Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2006-04-16

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
01:05 cmarcelo joined perl6
01:08 Limbic_Region joined perl6
01:18 wolverian heh, a friend is none too happy that perl still isn't ditching the old dual type system in favour of return type mmd :)
01:24 wolverian he hates implicit flattening, because he thinks it makes people too lazy to really think about their data structures
01:25 frederico joined perl6
01:36 enantiodrome joined perl6
02:11 PolettiX joined perl6
02:13 PolettiX joined perl6
02:16 wolverian > 2/5
02:16 wolverian er
02:16 wolverian ?eval 2/5
02:16 evalbot_9946 2/5
02:16 wolverian ?eval 2/4 + 2/3
02:16 evalbot_9946 7/6
02:24 PolettiX joined perl6
02:26 PolettiX is now known as froh-doh
02:49 frodo72 joined perl6
02:49 r0nny joined perl6
02:49 frodo72 is now known as PolettiX
02:52 PolettiX is now known as frodo72
02:53 mako132_ joined perl6
03:08 svnbot6 r9962 | geoffb++ | Added a new question directly to Capture.pod as per Audrey's request
03:09 geoffb That was surreal.  I committed, turned around to do something else, and my computer was beeping at me with a new message addressed to me.  Of course, it was just the svn bot ....
04:31 coumbes joined perl6
04:38 PerlJam geoffb: I think the example is flawed.
04:38 geoffb PerlJam, fair enough.
04:41 PerlJam What's the thing called that Captures map to in a subroutine call?  Is there a "Signature" object?
04:42 PerlJam Captures are about arguments, what deals with parameters?
04:50 cmarcelo left perl6
04:53 gaal PerlJam: yes, it's called precisely that.
04:54 pmichaud hello, pj
04:55 PerlJam hey pm.  I see you've been busy :)
04:57 pmichaud yes,  unfortunately I took a detour today and refactored the operator precedence parser a bit
04:57 pmichaud (perhaps a bit much)
04:57 pmichaud otoh, the opp spec is much easier to read than the PIR code I used to have
05:00 PerlJam OPTable.pir looks *nothing* like I remember it  :)
05:03 PerlJam Then again, most things look nothing like I remember them. (in parrot and in pugs)
05:03 PerlJam Things change quickly
05:07 gaal geoffb: I think you're on to something :)
05:07 gaal (r9962)
05:09 PerlJam gaal: Just that the example is wrong.
05:11 gaal there may be something missing in the API, since you can't introspect optional bindings from without the function that declares them.
05:13 gaal iow: if you write very introspective code that takes functions and Captures, you have to query both if you want to know which variables will be in scope inside the function.
05:14 gaal which probably even makes sense, but could be a gotcha.
05:14 gaal I don't know if the association between Captures and Signatures is fully fleshed out yet.
05:14 pmichaud perljam: the biggest place to see the change is in languages/perl6/lib/grammar_optok.pge (which defines the tokens for the operator precedence parser)
05:15 pmichaud perljam: (svn up to get r12269)
05:16 PerlJam gaal: It seems to me that Signatures are templates, and Captures fill them and that's about all there is to it.
05:17 PerlJam gaal: though I'm not sure if there shouldn't be an object that represents a "filled in" Signature (i.e. the combination of Signature and Capture)
05:17 PerlJam (and boy Capture sure seems like a weird name!)
05:17 pmichaud perljam:  so far nobody has come up with a better one, I fear
05:20 gaal PerlJam: @_ from within a function :)
05:20 gaal I don't think it necessarily deserves a class of its own.
05:20 gaal because it behaves like a Capture
05:22 gaal $work &
05:22 svnbot6 r9963 | gaal++ |  r10007@sike:  roo | 2006-04-16 08:21:25 +0300
05:22 svnbot6 r9963 | gaal++ |  * FAQ::Capture - trivial cosmetic reflow to keep long lines
05:22 svnbot6 r9963 | gaal++ |    from being eaten on the web rendition.
