pugscode.org | temporary feather address: perl6.wall.org | sial.org/pbot/perl6 | ?eval [~] <m oo se> | We do Haskell, too | > reverse (show (scanl (*) 1 [1..] !! 4))
Set by allbery_b on 24 March 2007.
QtPlatypus Is my's scope "THings that follow it" or "Things that follow it and are to the right of it?" 00:02
In other words. 00:03
my $sub = sub { $sub->() }; # What $sub gets called? 00:04
00:12 aukjan|gone joined
svnbot6 r15871 | lwall++ | compensate for pugs not parsing '{' and '}' correctly in rules 00:18
TimToady QtPlatypus: that's a syntax error in Perl 6, and in any case the answer depends on whether you're talking about Perl 5 or Perl 6. (Hint: Perl 5 does it wrong. :) 00:19
geoffb ?eval my $sub = sub ($a = 5) { say $a }; my $sub = sub { $sub->(4) }; $sub->(3)
evalbot_r15831 Error: ␤Unexpected "$sub"␤expecting "=", "::", term postfix, operator, "}", context, ":" or "("
geoffb ?eval my $sub = sub ($a = 5) { say $a }; my $sub = sub { $sub.(4) }; $sub.(3) 00:20
I'm guessing recursion just occured.
evalbot_r15831 pugs: out of memory (requested 1048576 bytes) 00:21
geoffb yep
TimToady indeed, that's what you just asked it to do
in any case the two "my $sub" declarations declare the same variable.
geoffb Actually, I was hoping to get a redecleration eror
er, error
TimToady not impl 00:22
geoffb nodnod
QtPlatypus TimToady: I was asking about Perl6
TimToady but to clarify to QtPlatypus, in Perl 5 "my $sub" is not introduced till after the statement, whereas 00:23
in Perl 6 the $sub is introduced immediately, so is visible in the body.
QtPlatypus So I should be able to do something like this in perl6 (Syntax correct fix if needed)
geoffb I wish there was something I could do to help audreyt ... Haskell just doesn't ~~ my coding style
[particle] ...join parrot... 00:24
;)
TimToady QtPlatypus: yes, that sort of thing should work fine in P6 (modulo the infinite recursion).
QtPlatypus my @fib = gather { take 1; take 1; my $i = 0; while (1) { take @fib[$i++] + @fib[$i] } };
?eval my @fib = gather { take 1; take 1; my $i = 0; while (1) { take @fib[$i++] + @fib[$i] 00:25
evalbot_r15831 Error: ␤Unexpected "@fib"␤expecting "=", operator, ":", ",", "}", context or "("
Aankhen`` ?eval my @fib = gather { take 1; take 1; my $i = 0; while (1) { take @fib[$i++] + @fib[$i] } } 00:26
TimToady it might need to be a := instead
QtPlatypus ?eval my @fib := gather { take 1; take 1; my $i = 0; while (1) { take @fib[$i++] + @fib[$i] } }; say @fib
Aankhen`` Um.
geoffb [particle]: things may have changed, but last I looked adding features to languages/perl 6 was not terribly approachable ... and I can only spend time in small slices. Not enough to go deep and study the existing code for a week.
QtPlatypus TimToady: Why?
TimToady = is probably going to try to be eager
Aankhen`` QtPlatypus: You might want to stop the loop at some time.
00:26 whooosh joined
Aankhen`` ?eval my @fib = gather { take 1; take 1; my $i = 0; while ($i < 10) { take @fib[$i++] + @fib[$i] } } 00:26
QtPlatypus ?eval my @fib := gather { take 1; take 1; my $i = 0; while (1) { take @fib[$i++] + @fib[$i] } }; say @fib[5]
Aankhen`` ?eval my @fib := gather { take 1; take 1; my $i = 0; while ($i < 10) { take @fib[$i++] + @fib[$i] } }; @fib.say 00:27
[particle] geoffb: i'll keep you in mind for when we get Test.pm passing
geoffb [particle]: do.
[particle] then you should be able to do more coding in perl 6
Aankhen`` Bah.
QtPlatypus thinks we have gilled eval bot?
Aankhen`` :-(
geoffb I'm just *itching* to convert a buttload of P5 to P6, but too many features are currently not impl
Aankhen`` ?eval 1
Looks like it. 00:28
TimToady probably an infinite loopish thing, so needs to exceed CPU limit
Aankhen`` 80,69404230pugs> my @fib := gather { take 1; take 1; my $i = 0; while ($i < 10) { take @fib
80,69404230[$i++] + @fib[$i] } }; @fib.say
80,69404230110000000000
TimToady I suspect lazy array specs aren't implemented yet either. 00:30
moritz is 1..* a lazy list or a range? 00:31
geoffb In fact, a long time ago I had to add source filtering to some of my most complex P5 code to bring a little of the P6 syntax. Because dozens of lines containing 'my $self =shift;' and '$self->foo' just drive me insane.
TimToady moritz: it's both, presumably.
moritz TimToady: ok ;-) 00:32
geoffb TimToady++ # Now, less insane. Or rather, less verbose in his insanity.
moritz @karma TimToady 00:33
lambdabot TimToady has a karma of 27
01:08 nipotaway is now known as nipotan 01:12 dolmans joined 01:29 nipotan joined 01:35 avar joined 01:36 avar joined 01:47 Polar joined 02:34 amnesiac joined 02:59 mako132_ joined 03:01 Eidolos_ joined 03:10 Eidolos_ joined 03:15 perlmonkey2 left 03:26 ccc8 joined 03:32 ccc6 joined 03:34 justatheory joined 03:37 SubStack joined 04:05 esstone joined 04:19 geoffb joined 04:24 ortmage_ joined
cj anybody done any work on compiling to .NET bytecode yet? 04:24
04:49 agentz joined
agentz @tell audreyt ~audreyt/pugs/ on feather is no longer auto-updated? yes, i see svnbot6 is still working, but your pugs_version.h has shown r15831 for days and ~audreyt/pugs/t is obviously out of date according to the snippets in perlcabal.org/syn/S04.html 04:54
lambdabot Consider it noted.
