audreyt agentzh: I did translate it 03:02
www.pugscode.org/talks/yapc/slide5.html
lambdabot Title: , tinyurl.com/f5y8z 03:03
audreyt it's part of Pugs :)
agentzh audreyt: thanks!
audreyt: it's wonderful... :D
svnbot6 r13719 | audreyt++ | * adjust smoe tests that fails due to $.foo(1,2,3) parsing
audreyt thanks :) I did spend a lot of time on the translation of that...
bbiab 03:04
03:12 rashakil joined
mugwump huffmanises his keymap 03:19
03:23 b00t joined
svnbot6 r13720 | agentz++ | [util/smartlinks.pl] 03:30
r13720 | agentz++ | - renamed globals $count and $broken_count to
r13720 | agentz++ | $link_count and $broken_link_count.
03:38 mako132_ joined 03:59 pfenwic1 joined 04:02 Eidolos joined 04:24 pfenwic1 is now known as pfenwick 04:25 SubStack joined 04:52 pfenwick left, Aankhen`` joined 05:00 Aankhen`` joined 05:30 mdiep_ joined 05:33 bsb joined 05:44 hikozaemon joined 05:55 yts joined
yts hello~ 06:02
TreyHarris hello, yts 06:06
did you get a commit bit?
yts <msg TreyHarris i have learned a little Haskell and i have interest in Perl6, how could i joini to develop pugs
<msg TreyHarris commit bit? 06:07
TreyHarris yts: have you downloaded the pugs source from svn?
yts: you can talk to me on the channel, it's fine, and if you want to send a private message, you use "/msg", not "<msg" 06:08
yts ohoh
06:13 agentzh joined
TreyHarris hi agentzh 06:13
agentzh hi Trey
here're the slides for this evening's talk: perlcabal.org/agent/slides/naming/naming.xul 06:14
lambdabot tinyurl.com/hbtly
agentzh i used audreyt and TimToady's photos there (with permission). :)
it's a "baby talk", but has some atoms of Perl 6. :)
comments welcome. :D 06:15
TreyHarris ooooh... can you really use "but non" junctions to mix-out behavior? 06:16
or was that poetic license? :)
s/but non/but none/
agentzh Trey: oh, i was not aware of that. 06:17
what's the correct syntax?
TreyHarris i don't know if syntax exists... you said "Silver $.mast but none (Oar::Shaven, Sail);" 06:19
i didn't know if that were possible
agentzh that piece of code is from the pugs tree. 06:20
gaal Help me find a Chinese name! I think "bu" means wave, so maybe something based on that?
agentzh bu?
gaal is enjoying agentzh++'s slides
svnbot6 r13721 | malon++ | smokeserv-server.pl - more broken link avoidance
r13721 | malon++ | * don't make SYN links if we couldn't retrieve the synopses
r13721 | malon++ | * correct r13435, unbreaking links to t/*.t files from TAP HTML matrices
agentzh gaal: glad you like it. :)
gaal agentzh: my name means wave, in Hebrew
agentzh ah!
gaal as in, of the ocean 06:21
agentzh i think this one is appropriate: ōæ½xC0ōæ½xCB
gaal doesn't show up right here 06:22
TreyHarris agentzh: ah, i thought you wrote it, i never looked at the poetry subdir before
agentzh heh, i didn't write it. :)
or maybe i should make it obvious.
it's copyrighted by pugs, i think. :) 06:23
gaal: i'll work on your chinese name for you. :D 06:24
gaal yay, thanks :)
agentzh it must be cooler than ingy's. :)
yts i have use "svu checkout svn.openfoundry.org/pugs/"
lambdabot Title: Revision 13721: /
yts and then i download the source code already 06:25
TreyHarris yts: good. msg me your email address and i'll send you a commit bit
agentzh s/svu/svn/
TreyHarris unless you want spam, don't say it on the channel :)
yts agentzh : typo sorry ^^""
agentzh yts: :)
yts: so what's the question? 06:26
TreyHarris agentzh: you logged in midstream
agentzh Trey: ah, i see.
TreyHarris ?eval 1 06:27
06:27 evalbot_r13710 is now known as evalbot_r13721
evalbot_r13721 1 06:27
TreyHarris ?eval my $x; Int $x;
evalbot_r13721 Error: Unexpected "$" expecting comment, "(", ":", operator, statement modifier, ";" or end of input
TreyHarris I thought so. I don't think that poem is legal Perl 6 anymore. Too bad, as it's lovely
yts this is the first time i use SVN, so i will read SVN doc now :)
TreyHarris yts: what is your email address? i am happy to walk you through your first commit 06:28
gaal agentzh: the word "consistency" should probably make its way into those slides 06:29
TreyHarris yts: the docs distributed with svn are not as helpful for beginners as this i think: svnbook.red-bean.com/
lambdabot Title: Version Control with Subversion
agentzh gaal: yeah, you're absolutely right! 06:30
maybe something like "do the same thing everywhere"
TreyHarris that's not consistency, that's obliviousness :) 06:31
agentzh oh
i'll add notes on consistency to the slides. 06:32
gaal++ 06:33
gaal "The problem with being consistent is that there are lots of ways to be consistent, and they're all inconsistent with each other." -- TimToady++
agentzh great quote! 06:34
TreyHarris gaal: did you get your answer as to t1, etc., in haskell messages? i didn't see it in scrollback and was curious too
gaal nope
TreyHarris oh, this is just awful. a project at $job requires me to pull two SNMP tables rooted at different points in the tree. both tables are indexed to a third. but that table--all three actually--keeps changing (it is maintained in alphabetical order and has items added/removed regularly). so at no point in time can i be sure that the first table's nth item matches the other two's nth items. :-/ 06:42
unless snmp has some transactional or atomic aspects that aren't exposed by Net::SNMP and that i'm not seeing in the RFC.... 06:43
06:43 marmic joined
gaal no locking, eh? 06:43
TreyHarris not that i can see
i think clearly this MIB tree was designed with low-turnover tables in mind 06:44
gaal (even if you had had locking, nobody said you could use it - other users of the device probably don't expect waits)
06:45 iblechbot joined
agentzh gaal: i've updated the slides according to your suggestion. (see slide 16 and 17) thanks! 06:47
gaal whee
agentzh gaal: how about this name: perlcabal.org/agent/slides/naming/gaal.xul ? 06:55
lambdabot tinyurl.com/hh8do
TreyHarris heh, i'm beginning to think that yts is just going to join, ask for a commit bit and leave, twice a day from now on :)
agentzh the first glyph has similar pronunciation as gaal 06:56
gaal: while the last two means "making waves" in chinese.
gaal hee :)
audreyt actually, "rabble rouser".
gaal what's the pinyin for it?
agentzh audreyt: sort of. :)
audreyt ge3 xing1-lang4 06:57
agentzh ge xing lang
audreyt where xing1-lang4 is short for xing1-feng1-zuo4-lang4, "rabble rousing"
06:57 ruz joined
agentzh audreyt: lol 06:57
gaal recursive definition? wow
agentzh i was just about to say "gaal, let's see what audreyt would say on it" 06:59
too sad "xing lang" has that meaning...
chinese is a scary language, like perl. :)
i has thought about "da4 lang4", "jing1 lang4"... 07:00
TreyHarris is jonathan lang on this channel?
speaking of langs :) 07:01
agentzh maybe "jing1 lang4" is better?
gaal yes, maybe I shouldn't take a name I can't pronounce
agentzh fixing...