06:11 ardder joined perl6
06:17 svnbot6 r9964 | gaal++ |  r10009@sike:  roo | 2006-04-16 09:13:42 +0300
06:17 svnbot6 r9964 | gaal++ |  * FAQ::Capture - method delegation. hw !! 3
06:57 nnunley joined perl6
07:02 Quell joined perl6
07:06 gaal audreyt: ping
07:12 q[uri] joined perl6
07:13 Aankhen`` joined perl6
07:21 elmex joined perl6
07:31 _mjk_ joined perl6
07:38 drrho joined perl6
07:42 ghenry joined perl6
07:53 cognominal joined perl6
08:04 cognominal audreyt++ # your blog is invaluable
08:23 manbou joined perl6
08:35 larsen joined perl6
08:39 xinming So $x = @y does not mean $x = \@y anymore. What does it mean then?
08:40 xinming anyone would explain this for me? I don't understand what It said in audreyt's blog... :-/
08:47 gaal xinming: reading it together with the question preceding it may help
09:03 iblechbot joined perl6
09:21 _bernhard joined perl6
09:50 meppl guten morgen
09:57 chris2 joined perl6
10:34 larsen joined perl6
10:51 siosiosios joined perl6
10:54 * ruz coudln't build pugs :( http://paste.husk.org/5518
11:00 arcady that's an odd message
11:00 arcady have you tried a clean build and the like?
11:24 TreyHarris ?eval ~localtime
11:24 evalbot_9946 "<obj:Time::Local>"
11:25 TreyHarris that's odd.  shouldn't localtime stringify into a human-readable date?
11:25 TreyHarris ?eval +localtime
11:25 evalbot_9946 Error: cannot cast from VObject (MkObject {objType = (mkType "Time::Local"), objAttrs = <ref>, objOpaque = Nothing, objId = 52}) to Double (VNum)
11:28 ayrnieu ?eval "" ~ localtime
11:28 evalbot_9946 "<obj:Time::Local>"
11:31 gaal TreyHarris: it probably should, but the coercion hasn't been written yet.
11:32 gaal in the mean time you can do
11:32 gaal ?eval localtime().perl
11:32 evalbot_9946 "\\Time::Local.new((\"day\" => 16), (\"hour\" => 13), (\"is_dst\" => bool::true), (\"min\" => 33), (\"month\" => 4), (\"picosec\" => 138780000000), (\"sec\" => 3), (\"tz\" => 7200), (\"tzname\" => \"CEST\"), (\"wday\" => 1), (\"yday\" => 105), (\"year\" => 2006))"
11:42 siosiosios is now known as DaGo
12:18 nnunley joined perl6
12:23 chris2 joined perl6
12:24 iblechbot_ joined perl6
12:27 dakkar joined perl6
12:34 larsen joined perl6
13:09 cdpruden joined perl6
13:13 KingDiamond joined perl6
13:41 boobs joined perl6
13:41 boobs Yo !
13:47 larsen joined perl6
13:52 enantiodrome left perl6
13:57 cmarcelo joined perl6
13:58 ruz joined perl6
14:01 pmurias joined perl6
14:01 pmurias hi
14:18 boobs I just want fuck bad bitches
14:43 larsen joined perl6
14:48 Odin- joined perl6
14:48 justatheory joined perl6
14:50 Khisanth joined perl6
15:04 weinig joined perl6
15:09 ruz arcady: make clean didn't help, I found this http://www.nntp.perl.org/gro​up/perl.perl6.compiler/1327, such a change hides (fixes?) error, but later I get another similar
15:15 gaal ruz: with the Syck stub, or another one? the patch described in that post is applied in the makefile.
