Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2008-01-19

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:21 Schwern joined #perl6
00:32 jjore joined #perl6
00:35 ashleyb_ joined #perl6
00:36 stevan_ joined #perl6
00:36 _sprocket_ left #perl6
00:49 takanori joined #perl6
00:57 pugs_svn r19582 | lwall++ | [fudge] fudging everything that needs fudging
01:39 penk joined #perl6
01:42 meppuru good night
01:52 thoughtp1lice joined #perl6
02:23 pugs_svn r19583 | lwall++ | [fudgeall] depugsify and delinuxify
02:28 alester joined #perl6
02:56 ajs_home joined #perl6
02:59 japhb joined #perl6
03:15 jferrero joined #perl6
03:29 pugs_svn r19584 | particle++ | [fudgeall] delinuxify
03:29 pugs_svn r19585 | particle++ | [fudge] prettify usage
03:36 stevan_ joined #perl6
03:47 ashleyb joined #perl6
03:57 pugs_svn r19586 | lwall++ | [t/spec] Some missing colons
04:11 [particle] TimToady: does skip require a block?
04:16 [particle] nm, i'm working on it
04:23 orafu joined #perl6
04:23 pugs_svn r19587 | particle++ | [fudgeall] windows doesn't seem to want globbing help
04:33 pugs_svn r19588 | particle++ | [t/spec] add skip markers for rakudo
04:35 alester Adriano's articles indexed here: http://www.perlfoundation.org/per​l6/index.cgi?perl_6_microarticles
04:35 lambdabot Title: Perl 6 microarticles / Perl 6, http://tinyurl.com/2t5546
05:01 ingy hi [particle]
05:01 [particle] hey ingy
05:01 ingy did that meeting ever end? :)
05:02 [particle] no, i'm still waiting to do my talk :)
05:02 ingy haha
05:02 ingy the 40 minute talks were... well
05:03 ingy a little long for a 10 minute talk
05:03 [particle] yeah, i think next time we need a gong
05:04 ingy [particle]: did you ever notice that pugs is an anagram of spug?
05:04 [particle] haha, no
05:04 ingy P.U.G.S. Perl Users Group, Seattle
05:05 ingy I just noticed it myself
05:06 ingy maybe we'll have to start SHUG
05:07 * [particle] SHRUGS
05:07 beppu Just out of curiosity, where has audreyt been?
05:07 [particle] curiosity killed the cat
05:07 ingy I keep her in a jar.
05:07 [particle] told ya.
05:08 beppu Let her post a blog entry once in a while.
05:08 ingy she's been begging me to switch her into a par.
05:08 ingy can't stand that java.
05:09 [particle] har
05:09 beppu Are you trying to kill her?
05:10 [particle] audrey kept reality in her system tray for a long time, and when she restored it, it consumed all cycles and memory
05:10 ingy She's draws a good crowd in my museum of programming curiosities
05:11 ingy Don't worry. I feed her bits and bytes.
05:12 ingy Reality bytes.
05:13 beppu Reality is a funny thing.
05:13 ingy [particle]: we should hang out and hack sometime
05:14 [particle] ingy: that'd be cool
05:15 ingy I think after Feb 3rd would be good
05:15 [particle] that's better for me too
05:15 [particle] i leave for dc in the morning
05:15 ingy :)
05:17 Auzon pugs: my @test = 1 .. 10; say "sorted!" if [<] @test;
05:17 exp_evalbot OUTPUT[sorted!␤]
05:17 Auzon pugs: my @test = 1 .. 10, 6; say "sorted!" if [<] @test;
05:17 exp_evalbot RESULT[undef]
05:52 pugs_svn r19589 | particle++ | [fudgeall] ignore previously fudged output files
05:57 BinGOs joined #perl6
06:07 jjore joined #perl6
06:13 lambdabot joined #perl6
06:17 TimToady [particle]: skip does not require a block
06:17 [particle] figured that out, thanks
06:17 ingy hi Tim
06:17 TimToady and you can leave out the 1 on the "1 skip"
06:18 ingy or do you prefer Mr Toady?
06:18 TimToady if you say 3 skip a block only counts as 1
06:18 TimToady I prefer "hey you!"
06:18 [particle] aha
06:18 Auzon Hue?
06:18 ingy hi hey you
06:18 Auzon :P
06:19 TimToady hey me
06:19 allbery_b "yo"?
