pugs.blogs.com | pugscode.org | pugs.kwiki.org | paste: sial.org/pbot/perl6 | <stevan> Moose... it's the new Camel ":P | .pmc == PPI source filters!
Set by Alias_ on 16 March 2006.
mugwump Moose is alive and well, it even has roles support 00:00
I'm developing a model for standard perl collections with it
Actually this needs generics/higher order types/parametric types to work *really* well
Those I plan to hack in via obscure package names and UNIVERSAL::isa hacks 00:01
00:01 Quell joined
mugwump (so that, for instance, HASH->isa("Coll of (Str, Any) where Unique(0)") 00:01
It makes sense to defer exact design decisions on what .meta looks like at least until it is properly metacircular, ie self-defining. Possibly even further than that. 00:03
Toaster (sorry, got distracted by toast)
right - is there an intention to somehow wrap Moose back into p6 once it's done?
or is it another prototyping-perl6-concepts-in-perl5 doofer? 00:04
mugwump With good availability of meta-objects, language transformation gets even easier.
For instance, you could easily take Class::MOP objects (the model behind Moose) and generate Perl 6 metamodel objects with them 00:05
then it's just a matter of converting your code fragments, if you care
Toaster right
sounds like i need to go read one of those vast books or thesises on metaobject models :) 00:06
Toaster heads librarywards
mugwump it's certainly a confusing area to research, just ask stevan_ :) 00:07
but the concept is quite simple. an object that describes the class. Change the Class object and (if possible) you mutate the class
the confusion arises in questions such as, "what does $object.meta.meta return?". There's also a chicken and egg problem. 00:08
but that's ok, I used to live with chickens. The answer is there are always eggs and chickens. 00:09
FurnaceBoy_ marvels 00:20
TimToady meta cops out 00:23
00:31 mako132_ joined 00:42 jabbot joined 00:57 Odin-LAP joined 01:13 stevan__ joined 01:14 stevan__ is now known as stevan
stevan Toaster: I suggest "The Art of the Meta Object Protocol" as a good place to start 01:15
mugwump: $object.meta returns an instance of Class which describes $object, and $object.meta.meta also returns an instance of Class, but this one describes the class Class
for instance with Moose, ... Moose::Object->meta->meta->isa('Class::MOP::Class')
and ... Moose::Object->meta->meta->name eq 'Class::MOP::Class' 01:16
mugwump This knot tying is not necessarily forced, though - on a particular implementation you might hit the bottom 01:18
stevan mugwump: yes
knot tying is just one way to do it
either that or you just decide how deep the turtles go,. and stop there
knot tying tends to not work as well in more static languages
but more dynamic languages (Perl, Haskell, etc) tend to handle it just fune 01:22
01:35 jserv-- joined 01:39 revdiablo joined
clkao 4/win 22 02:09
02:10 pjcj joined 02:31 mako132_ joined 02:37 KingDillyDilly joined
KingDillyDilly PerlJam wrote "Soon I will echo Merlyn's lament that Perl 6 may be too bizarre to explain." I searched the #perl6 logs a bit, and I searched the web a bit, and couldn't find such a quote. Any tips on where to look? 02:39
(for the cons section of a pros and cons section that I might add to Wikipedia) 02:40
audreyt KingDillyDilly: maybe www.nntp.perl.org/group/perl.perl6....uage/25160 02:43
KingDillyDilly That would require some research on my part since I don't know if those two terms Merlyn's complaining about will both be used in Perl 6. 02:46
azuroth this red black trees in haskell... it's making my brain explode 02:47
KingDillyDilly I remember you praising Merlyn for "environmental" a few days ago. It reminded me of when I searched Google for "environmental variables" because that's the term I learned from Learning Perl (unless it was my Javascript book) and I couldn't find what I wanted...