15:16 xinming joined perl6
15:25 ruz gaal: Syck
15:26 ruz ok, now I tried to compile without embedding parrot or perl5
15:26 ruz and it compiles with Beau's patch
15:27 ruz but when I turno embedding I get errors, I could share them if you want
15:31 autark joined perl6
15:35 xinming1983 joined perl6
15:40 gaal ruz: please rm -rf dist/src/Data/Yaml and 'make' again
15:40 gaal no need for clean yet
15:40 gaal eh, though switching from embed to non-embed does generally need clean
15:41 gaal (you can probably get away without it by very selective cleaning of dist/, if you know exactly what you're doing.)
15:41 stclare joined perl6
15:48 xinming joined perl6
16:03 lichtkind joined perl6
16:04 lichtkind hello
16:04 lichtkind i found some little bugs in various docs like http://dev.perl.org/perl6/list-summa​ries/2006/p6summary.2006-02-07.html
16:05 dakkar lichtkind: that's not a doc... that's a mailing list summary...
16:06 gaal and it's from ancient history :-)
16:06 cdpruden joined perl6
16:07 lichtkind :)
16:08 lichtkind but nevertheless in http://svn.openfoundry.org​/pugs/docs/getting_started i read that there is actually no real perl6 tutorial
16:10 dakkar lichtkind: never seen one... want to write one?
16:11 lichtkind im actually write one
16:11 lichtkind am on chapter 5
16:11 lichtkind thats why oim here all days :)
16:12 lichtkind the downside for you is maybe that this tut is for the german perl-community.de aka de.perl.org
16:12 lichtkind but i would translate it
16:12 lichtkind when im done :)
16:13 lichtkind if you know german its on http://wiki.perl-community.de/bin​/view/Wissensbasis/Perl6Tutorial
16:13 mako132_ joined perl6
16:15 lichtkind but i need to lern a lot more currently its too unacurate to be semiofficial
16:19 lichtkind (1,2)Â¥(3,5)Â¥(9,8) == (1,3,9,2,5,8) ?
16:20 dakkar ?eval zip(zip([1,2],[3,5]),[9,8])
16:20 evalbot_9946 is now known as evalbot_9964
16:20 evalbot_9964 (1, 9, 3, 8, 2, undef, 5, undef)
16:20 dakkar ?eval zip(zip([1,2],[3,5]),[9,8,7,6])
16:21 lichtkind dont can evalbot Y?
16:21 evalbot_9964 (1, 9, 3, 8, 2, 7, 5, 6)
16:21 dakkar ?eval (1,2)Â¥(3,4)
16:21 evalbot_9964 Error:  unexpected "\\" expecting term postfix, "#", operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
16:21 dakkar IRC quoting problems, I suppose...
16:22 lichtkind ?eval (1,2)Y(3,5)Y(9,8)
16:22 evalbot_9964 (1, 3, 9, 2, 5, 8)
16:22 lichtkind yeah
16:22 dakkar ?eval zip([1,2],[3,5],[9,8])
16:23 evalbot_9964 (1, 3, 9, 2, 5, 8)
16:24 lichtkind now i understand i always thought you can place zip between like Y
16:25 gaal lichtkind: infix:<Y> ::= &zip
16:25 lichtkind thanks :)
16:31 SamB joined perl6
16:31 svnbot6 r9965 | audreyt++ | * Capture.pod: expands the examples; revamped the section on single/multi dispatch versus method/sub calls
16:34 jserv-- joined perl6
16:38 weinig is now known as weinig|out
16:42 FurnaceBoy joined perl6
16:55 audreyt TimToady: pmichaud and I came up with an observation on a chunk in S02
16:55 audreyt    macro quote:<qX> (*%adverbs) {...}
16:55 audreyt Note: macro adverbs are automatically evaluated at macro call
16:55 audreyt time if the adverbs are included in the parse.  If the adverbs are
16:55 audreyt to affect the parsing of the quoted text of the macro, then the text must
16:55 audreyt be parsed by the body of the macro rather than by an C<is parsed> rule.