05:03 agentz left 05:11 devogon joined 05:28 the_dormant joined 05:34 rfordinal joined 05:49 mjk joined 06:00 tokuhirom is now known as toku_away
pasteling "evalbot_r15871" at 194.145.200.126 pasted "Pugs build failure" (329 lines, 18K) at sial.org/pbot/24110 06:20
06:21 evalbot_r15871 joined 06:23 toku_away is now known as tokuhirom
nothingmuch masak: i stopped in the middle but the plan was to make tarballs for every chapter based on the source code history 06:23
masak: you can make a branch in which you pull the code patches incrementally 06:24
but the project was abandoned half way when I realized that forth is too beautiful to need a parser or anything interesting in haskell ;-)
the thing is that forth only needs a very small bootstrap, just enough to get the ':' and ';' words working 06:25
and you can even define these words in forth
and sort of hack it in using a bootstrap compiler
once you have that, you can just run a prelude, and when you're done loading that, save the memory image
that's a typical forth system
the interpreter could have been written in haskell, but I had intended it to be parrot 06:26
however, I found learning parrot an investment too big for my curiousity levels
masak: </rant>
gaal nothingmuch: moose 06:27
nothingmuch moose! 06:28
didn't go rowing today 06:29
well, i did *go*, i just didn't row
i was unsure whether or not we had actually made an appointment
so I just went
but it turns out we didn't finalize it
gaal oh :(
nothingmuch nevermoose, I had a nice morning bike ride
gaal yeah, that path's great. 06:30
nothingmuch how was yours?
gaal short but sweet.
06:32 ortmage_ joined
specbot6 r14368 | pmichaud++ | Add leading whitespace to lines so that the table 06:35
r14368 | pmichaud++ | actually ends up looking like a table instead
r14368 | pmichaud++ | of wrapping the text onto a single line.
r14369 | larry++ | Reorganization suggested by Jonathan Lang++
06:43 franck__ joined 06:55 dduncan joined 06:59 marmic joined 07:08 SubStack joined 07:13 iblechbot joined 07:17 silug joined 07:22 jisom joined 07:42 xinming_ joined 07:49 ortmage__ joined 07:53 pnu joined 07:55 frobnitz joined 07:59 the_dormant joined
svnbot6 r15872 | franck++ | r21212@franck-mbp: franck | 2007-04-10 09:58:47 +0200 08:03
r15872 | franck++ | ported File::Util to p6
08:06 SCalimlim joined 08:08 buetow joined 08:38 andara joined 09:10 Belaf joined 09:28 ruoso joined 09:41 tuna joined 09:59 f0rth joined 10:06 elmex joined 10:56 chris2 joined 11:01 xinming joined 11:05 lichtkind joined 11:22 nipotan is now known as nipotaway 11:27 TimToady joined 11:31 dduncan left
masak nothingmuch: anyway, I'm grateful for what you did write, and like it so far @ 04.kwid 11:33
11:37 lisppaste3 joined 11:38 yiwin joined 11:41 mako132_ joined 11:48 fglock joined 11:50 the_dormant joined 11:51 mjk joined 11:53 FurnaceBoy joined
nothingmuch masak: cool 11:55
12:01 Zaba joined
Zaba hi all 12:01
?eval [~] <m oo se>
12:01 evalbot_r15871 is now known as evalbot_r15872
evalbot_r15872 "moose" 12:01
Zaba heh
12:01 nekokak_ joined
Zaba is there an approx. release date for perl6? 12:02
12:04 iblechbot joined
moritz Zaba: christmas ;-) 12:05
12:05 Lunchy_ joined
Zaba moritz, o'rly? 12:06
12:06 prly_ joined, MikeJS__ joined
moritz Zaba: well, we don't know which year's christmas ;-) 12:06
Zaba oh heh
Khisanth o'reilly!
moritz Zaba: and once perl6 is released, _ever_ day is christmas ;-)
Zaba :D
Khisanth then it will mutate into Lisp 12:07
Zaba nooo
lisp--
moritz "Perl 6 is just another dialect of Lisp" -- larry wall ;-)
Zaba =/ 12:08
well 12:09
i don't like lisp, and i don't see perl any similiar to it
Khisanth well even when that is true it is often misleading :)
moritz Zaba: that's the beauty of perl: it's everything to eveyone ;-)
Zaba and will i have to write in p*thon is perl will become lisp? ;-) 12:10
moritz doesn't even know lisp ;-)
Zaba (and about python, the only thing i *really* miss in perl is interactive mode
)
12:11 xinming joined, rashakil joined
Khisanth pugs already has that! 12:11
moritz Zaba: well, pugs has an interactive mode
Zaba nice
Khisanth and there is or at least someone is working on a module that provides that for perl5
moritz Zaba: run.pugscode.org/ if you want to test it without installing
lambdabot Title: Run Perl 6 Now -- in your browser! 12:12
12:12 Limbic_Region joined
Khisanth you can just use evalbot for that :P 12:12
moritz Khisanth: it's a bit too annoying if you want to test many features 12:13
Khisanth never found much need for an interactive mode 12:14
Zaba it doesn't handle ansi escapes ._.
e[32;01mTest 12:15
is pugs the only implementation?
moritz Zaba: no, but the most complete one 12:16
Zaba: there is v6 and kp6/mp6 in the pugs repository
Zaba and parrot?