Much later on I discovered that "environment variable" is the more common term. 02:48
I'm with Merlyn on this one. 02:49
azuroth google is too smart; search for -ental and it includes -ent results
KingDillyDilly limits himself to two lines in chat since discovering he was truncated in the log 02:58
Maybe I'll forget about the bizarre remark 03:01
audreyt azuroth: www.cs.kent.ac.uk/people/staff/smk/...ck/rb.html ? 03:03
I've just submitted the title of my talk in November's YAPC::SouthAmerica 2006 03:10
"Golfing Perl 6"
I have no idea what I'll be talking about, but it sounds better than "Deploying Perl 6" imho :) 03:11
03:14 penk left
azuroth audreyt: yeah, the second one 03:15
audreyt Okasaki's book can help 03:17
KingDillyDilly I'd have to check the dictionary to make sure what's meant by "deploying" Perl 6. Releasing the language? Switching to it to program with it? 03:18
03:19 particle_ joined, neoesque joined
audreyt it means distribute, install, and maintaining Perl 6 programs in a place with programs already written in other languages, specifically Perl 5. 03:19
azuroth 'purely functional data structures'?
audreyt azuroth: yup.
azuroth cool. I'll have a look for it
audreyt KingDillyDilly: it mostly means "what I've learned from putting some perl 6 code in $job" 03:20
TimToady I wonder what I should talk about...
azuroth talk about me...
TimToady I'm tired of talking about me... 03:21
KingDillyDilly www.m-w.com/dictionary/deploy "to spread out, utilize, or arrange..."
azuroth which is why there's me instead!
audreyt KingDillyDilly: yup, that's the meaning
or, more specifically 03:22
en.wikipedia.org/wiki/Software_deployment
KingDillyDilly Could mean developing the language, or could mean learning Perl 6. If I were a programmer rather than curious about development of the language (which I am) I might not go to your talk if I thought it was about the former.
audreyt which is why I hope people will read the abstract instead of just the title :D 03:23
(and there is already merlyn+bdfoy's "Learning Perl 6" in the same conference)
KingDillyDilly Do you use those artsy photos of yourself in promos, or something more business-like? 03:25
audreyt TimToady: I enjoy very much the natural-language parts in your talk
in particular on translation
KingDillyDilly I enjoy the screen saver pictures.
audreyt yup, me too 03:26
KingDillyDilly: I don't have "business-like" photo at all
FurnaceBoy_ deploying != learning
KingDillyDilly Ok, then "utilizing"
FurnaceBoy_ also, "environment variable" is very well established and specific term, cannot be changed (dates back 35 years)
but apart from that, "hi." 03:27
KingDillyDilly I agree. It's more common than "environmental variable" and I wish it were the only term.
FurnaceBoy_ env-al seems to be confusing.
but when you said, "I'm with merlyn," I thought you were promoting env'al 03:28
I'm sure the confusion was accidental
KingDillyDilly I'm with Merlyn because I agree that "If they both have "env"-ish stuff in the name, let's use the same name for both. If they are far enough apart that they should be distinct, call them something different."
FurnaceBoy_ yes 03:29
audreyt considering how (conceptually)it's passed thru the call stack 03:30
maybe we can call them "envoy variables"
FurnaceBoy_ has a case of variable envy
audreyt FurnaceBoy++
spinclad 'deploying' is more 'putting into place so it's usable' than 'utilizing' or 'making use of' directly, in my experience 03:31
audreyt envy $me of Gucci;
03:31 justatheory joined
spinclad $audreyt is $Container of Gucci ? i do envy! 03:32
a Case of Gucci, even
audreyt actually no... mostly Dior nowadays.
though the :shape of container of envy_me is really pretty. 03:33
FurnaceBoy_ deploy -> "use in anger" ? :)
audreyt FurnaceBoy_: where did you get that definition? :) 03:34
FurnaceBoy_ /me
spinclad deploy -> 'set out for use', in anger or otherwise? 03:35
FurnaceBoy_ pretty much
although,
people do associate it with "real paying work"
if you get my drift. 03:36
"put in front of customers"
battle-ready
spinclad well, sure, i've deployed lots of stuff and gotten $pay ed
FurnaceBoy_ 's metaphors are overheated today
I;'m sure context will make the right word obvious :) 03:37
spinclad it does sound a bit like the opposite of 'employ', though
FurnaceBoy_ actually, it's a synonym in some senses. ;) 03:38
03:39 aufrank joined
FurnaceBoy_ there is at least one sense of deploy that means exactly employ 03:39
spinclad 'deploy for employment'
FurnaceBoy_ for continued employment, deploy Perl6
To enjoy your employment, deploy
audreyt it's all just a ploy... 03:40
FurnaceBoy_ feels a limerick coming on
"...it's not just a toy!"