TreyHarris how can i change some utf8 text into a string of \x... literals? 07:02
gaal (the third tone is tricky for uneducated folks like myself) 07:03
audreyt I think xing lang is fine
rabble rousing is a good thing :)
agentzh audreyt: yes?
audreyt: glad you think in the same way as me. :) 07:04
gaal "svn"
agentzh maybe we should just the chinese family name for gaal?
gaal has difficulty pronouncing ge3 07:05
maybe we should simply get rid of the family name entirely?
gaal maybe with some coaching. heh, february's approaching
agentzh xing1 lang4 is good enough.
xing1 lang4 reminds me of www.sina.com.cn ... 07:06
lambdabot Title: ōæ½xD0ōæ½xC2ōæ½xC0ōæ½xCBōæ½xCAōæ½xD7Ņ³
audreyt I like ge3. gao1 can work (as in clkao), but gao1xing1 is a phrase and unsuitable here 07:07
heh, that's xin1 lang4
TreyHarris audreyt: is jonathan lang correct? only roles may be abstract, not classes or packages (i think he means modules)?
audreyt not xing1 lang4
abstract, as in parameteric?
class Foo[::t] ? 07:08
TreyHarris audreyt: no, as in uninstantiable, meant only for is-ing
ajs_home I would read that as "abstract is in class A {...}" which I think A12 said was legal.
audreyt oh. you can instantiate a role
in which case it creates an anonclass that trivially does it
TreyHarris *may* be... i'm talking about the contrapositive of that
audreyt but if you say .WHAT, you don't get back the role 07:09
I mean, .HOW
that is, a .HOW metaobject is never a role
TreyHarris can you create a class that is uninstantiable until it is subclassed and filled in?
07:09 snowstalker joined
audreyt well, you can always trap things in new/CREATE/BUILD 07:09
but that'd be dynamic runtime checks
I don't think there's a declarative way to do that 07:10
agentzh audreyt: i have some difficulty to differentiate xing and xin due to my chinese dialect. ;)
TreyHarris ok, like Perl 5
audreyt agentzh: I think "gao1 lang4" is good enough 07:11
literally, "high wave"
agentzh yeah, concise too.
TreyHarris i'm very fond of factoring out my common "private" code into abstract superclasses. roles are wrong for that, because roles are about "public" behavior mostly
audreyt that is not so
TreyHarris: "our method !foo" in a role can mixin into classes 07:12
that use it
yet unconsumable by public
agentzh gaal: perlcabal.org/agent/slides/naming/gaal.xul 07:13
lambdabot tinyurl.com/hh8do
TreyHarris audreyt: yes. but would a role's BUILD be visited at a doer's instantiation time?
agentzh gaal: that is gao1 lang4
gaal heh, except i'm not the tallest of hackers around here :) 07:14
audreyt TreyHarris: sure, as long as the doer doen't define BUILD
07:14 zakharyas joined
gaal was i totally wrong about bu- being a particle for wave words? weird. 07:15
agentzh gaal: lol
audreyt or, if both cooperate and define BUILD as multi, that will cause both be visited
agentzh gaal: bu4 means nope
audreyt gaal: gao1 is a common family name :)
agentzh gaal: very common, actually. much more common than mine (zhang1).
gaal and i get cl as a relative? neat.
audreyt and in .cn everyday speech (but not .tw), "gao1" also means "high" in english's other meaning
as in excited 07:16
agentzh lang4 means waves.
*nod*
gao! shi zai shi gao!
gaal: if you like it, i'll fix my slides using your shiney new chinese name. :) 07:17
TreyHarris audreyt: hmm. i have many classes whose new() says something like "my $proto = shift; my $class = ref($proto) || $proto; croak "Attempt to instantiate abstract class" if $class eq __PACKAGE__; # proceed with common object instantiation" 07:18
with the idea that those that use it as a base will call NEXT in their own new (if they want to override something)
gaal sure! does it work in Chinese to do { temp $name = "gao1 xing1 lang4" } when I'm feeling subversive? 07:20
TreyHarris so the perl 6 words work well for the concept: i create such uninstantiable classes when the instantiable class *is* a type of the uninstantiable class, rather than simply being a class that *does* what the uninstantiable class does
agentzh gaal: well, it worked 20 years ago in china. :) 07:23
or maybe 30 years ago. 07:24
gaal good then. thanks! :)
agentzh hmm...i've found gao1 lang4 very cool! 07:25
audreyt++
i'll tell the origin of gaal's chinese name to the audience tonight. :) 07:26
ajs_home TreyHarris: I replied to your mail. I think you're confusing yadda methods with interfaces. A role is only an interface when all of its methods are undefined per S12....
gaal :)
TreyHarris audreyt: I thought that the ordinary way to declare BUILD was as a submethod? Then the default new calls all the BUILDs for you, right? 07:27
agentzh gaal: fixed. see naming.xul slide 15 and get a feeling of it. :) 07:28
gaal yay :) agentzh++ audreyt++ 07:29
"slothlike" doesn't translate succinctly, does ir
*it
agentzh gaal: hmm 07:30
gaal I actually have no idea - it just looks long 07:31
TreyHarris ajs_home: you're referring to S12:851? I read that as an operational observation, not as a definition of an "interface". It seems absurd to me that if you have an interface role with "method is { ... }", that you aren't allowed to declare "method isn't { ! .is }" without no longer being an interface
agentzh gaal: i can't find a better way to translate that word. :(
gaal: however, it reads well to chinese folks. :)
gaal: although a bit long. :D
gaal "sloth" is an animal. maybe use the chinese word for it? though that might not carry the English/Christian connotations. 07:32
okay, whatever works :)
07:32 rashakil joined
gaal I obviously can't criticize chinese. Was just making an observation. 07:32
agentzh gaal: thank you :) 07:33
if i mention the sloth animal, the chinese translation will be much longer. :)
gaal lol
agentzh gotta run to class, bbiab & 07:35
07:35 agentzh left 08:02 bsb left 08:06 Psyche^ joined 08:14 Psyche^ is now known as Patterner
TreyHarris gaal, audreyt: I'm still working through YAHT, and I created a data type that =, as one of its guards, 'Nothing'. But when I try to compile, I get, "Ambiguous occurrence `Nothing'... It could refer to either `Main.Nothing', defined at MyTest.hs:6:21, or `Data.Maybe.Nothing', imported from Data.Maybe at MyTest.hs:4:0-16" 08:33
if i change my Nothings to "Data.Maybe.Nothing", it tells me "Not in scope: data constructor `Data.Maybe.Nothing'
08:35 crem_ joined
TreyHarris oh, well. 08:35
audreyt are you importing Data.Maybe?