06:19 TimToady a statement counts as starting at a line with "is" et al, and terminating with a line with  ; (and optional comment)
06:20 pugs_svn r19590 | particle++ | [t/spec] skip some unicode tests for rakudo until we can guarantee folks have icu installed
06:20 TimToady I finally got my pugs smokes working again by skipping the tests that spat out malformed utf-8
06:21 TimToady we probably need another verb in fudge like "inline" so we can just insert code to call "skip_rest" and such
06:22 allbery_b pugs still losing for me.  I'm not sure if it's the darcs ghc or the new cabal, but the package.conf-mangling wrapper for ghc that pugs uses is failing to find any installed packages (like, say, bse)
06:22 allbery_b er, base
06:22 TimToady I made fudged files end with exit(1) so they come out "dubious" despite passing :)
06:23 TimToady what version of ghc?
06:23 TimToady (I use svn so can't help with darcs)
06:23 allbery_b 6.9-* from darcs
06:23 TimToady yow
06:23 allbery_b (the equivalent of cvs/svn HEAD)
06:24 TimToady still at 6.6.1 meself
06:24 [particle] TimToady: rakudo is working with fudge now, so it's full steam ahead
06:24 TimToady [particle]++
06:24 allbery_b guess I'll try to find a 6.6.1 package for ppc then
06:25 allbery_b (have 6.6.0 and recent 6.9s)
06:25 TimToady It's actually quite fun to watch rakudo following in pugs footsteps wrt testing and development
06:26 TimToady it's a concrete example of the good kind of cross-fertilization
06:26 TimToady and a sign that we can use our differences as a strength rather than a weakness
06:27 TimToady anyhoo, thanks for helping with fudge
06:29 TimToady zzz &
06:33 [particle] t\spec\S29-str\ucfirst.rakudo
06:33 [particle] Non-zero exit status: 1
06:33 [particle] lovely of you to create fudge in the first place. all i had to do was complain properly, and there it was!
06:35 TimToady you either want to drop the 2 or the block there
06:36 TimToady the block only counts as 1 skip
06:36 [particle] i'll fix that up
06:36 [particle] just good to know it's running the .rakudo files now :)
06:36 [particle] and giving exit(1)
06:37 TimToady probably still need a way to say the following should be multiplied by N for loops and such
06:38 TimToady maybe I should force the count on a block to be interpreted that way
06:38 TimToady but I still think it's an attribute of the block, not the skip command
06:38 [particle] as long as fudge handles the preprocessing, it's all good
06:38 TimToady since it's shared by all platforms
06:39 [particle] i suppose fudge could look ahead and generate the block count
06:39 TimToady maybe a #?all: scale 5
06:39 TimToady or some such
06:40 TimToady I'll think about it overnight.
06:40 [particle] i'm sure you'll come up with something. you always do.
06:40 TimToady meanwhile, it really doesn't matter so much if the count gets off
06:40 [particle] not for us yet, anyway. i just want to be able to get the files parsing properly
06:41 TimToady well, didn't get a nap today, so really need to thunk...
06:41 TimToady zzz really &
06:41 [particle] enjoy
06:44 jisom joined #perl6
06:55 pugs_svn r19591 | particle++ | [t/spec] fix some incorrect test markers for rakudo
07:05 [particle] TimToady: skip throws off the line count without a block:
07:05 [particle] #?rakudo: 2 skip "unicode"
07:05 [particle] skip(1,"unicode"); # { is(uc("ß"), "SS", "uc() of non-ascii chars may result in two chars");
07:05 [particle] #  }
07:06 [particle] that extra # } is trouble. probably don't even need # { and # } for single lines
07:06 [particle] bed &
07:08 alester joined #perl6
07:08 pugs_svn r19592 | putter++ | [kp6] Emit/Ruby.pm: a minor bit of transliterating
07:10 lisppaste3 joined #perl6
07:23 iblechbot joined #perl6
07:29 pugs_svn r19593 | particle++ | [t/spec] rakudo markers
07:47 BinGOs joined #perl6
07:59 Aankhen`` joined #perl6
08:16 kst joined #perl6
08:21 agentzh joined #perl6
08:29 meppuru joined #perl6
08:31 iblechbot joined #perl6
08:37 barney joined #perl6
09:09 devogon joined #perl6
10:13 ludan joined #perl6
10:47 zhuzi joined #perl6
11:00 nothingmuch joined #perl6
11:22 chris2 joined #perl6
11:50 ruoso joined #perl6
12:01 fglock joined #perl6
12:03 fglock @tell obra re test prioritization, this looks interesting  http://mapstext.unl.edu/public/prioritization/
12:03 lambdabot Consider it noted.