To enlarge your endowment, deploy (Perl6 spam)
azuroth sounds a bit like a KOMPRESSOR song to me 03:41
aufrank hello all!
FurnaceBoy_ well it needs work.
FurnaceBoy_ phones session musicians
aufrank audreyt: how's the AIX gcc porting going?
FurnaceBoy_ aufrank: ola 03:42
audreyt aufrank: it works
spinclad .oO{ no, no, gcc already runs on AIX }
audreyt right, it's GHC
FurnaceBoy_ rewrote to GHC
audreyt aufrank: stage1 was completed yesterday, so I was able to compile openagp; I left stage2 running overnight
it's already past the typechecker when I left office at 8pm last night
so it should be completed by now 03:43
s/openagp/openafp/
aufrank :)
audreyt it's interesting that Dior is also the father-in-law of Earendil the mariner 03:44
aufrank silmarillion++
spinclad but that one wasn't a Christian 03:45
audreyt no, I don't know any Half-elven Christians
03:46 FurnaceBoy_ is now known as FurnaceBoy 03:47 pjcj joined
audreyt (though there is an elven Celebrian) 03:48
spinclad and the fields of Celebrant (or i misremember) 03:49
audreyt that's correct; the field is named after the Celebrant river 03:51
audreyt blames TimToady for inflicting this strange Tolkien addiction 03:52
TimToady blames Tolkien... 03:54
spinclad blames Ilúvatar 03:55
Kattana demands support for tengwar charsets 03:56
audreyt hands Kattana ConScript fonts 03:58
U+E001 .. U+E07F iirc 03:59
er, sorry, U+E000.
Kattana lol, I should have known. 04:01
04:03 lisppaste3 joined
FurnaceBoy blames the Kalevala 04:03
KingDillyDilly Since there's a Wikipedia page for Perl 6, maybe I should create one for Python 3000. 04:12
(. o O) 04:13
04:14 stevan_ joined
FurnaceBoy at least due date for that is contained in its name :) 04:15
unlike PerlChristmas
err, deployment date?
FurnaceBoy joins the undeployment queue
KingDillyDilly It seems to be on approximately the same track as Perl 6. 04:16
FurnaceBoy I guess you could say that the long tail of Sourceforge projects are "undeployed".
that's pretty damning for a program.
Like an unseeded torrent
a soul in limbo, disembodied 04:17
KingDillyDilly thinks the same in #python, gets a shrug from Jerub, thinks it's Juerd, not sure. 04:20
FurnaceBoy don;t wait for approval, just do it if ya wanna 04:21
spinclad pretty sure Jerub =!= Juerd
KingDillyDilly Yeah, but all I know is there's an enumerate function. I think it has something to do with numbers.
FurnaceBoy that's all you know about Python 3000 ? 04:22
there are a bunch of PEPs. I came across a trove of info recently, can't be hard to find.
KingDillyDilly Actually, that may just be Python.
I'd just do a bunch of copy and pasting with some rewording. Like my High School book reports. 04:23
I did a good job on Wikipedia's en.wikipedia.org/wiki/Comparison_of..._languages with that method. 04:25
azuroth C++ isn't _that_ weak, is it? 04:27
KingDillyDilly 3 is it's rank. See the footnote.
1 is most popular.
azuroth I mean, in type safety 04:28
FurnaceBoy it's rather strict, as they go...
mathematica ... hmmm!
KingDillyDilly Don't know. That was carried over from en.wikipedia.org/w/index.php?title=...d=42439602 04:29
FurnaceBoy influences column looks fairly right
KingDillyDilly I didn't create the entire chart.