moritz Zaba: and "onion" = perl6 on parrot in the parrot repos
Zaba: parrot is a virtual maschine, designed as perl6 backend... 12:17
12:17 buetow joined, fglock joined
Zaba heh 12:17
moritz and pugs did support compiling to parrot code
Khisanth also "Failed to create a new terminal", only one visitor at a time?
moritz that's currently bit rotted
Khisanth after selecting development version 12:18
moritz same for me 12:20
12:21 tokuhirom joined 12:32 mdiep joined 12:36 Belaf_ joined 12:43 jisom_ joined 12:44 riffraff joined
riffraff hi 12:44
moritz hi riffraff ;-) 12:45
12:53 the_dormant joined, rfordinal joined
Zaba > reverse (show (scanl (*) 1 [1..] !! 4)) 12:57
lambdabot "42" 12:58
Zaba aha
and the point is?
riffraff Zaba, read douglas adams 13:00
spinclad ? 4*3*2*1 s'tahw
Zaba riffraff, ?
?eval 4*3*2*1
spinclad [what's 6*9]
evalbot_r15872 24
spinclad ?eval 6*9
moritz Zaba: "the hitchhiker's guide to the galaxy" ;-)
Zaba spinclad, 54
evalbot_r15872 54
Zaba moritz, nooo i don't like it 13:01
spinclad eval gets it ?wrong?
riffraff well, it is humor based n his novels, you can't get it until you read those :)
Zaba ?eval print "I'm ok"
evalbot_r15872 OUTPUT[I'm ok] Bool::True
Zaba :D
lumi ?eval :13(6*9)
evalbot_r15872 69
Zaba 0,0
lumi I think it doesn't work 13:02
Zaba ?eval 9*6
evalbot_r15872 54
Zaba 0,.
lumi ?eval :13(54)
evalbot_r15872 69
Zaba ?eval 6*9
evalbot_r15872 54 13:03
spinclad ?eval :13(10_
evalbot_r15872 Error: ␤Unexpected ":13"␤expecting "::"
spinclad ?eval :13(10)
evalbot_r15872 13
moritz what's :13(10)? a pair 13 => 10? 13:04
andara Khisanth: No, but the development version is broken for PUGS_SAFEMODE=1 13:05
13:07 qu1j0t3 joined, jweyrich joined 13:10 fglock joined
spinclad ?eval :13(100) 13:10
evalbot_r15872 169
spinclad read a number in base 13
don't think you can write 13 => 10 as an adverb phrase 13:12
?eval :13(5+5) 13:13
evalbot_r15872 13
spinclad number -> string -> number
moritz ?eval :2(1001) 13:16
evalbot_r15872 9
13:17 jweyrich left
moritz ?eval :13(20) 13:17
evalbot_r15872 26
spinclad ?eval :13(1c) 13:18
evalbot_r15872 Error: ␤Unexpected ":13"␤expecting "::"
spinclad ?eval :13('1c') 13:19
evalbot_r15872 25
spinclad now we need the opposite 13:20
write a number in base 13
moritz spinclad: I think that can easily be done in a library function 13:21
no need to put that into the core
Zaba ?eval `ls` 13:23
evalbot_r15872 Error: Unsafe function 'Pugs::Internals::runShellCommand' called under safe mode
Zaba heh
i would except that
fglock a recent discussion on regex algorithms: lambda-the-ultimate.org/node/2064 13:28
lambdabot Title: Regular Expression Matching Can Be Simple And Fast | Lambda the Ultimate
13:31 vel joined 13:32 qu1j0t3 left
[particle] ?eval :10(:13(5+5)) 13:37
evalbot_r15872 13
spinclad ?eval :13(1*10**2) 13:42
evalbot_r15872 169
spinclad oddly, this meets the spec
oop no it doesn't 13:43
pugs specbug
should be 100
[particle] i think it should be 100, too
spinclad ?eval :13(1*:13(10)**2) 13:44
evalbot_r15872 256
13:46 falseep joined
spinclad the arithmetic -> number coerce-to string -> number is wrong 13:46
?eval :13(1C) 13:47
evalbot_r15872 Error: ␤Unexpected ":13"␤expecting "::"
Zaba ?eval 10^100 13:48
evalbot_r15872 (10 ^ 100)
Zaba >_<
?eval (10 ^ 100)
evalbot_r15872 (10 ^ 100)
Zaba ?eval print (10 ^ 100)
spinclad that's an xor-junction 13:49
moritz Zaba: use **
Zaba ah
?eval print (10 ** 100)
evalbot_r15872 OUTPUT[one(VInt 10,VInt 100)] Bool::True
OUTPUT[10000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000] Bool::True
Zaba yayay!
?eval 10 ** (10 ** 100)
evalbot_r15872 Inf
Zaba o.o
googolplex will be it!
spinclad you can't have everything, where would you put it? 13:50
Zaba ?