if you are, then you have to import it qualified 08:36
import qualified Data.Maybe;
TreyHarris yes... program starts "module Main; where; import Data.Maybe"
audreyt so import it qualified will fix it 08:37
TreyHarris ahh. i guess YAHT left that out
all their examples just have bare "Nothing" 08:38
08:38 yts joined
audreyt yts: heya. did you get your commit bit? 08:40
ifso, please add yourself to the AUTHORS file and commit it to test the commit
TreyHarris audreyt: no, yts hasn't given me his/her email yet
yts: please do :) 08:41
audreyt: in my ghc source tree, where can I find Data.Maybe's source? i must be dumb 08:45
audreyt libraries/base/ 08:46
Data/Maybe.hs
TreyHarris blushes 08:48
oops! i was in your binary release tree, that's why i couldn't find it
audreyt that explains it :)
yts how could i check the commit bit, just use "svn commit"? 08:49
TreyHarris yts: have you sent me your email address? 08:50
audreyt yts: what's your email address? 08:51
TreyHarris yts: got it, thank you 08:52
yts TreyHarris:i can't send msg to you 08:53
TreyHarris yts: you can now
08:54 evil3y3 joined
TreyHarris yts: I am reading your messages 08:54
are you not seeing mine?
yts :TreyHarris because what i type to you didn't show on screen 08:56
TreyHarris yts: it's very strange. i have been getting your messages fine
but apparently you are not getting mine
but you are seeing what i type on the channel 08:57
in any case, check your email
yts F00
audreyt TreyHarris: did you identify to freenode?
TreyHarris audreyt: yes, and i turned filtered off 08:58
08:59 b00t joined 09:07 buetow joined, rodi joined
gaal TreyHarris: but what do you means as one of its guards? 09:16
09:16 kanru joined
gaal if you have data MyDatatype = OneVariant | Nothing, then you have a clash 09:16
TreyHarris gaal: yes, I see that now
gaal because Maybe's nothing constructor is in scope. 09:17
TreyHarris I didn't realize that $type eq $constructor is merely convention in Haskell
but now I'm clearer :)
gaal ah, I just noticed audreyt's 'import qualified' hint
09:17 penk joined
gaal it's not always a very good convention 09:17
09:22 rashakil joined 09:23 elmex joined
yts i have ghc-6.4.2 command, but when i want to perl Makefile.pl 09:27
it says "Cannot find GHC 6.4.1 or above from path (we have 6.4)."
i type " alias ghc='ghc-6.4.2'; perl Makefile.PL" but didn't work .. 09:28
09:28 ruoso joined
TreyHarris yts: aliases are a shell convention, Makefile.PL can't see them 09:29
if you type "unalias ghc; ghc --help" what do you get? 09:30
09:32 chris2 joined
yts -bash: unalias: ghc: not found 09:37
Usage:
ghc-6.4
[command-line-options-and-input-files]
To compile and link a complete Haskell program, run the compiler like
so:
ghc-6.4 --make Main
TreyHarris yts: that's very strange 09:38
unalias is a bash builtin 09:39
i can't figure why it would not let you do it
yts -bash-3.00$ unalias
unalias: usage: unalias [-a] name [name ...]
-bash-3.00$ unalias ghc
-bash: unalias: ghc: not found
-bash-3.00$
but i can use "ghc"
TreyHarris oh, ok. 09:41
type "ghc --version"?
yts The Glorious Glasgow Haskell Compilation System, version 6.4
my system has two ghc version 09:42
TreyHarris that's not 6.4.2
yts one is 6.4 and the other is 6.4.2
TreyHarris yts: ok... are they both installed in the same directory?
just export GHC and set it to the 6.4.2
and then run perl Makefile.PL again 09:43
fyi, you will need a 6.5 version of GHC in order to build Pugs fairly soon (days? weeks?)
yts but in GHC web, it only release 6.4.2 @@ 09:46
TreyHarris yts: www.haskell.org/ghc/dist/current/dist/ 09:48
lambdabot Title: Index of /ghc/dist/current/dist
TreyHarris 6.5 is pre-release
yts but now i work in workstation...i think i will install 6.5 later when i install linux in my labtop 09:51
Juerd XY problem? 09:53
perlbot: XY problem?
perlbot They ask how to do X, but that's because they really want to do Y but think X will get them there and most of the time they are WRONG
TreyHarris Juerd: to what are you referring? :)
Juerd TreyHarris: Nothing. Needed to paste that line :) 09:54
Elsewhere
TreyHarris ah. you can priv message perlbot, fyi
Juerd Useful. Thanks. 09:55
yts i have a make pugs error, is anyone who can help me ? 10:27
rodi depends on the error :) 10:32
yts configure: Compiler flavor: GHC 10:34
configure: Compiler version: 6.4.2
configure: Using package tool: util/ghc-pkg-wrapper
configure: Using ar found on system at: /usr/bin/ar
configure: No haddock found
configure: No pfesetup found
configure: Using ranlib found on system at: /usr/bin/ranlib
configure: Using runghc found on system at: /chalmers/sw/sup/phc/b/binl/runghc
configure: Using runhugs found on system at: /chalmers/sw/sup/phc/b/binh/runhugs
configure: Using happy: /chalmers/sw/sup/phc/b/binh/happy
configure: Using alex: /chalmers/sw/sup/phc/b/binh/alex
configure: Using hsc2hs: /chalmers/sw/sup/phc/b/binl/hsc2hs-6.4.2
configure: No c2hs found
configure: No cpphs found
configure: No greencard found
Preprocessing library Pugs-6.2.12...