FurnaceBoy I guess it's been WikiWatched for a While 04:30
azuroth would I be wrong to change C++ from "static, weak" to "static, strong"?
FurnaceBoy wouldn't object
FurnaceBoy didn't really intend the pun
audreyt C++ allows explicit cast, no?
mugwump doesn't Java, too ? 04:31
spinclad yes, that weakens its rating
[C++]
audreyt but Java throws exception, while C++ silently permits it, iirc
(on incompatible types)
FurnaceBoy true
azuroth hmm, I guess so. but it's explicit, and it's not like 'int a = "foo"' works...?
FurnaceBoy that might explain the rating
azuroth "static, weak (but far stronger than C)"? 04:32
FurnaceBoy LOL. 04:33
azuroth ;-p
mugwump Someone here says that (SomeClass)blahObject works in Java
FurnaceBoy "and especially much much stronger, hard to overstate really, than K&R C"
"...not to mention BCPL"
spinclad the world an array of int 04:34
KingDillyDilly cgibin.erols.com/ziring/cgi-bin/cep...ey=C%2B%2B "The more recent C++ standards do support safe casts, but this feature is not yet universally available or employed."
audreyt mugwump: but that only works if it is really of that class or has been special-cased to handle that; it's an exception otherwise
KingDillyDilly That page was last updated in 2000.
azuroth what, dynamic_cast?
FurnaceBoy well, dynamic_cast works around the underlying weakness of C++ 04:35
mugwump right, and presumably you can't resume the exception
or can you?
04:36 jabbot joined
mugwump I suppose not, if you get caught by a caller, but possible if you get pushed to an exception handler and can return an appropriate object 04:38
KingDillyDilly notices the word "employed" in his quote 04:44
azuroth it's not exactly a common thing you need to do, I don't think.. 04:46
PerlJam it certainly is a different crowd here now than when I'm usually paying attention 04:54
azuroth what, you expect us to talk about perl? 04:57
pft
FurnaceBoy some days it's the Perl, some days it's the swine. 05:11
PerlJam azuroth: this being #perl6 I expect people to talk about almost anything 05:13
FurnaceBoy we're pathologically eclectic
KingDillyDilly PerlJam: has Merlyn really said that Perl 6 may be too bizarre to explain? 05:14
05:16 kanru joined
PerlJam KingDillyDilly: not in those words, no. He has expressed some worry over how to explain perl6 to someone though. 05:16
KingDillyDilly I'm wearing my journalist hat. I need trash talk.
PerlJam KingDillyDilly: you need to get some context on that though. Read the lines before I said anything about merlyn 05:17
KingDillyDilly: "journalist"? Where will you journal?
KingDillyDilly Wikipedia, in Perl
's con section.
(Perl 6's yet to be created con section) 05:18
FurnaceBoy well at least it's editable :)
unlike yesterday's NYT
PerlJam no more banging on the monk's door? :)
KingDillyDilly Wikipedia has a larger door and gets more visitors. 05:19
PerlJam that's for sure.
FurnaceBoy mentions runs on MySQL :) 05:22
05:25 avar joined
KingDillyDilly I still have some quotes about Perl 6 trying to get emocons to compile though. 05:26
I'm scraping for trash because I figure praise will be a lot easier to find when I get to the pros part.
FurnaceBoy the problem has been mentioned before, that the trash (Fish's article) is meaningless so far, because there's nothing to critique yet. 05:28
wait for 1.0 and the bad reviews on opening night.
KingDillyDilly (emoticons)
FurnaceBoy well people said that about Perl 5, 4, 3...
Perl 6 just does it more properly like. 05:29
KingDillyDilly Unfortunately, I'm not inclined to give Perl 6 the benefit of the doubt, but I'll still be reasonably fair...considering 05:30
05:34 jabbot joined
KingDillyDilly I use those "emoticons" maybe more heavily than most because of my regexes. It's the non-regex emoticons that make me feel lost, but maybe that's just because I've been reading posts on a this channel instead of tutorials. 05:35
FurnaceBoy this channel will not clarify much :)
to an end user 05:36
KingDillyDilly I hate that show French in Action too. Learning by observation as opposed to instruction is innefficient. 05:38
FurnaceBoy I believe in immersion. 05:39
applies to computer languages too.
you only get good at something you do every day, that's what I believe.
PerlJam If the end user asks pertinent questions that are specific, the channel could clarify some things 05:40
FurnaceBoy and after 5 years of doing something every day, you can consider yourself semi-proficient.
PerlJam FurnaceBoy: suddenly this sounds like the intro to a porno or something.
FurnaceBoy PerlJam, it could. I was thinking of 'passive listening' to the channel being unenlightening for tutorial purposes. ;) 05:41
KingDillyDilly FurnaceBoy: It doesn't have to be that way if the language isn't among the most difficult and if the instructions are top notch.
05:41 neoesque joined
FurnaceBoy true, you can 'get started' quicker, but Perl6 may not be the best 'quick start' language. or maybe it will be. hard to say. 05:41
PerlJam KingDillyDilly: you consider perl6 to be very difficult?
FurnaceBoy I just had to listen to the caterwauls of an MS-ite installing Gentoo.
KingDillyDilly PerlJam: I don't really know. I just get the impression that it will be even more difficult to learn than Perl, which is supposedly harder than average. 05:42
audreyt can certainly attest that English is much harder to learn than Perl. 05:43
(I started learning the two at around the same time) 05:44
aufrank audreyt: and that took you about 22 days, right? 05:45
audreyt and I still get plurals all mixed up. I much prefer the prefix "@-" sigil over the suffix "-s" sigil
aufrank =P
audreyt aufrank: nope ;)
PerlJam audreyt: such is the problem with organic growth. You get all sorts of odd protuberances.
audreyt but english is also useful :)
FurnaceBoy English must be learned by rote.
(mostly) 05:46
for native english speakers, they only improve by immersion themselves.
constant input
TV doesn't cut it :)
but TimToady's the linguist... 05:47
I'm just an end user...
KingDillyDilly Has anyone who's writing Perl 6 documentation ever taken a technical writing course? 05:48
PerlJam KingDillyDilly: why would a linguist care about technical writing? ;-)
audreyt asavige was quite versed in technical writing 05:49
and merlyn is also reputable for it :)
audreyt looks forward to the YAPC::NA "Learning Perl 6" course
KingDillyDilly critiqued Learning Perl on Perlmonks 05:50
audreyt sure, but well, I still appreciate merlyn's technical writing 05:51
PerlJam criticizing is easy to do. What's hard is creating something.
KingDillyDilly As Wassercrats, not the anonymonk who acts like me.
audreyt (as both translator and instructor)
KingDillyDilly A better index is easy. Maybe I'm unique in how I learn though. 05:52
FurnaceBoy PerlJam++
KingDillyDilly One day I'm going to take a printout of my critique to a bookstore to see if the new edition of Learning Perl was improved per my suggestions. 05:54
If so, I'll write to O'Reilly to ask for a free copy since I can't buy one due to my boycott. 05:56
FurnaceBoy you could put that time into your own creation...
KingDillyDilly I use my skill in explaining in other things. 05:58
In my criticism, for example. 05:59
06:00 elmex joined 06:04 codebreaker_ joined
KingDillyDilly Copycat: www.geocities.com/boycott_o_reilly/ 06:05
PerlJam heh, somehow I think that boycott is of a different nature 06:06
Do you not purchase parts from O'
O'Reilly auto parts?
KingDillyDilly I'd try to avoid it, but I haven't had occasion to since my boycott began. 06:07
Yes, I'm boycotting the auto parts place too. "Your browser does not have JavaScript enabled. Most of our site will not function correctly" 06:09
I hate that.
arcady hah! 06:21
where I go to school, if I use firefox, the webapp used for signing up for classes just throws an exception and dies
FurnaceBoy ouch. 06:24
hit them with a cluestick.
arcady or just rewrite the whole damn thing in a more sane way 06:26
I'd've used perl 6, if it were remotely ready for that kind of use
06:48 r0nny joined 06:50 iblechbot joined 06:54 KingDillyDilly left, drrho joined 07:33 marmic joined 07:44 Aankhen`` joined 07:53 Ymmv joined 08:15 kane_ joined
svnbot6 r10222 | scw++ | Pugs::Grammar::MiniPerl6 08:19
r10222 | scw++ | * Combination of function application and anding parsed.