12:11 lichtkind joined #perl6
12:11 lichtkind i have problems witz the perl 6 wiki
12:41 fglock how about adding severity/criticality to at least some of the tests?
12:41 fglock left #perl6
12:49 integral hmm, this month's MSDN magazine has an article on writing compilers from .NET (the hard way with manually written lexers and parsers :-)
13:14 ebassi joined #perl6
13:48 njbartlett__ joined #perl6
14:04 alester joined #perl6
14:06 stevan_ joined #perl6
14:10 meppl http://search.cpan.org/~audreyt/
14:10 lambdabot Title: &#9787; &#21776;&#40179; &#9786; - search.cpan.org
14:10 meppl beautiful picture there
14:11 meppl lambdabot, nice try
14:33 stevan_ joined #perl6
14:41 jferrero joined #perl6
14:42 icwiener joined #perl6
14:44 stevan_ joined #perl6
14:47 stevan_ joined #perl6
14:55 kanru joined #perl6
15:00 soisoisoi joined #perl6
15:05 penk joined #perl6
15:06 wknight-away joined #perl6
15:13 alester joined #perl6
15:22 pugs_svn r19594 | ruoso++ | [smop] adding a new test before starting to implement the interpreter logic
15:25 ruoso fgl
15:25 ruoso oops
15:35 __Ace__ joined #perl6
15:36 stevan_ joined #perl6
15:37 stevan_ joined #perl6
15:40 NathanE joined #perl6
15:50 stevan_ joined #perl6
15:51 stevan_ joined #perl6
15:53 pugs_svn r19595 | particle++ | [t/spec] rakudo skip markers
15:57 ruoso I've realised that delaying the gc implementation to the oo system hidden behind the responder interface brought me an unexpected important feature... I can have immortal values that are not subject to gc at all..
15:57 ruoso the constant identifiers are some of them...
15:59 TimToady hmm, does that work for native types, I wonder...
16:02 ruoso TimToady, in theory, it can
16:02 ruoso because native types cannot have their representation changed
16:02 ruoso so we can just assume to know how they are
16:02 wolverian http://www.theonion.com/content/news​/earthquake_sets_japan_back_to_2147 quite OT, but funny, and I was led here via charles stross's blog, who has a section for perl on his site, so it's not quite completely OT, right..
16:03 lambdabot Title: Earthquake Sets Japan Back To 2147 | The Onion - America's Finest News Source, http://tinyurl.com/yt4egq
16:04 ashleyb joined #perl6
16:06 ruoso TimToady, you think having some constant ints pre-defined would have any impact?
16:06 ruoso like.. everytime you use a 1, you can use the constant 1 value?
16:07 TimToady well, you need to have a representation of the bits somewhere, and there's little point in duplicating if you don't have a machine code that inlines and forces indirection
16:07 TimToady instead
16:08 TimToady and as long as you keep your readonlyness straight, it works
16:09 * ruoso interesting the fact that when I was putting the tags for smop in ohloh and it listed squeak as a related project...
16:09 ruoso TimToady, ints are readonly by definition
16:09 ruoso there's simply no way of "legally" changing its values
16:10 TimToady well, it was easy enough to change them in Fortran
16:11 ruoso I mean, the pointer is there, the structure is known, but the API doesn't expose it...
16:11 ruoso I mean, not even the typedef is exposed to the public api
16:11 ruoso it's all design by contract
16:14 ruoso It's almost void* everywhere... it's SMOP__Object* everywhere...
16:18 ruoso hmmm...
16:19 ruoso ohloh really thinks smop is related to smalltalk... the two top related-by-tag projects to smop are squeak and smalltalk yx...
16:24 jhorwitz joined #perl6
16:42 netix joined #perl6
16:42 netix Hi, I know it must be a borring question but, do you know any date about perl 6 release ? On Internet I saw "alpha in the begin of 2007 and maybe a release at next christmas"
16:44 TimToady the Internet is a big place
16:45 icwiener :)
16:47 TimToady it's possible that there are universes in which Perl 6 is done already, and maybe the Internet leaks between universes.
16:47 netix TimToady> like ONE with jet lee ? :)
16:48 TimToady sorry, he's rocket lee in that universe
16:49 TimToady it's a strange quirk of english that jets aren't rockets, but rockets use jet propulsion
16:50 netix TimToady> anyways, any idea of a period for release ?
16:50 TimToady we'll release as soon as we get up to about 40,000 passing regression tests.