Setup: Error: file does not exist: "/chalmers/sw/sup/phc/b/binl/hsc2hs-6.4.2"
Build failed: 256 at util/build_pugs.pl line 344.
make: *** [pugs] Error 2
i have hsc2hs command, but i don't know why it find hsc2hs-6.4.2... 10:35
rodi you can set the HSC2HS environment variable, and it should fix this error (assuming everything is there, compatible, etc.) 10:40
also, if you need to paste a lot of text in the channel, you can use the paste bot from the channel topic (sial.org/pbot/perl6)
that way not everyone has to look at it :) 10:41
currently, the way that Makefile.PL finds hsc2hs is to simply substitute "hsc2hs" in for "hsc", which appears not to work on your system. 10:42
What system is it, incidentally?
10:43 nothingmuch joined
yts i think is Red Hat 10:44
lambdabot yts: You have 1 new message. '/msg lambdabot @messages' to read it.
nothingmuch moos
lambdabot: does anybody love me?
*sob*... no messages 10:45
rodi @tell nothingmuch mooooooooose! 10:46
lambdabot Consider it noted.
nothingmuch lambdabot: does anybody love me now? 10:48
rodi: thanks for trying ;-)
lambdabot nothingmuch: You have 1 new message. '/msg lambdabot @messages' to read it.
theorbtwo lambdabot: Spread the love, baby.
nothingmuch lambdabot: yay! i know somebody cared
@messages
lambdabot rodi said 2m 27s ago: mooooooooose!
nothingmuch @tell rodi yeah! moose are awesome 10:49
lambdabot Consider it noted.
rodi how about me? 10:50
lambdabot rodi: You have 1 new message. '/msg lambdabot @messages' to read it.
rodi w00t! m00s!
@messages
lambdabot nothingmuch said 1m 37s ago: yeah! moose are awesome
rodi oh man, that's great!
nothingmuch @moosages
lambdabot You don't have any new messages.
rodi LOL
nothingmuch =D
@mĆøĆøsages 10:51
lambdabot Unknown command, try @list
nothingmuch no unicode, eh ;-)
rodi must be a perl 5 bot ;)
nothingmuch perl 5 has great unicode support, don't be mean ;-)
rodi hugs lambdabot
errr....
rodi hugs perl 5 10:52
yts rodi: i have set $HSC2HS, but it didn't work @@ 10:57
rodi: i have set $HSC2HS, but it didn't work @@ 10:58
hih
kjslkd
jksld
jslkdf
jksldf
10:59 yts joined
yts hi 10:59
hihi
hihi 11:00
jslkfj
ksdjf
11:00
yts sjkdl 11:00
jsldf
jksldjf
skldfj
rodi huh.
I guess yts was sad. 11:01
11:02 yts joined
yts hi 11:03
11:04 foo\ joined, TimToady joined 11:06 rashakil joined 11:09 yts left
rodi Check your feather email | spec.pugscode.org | paste: sial.org/pbot/perl6 | pugs.blogs.com | www.treehugger.com/files/th_images/paradigm.jpg 11:12
11:48 yts joined
yts Setup: Error: file does not exist: "/chalmers/sw/sup/phc/b/binl/hsc2hs-6.4.2" 11:49
but i have hsc2hs in different directory.. how should i do
audreyt yts: hi 11:51
export HSC2HS=/path/to/hsc2hs 11:52
yts i have set it but it didn't work 11:53
audreyt can you nopaste your build log?
using sial.org/pbot/perl6
lambdabot Title: sial.org Pastebot - pasteling
pasteling "yts" at 193.11.195.34 pasted "*** Finished building dependen" (41 lines, 2.3K) at sial.org/pbot/20051 11:55
audreyt so you did set export? 11:58
what did you set it to?
yts ya
-bash-3.00$ export HSC2HS=/chalmers/sw/sup/phc/b/binh/hsc2hs
audreyt investigating 11:59
did you rerun configure after the export? 12:02
"make" will not pick the setting up
you need to configure first
yts ohoh! 12:03
audreyt since it's probed and then wrote to makefile
yts i see~ 12:04
12:04 jabbot joined
yts thans a lot :p 12:04
audreyt no prob :)
12:32 rashakil_ joined 12:40 rodi joined 12:42 Yappo is now known as \1 12:46 \1 is now known as root 12:47 root is now known as rot 12:49 rot is now known as Yappo 13:00 agentzh joined
agentzh whee, the talk was a big success... 13:01
audreyt yay!
agentzh the audience is keeping saying "wow" when i was introducing audrey to them.
audreyt *smile*
agentzh and the students laughed a lot when i was talking about gaal's chinese name. ;-) 13:02
now every student in my C.S. department knows larry, audrey, and gaal. yay!
audreyt: a lot of my friends have say you're very charmful in that picture. 13:03
*have said
;)
my instructor said himself was about to give up his C.S. career but my talk had changed his mind and gave him inspiration and determination. 13:05
i think that is the power of #perl6! 13:06
audreyt I'm very glad to hear that :) 13:09
agentzh++
agentzh audreyt++ #perl6++ larry++ (everyone here)++
audreyt leo_: ping 13:10
chip: ping
agentzh sleep & 13:14
13:15 agentzh left 13:16 yts joined 13:19 discordja joined
audreyt gets weird bus errors on embedding pugs again :/ 13:22
I mean, embedding parrot
otoh, the find_global has gone away and we need to either fix that
or remove parrot embedding altogether 13:23
@tell leo_ is this a sensible find_global replacement? void* Pugs_get_base_global(Parrot_Interp i, STRING *ns, STRING *sym) { return Parrot_get_global( i, Parrot_get_namespace_keyed_str(i, i->root_namespace, ns), sym); } 13:24
lambdabot Consider it noted.
13:27 dakkar joined 13:30 kanru joined 13:34 cdpruden joined
gaal agentzh: ah, I meant bo (ę³¢) earler 13:38
13:40 rashakil joined 13:42 cjeris joined 13:44 xinming_ joined 14:10 hexmode joined 14:13 weinigZzZ is now known as weinig 14:14 ofer0 joined 14:15 count_leto joined
count_leto hi :) 14:15
14:15 jferrero joined 14:19 vel joined 14:27 awwaiid joined 14:31 buetow joined 14:44 fglock joined 14:45 fglock left 14:51 etzel joined 15:09 weinig joined 15:16 salty-horse joined
salty-horse Heinlein? I'm just reading that book! www.perl.com/2006/09/21/graphics/x63.jpg 15:17
lambdabot tinyurl.com/gmnrn
gaal yeah, that's a Heinlein quote. 15:18
salty-horse I read that part a only week ago :)
count_leto hey... could anybody tell me why pugs doesn't want to interpret my smartmatch operator? 15:25
[particle] post the code 15:26
count_leto ahm... where?