08:35 jserv-- joined 08:42 q[uri] joined 09:47 xinming joined 09:53 Odin-LAP joined 10:10 Aragone is now known as Arathorn 10:13 lichtkind joined 10:14 nnunley joined 10:31 Daveman joined 10:38 codebreaker_ joined 10:52 xinming joined
xinming what the ip for perlcabal.org is please? DNS server seems to be down... 10:53
Arathorn 194.145.200.126 is feather - isn't perlcabal hosted on the same box? 10:54
xinming Arathorn: thanks, works now... 10:56
11:03 neoesque joined
Arathorn np 11:03
11:03 chris2 joined, TimToady joined 11:04 nwc10 joined 11:10 TimToady joined 11:24 Maddingue joined 11:27 Grrrr joined
azuroth if a StockItem is actually a stock type, what should we call {x items of a specific stock type} 11:28
11:36 mako132_ joined 11:57 stclare left 11:58 mako132_ joined 12:04 jsiracusa joined 12:06 iblechbot joined 12:09 Limbic_Region joined 12:14 flounder99 joined 12:23 stclare joined 12:25 nwc10 left 12:33 xerox joined
lichtkind theorbtwo please get tsunami one remix of little fluffy clouds 12:44
12:44 neoesque joined
theorbtwo lichtkind: Huh? 12:49
lichtkind theorbtwo The Orb - Little Fluffy Clouds (Adam Freeland Tsunami One Remix).mp3 12:50
theorbtwo i love it
theorbtwo Ah. 12:51
I'm not named after the band.
lichtkind i thought you said that
theorbtwo I'd never heard of the band at the time I started using the name "The Orb" -- indeed, it would have been around the same time they were named. 12:52
azuroth I never even thought to read it as three separate words before. it reminded of theodore somehow.
theorbtwo I get that surprisingly much. 12:53
12:56 merlyn joined
Limbic_Region oh - FYI, rafael has added "state" variables to bleed perl. He is also using a p5 varation of examples/cashiers.p6 as a test case - go Pugs 12:56
rgs yes. 12:57
and also, I'm her e:)
theorbtwo Wow!
rgs Limbic_Region++ # sends patch ideas
theorbtwo wonders when 5.10.0 will be out.
rgs too.
Limbic_Region oh, cool - except I think the example could use a bit of refining 12:58
Limbic_Region is testing now
theorbtwo I suppose asking more questions along that line would be both OT and annoying.
rgs refining patches welcome
theorbtwo: /me wonders when pugs 2*pi will be out. :) 12:59
Limbic_Region rgs - I guess it is just a matter of taste. To be closer to the p6 example, $amount belongs inside each of the anonymous subs not the pseudo-object constructor 13:00
but it works the same
so rgs++
rgs Limbic_Region: in Perl 5, if $amount is not declared outside, those are not closures, so the point of the test is a bit missing 13:01
Arathorn state is for lexical variables which last for the life of the closure, right?
Limbic_Region rgs - they don't need to be closed over 13:02
let me provide an example
rgs right
but the test is testier that way ! :)
(meaning, less redundant with previous tests)
pasteling "Limbic_Region" at 129.33.119.12 pasted "For rgs - a closer bleed perl example of examples/cashiers.p6" (20 lines, 409B) at sial.org/pbot/17037 13:06
rgs Limbic_Region: yes, that was my 1st implementation in perl 5 of the cashier example
(and it works :)
Limbic_Region anyway - just good to see that Pugs is in a position to be blazing the way for p5 13:08
theorbtwo Roughly how many posts/day does p5p get?