16:56 audreyt TimToady: it seems that having the macro body do the parsing is very klugy (and unspecced)
16:56 mako132_ joined perl6
16:56 audreyt as it must receive out of band info about parse state and return updated position as an out-of-band info in addition to the AST.
16:56 audreyt TimToady: an arguably far cleaner way is to make lexicals in the subroutine parameter svisible to the rules in the "is parsed" trait
16:56 saorge joined perl6
16:57 audreyt macro quote:<qX> (:$adv1, :$adv2) is parsed /...some rule that makes use of $adv1 and $adv2.../ { ... }
16:57 audreyt TimToady: sounds sane to you? if yes, S02 can get updated :)
16:58 gaal audreyt: did you see the method delegation question in FAQ::Capture?
16:58 gaal it's bound to be an actual faq
17:03 dakkar how can a macro take parameters? by what rule would they be parsed?
17:03 * dakkar starts looking at synopses
17:05 _bernhard joined perl6
17:10 stclare left perl6
17:10 audreyt gaal: yes, see the commit
17:14 svnbot6 r9966 | audreyt++ | * Capture.pod: postulate that
17:14 svnbot6 r9966 | audreyt++ |     $arg = \($inv: $pos);
17:14 svnbot6 r9966 | audreyt++ |     $obj.meth(*$arg)
17:14 svnbot6 r9966 | audreyt++ |   means
17:14 svnbot6 r9966 | audreyt++ |     $obj.meth($pos)
17:14 svnbot6 r9966 | audreyt++ |   instead of an error.
17:15 audreyt TimToady: hopefully you are okay with that (which agrees with how scalars behave, if we treat the inv as the scalar slot in the fat typoglobby capture) :)
17:15 * audreyt waves goodnight &
17:16 gaal audreyt: great, I like that semantic. night and thanks!
17:18 audreyt :D
17:28 dakkar argh... the 'quote' default parsing rule parses adverbial arguments right after the starting token? so that an eventual 'is parsed' trait on a 'quote'-type macro starts seeing characters only after the adverbs?
17:33 audreyt dakkar: well, it's just speculative at this point... one may also do
17:34 audreyt macro quote:<qX> ($adv1, $adv2, $body is parsed /.../) {...}
17:34 audreyt provided that per-parameter "is parsed" is allowed.
17:35 dakkar I'm having difficulty figuring out how the parser will manage...
17:35 audreyt but I'd like things to stay simple, and defaulting quote: to parse adverbs uniformly seems like a win
17:35 dakkar yes, definitely
17:35 audreyt as indeed per-parameter is-parsed is quite difficult.
17:36 bsb joined perl6
17:36 audreyt TimToady: last thing... would you mind if all the "invocant parameters" in multi dispatch is renamed to "tiebreaking parameters" or some other name?
17:36 audreyt especially because we allow multilevel
17:36 audreyt "multi-level tiebreaking" makes sense
17:36 enantiodrome joined perl6
17:36 audreyt but "multi-level invocation" really doesn't, at least to me
17:37 audreyt with this change, "invocant" can go back to mean what it means in p5, namely "self-tobe"
17:37 audreyt (more details on my journal entry... bbl :)) &
17:45 Daveman :)
17:47 larsen joined perl6
18:17 cm joined perl6
18:26 pasteling "ruz" at 213.251.192.51 pasted "error with PUGS_EMBED="perl5" make" (19 lines, 1.6K) at http://sial.org/pbot/16763
18:28 gaal ruz: did you run 'perl Makefile.PL' with that env set?
18:30 ruz gaal
18:30 ruz yep
18:31 ruz full command is: 'make clean; PUGS_EMBED="perl5" perl Makefile.PL && make unoptimized'
18:32 pasteling "ruz" at 213.251.192.51 pasted "error with 'PUGS_EMBED="parrot" perl Makefile.PL && make'" (4 lines, 220B) at http://sial.org/pbot/16764
18:33 ruz I just paste it for those who knowns what is he doing :) cuz I don't yet
18:34 gaal does that env propagate to the second command?
18:35 gaal better export the setting and be safe.
18:35 gaal what shell are you using?
18:35 ruz no doesn't propagate I think
18:35 ruz bash
18:36 gaal I'm not sure, but it could be you need it set at make time and not just make generation time.
18:37 * ruz 's trying
18:43 nnunley joined perl6
18:44 ruz don't help
19:11 cognominal audreyt: why editng a FAQ, editing the synopsis would not be saner?
19:18 cognominal I have not read the backlog. but the Capture entity sonds like the siva pmc, I have proposed month ago. Is this intended to modelise tree nodes like ASTs or XML nodes as well?
19:21 cognominal sorry I did not notice that the synopsus has been edited as well.
19:21 cognominal audreyt++; TimToady++
19:24 boobs left perl6
19:30 marmic joined perl6
19:35 mako132_ joined perl6
19:40 Odin- joined perl6
19:47 elmex joined perl6
19:53 KingDiamond joined perl6
19:55 elmex joined perl6
20:02 elmex joined perl6
20:04 elmex joined perl6
20:07 elmex joined perl6
20:09 elmex joined perl6
20:25 pasteling "ruz" at 213.251.192.51 pasted "[PATCH] for compiling with embedded perl5" (19 lines, 742B) at http://sial.org/pbot/16766
20:27 gaal hey ruz, sorry for not responing eariler, was away. got a commit bit?
20:27 gaal if not, /msg me your email please
20:28 gaal or just say it here if you aren't registered on freenode and can't /msg
20:30 obra gaal: msging you his address
20:31 gaal ruz, thanks for the patch, an invitation letter is en route to your mailbox.
20:31 gaal please commit :)
20:31 gaal also, don't forget to add yourself to AUTHORS
20:31 gaal and welcome aboard :)
20:33 gaal and now if I can get up without distubing the cat asleep on my lap, I will go to sleep myself. see y'all :)
20:33 bsb 'night gaal & foo
20:35 FurnaceBoy n8
20:40 elmex joined perl6
20:53 elmex_ joined perl6
21:08 mako132_ joined perl6
21:14 KingDiamond joined perl6
21:16 buu Hello happy people. Anyone happen to know off hand how I can increment a perl global from the perl source?
21:23 avar use the eval routines in perlapi;)
21:23 avar arg, #perl6
21:25 buu =[
21:28 dduncan joined perl6
21:33 justatheory joined perl6
21:55 buu Arfg. Fucking perl.
21:58 pmurias sleep&
22:07 vel joined perl6
22:13 mako132_ joined perl6
22:16 b_jonas joined perl6
22:23 avar joined perl6
22:27 buu Someone come debug my fucking perl.
22:28 arcady perl's not for fucking!
22:28 arcady maybe that's why it doesn't like you
22:31 buu arcady: =[
22:31 buu SvIV_set(foo,0); is segfaulting
22:31 buu I'm trying to understand why.
22:45 mako132_ joined perl6
22:47 b_jonas left perl6
23:01 FurnaceBoy_ joined perl6
23:18 elmex joined perl6
23:20 weinig|out is now known as weinig
23:29 amnesiac joined perl6
23:31 frodo72 is now known as froh-doh
23:33 khisanth_ joined perl6
23:35 ruz gaal: I havn't got any message from you.
23:37 fglock joined perl6
23:41 ruz ? 0 or say "woot";
23:41 ruz evalbot_9964: ? 0 or say "woot";
23:42 ruz ?eval 0 or say "woot";
23:42 evalbot_9964 is now known as evalbot_9966
23:42 evalbot_9966 OUTPUT[woot ] bool::true
23:46 Khisanth joined perl6
23:55 svnbot6 r9967 | fglock++ | PCR/PCP - minor fixes
23:55 svnbot6 r9968 | fglock++ | PG-P6 - parses the first 10 files in t/01-sanity

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

Perl 6 | Reference Documentation | Rakudo