well
let's give the bot another 100**100 TB ram and seeif it can calculate that in a year then
and we'll maybe get $googolplex if we will discover it's possibleto store it in mem :P 13:51
moritz Zaba: it's easier to implement it algebraicly ;-) 13:52
spinclad lazy arithmetic would be nice here
14:08 kanru joined 14:10 Casan joined 14:21 perlmonkey2 joined 14:23 rlb3_work joined
TimToady ?eval :13<1C> 14:24
evalbot_r15872 25
TimToady :13(1C) doesn't work because 1C is not a valid token 14:25
Zaba ?eval :10(1C) 14:26
evalbot_r15872 Error: ␤Unexpected ":10"␤expecting "::"
Zaba ?eval :10(10) 14:27
evalbot_r15872 10
Juerd ?eval :13(1B) 14:32
evalbot_r15872 Error: ␤Unexpected ":13"␤expecting "::"
Juerd ?eval :13<1B>
evalbot_r15872 24
Juerd I don't understand why this isn't 13 => "1B"
TimToady the difference is that :13() is parsed as a function with an argument while :13<> is parsed as pair with a string argument. 14:34
?eval :foo(1+2)
evalbot_r15872 ("foo" => 3)
TimToady ?eval :foo<1+2>
evalbot_r15872 ("foo" => "1+2")
TimToady but the () form must have a valid expression inside, and 1B ain't one. 14:35
but the :<digit> form is forced to be radix notation, not pairs 14:36
you have to use fatarrow if the left side is a number. 14:37
(and you want a pair)
14:37 amnesiac joined 14:39 bonesss joined
TimToady that's just part of the generalization that adverb notation only allows identifiers on the left; any other expression there needs to use => instead, including 13 => '1B' 14:40
14:43 xinming joined
TimToady hmm, we could possibly allow $num.^13 to go the other way. 14:43
might be too confusing with metamethods though 14:44
masak so where does the '24' come from? 14:45
[particle] 5.^13
masak ?eval :13<1B>
evalbot_r15872 24
TimToady 13 + 11
masak ok 14:46
lumi ?eval 13:<42>
evalbot_r15872 Error: ␤Unexpected ":<42"␤expecting "_", fraction, exponent, term postfix or operator
lumi Erk
?eval :13<42>
evalbot_r15872 54
masak is there anything special in the <...> that forces hexadecimal?
TimToady no
[particle] string context
lumi masak: qw
masak a bit quizzical 14:47
TimToady :13['42']
masak seems a lot of constructs conspire here
TimToady: yes...
TimToady ?eval :13['42']
evalbot_r15872 42
TimToady er...
masak right
this is getting simpler and simpler :)
TimToady :13{'42'} 14:48
?eval :13{'42'}
evalbot_r15872 54443306934741
masak oh...
moritz nice one ;-)
masak I see
lumi Now all becomes clear
masak not
yes, I'm glad you provide clear examples
TimToady looks like <> is hardwired, not generalized to the other subscript forms
but then that's probably not specced, so it's making up random numbers. 14:49
:)
masak back to :13<1B>
what happens, step by step?
it's a pair, yes?
TimToady parsed as one initially
masak like 13 => '1B'
TimToady so '13' => '1B'
masak yes
whence 24? 14:50
TimToady then because of : and because of it being a number and not and identifier
it's forced to mean radix conversion
masak <1B> is?
TimToady 1B in base 13 is 13 + 11
masak aha
TimToady 1A would be 13 + 10
masak for some reason I thought the radix conversion went the other way 14:51
and it confused me that the radix was in (what looked like) hex
TimToady no, it's for specifying the number in whatever base is convenient.
mostly for literals, but the functional form fell out of it.
masak so :13<1B> simply means "1B parsed as a base-13 number"
TimToady ?eval :13('1B') 14:52
evalbot_r15872 24
TimToady yes
masak understands
thx
14:52 stevan__ joined
moritz ?eval my $b = 5; :$b('11') 14:52
evalbot_r15872 Error: Cannot cast from VList [VStr "b",VInt 5] to Pugs.AST.Internals.VCode (VCode)
TimToady have to use the radix() function for that. 14:53
moritz ok ;-)
14:53 rindolf joined 14:54 jerrylee joined 14:55 buetow joined
jerrylee how's everything going on perl6 now? 14:57
moritz well, it's evolving ;-) 14:59
masak at quite a steady pace, actually
15:00 yxam joined
TimToady there's a steady baseline pace defined by the tortoises, with occasional bursts of speed defined by the hares. 15:00
we don't expect the tortoises to go fast, nor do we expect the hares to go constantly. 15:01
masak true. hacker-like creativity is a bit like rocket fuel in that sense 15:02
[particle] TimToady: don't forget about the mooses
TimToady mooses generally don't go at all
[particle] but they're nice to look at
jerrylee the development is mainly in parrot?
I have been waiting it for about 5 years ;) 15:03
TimToady that's okay, I've been waiting for 6.5 years. :)
jerrylee TimToday: :D cool
and I am still waiting 15:04
masak jerrylee: have you tried pugs lately?
TimToady basically, parrot is working bottom up, and pugs is working top down, and someday they'll meet in the middle.
masak ?eval "hello from pugs"
evalbot_r15872 "hello from pugs"
15:04 yxam left
jerrylee masak: latest version? not yet 15:04
i will go take one 15:05
[particle] how close is 6.28 release?
masak [particle]: guess it depends mostly on the hares right now
15:06 xinming joined
jerrylee well, run it in browser is really cool 15:06
masak jerrylee: writing small scripts is cool too 15:07
the holes are only noticeable once you start toying with OO
jerrylee masak: interesting thing, is it a really pug machine behind?
or just some cgi? 15:08
[particle] yes, the hampsters weren't cute enough
y/p//
masak jerrylee: not sure what you mean by that
it's pugs that's running there through that web page, yes
but it's also some cgi
jerrylee masak: then it's really cool 15:09
masak of course
that's what we do here
:)
jerrylee pugs> :e print 'hello'
Loading Prelude... done.
helloVBool True
this is really strange ;)
masak jerrylee: you probably don't want the ":e"
TimToady ?eval say 'hello' 15:10
evalbot_r15872 OUTPUT[hello␤] Bool::True
masak ?eval 'hello'
evalbot_r15872 "hello"
jerrylee ha, i got it
TimToady jerrylee: would you like a commit bit? 15:11
jerrylee all features of perl6 are loaded?
TimToady all the implemented features... :/
jerrylee TimToady: commit bit?
masak jerrylee: writing permissions for the pugs project
TimToady to be able to check in fixes/tests to pugs
masak most people in here have one 15:12
TimToady we hand 'em out like candy
jerrylee TimToady: sounds good thing to try
masak that's the spirit!