rgs there are lots of patches lately 13:09
probably 30 posts a day or so 13:10
theorbtwo Hm, that's not bad. 13:12
13:14 jiing__ joined 13:15 jiing___ joined, weinig joined
theorbtwo subscribes to perl5-porters, unsubscribes from qemu-devel 13:15
rgs woot 13:16
clkao audreyt: are you coming to yapc::eu? 13:18
13:18 jiing_ joined
theorbtwo Hm, is there anything stopping us from allowing %foo {'bar'}, other then paren-less if? 13:19
(Assuming we allow the parser look-ahead to the next non-whitespace char. 13:21
Limbic_Region isn't that what abigail was bitching about a couple of years ago 13:23
that if you wanted to do that - you need the . to let the parser know what you were doing? 13:24
rgs yes. abigail was also bitching about print ("foo") emitting warnings in perl 5
Limbic_Region well, he can't be right all the time now can he
;-)
Arathorn what is the ambiguity in 'print ("foo")', ooi? 13:26
(and why doesn't that warning also show for 'print ("foo")'?
azuroth Arathorn: what's the difference? 13:27
rgs print ($x + $y) + $z
theorbtwo Which, as abigail pointed out, is an addition in void context. 13:28
I do that all the time, but only because my fingers are programmed to type perl -le instead of perl -wle
rgs but the keystroke you save on avoiding w is wasted on a good whitespace ! 13:29
Arathorn azuroth: oops - this client munges whitespace - the second example had a doublespace between print and ( 13:30
azuroth ahh
that makes me hurt... 13:31
theorbtwo Arathorn: Because the code that creates the warning is a bit on the buggy side. It should probably be fixed or removed entirely, but apparently nobody has done either yet, possibly because of not being able to decide which should happen.
OTOH, in perl6 they mean completely different things.
Arathorn fair enough. 13:32
13:35 KingDiamond joined 13:36 shachaf joined 13:37 manbou joined 13:39 kolibrie joined 13:50 cdpruden joined 13:56 justatheory joined 13:59 eric256 joined 14:30 aufrank joined 14:44 FurnaceBoy joined 14:53 clkao joined 15:24 vel joined 15:56 jserv-- joined 16:35 amnesiac joined
svnbot6 r10223 | scw++ | Pugs::Grammar::MiniPerl6 16:37
r10223 | scw++ | * Function application with parameters supported
16:52 wilx joined 16:54 particle_ joined, Aankhen`` joined
svnbot6 r10224 | fglock++ | RuleInline-more.pl - additional rule/token primitives, to be merged into RuleInline.pl 16:58
17:03 fglock joined 17:06 aufrank joined 17:20 dduncan joined 17:24 miyagawa joined 17:27 FurnaceBoy is now known as PM 17:35 PM is now known as FurnaceBoy, ghenry joined 17:39 FurnaceBoy is now known as FB|afk
aufrank ?seen audreyt 17:40
17:57 DaGo joined 18:02 ruoso joined 18:07 justatheory joined 18:19 xinming joined 18:20 iblechbot joined
fglock SoC suggestions answered - I hope I'm not over-promising 18:33
18:36 Aankh|Clone joined
aufrank fglock: what are you proposing? 18:36
fglock basically, a rewrite of Perl6-Value/Perl6-Container for integration with Pugs::Compiler::Rule and the other Perl6-on-Perl5 tools 18:39
aufrank neat!