[particle] see title
sial.org/pbot/paste
pasteling "count_leto" at 84.173.145.196 pasted "test for getting into perl 6" (9 lines, 132B) at sial.org/pbot/20057 15:27
[particle] $result = $text ~~ m/Test/; 15:28
count_leto I've found that in a tut
[particle] where? 15:29
count_leto faq.perl-community.de/bin/view/Wiss...alKapitel4
lambdabot Title: PerlCommunityWiki ōæ½xBB Wissensbasis ōæ½xBB Perl6TutorialKapitel4, tinyurl.com/jvqu5
[particle] you should probably correct the example in the wiki, then 15:31
count_leto well, [particle], your version causes my pugs to complain about an 'unexpected "m"'...
PerlJam count_leto: perhaps you need a newer pugs?
count_leto I installed yesterday (maybe 2 days ago... cant remember) ...
but from a debian-package. 15:32
pasteling "[particle]" at 144.81.84.167 pasted "my pugs session log" (38 lines, 1.5K) at sial.org/pbot/20058
[particle] take a look at that log, count_leto
...and what version does 'pugs -v' give you? 15:33
PerlJam installs ghc again so he can build pugs
count_leto 6.2.10
[particle] latest is 6.2.12, .10 is pretty old 15:34
count_leto well, the error at the beginning is quite the same that occurs on my machine.
audreyt "$text ~~ $result m/Test/;" is not valid syntax anyway
count_leto hm... 15:35
I'll install version .12 and complain again later, mkay? ;-)
audreyt that part is version independent :) 15:36
count_leto that doesn't make my problem smaller... I think
after correcting a typo and using [particle] 's code snippets, I get '*** Cannot parse PGE: Test' 15:37
pugs leaves me questioning my mental sanity 15:38
audreyt ok... one thing at a time
first. can you give me your email address so I can send you a commit bit? :)
second, install "subversion" and "ghc6" via apt, and run "svn co svn.openfoundry.org/pugs ; cd pugs ; perl Makefile.PL ; make"
lambdabot Title: Revision 13721: / 15:39
gaal libperl-dev also
audreyt once you got the invitation mail, while waiting for pugs to build, edit the AUTHORS file and "svn ci AUTHORS" to make sure the commit works
15:40 weinig is now known as weinig|coffee
count_leto that all sounds great, audreyt and I thank you for helping me... but what is it I'm installing at the moment? :) 15:41
gaal count_leto: it's the latest code from source control 15:42
since we haven't had a release in a while
(though one is coming up soon)
but this is the best way to keep yourself up to date.
ghc is the compiler used to build pugs itself. 15:43
subversion is a revision control system (like, but better than, cvs)
count_leto okay...
audreyt and the reason that the latest revision is relevant is that "Cannot parse PGE" is gone from the set of possible errors 15:44
and so your sanity may be preserved a bit better :)
count_leto not that I'm questioning it anyway, hm? ;)
in which directory should i run the commands ("svn co ..." pp.) 15:45
gaal it will create a subdir called 'pugs' wherever you run it.
so if you have an ~/src/ directory you like to keep projects in, that's a good place. 15:46
but the general answer "it doesn't matter"
count_leto thanks.
*g*
gaal when you use subversion, btw, you can mv the pugs directory to other places 15:47
audreyt sanity is a controlled substance around here
gaal periodically, run 'svn up' at the top level to update your sources. this will be much faster than your current full checkout.
audreyt we need to caliberate your sanity level to touch insanity but not beyond :)
count_leto lol 15:48
thanks again, gaal
gaal sure! 15:49
audreyt gives up on repairing parrot embedding
the packfile segment init sequence doesn't work anymore :/
maybe when leo/chip is around tomorrow
I need to sleep now *wave* 15:50
gaal I have to go now too *lang4*
count_leto cu
count_leto wonders what lang4 could mean 15:52
after long contemplation, I found the answer... 15:56
audreyt "wave" in chinese
gaal it's a silly lang4 joke
audreyt count_leto: I fixed the wiki, btw. 16:06
count_leto that's great, audreyt :) 16:07
16:08 weinig|coffee is now known as weinig 16:24 fglock joined 16:29 rashakil joined 16:30 justatheory joined 16:31 lanny joined, swaroop joined 16:32 snowstalker joined 16:33 snowstalker joined 16:34 Psyche^ joined
TreyHarris count_leto: have you gotten pugs built yet? 16:37
16:38 Odin-LAP joined, kanru joined
count_leto lol no, TreyHarris ... still compiling ;-) 16:39
count_leto hasn't got a high-end PC 16:40
TreyHarris when did you start the compile? what speed is your machine?
16:40 tewk_ is now known as tewk
count_leto (unfortunately) 16:40
16:40 snowstalker is now known as snowstalker_ 16:41 snowstalker_ is now known as snowstalker
TreyHarris sure, many of do not :) 16:41
count_leto iirc it's a 266 MHz
TreyHarris but i just wanted to give you an idea how long to expect
ah... it may be 3 hours or more 16:42
count_leto root 20785 0.0 0.0 3276 136 pts/3 S+ 18:14 0:00 util/runcompiler
its running for almost half an hour now
so I might just as well take a shower, go out and drink some beer with some friends, come back and 10 mins after returning, its done? 16:43
TreyHarris yes probably :)
count_leto lol
thats awesome
i was so eager to start hacking some Perl v6... 16:44
.o(YOU STILL ARE)
count_leto is trying to convince himself
I love the idea of using uni-encoded source
16:45 BooK_ joined 16:46 Psyche^ is now known as Patterner
TreyHarris i was in the same boat until a couple weeks ago, very slow machine 16:46
but last week i got a brand new machine, compile takes 7 minutes now :)
count_leto well i do have a 1.x GHz Laptop... but the fan decided not to like me, and... died 16:47
so im having a laptop that boots up a bit, then says 'critical temperature reached: 105 C. Shutting down...'
argh
TreyHarris ouch!
count_leto but it's just the cable, so im going to try to sold it 16:48
lanny Water cool it. /nod
lambdabot lanny: You have 1 new message. '/msg lambdabot @messages' to read it.
lanny lambdabot: @messages
lambdabot agentzh said 1d 10h 52m 41s ago: i've sent a mail to p6c. hopefully iblech and malon will fix the smoke server soon. thanks for the catch. :)
count_leto lanny: that would be a great idea
lanny Yeah. It's tough on the warranty though. 16:49
count_leto I tried to ice cool it (with ice packed in a plastic bag), but i think i didnt hit the 'hot-spot'
lanny I coffee cooled mine the other day. :(
TimToady just tape a vacuum cleaner to it.
TreyHarris hi TimToady!
TimToady what's cookin' (besides the laptop)?
count_leto TimToady: that's *not* funny :/ 16:50
TimToady believe me--I empathize. 16:51
TreyHarris count_leto: check your email, you should have instructions for creating your svn login
TimToady but I often make jokes about things that aren't funny...
lanny It's hard to make jokes about things that are funny. They don't need it.
TreyHarris true. monkeys. *giggle* you don't need to make jokes about monkeys. you just point out monkeys... 16:52
TimToady most comedians tread the fine line between laugther and tears
Obviously I need to figure out how to cry when I see monkeys... 16:53
maybe if they're flying butt monkeys... 16:54
TreyHarris TimToady: maybe if one bit you...
TimToady: but isn't that how Linux got its penguin?
lanny Flying butt penguins?
TimToady What You Don't Know About Tux 16:55
count_leto TreyHarris: I'll check later, maybe... got a load avg of about 4 atm
16:55 kanru joined 17:00 frederico joined 17:01 ofer0 joined
lanny Hmm. If Elbereth affixed the Silmiril as a lantern and banner to the mast wouldln't the ship have a mast? 17:02
17:03 rindolf joined
rindolf Hi all. 17:03
ingy: here?
lanny Ah. Maybe it's the prow she set it to and thus the shining.
rindolf Hmmm... ingy's been idle for 59 hours. 17:04
ingy rindolf: nope
back on Tuesday
rindolf ingy: hi.
ingy: I commited the fix to IO::All.
ingy I really am only on for a minute
thanks
I'll release it Tuesday 17:05
rindolf ingy: OK.
ingy later! &
rindolf ingy: bye
ingy: thanks.
17:06 ofer0 joined
pasteling "lanny" at 128.107.248.220 pasted "op0RoundRobin" (13 lines, 556B) at sial.org/pbot/20060 17:06
lanny Can anyone suggest how I can take the Val from op0Each and get the [Vscalar] I need? 17:07
17:07 vel joined
TimToady lanny: I think it wouldn't describe it as "silver mast" if it were trying to claim the ship didn't have one... 17:14
unless JRRT was just looking for a filler word...
lanny It's a stretch but all elven ships might have had silver masts. But yeah. 17:15
count_leto allright, I'm getting out of here. thanks for your assistance. :-)
17:15 count_leto left
TreyHarris lanny: silver is a color or character as well as a substance. it could be painted, or glinting in the light.... 17:15
lanny The placement of the ';'s almost has me thinking the oar, sail, mast clause is just an aside though. 17:16
cjeris silver could be used figuratively to describe the color of birch or other pale wood?
TimToady I'm not sure elves believe in paint...
lanny Or Malorn (sp?) 17:17
Been too long. Need to re-read the books although not sure I can get up enough energy for the Silmirillion.
cjeris many woods are a silvery gray color after enough weathering, anyway. 17:18
lanny ooh. Scary. Might have stumbled onto a good coding of roundrobin()
Sure wish I *knew* what I was doing instead of just casting around until stuff works. :/
nothingmuch gaal: benewed on your english name =) 17:20
err 17:21
freudian slip
i meant engrish ;-)
TreyHarris TimToady: one of my coworkers waved me over the other day and showed me an essay from Paul Graham's last book, where he shows why ruby and lisp are better than perl because creating a closure-based accumulator takes three fewer syntactic elements in those two languages than in Perl 5. the coworker wanted to know how it would be in Perl 6. you'll be happy to know that placeholders and blocks always being closures has, according to graham at least, now ...
... made Perl 6 as powerful as Ruby and Lisp. :)
nothingmuch fwiw, it's cool that lang4 is a syn for gaal in heb ;-)
lanny Hmm. I always thought the partition was if a language *could* create a closure-based accumulator. Would seem trivial to say Perl 6 now has builtin $*CLOSURE_BASED_ACCUMULATOR.(3). We win! 17:23
TreyHarris lanny: i agree, i think it's silly. but graham's assertion seems to be that the less syntactic elements required to do any given thing, the more powerful the language. this follows i suppose from physics' definition of "power"... you wield more force with each keystroke... 17:25
TimToady that's like "there's only one measure of IQ"
someone needs to write "The Mismeasure of Programming Languages"
rindolf Hi nothingmuch 17:26
nothingmuch hi rindolf
what's up?
rindolf nothingmuch: G'mar Hatima Tova.
nothingmuch: I'm fine. Working on Test::Run.
nothingmuch: worked on Yapcom previously.
nothingmuch gabor's hackathon?
rindolf nothingmuch: what's up with you?
nothingmuch recovering HD
(well, now i'm done copying, but i'm still installing my env) 17:27
17:27 elmex joined
nothingmuch and i finished packing for my hike 17:27
rindolf nothingmuch: I see. 17:28
nothingmuch: I had some hard-disk problems too.
nothingmuch of the software or hardware variety? 17:29
17:30 mdiep joined
lanny TreyHarris: and in the face of macros it really is an either/or and not of degree since you can reduce the syntactic elements as far as desired. 17:32
17:37 DaGo joined
rindolf nothingmuch: hardware. 17:38
nothingmuch: or at least I think it was hardware.
nothingmuch mine too =( 17:40
rindolf nothingmuch: I blogged about it.
nothingmuch i didn't ;-)
rindolf nothingmuch: OK. :-)
18:03 rashakil_ joined 18:18 beppu_ joined
lanny svnbot seems dead. 18:19
18:29 yts joined
yts it's been a long time to build pugs... 18:30
rindolf yts: you mean that it took a long time to build it? 18:33
avar rindolf: that's still ambiguous:) 18:47
rindolf avar: why?
gaal yts: "make fast" instead of "make"
rindolf Hi gaal 18:48
yts i have stay in " # Actual: undef
t/examples/all_parse....................................ok 32/160
" for a long time
rindolf gaal: Gmar Hatima Tova.
gaal nothingmuch: pinyin /.*ng$/ is not English /.*ng&/
nothingmuch .ng?
gaal s/&/$/
"lang" doesn't sound like "language"
nothingmuch i just meant that גל is wave
גלים בים
avar rindolf: It can refer to make(1) time or how long it took to program it:) 18:49
gaal yts: you're in the tests now, not in pugs build proper
so if you want to start playing around, you can interrupt make and start right away.
rindolf: thanks.
yts i see
rindolf gaal: what's up besides?
yts but why it test so long @@ 18:50
nothingmuch gaal: kapish
gaal yts: pugs isn't particularly tuned for speed yet. we have done a few rounds of optimizations, but not nearly as many as perl5 had undergone throughout its existence. 18:51
TimToady yts: it's a big test. It is trying to compile all the examples just to see if they compile.
gaal rindolf: uh, not much. 18:52
rindolf Hi TimToady
gaal: OK.
TimToady hi
yts thanks
rindolf TimToady: what's up? 18:53
TimToady: I enjoyed your "State of the Onion 10"
I tried to slashdot it, but it wasn't published. Eventually chromatic got it slashdotted, though. 18:54
TimToady yts: you can do a ps for 'pugs' and see which one it thinks it's working on.
yts i have not make install yte 18:55
typo. yet
TimToady don't need make install
gaal yts: you don't have to.
TimToady rindolf: thanks
yts @@""
gaal I think I ran make install only once or twice ever.
(for pugs that is...)
TimToady in fact, 'make install' is rather unrecommended right now. 18:56
rindolf TimToady: no problem.
yts why not make install @@, it a little weirt
gaal yts: can you explain how to read those smileys, which I suppose are frowneys?
18:56 lanny joined
lanny ?list(1) 18:57
lambdabot Unknown command, try @list
yts ok, i won't type this again
sorry
lanny ?eval list(1)
18:57 evalbot_r13721 is now known as evalbot_r13730
evalbot_r13730 1 18:57
TimToady ?eval List(1) 18:58
evalbot_r13730 ::List
lanny Shouldn't that be (1,)
gaal no, I was curious! I never saw those before.
lanny: did you find the answer to your haskell question?
TimToady list() just puts the 1 into list context.
which the 1 ignores.
lanny Yep. Well more stumbled across it. Implemented roundrobin() and cat()
aha
Tricksy.
ajs_work TimToady: A while back you said that we were abandoning term constants where possible. Did you mean to include the math constants such as pi and e? Should they be $pi and $e respectively? 18:59
yts @@ ----> which means i feel confused
gaal what are roundrobin and cat?
lanny S29/Container
roundrobin is like each but drops the undefs. cat just concatenates its sliced arguments
TimToady originally from S09 and feed operators
so we can have useless-use-of-cat awards in Perl 6 19:00
lanny ?eval cat(1)
evalbot_r13730 (1,)
gaal roundrobin's a bit like zip*
19:00 glasser left
gaal but flattened. 19:00
lanny Well zip returns tuples (really arrays?)
?eval roundrobin(1) 19:01
evalbot_r13730 (1,)
gaal rr = cat . zip ?
lanny ?eval each(1;1..2;1..3)
evalbot_r13730 (1, 1, 1, undef, 2, 2, undef, undef, 3)
lanny ?eval roundrobin(1;1..2;1..3)
evalbot_r13730 (1, 1, 1, 2, 2, 3)
gaal yts: cute! 19:02
TimToady ajs_work: obviously, those should be Ļ€ and ā„Æ ā˜ŗ 19:03
ajs_work but should they be $? and $?? Actually, I'm not sure about ?, but certainly ? was already in API/Math/Basic.pod 19:04
TimToady in general, I'm not worried if a few things like "pi" and "e" sneak through, since people won't be confused about whether they take arguments.
ajs_work What about Math::Basic vs. Num vs. Int, etc. Should I migrate the API doc to Num and (where integers are involved) Int? 19:05
[particle] used to use an exception function named e() frequently
ajs_work [particle]: If it took parameters, you're fine ;) 19:06
I'm still struggling to figure out some of the Array/Seq/List delineation, and I want to know what your thinking is on Int/Num/Math before I check in any of that 19:07
TimToady I don't believe in Math 19:08
it's not a type
ajs_work I think Ron intended it to be a library that provided the general math functions, but if we're going with types as the guiding rule for finding functions in S29, then I'll keep that in mind for the others too, and re-name / break up Math::Basic API 19:09
Ahem. Rod that is
I don't know if we can do that as easily for the stuff that's currently in OS, though. I'll think about that 19:11
araujo TimToady, hah
19:11 mdiep joined
TimToady the OS can be viewed as a (singleton?) object. 19:12
ajs_work heh, well I suppose so ;)
TimToady I'm not sure what a Math object is...
ajs_work I think that's a chalkboard.
yts i have already make ...how should i run pugs ? 19:15
TimToady ./pugs
yts ohoh 19:16
TimToady was that o-hoh or oh-oh?
yts oh-oh~
i see :p 19:17
TimToady oh-oh is bad
yts maybe i can type oh-oh!
ajs_work heh
yts sorry, i just use as chinese
lanny Hmm. Tempted to make eqv work for tests by sub infix:<eqp> ($a, $b) { $a.perl eq $b.perl; }. It seems faster than eqv as well.
Not sure it's a robust solution for the long term though. 19:18
TimToady well, eqv is defined in terms of serialization, but the hope is that both sides can be serialized lazily so it falsifies without having to completely serialize both sides first. 19:19
it's also not entirely clear how much canonicalization can go on
but that's why eqv has optional arguments to specify what you mean. 19:20
lanny A telling points. Guess I'll live with :depends<eqv> for now.
TimToady and why the semantics of eqv are closely bound with cmp and sorting.
and why eqv should optionally take a signature saying which parts you want to compare. 19:22
hmm, a sig for sorting could use :1($foo), :0($bar) to sort on bar before foo... 19:23
unless we want to reserve that for radix conversion in binding somehow. 19:24
lanny The signatures as Comparitors for sort() didn't make a lot of sense to me.
19:24 mdiep joined
lanny Sorry. signature as Criterion 19:25
TimToady signatures are good at picking out and naming bits of things. they're less good at specifying an order different from the positionals...
so maybe if you want that you go ahead and write the corresponding body. 19:26
sigs as Criteria is just for the cases where it can write the body for you.
19:26 theorbtwo joined
lanny So are we talking: sort :(Str $name, Num $val), @array_of_KeyVal_objects; ? 19:27
TimToady there's some potential confusion there though with :($a,$b) looking like an old-fashioned comparator
lanny I think I'm missing how signatures get mapped to values. 19:28
TimToady so maybe the notation needs to clarify that it's binding the sig to a single $a arg.
that's what binding is.
lanny Is that binding still to be spelled out in the Synopses? I'm looking at hits of 'signature' and 'signature binding' across them but don't seem to be finding anything appropriate. 19:32
TimToady see S06:1171. this is a slight generalization of that idea. 19:34
lanny Ok. 19:38
TimToady basically, when you bind a signature the binder can pretend most objects are Captures and do something sensible to name the bits.
the generalization here is that sort will take the named bits and assume you just want them sorted in the order named, each according to its own type. 19:39
so numeric vs string automatically falls out.
for instance.
"is descending" does not fall out of the type info though.
lanny Right. I figured that was the idea. I'm just thinking how to add it into the sort() I've got.
TimToady probably need to way till Captures/Signatures are fully hacked in. 19:40
*wait
lanny There are some other little questions on sort() that came out of trying to implement it.