TimToady msg me your email and we'll add you
jerrylee TimToady: I might need some candy
moritz you virtually can'T hang around here longer than 5min without getting one ;-)
masak it's such a wonderful system 15:13
most people who are in the least interested are committers
jerrylee how to register freenode?
I cannot send private message
riffraff write /msg nickserv help 15:14
moritz and it works... at least it encouraged me to do a few commits
jerrylee riffraff: thanks, trying now
riffraff I think to register is /ns register email paswsord or something like that
jerrylee TimToady: did you get that? 15:15
rindolf Hi TimToady , moritz
moritz hi rindolf ;-)
rindolf moritz: what's up? 15:16
An Israeli friend told me "sup with you?" on IM.
It sounded strange.
TimToady yes, commit bit is on its way
moritz rindolf: I'm writing a german webpage on perl6 ;-)
rindolf His English is relatively bad.
moritz: oh nice.
TimToady jerrylee: it is customary to add yourself to AUTHORS as first checkin to see if it all works. 15:17
jerrylee TimToady: no problem 15:18
andara hi all
moritz hi andara ;-) 15:19
andara moritz: do you have an account on feather?
jerrylee svn-ing now 15:22
moritz andara: no
rindolf moritz: I've invested some time on perl-begin.berlios.de/ - also a bit today. 15:23
moritz: I hope I can give it a more official status soon.
lambdabot Title: Perl Beginners' Site
andara moritz: shame. I'm looking for someone who could build pugs on feather for me.
moritz andara: and do you think I'am qualified? ;-) 15:24
andara moritz: why not? you have a commit bit :-) 15:25
moritz andara: ;-)
andara: I have debian packages, but I don't know if you can install them on feather...
andara: they install globally, eg in /usr/bin/, not the kind of thing you want on a multi user maschine 15:26
andara moritz: I know how to build pugs just fine, but because I host runpugs, my account is ulimit'ed
moritz: so I can't build pugs while running runpugs :-( 15:27
jerrylee pugs is really a big thing
moritz andara: maybe you should ask Juerd for another account
andara: that might be the cleanest solution
andara moritz: I agree.
masak jerrylee: yes. and a lot of things in it already work 15:28
andara moritz: BTW, are you still in Heriot-Watt?
moritz andara: yes
masak ie they implement perl6 correctly
andara moritz: I'm at Glasgow
moritz andara: hey, cool
jerrylee masak: it's still svn=ing, maybe my internet connection is too slwo
masak jerrylee: maybe, or maybe pugs takes a while to download regardless of connection 15:31
but only the checkout, of course
jerrylee it looks so ;)
masak subsequent updates are much speedier 15:32
jerrylee masak: it's not running on windows, is it? 15:33
TimToady jerrylee: for where it runs and how well, see m19s28.vlinux.de/cgi-bin/pugs-smokeserv.pl? 15:34
lambdabot Title: Pugs Smoke Reports
TimToady looks like it's working as well on Windows as on Linux
(look at the repository lines, not the release lines)
15:34 DarkWolf84 joined
jerrylee can i make it on windows? 15:36
TimToady if you have ghc installed, yes 15:37
[particle] TimToady: pmichaud has been converting pge tests to use active comments (# todo). based on this work, we're thinking about some extensions to the syntax
for an example of what's done, see svn.perl.org/parrot/trunk/t/compile..._charclass
lambdabot tinyurl.com/292jz2
[particle] near the bottom, there's a bunch of todo for pugs, all the same. a block syntax would be prettier 15:38
C<<# todo :pugs<feature> {>> paired with C<<# todo :pugs<feature> }>> maybe? 15:39
TimToady could get confusing if different platforms want different blocks...
[particle] or, make block syntax TODO
uppercase would make it similar to p5 15:40
as you say, you definitely need some way to distinguish between overlapping todo blocks
TimToady similarity to p5 is not necessarily a recommendation around here. :)
[particle] it'd be nice to have the block be easily visually distinguished from non-block 15:42
i think it should be different near the front, and not near the back
# todo-block-begin :pugs<feature> seems too verbose 15:43
TimToady I'm not yet convinced the blocks are needed. There's something to be said for ugly when you want to motivate people to untodo things. :)
masak jerrylee: how's the downloading going? 15:44
jerrylee masak: i am now learning how to build it on my xp pc
masak: all downloaded 15:45
58MB, hweh
[particle] TimToady: how about # skipall :parrot<not yet implemented> 15:47
jerrylee is wondering why GHC is so big
masak jerrylee: you can still make that AUTHORS file commit, you know
even without building pugs 15:48
[particle] hurry, before your bit runs out!!!! ;)
masak :)
[particle] TimToady: i'm not sure about skipall, but i'd like to see # spec
jerrylee masak: oh haha ;)
[particle] have any thoughts about that? 15:49
jerrylee masak: how shall i do that is fastest?
TimToady [particle]: what does it do?
masak jerrylee: add yourself to AUTHORS using your favorite editor 15:50
and then do "svn ci AUTHORS"
that's "svn commit AUTHORS" -- either works 15:51
jerrylee masak: ok, i will make it happen asap
masak jerrylee: swell
TimToady jerrylee: on the other hand, don't let the gang give you grief. :)
jerrylee TimToady: :D 15:52
i am beating them
TimToady that will not, of course, prevent them from giving you grief. :) 15:53
masak in fact, beating a gang has been known to make it give even more grief
not this gang, though
we just assimilate 15:54
TimToady we're all either thick-skinned or thin-skinned around here as the need arises. :)
masak exactly :) we don't take offense so easily, because we evaluate insults lazily 15:55
[particle] TimToady: a new smartlink syntax, eg. # spec PDD15<Role PMC API>
(i forgot a colon) 15:56
jerrylee ;)
TimToady I'd try to make the syntax as close to the current smartlink syntax as possible. Basically these comments are standing in for what would be in the .t file, except that the .t file is using a data file of some sort. 15:57
and in fact I'd like to get the :todo entries out of the tests in the .t files as well. 15:58
[particle] ok. then we need a new pod smartlink syntax
TimToady: you were the one who argued *for* them last time around, you big fence-sitter :P
TimToady I think currently they're using L<DOC/section/searchstring>
or some such
[particle] yes, which isn't even valid p5 pod syntax (but an easy extension) 15:59
TimToady argued for which? I was arguing for todo() on a separate line back then too.
[particle] oh, i missed the colon in your message
TimToady something that can in theory just be visually grepped out when you aren't interested
jerrylee is doing svn cleanup 16:00
[particle] so, you'd like to see # todo in .t files?
TimToady oh, okay
either # todo or todo(...) as a function call
masak jerrylee: why? 16:01
[particle] i thought todo/skip/spec looked nice together
TimToady todo(:pge<feature>) could work
jerrylee masak: ha, I used cygwin to do svn, it failed ;)
TimToady using todo() rather than # todo has the advantage of being processable without preprocessing the .t file 16:02
jerrylee masak: i have cygwin in windows as linux emul
TimToady smart comments would require a pass over the text
[particle] right, in data files, # todo works better
TimToady important thing is to get it out to a separate line.
[particle] ...so you don't have to execute the data file
TimToady maybe we could have skip and unskip 16:03
jerrylee has update AUTHORS, new build is 15873 ;)
svnbot6 r15873 | jerrylee++ | Added Jerry Lee
masak wei!
TimToady sugoi!
jerrylee sorry for my slow motion ;) 16:04
TimToady no problem, most of us here are tortoises
masak yes. welcome
TimToady the hares are mostly sleeping right now
or should be... 16:05
jerrylee :) interesting
[particle] jerrylee: f􏿽xE1ilte 16:06
jerrylee i guess tomorrow I should try to build it
irish?
[particle] yep, irish 16:08
16:08 xinming joined, falesca joined
TimToady it ain't japanese... 16:08
masak jerrylee: good luck. that windows thing sure sounds like a pain
masak curses under his breath for missing obvious pun moment
TimToady if you have trouble look up Lanny or Limbic_Region
jerrylee no problem
[particle] you can also use the many bots hanging out here 16:09
@tell jerrylee hello
lambdabot Consider it noted.
[particle] jerrylee: say something
16:09 xinming joined
[particle] @tell particle hello 16:09
lambdabot Consider it noted.
moritz ?eval say "something" # ;-)
[particle] hee
evalbot_r15872 OUTPUT[something␤] Bool::True 16:10
jerrylee bots?
lambdabot jerrylee: You have 1 new message. '/msg lambdabot @messages' to read it.
[particle] see
masak @tell evalbot_r15873 hello
lambdabot Consider it noted.
moritz ?eval 1
16:10 evalbot_r15872 is now known as evalbot_r15873
evalbot_r15873 1 16:10
lambdabot evalbot_r15873: You have 1 new message. '/msg lambdabot @messages' to read it.
masak he he
16:10 xinming joined
moritz ?eval "@messages" 16:10
evalbot_r15873 "\@messages"
[particle] @tell evalbot_r15873 ?eval fork while fork;
lambdabot Consider it noted.
jerrylee very interesting bot 16:11
Juerd Sigh :)
masak [particle]: now that's just eeevil
[particle] :)
masak &
TimToady ?eval fork
evalbot_r15873 Error: Unsafe function 'fork' called under safe mode
lambdabot evalbot_r15873: You have 2 new messages. '/msg lambdabot @messages' to read them.
16:17 prly joined
Zaba dude 16:22
it's evil >_>
it can't read its messages
16:22 prly__ joined
[particle] somebody will just have to commit something 16:22
Tene ?eval @moosages
evalbot_r15873 []
Tene ?eval \@moosages
evalbot_r15873 []
Tene ?eval "@moosages" 16:23
evalbot_r15873 "\@moosages"
jerrylee ?eval
evalbot_r15873 No code to eval given.
jerrylee ?eval "say hi"
evalbot_r15873 "say hi"
jerrylee ?eval sleep
evalbot_r15873 Error: ␤Unexpected end of input
jerrylee ?eval exit 16:24
evalbot_r15873 Error: Unsafe function 'exit' called under safe mode
16:27 prly___ joined 16:28 ludan joined
jerrylee ?eval say hi while (1) 16:29
evalbot_r15873 Error: No such subroutine: &hi
jerrylee ?eval print while (1)
what's going on
DarkWolf84 while(1) ^^ 16:30
jerrylee ?eval fork
evalbot_r15873 OUTPUT[
DarkWolf84 thats cruel :)
evalbot_r15873 Error: Unsafe function 'fork' called under safe mode
[particle] pokes evalbot_r15873
jerrylee but it did works
maybe it's no safe enough ;) 16:31
16:32 lichtkind_ joined
jerrylee ? eval print "hi" 16:32
?eval print "hi"
evalbot_r15873 OUTPUT[hi] Bool::True
jerrylee is really sleepy 16:33
time to say goodnight :)
moritz jerrylee: night ;-)
jerrylee bye guys :)
bye moritz 16:34
16:38 bernhard joined 16:40 franck__ joined 16:41 justatheory joined 16:47 DarkWolf84 joined 16:49 kane_ joined 16:50 Lunchy_ is now known as Lunchy 16:54 ortmage_ joined 17:16 toshinori joined
Limbic_Region @seen audreyt 17:39
lambdabot audreyt is in #perl6 and #haskell. I last heard audreyt speak 11h 21m 44s ago.