when you say 'suggestions answered' does that mean you proposed something in public somewhere? 18:40
fglock there is a place for people to make contributions in the SoC page - I think only mentors can access that 18:41
aufrank fglock: did you email with your mentor before you submitted the proposal, or was the proposal the first step? 18:48
18:52 shachaf joined 18:54 FB|afk is now known as FurnaceBoy
fglock aufrank: I discussed it here, and then I mailed some people for suggestions - but I don't know who is the actual mentor yet 18:54
aufrank I see
18:55 masak joined
fglock later & 18:58
18:58 fglock left
aufrank bye! 18:58
19:12 Aankh|Clone is now known as Aankhen`` 19:21 hugues23 joined 19:48 walter joined 19:54 larsen joined 19:58 Steve_p joined
Steve_p seen TheDamian? 19:58
jabbot Steve_p: I havn't seen TheDamian, Steve_p
Steve_p Hmmmm 19:59
20:09 DaGo joined 20:11 froh-doh joined 20:23 bernhard joined 20:41 DaGo joined 20:42 geoffb joined 20:46 f_ounder99 joined 21:12 justatheory joined 21:14 frederico joined 21:30 kanru joined 21:40 marcus_ joined 21:51 GameZMember337 joined 21:59 clkao joined 22:14 awwaiid joined
mugwump TimToady, audreyt : S06 typo: 'subset' under Polymorphic types should be 'subtype', no? 22:22
TimToady no, we renamed it "subset" because people think subtype means derived type, and subsets sets are designed to violate Liskov subst. 22:26
22:29 rollie joined, rollie left
mugwump ok. Hey, is there syntax for higher order/parametric types? I remember something about 'is shape', I'm just searching the synopses for it now 22:30
I see S09 using it for multidimensional arrays, which is a bit different 22:31
TimToady parametric types are normally declared as roles--see S12. 22:32
mugwump ah, I see
TimToady shaped arrays are a particular sort of parametric type.
mugwump yes. and multidimensional ones are actually nested parametric types, I think. With a nice syntax of course 22:33
ie, my int @foo[^3;^3;^3] is something like Array of (Array of (Array of int where { .items == 3}) where { .items == 3 }) where { .items == 3 } 22:34
theorbtwo is certianly glad for the nice syntax. 22:35
mugwump so, is Array defined something like;
TimToady well, it's more like "this dimension's type is 0..2
mugwump role Array[::X] { ... } 22:36
?
and Mapping: role Mapping[::X, ::Y] { ... }
Can I make it a list? 22:39
TimToady cat you make what a list?
*can
mugwump role Relation[ ::X, ::Y, ... ] { ... } 22:40
22:40 aufrank joined
TimToady presumably one could use a slurpy of some sort. 22:40
it's just a signature, as far as I know. 22:41
mugwump ooo that's interesting indeed
TimToady at worst you say role Relation[ \$Capture ] {...} and decode the capture by hand... 22:42
mugwump But could I write : role Tuple[ *::T ] { has @.items of T } 22:43
?
TimToady I don't know. that looks more like a slurpy scalar than a slurpy array. 22:44
Type names are first class things, so you can collect them in an ordinary *@types array. 22:45
mugwump cool 22:46
TimToady but how you'd then use that to declare @.items I'm not sure.
arrays are generally considered to be of homogenous type. 22:47
Seems like the Tuple[] declaration is just trying to dup a class declaration with separate attributes... 22:49
theorbtwo TimToady: Why not just has @.items of any(@types)?
TimToady I don't think that's the customary meaning of Tuple... 22:50
mugwump no, certainly not what I'm after ... :)
theorbtwo I was just looking at the small problem.
TimToady classes and roles look like declarations, but they're really just calling metaclass methods underneath when you say "has" and such. 22:52
mugwump ssh don't say that too loudly TimToady ;)
TimToady Perl will remain a dynamic language... :) 22:53
the point, though, is that with customary forms of declaration, customary forms of analysis can be brought to bear.
BEGIN is just an inside-out eval. 22:54
both of them are last-ditch escape valves.
use of either of them usually indicates a missing language feature.
mugwump nods
TimToady and usually the language feature can be added with much better control. 22:55
symbolic links instead of eval...
use instead of BEGIN
mugwump parsetree macros instead of source filters... ;)
TimToady and the "official" language can encompass those dynamic constructs that
can keep track better, like macros, yes. 22:56
so we can keep the ideal that Perl is really a suite of languages, not a single language.
But hopefully still enable tools to work consistently.
mugwump Perl not single language, it rules! 22:58
TimToady well, we still have to prove it can be done...
23:13 SamB joined 23:45 buu joined
azuroth gee, it's surprisingly early. 23:46
meppl gute nacht

Powered by Cro and the Raku® Programming Language.

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

Thank you!