TimToady I've got a few minutes while I eat lunch over my laptop. 19:41
if they're heavy questions, though, p6l is fine.
lanny For instance a Pair Criterion is composed of a KeyExtractor and Comparitor. But those can take descending, insensitive. So could you have ( { +$^a} is descending => { ~$^a cmp ~$^b} is insensitive)? 19:42
I should put them to p6l. I'm sure you could resolve em but that would also give a wider view and record them on something a bit less transient.
TimToady I'm not sure the "return a number or string" approach is quite right, since we'd like to have that info at compile time. 19:43
and while we could inspect the body, it's not guaranteed to be there yet. 19:44
lanny Yeah. But doing it all in Haskell right now is way beyond my level and doing it in Perl 6 wasn't.
The core algorithms are working when I test stuff against Code for KeyExtractors and Comparitors
TimToady the advantage of the sigs is that you can see the types at compile time
lanny nods 19:45
TimToady but there's no reason it can't fallback to runtime, I guess.
lanny I figured something like sort would end up deep in the language instead of up at the Prelude
TimToady hard to say--depends on how powerful P6 actually ends up.
we are making Captures and Signatures visible, after all.
well, need to commute... 19:46
lanny Laters.
theorbtwo (larry + work) == (work + larry) 19:47
TimToady afk &
yts i have build pugs and i can run pugs now. it's now a little like hugs... 20:01
20:01 Limbic_Region joined
lanny Not surprising since it's built on top of Haskell. 20:02
yts so is there any thing i can do for pugs? 20:04
lanny Lots. :)
yts i learn a little Haskell,
lanny Look through the source directories and read any READMEs or TASKS.
I constantly go back to this search.cpan.org/~autrijus/Perl6-Pug...c/Hack.pod 20:05
lambdabot Title: Pugs::Doc::Hack - How to hack on Pugs - search.cpan.org, tinyurl.com/km8q3
lanny The easiest thing to do is probably write tests and/or put smartlinks in the tests.
[particle] the algorithm is simple: 1 while do_something_useful( find_something_interesting() ); 20:06
lanny Also it's not too bad to add simple operators that aren't in the language yet. The monads can be a bit daunting at first but if you understand the IO monad it's easy to see what's going on. 20:07
At least that's how it's been for me.
yts Hm...i think i should play perl6 first :p 20:08
lanny Just do that until you bump into something you want that isn't implemented. It's very motivational. :) 20:09
yts :) 20:10
my $gen = @*ARGS[0]; #what's this mean? take the first argument? 20:13
20:13 theorbtwo joined
yts why does it use *? 20:13
nothingmuch see you guys on tuesday
lanny yts: You probably want to do a once-over read of all the Synopses. feather.perl6.nl/syn/ 20:15
lambdabot Title: Official Perl 6 Documentation
lanny A lot of the syntax is covered there.
yts ok~
lanny If you want send me a /msg and I'll answer some questions to get you started.
theorbtwo or spec.pugscode.org/
lambdabot Title: Official Perl 6 Documentation
theorbtwo (Which is likely really the same page with a nicer URL.) 20:16
20:29 mdiep joined 20:30 weinig is now known as weinig|bbl
ajs_work I wonder... would it be a good thing to add the various Unicode functions (such as the four forms of normalization) to S29's string operations along with uc, lc, etc? 20:32
20:35 justatheory joined 20:37 rashakil joined
ajs_work I'll add some examples and see what people think 20:39
TimToady ajs_work++ 20:40
yts what's the meaning of multiparadigmatic 20:41
ajs_work yts: utilizing multiple paradigms at once. For example, both functional and object oriented programming paradigms. 20:44
20:49 rashakil joined
TimToady it means not being able to make up your mind what kind of programmer you want to be. 20:51
yts TimToady: why you can't make up your mind to a specific programmer 20:55
20:56 weinig|bbl is now known as weinig
TimToady if I knew the answer to that, I'd be a different person 20:56
21:00 larsen joined 21:03 SubStack joined
[particle] i think timtoady has decided to be all things to perl people :) 21:10
...or is that perl things to all people? 21:12
yts why perl6 define a lot of ways...to have comment... 21:22
[particle] there's more than one way to do it
yts i know it but don't you think it's too much? 21:23
[particle] i'm american, i'm not sure i understand the phrase "too much." 21:24
yts i am reading "feather.perl6.nl/syn/S02.html" now, i found it contain many ways to write comment 21:25
lambdabot Title: S02
yts but why perl6 need so many ways just for comment
21:28 lanny joined 21:50 Ikarus joined
Ikarus anyone remember the operand chart drawn for Perl 6 and if it is still availible somewhere ? 21:50
21:56 dolmen joined
Ikarus hrm, it is from 2004, wonder how valid it still is 21:56
Anyone care to look ? www.ozonehouse.com/mark/blog/code/P...Table.html 21:57
lambdabot Title: Mark Lentczner's Journal, tinyurl.com/ysxg6
22:04 etzel joined
SamB ouch 22:06
22:06 diotalevi_ joined
SamB you know you have too many operators when you have to put them in a periodic table... 22:07
lanny Yes, and too many verbs when you have to have a dictionary. 22:09
SamB yeah, well, I mean, infix ops are all well and good, but I like to keep them in modules ;-) 22:10
lanny Ikarus: looks like mostly syntax changes. Most of the operator concepts look there. Not surprising since the concepts got defined (or were known) first-ish.
Ikarus lanny: hmmm, any chance you could point me to a document to update it from 22:11
I actually like the format of the table
lanny heh. feather.perl6.nl/syn/
lambdabot Title: Official Perl 6 Documentation
lanny Although you could probably go a long with with pugs/docs/Perl6/Overview/* 22:12
svn.openfoundry.org/pugs/docs/Perl6/Overview/ 22:14
lambdabot Title: Revision 13736: /docs/Perl6/Overview, tinyurl.com/frbrv
Ikarus Hmmmm, I should make a version for web use, with hoovering on a operand gives you the docs for it 22:26
22:37 Limbic_Region joined 22:42 rashakil_ joined 22:43 diotalevi_ joined
lanny @tell markstos re: t/unspecced/sort.t -- Was the place suggested when I asked in channel. More than willing to move it to a better place if one is suggested. (I was aiming for Prelude but sort() still seems a moving target.) 22:51
lambdabot Consider it noted.
TimToady @tell audreyt my new pet peeve is that "No compatible subroutine found" doesn't answer the question: "Compatible with what?" 22:52
lambdabot Consider it noted.
22:54 lambdabot joined 22:57 justatheory joined 23:11 lidden joined 23:29 statico joined 23:34 shizu joined 23:37 jetscreamer joined 23:46 wkh joined 23:48 mjk joined 23:50 MacVince joined 23:53 mjk joined 23:57 Psyche^ joined

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!