17:42 Psyche^ joined 17:44 jamhed joined
svnbot6 r15874 | fglock++ | - updated v6/yapc-sa-hackathon.txt 17:52
17:53 explorer joined 17:54 Psyche^ is now known as Patterner 17:57 fglock joined 18:05 xdg joined
spinclad ?eval :13<1*:13<10>**2> 18:05
18:05 evalbot_r15873 is now known as evalbot_r15874
evalbot_r15874 406616 18:05
Limbic_Region reverse (show (scanl (*) 1 [1..] !! 4)) 18:06
grrr
moritz Limbic_Region: I think you need a '> ' in front 18:07
Limbic_Region moritz, no - it is this terrible IRC client I am forced to use
spinclad ?eval :13<113102>
evalbot_r15874 406616
Limbic_Region but c'est la vie 18:08
moritz oui, c'est vrai ;-)
spinclad so pugs doesn't have the specced exponential form... 18:09
18:13 jisom joined 18:29 the_dormant joined
svnbot6 r15875 | lwall++ | Don't need to indirect stoppers when <?before($stop)> can take arg directly. 18:35
r15875 | lwall++ | Teach cheater to translate $.expect* calls to subrule calls.
pasteling "evalbot_r15874" at 194.145.200.126 pasted "Pugs build failure" (15 lines, 593B) at sial.org/pbot/24119 18:38
TimToady strange, builds here 18:40
moritz usually a "make realclean" will do ;-) 18:41
jisom anyone have any clue as to how to get ghc running on freebsd/amd64 18:52
TimToady looks like clkao was doing smokes on freebsd back in October 18:54
but he's probably in bed right now 18:55
jisom there's an x86 version in ports for freebsd, but it's i386 only(probably a bootstrapping problem) 18:57
18:57 yiwin left
TimToady on the bright side, the smokes look like 64bit 18:57
says archname: i386-freebsd-64int
perhaps you will be doing "original research" 18:58
jisom eh, doubtfully, but maybe independent research 18:59
but that's worse 19:00
gaal jisom: funny, there's a thread on that on ghcusers just now
[email@hidden.address]
the short of it, in libraries/base/Text/Regex/Posix.hs change 19:01
- nsub <- ((\hsc_ptr -> peekByteOff hsc_ptr 4)) regex_ptr
+ nsub <- ((\hsc_ptr -> peekByteOff hsc_ptr 8)) regex_ptr
if you're building your own ghc
but look out for further patches.
jisom yeah, it's fairly high up on google for "ghc freebsd amd64" actually.... 19:02
gaal (and a binary...)
TimToady fix one hardwired constant with another...cool... :/
gaal TimToady: it's a generated constant, but the unfortunate thing is that it's generated on a different machine 19:03
TimToady that would be a problem
jisom so the cross compiler part doesn't really do enough cross compiling?
TimToady but you'd think the Haskell folks would be about the last to fall into the trap of too little abstraction. :)
gaal "This will be fixed as a side effect of hackage.haskell.org/trac/ghc/ticket/1160." 19:04
19:04 ortmage_ joined
lambdabot Title: #1160 (drop regex-posix in favour of regex-tdfa?) - GHC - Trac 19:04
gaal jisom: the compiler doesn't see it, it's in hsc2hs
TimToady tdfa? to-die-for algorithm?
jisom then I'll get to look into the other issues with freebsd/amd64, which might involve freebsd using amd64 instead of x86_64 19:05
xdg quick question: how is v6.pm going?
19:07 the_dormant joined 19:08 the_dormant joined
TimToady xdg: quick answer: fglock has mostly been working on mp6 and kp6 recently. 19:08
xdg Could I please get a translation of mp and kp? 19:09
TimToady miniperl and kindaperl
subsets intended for bootstrapping 19:10
xdg watches the Perl 6 learning curve get steeper...
TimToady you don't have to learn them unless you're into bootstrapping. :)
xdg I was going to mention v6.pm on Win32 for my Vanilla Perl talk, but I may not if development on it is stalled or if its irrelevant 19:11
TimToady well, the main problem with v6 is the same as all the other current implementations, which is that they're not using a parser written in p6. 19:13
fglock hi
TimToady I'll let fglock answer for the rest of it... :)
fglock :)
v6.pm got some maintenance lately, but it needs a cpan upload 19:14
TimToady to what extent is v6 blocking on issue in the p5 runtime?
fglock it currently depends on an update of Sub::Multi
TimToady *issues
fglock it's not really blocking; I plan to replace parts of v6 with kp6 routines 19:15
maybe depending on the yapc::sa hackathon results
19:15 prism joined 19:17 polettix joined
fglock the long-term idea is to start sharing more code between kp6 and v6, until v6 and 6-on-6 become almost the same thing 19:17
xdg fglock: thanks for the overview. So if I mention in in a Win32 talk, it's still a relevant project, then...
*it 19:18
fglock xdg: mp6 also runs on perl5; but it is written in a subset of perl6
xdg I feel doomed to step into the Escher and ask what parses mp6 19:19
fglock re using a parser written in perl6, it is in the works; but currently it is not worth the trouble, there are some things to do first 19:20
19:20 rhr joined
fglock xdg: mp6 parses mp6; see svn.pugscode.org/pugs/v6/docs/ 19:21
lambdabot Title: Revision 15875: /v6/docs
xdg And the mp6 emitter to backend "X" isn't written yet? 19:23
fglock it depends on X 19:24
by the end of the week there will be a hackathon, we plan to port mp6 to pugs (and possibly parrot) 19:26
xdg For which X are there emitters?