16:51 TimToady so you'd better get busy :)
16:52 netix lol... thanks for help
16:53 pugs_svn r19596 | ruoso++ | [smop] smop_interpreter.c. The instance that holds the current continuation. Some code done, but now I need to define the capture api
16:56 DarkWolf84 joined #perl6
16:56 ruoso TimToady, I know that the capture object is supposed to be used with prefix $, @ and % to get each value, but this seems a little too much overhead for the low-level...
16:57 ruoso do you think methods for getting explicitly the named and positional arguments should be part of the spec of the capture object?
16:58 TimToady to the first approximation, a capture is just a list of generators
16:58 TimToady some of which are marked as invocant or named, but to begin with they could just live in a list
16:59 TimToady and the methods could comb them out of the list on demand
16:59 stevan_ this might be of interested to you p6 folks, we just added rudementary support for attribute "traits" in Moose
16:59 stevan_ http://code2.0beta.co.uk/moose/svn/Moose/tr​unk/t/020_attributes/015_attribute_traits.t
16:59 lambdabot http://tinyurl.com/26s276
17:00 stevan_ basically it applys a set of roles to an anon-subclass of the attribute metaobject
17:00 stevan_ which should allow us to immitate some of the crazier p6 bits
17:02 ruoso TimToady, I'm not yet dealing with the high-level capture object... I'll use design-by-contract to implement a native capture type that is compatible with the high-level capture, but that can be implemented simpler than the high-level capture object
17:02 ruoso the only requirement is that both implements the same API
17:02 [particle] farewell &
17:05 ruoso TimToady, for instance, the low-level Capture won't require the values to be inside containers for it to fetch
17:05 ruoso as it will make avaiable a method for returning the actual value, instead of a container with the value
17:07 TimToady well, $ @ % & are essentially just methods too
17:07 ruoso yeah.. i know...
17:07 ruoso but they return containers
17:07 ruoso and that's an overhead I'd like to avoid
17:07 ruoso as I don't even have the containers yet in the low-level
17:08 TimToady I don't see why they have to return containers
17:08 ruoso I mean, I need a capture object to be able to initialize the Scalar prototype, for instance
17:08 TimToady they're just context
17:08 ruoso doesn't $ returns a Scalar?
17:08 TimToady no, it returns an item
17:09 TimToady and @ returns a list, not an Array
17:09 ruoso isn't item a kind of a simpler container?
17:10 ruoso like something I call FETCH?
17:10 TimToady sure, but 1.STORE fails
17:11 ruoso one way or another
17:11 ruoso it is a container...
17:11 TimToady only in the sense that everything is an object
17:11 ruoso exactly
17:11 TimToady and all objects are containers
17:11 ruoso hm?
17:12 ruoso int has a method FETCH?
17:12 ruoso that returns itself?
17:13 TimToady presumably FETCH would be a noop on a nonScalar item
17:13 ruoso ok... let me see if I got it straight...
17:13 TimToady you only need a Scalar when you have an lvalue, not an rvalue
17:14 TimToady $foo = 42 requires a Scalar on the left
17:14 TimToady if you say $foo := 42 then $foo isn't in fact a Scalar anymore
17:14 TimToady and $foo = 1 will fail after the binding
17:15 ruoso so Scalar is not really a class, it's more like a role
17:16 ruoso or even, an interface
17:16 ruoso an API actually
17:16 TimToady just like Array or Hash, but not needing a subscript
17:18 xinming joined #perl6
17:19 ruoso ok... I need to figure out how that impacts in the capture...
17:19 ruoso in the lowlevel, I have methods like
17:20 ruoso smop_new_capture_for_some_method(SMOP__Object invocant, SMOP__Object thefirstpositional, SMOP__Object thenamedfoo)
17:20 ruoso this would create a capture equivalent to
17:20 ruoso ($invocant: $thefirstpositional, :foo($thenamedfoo))
17:21 TimToady that's for creating an unbound call?
17:22 ruoso That's what will be put as the parameter to the SMOP_DISPATCH(interpreter, responder, capture) call
17:22 ruoso sorry
17:22 ruoso SMOP_DISPATCH(interpreter, responder, identifier, capture)
17:22 ruoso that's the deepest low-level thing
17:22 TimToady seems reasonable; the syntax is designed so that you can sort out the inf/pos/named at compile time
17:23 ruoso Ok... but which method identifiers should I use to fetch the values from the capture object?
17:23 ruoso $ would return the invocant
17:23 ruoso (I'm not doing any bindings in the low-level...