fglock currently perl5 (bootstrapped), parrot (incomplete)
I started some C work, but it doesn't look promising 19:27
xdg glances around the twisty maze of passages in despair 19:28
[particle] re: parrot (incomplete) ...i see that all too often :(
fglock I need to find out what is the current state of perl6-in-parrot, in order to write code for it 19:29
xdg fglock: thanks for walking me through all that. I've got to get back to real work now, but look forward to seeing how it all turns out. 19:30
kattana_ runs around a corner in the maze and bumps into xdg
fglock xdg: ok :)
[particle] fglock: feel free to join us on #parrot for answers 19:36
19:37 larsen_ joined
spinclad fglock: pmichaud mentioned in #parrotsketch today : 19:41
svnbot6 r15877 | franck++ | r21237@franck-mbp: franck | 2007-04-10 21:38:17 +0200
r15877 | franck++ | File::Find : follow (or not) symlink with follow attribute
r15877 | franck++ | File::Util : code for isbin filetype
spinclad 14:36:40 < pmichaud> * I also spent a fair bit of time examining fglock++'s "miniperl6" implementation in pugs, now that I understand what it's doing
14:36:43 < pmichaud> * I may see about taking a miniperl6-like approach with some of Parrot's translation tools
19:41 baest joined
spinclad 14:36:46 < pmichaud> * I also spent some time looking at generator/iterator ideas over the weekend so that I can get that part of perl6 implemented 19:41
19:54 rashakil_ joined
fglock [particle]: thanks; I'm installing a newer parrot 19:57
hmm - I hope we could join forces some day 19:59
anyway, if there is some reuse it's ok 20:00
20:00 ortmage_ joined 20:01 Aankhen`` joined
fglock TimToady: I found out I can implement regex "|" using the current <%hash> code 20:01
only doing some pre-processing
20:04 ortmage_ joined 20:07 geoffb joined
fglock spinclad: miniperl6 was actually meant for parrot 20:08
I wonder if pm would help with kp6
20:08 pjcj joined
fglock but it is nice to see he is interested :) 20:09
spinclad ask him -- i want to see you two working together
[particle] meetoo
20:10 Chatterman joined 20:12 Chatterman left 20:15 lichtkind joined 20:17 offby1` joined, diakopter joined 20:32 offby1` is now known as offby1 20:37 dduncan joined 20:41 c6rbon joined 20:58 pjcj_ joined 21:00 buetow joined 21:05 fglock left 21:20 Limbic_Region joined 21:29 larsen_ joined 21:45 lichtkind joined
lichtkind ?eval do{} 21:45
21:45 evalbot_r15874 is now known as evalbot_r15877
evalbot_r15877 undef 21:45
lichtkind ?eval do{1}
evalbot_r15877 1
lichtkind ?eval do{...}
evalbot_r15877 Error: Cannot cast from VUndef to Pugs.AST.Internals.VCode (VCode)
21:45 Polar joined
lichtkind ?eval sub arg{...} 21:45
evalbot_r15877 undef 21:46
lichtkind ?eval sub arg{}
evalbot_r15877 undef
lichtkind ?eval sub arg{!!!}
evalbot_r15877 undef
lichtkind ?eval sub arg{}; arg;
evalbot_r15877 undef
lichtkind ?eval sub arg{!!!}; arg;
evalbot_r15877 Error: !!! - not yet implemented
lichtkind ahhh
Tene ?eval sub arg{...}; arg; 21:47
evalbot_r15877 *** ... - not yet implemented␤ at <eval> line 1, column 9-12
21:51 FurnaceBoy joined 23:01 Odin- joined 23:02 Polar joined 23:10 perlmonkey2 left
prly___ ?eval 23:10
sub f1($a,$b){say "$a:$b"}; sub invo($f,*@args){$f(|@args)}; invo(&f1,2,5)
evalbot_r15877 No code to eval given.
prly___ ?eval sub f1($a,$b){say "$a:$b"}; sub invo($f,*@args){$f(|@args)}; invo(&f1,2,5)
evalbot_r15877 OUTPUT[2:5␤] Bool::True
prly___ ?eval sub f1($a,$b){say "$a:$b"}; class foo{has $.f; method invo(*@args){$.f.(|@args)}}; foo.new(f=>&f1).invo(2,5) 23:11
evalbot_r15877 OUTPUT[2 5:␤] Bool::True
prly___ there must be something I didn't understand properly 23:14
shouldn't they produce the same result ?
TimToady I'd think so. I can't even get it to compile in my pugs though... :( 23:23
23:24 mr_ank joined
TimToady says: *** Undeclared variable: $b at try line 1, column 22-30 23:24
Limbic_Region TimToady - do you know since what build? 23:25
Limbic_Region hasn't tried in a few days because there didn't appear to be any significant updates
TimToady no, I'm at 15875
Limbic_Region ok - will check my current pugs (which is built) 23:26
hrm, this doesn't look right
r15836
will attempt to build anyhow 23:27
23:27 meppl joined, marmic joined
Limbic_Region pugs built fine for me as of r15837 23:31
er
r15877
23:35 gnuvince_ joined
gnuvince_ > reverse (show (scanl (*) 1 [1..] !! 4)) 23:36
lambdabot "42"
gnuvince_ > scanl
lambdabot Add a type signature
TimToady hmm, it actually parses, it's the |@args that isn't binding right, so it seems to be roughly the same error that prly___ is getting. 23:38
except I'm getting a strictness error 23:39
and evalbot seems not to care about the unbound $b
would be nice if pugs didn't report an undefined parameter as undeclared though... 23:41
by the way, changing the example to use (*$a, *$b) makes it work.
?eval sub f1 (*$a, *$b) { say "$a:$b" }; class foo { has $.f; method invo (*@args) { $.f.(@args) } }; foo.new(f=>&f1).invo(2,5) 23:42
evalbot_r15877 OUTPUT[2:5␤] Bool::True
TimToady so it appears that |@args in a method call is what is hosed. 23:43
23:46 buetow joined 23:48 MikeJS__ is now known as MikeJS