17:24 ruoso @ would return a list to which I could ask for an item
17:24 TimToady .item .list .hash are the corresponding context methods
17:24 ruoso ok...
17:24 TimToady (and .slice is a variant of .list)
17:25 ruoso the question is... wouldn't it be nice to have a .named(identifier) and a .positional(index) methods?
17:25 ruoso to avoid the eventual proxy object?
17:27 TimToady well, sure, low-level binding needs to do whatever will be efficient
17:27 ruoso the thing is that I won't even bind in the low-level
17:27 ruoso all parameters are rw
17:27 TimToady and even when binding to *@_ and *%_, those proxies are only the *remaining* args
17:28 TimToady so shouldn't be confused with @($capture) or %($capture)
17:29 araujo joined #perl6
17:30 TimToady you do probably need to be careful that you have a way of preventing eventual assignment to a formal parameter that isn't declared "rw", even if the actual argument is "rw"
17:30 TimToady but it's just something to be bear in mind
17:31 ruoso ok.. but I'm working in deep low-level here.. so in fact they're not really rw... I just don't copy them...
17:32 TimToady that's fine; is how all computers really work anyway...
17:32 TimToady well, most of them
17:33 TimToady there have been computers with fine-grained ro/rw attriubutes
17:33 ruoso but this problem will probably be solved when I implement the Signature
17:33 ruoso and then I can bind the capture to the signature and that's all
17:34 TimToady sounds like you're on the right track
17:35 TimToady certainly you want the mechanism to be fast when it can be proven that a formal parameter cannot be written to
17:35 TimToady no point in protecting it if there's no threat
17:35 TimToady if protecting it implies overhead
17:36 ruoso ok.. so for the lowest-level calls, I will use .invocant(), .positional(index) and .named(identifier) to fetch the items of the capture
17:38 ruoso and I'll have to use a trick to be able to not use a capture to fetch the item from a capture
17:39 ruoso which is to pass the value itself as the capture to those methods
17:39 ruoso instead of another capture
17:39 ruoso like...
17:40 ruoso .invocant(|1)
17:40 ruoso .positional(|1)
17:40 ruoso i mean
17:40 ruoso but in fact it's something mor
17:41 ruoso my $a = SMOP__CAPTURE__OPCAPTURE__postional_new($capture, 1); $capture.positional(|$a);
17:42 ruoso where $capture.positional would refuse any other object type
17:45 ruoso so it probably should be named as something else...
17:48 TimToady have you thought about optimizing multiple dispatch yet?
17:48 ruoso only that one way to do it is to check for known responder interfaces
17:52 ruoso TimToady, is there any reason for Capture and Signature not being native?
17:53 ruoso I mean... it makes things so harder
17:53 TimToady depends on what you mean by native
17:54 ruoso by not being possible to change its representation
17:54 TimToady certainly captures and signatures are at the top level just values
17:55 TimToady it's only by indirection that you get generators or containers
17:55 ruoso If I could count on knowing for sure the C structure of a capture (like I can with int)
17:55 ruoso it would be much easier
17:56 TimToady Signature and Capture are listed as immutable in S02
17:56 ruoso but this doesn't mean that any thing can say true to .does(Capture)
17:57 TimToady well, sometimes you have to box values
17:57 ruoso I mean.. I can consider to be illegal for anything else to say true on .does(int)
17:57 TimToady and most value types are punning with an identically named role
17:58 TimToady well, in a sense there is no int type
17:58 TimToady the type of int is Int
17:58 TimToady but with restrictions on storage
17:59 ruoso I mean...
17:59 ruoso I can count on every int to have *my* responder interface
17:59 ruoso but I can't count on every Int to have the same responder interface
18:00 ruoso I can have a perl5 and a parrot value claiming to be Int
18:00 TimToady yes, Int is a role that can be mixed into make other RIs
18:01 TimToady but we can restrict Capture/Signature, or call 'em capture/signature, if we need to
18:01 ruoso pleeeeeeeeeeeease
18:01 meppl joined #perl6
18:01 ruoso having the possibility of a foreign object to pass parameters is overkill, imho
18:02 ruoso i mean, using a sv* as a capture...
18:02 ruoso for instance
18:02 TimToady you don't have to convince me--I want dispatch to be fast
18:03 ruoso so, can I consider the capture and the signature to be "native" in the same sense as "int"?
18:04 TimToady the lowercase ones
18:04 TimToady we still might like a Capture role that can mix
18:04 ruoso Ok
18:04 TimToady for instance, into Match objects
18:04 TimToady but then you just have to unbox to capture
18:04 TimToady if you use it that way
18:05 ruoso seems perfect.
18:06 ruoso and I think that for smop, the capture in SMOP_DISPATCH will always be the lowercase one... the language must unbox it before passing it to the method...
18:06 ruoso this way I don't have to unbox everytime...
18:07 ruoso or at least check if it needs to be unboxed
18:07 wolverian ruoso++ # I don't understand any of this but it is fascinating
18:07 ruoso heh
18:07 ruoso I'm still in the beach trying to pull my boot straps to get out of the water...
18:07 TimToady maybe the original boxed Capture goes through the invocant if you don't have one.
18:07 TimToady maybe not
18:08 TimToady bbl &
18:10 ruoso TimToady, that's the reason for the .^native(type) method, that I've declared in SMOP...
18:11 ruoso it actually shouldn't be called ^native, because that seems to be a method for the metaclass...
18:12 ruoso maybe it can be delegated to the metaclass if the responder doesn't intercept it
18:12 ruoso but the expected is that the responder interface intercepts it
18:26 jisom joined #perl6
18:29 pugs_svn r19597 | ruoso++ | [smop] Now we have the native capture type. This simplifies a *LOT* everything, as I can count on knowing the C structure of the capture in every call. TimToady++
18:33 Khisanth joined #perl6
18:33 khisanth_ joined #perl6
18:34 wknight-away joined #perl6
18:43 wtgee joined #perl6
18:49 lichtkind TimToady: are your shure that the  reduction syntax [\ ] is optimal?
18:53 ruoso for the record: http://www.perlfoundation.org/pe​rl6/index.cgi?smop_boot_sequence documents what will happen in the smop_init() call.
18:53 lambdabot Title: SMOP Boot Sequence / Perl 6
18:56 TimToady lichtkind: optimal in which dimension or set of dimensions?
18:57 TimToady (it's primarily optmized for the visual dimensionh)
18:57 lichtkind TimToady: honestly the syntax is ugly
18:57 TimToady it's an unusual thing to want to do
18:57 lichtkind i know weird things should look weird
18:58 TimToady and the triangle part of it is kinda pretty
18:58 lichtkind triangle ?
18:59 TimToady yes, the left vertical and the backslash suggest a triangle
18:59 TimToady [\,] 1..10
18:59 TimToady 1
18:59 TimToady 1,2
18:59 TimToady 1,2,3
18:59 TimToady 1,2,3,4
18:59 TimToady 1,2,3,4,5
19:00 TimToady almost the same slope even
19:00 ruoso brb &
19:00 lichtkind TimToady: triangle numbers i see , i recently know them; since we boost the perl ratings in project euler a bit
19:01 lichtkind TimToady: its about mathematic problems
19:01 lichtkind TimToady: and during this tasks i many times wished to have perl 6 today
19:01 TimToady well, you and me both :)
19:02 lichtkind :)
19:05 lichtkind TimToady: so this won't change ?
19:06 TimToady not likely unless somebody comes up with something spectacularly better
19:09 meppl joined #perl6
19:09 lichtkind TimToady: im really tempted
19:16 gbacon joined #perl6
19:31 simcop2387 joined #perl6
19:31 drrho joined #perl6
19:40 rindolf joined #perl6
20:16 alester joined #perl6
20:24 jjore joined #perl6
20:35 gbacon joined #perl6
20:35 kst joined #perl6
20:45 wolverian hmm. building a nested set of sets of sets ... Set(Set(), Set(1, 2, Set(3, 4, ...))) is okay, I suppose, but I'm wondering if there's a less pythonic way.
20:46 wolverian this isn't very relevant, really, since I don't foresee myself doing this in perl6. using maxima for now.. :)
20:46 wolverian though I don't like maxima's lack of nice operators
20:47 wolverian Set.new_recursively(@@(...))
20:48 wolverian %%() almost makes sense here :p
20:48 kst joined #perl6
20:58 wolverian as a side note, you can't put a set object inside another set in python
20:59 wolverian hm, ah, you can't put _mutable_ sets inside other sets
21:00 wolverian that makes sense :)
21:11 xinming joined #perl6
21:18 meppl joined #perl6
22:09 Schwern joined #perl6
22:15 lodin joined #perl6
22:25 alester joined #perl6
22:36 macroron joined #perl6
22:43 Auzon joined #perl6
23:15 justatheory joined #perl6
23:32 meppl good night

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo