Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2005-10-09

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:03 autrijus ok, "make test" should yield 100% now.
00:03 autrijus consider this a preliminary preflight
00:03 autrijus still awaiting rafl
00:03 autrijus and I can't stay up any longer
00:04 clkao i am doing smoke-js
00:04 autrijus cool, clkao++
00:04 clkao go for another sauna?
00:04 autrijus we'll release 6.2.10 tomorrow.
00:04 * autrijus waves &
00:04 autrijus nah too tired.
00:08 timos has quit IRC (Read error: 110 (Connection timed out))
00:15 timos has joined #perl6
00:15 scook0 autrijus: oh, I see my patch broke Prelude--I had a feeling something would go wrong
00:15 scook0 I'll look into it
00:18 autrijus cool.
00:18 autrijus journal up -- g'nite!
00:18 scook0 bye
00:24 dduncan I'm back ... but about to leave again ... I'll svn up in a few hours and smoke with precompiled Test.pm and lib.pm
00:33 elmex has quit IRC ("leaving")
00:37 joepurl has joined #perl6
00:43 wilx has quit IRC (Remote closed the connection)
00:48 wilx has joined #perl6
00:51 larsen has joined #perl6
01:10 cyclist38 has quit IRC (Remote closed the connection)
01:16 larsen has quit IRC ("later")
01:42 dolmen has quit IRC ("Fermeture du client")
01:45 Blicero has joined #perl6
01:57 Cryptic_K has joined #perl6
01:59 evalbot_7453 has quit IRC (Remote closed the connection)
02:00 evalbot_7454 has joined #perl6
02:02 svnbot6 r7454 | stevan++ | Perl6-MetaModel -
02:02 svnbot6 r7454 | stevan++ | * adding 2 pictures :
02:02 svnbot6 r7454 | stevan++ |    1) normal method disptaching
02:02 svnbot6 r7454 | stevan++ |    2) method dispatching with eigenclasses
02:25 justatheory has quit IRC ()
02:43 Cryptic_K is now known as K
02:49 Blicero has quit IRC (Remote closed the connection)
02:57 justatheory has joined #perl6
03:04 rafl_ has joined #perl6
03:04 rafl has quit IRC (Read error: 104 (Connection reset by peer))
03:05 joepurl has quit IRC (Read error: 104 (Connection reset by peer))
03:12 joepurl has joined #perl6
03:22 luqui has quit IRC ("Leaving")
04:24 tcliou has quit IRC (Connection timed out)
04:29 justatheory has quit IRC ()
04:47 b6s has quit IRC (Read error: 110 (Connection timed out))
04:48 timos has quit IRC (Read error: 110 (Connection timed out))
05:10 scook0 autrijus: I have a revised version of my Parser patch, but I might wait until after 6.2.10 in case it breaks other stuff
05:13 joepurl has quit IRC (Read error: 110 (Connection timed out))
05:14 Blicero has joined #perl6
05:19 xinming_Beijing has quit IRC (Read error: 104 (Connection reset by peer))
05:40 svnbot6 r7455 | qtplatypus++ | Adding a simple DFA.  This is in preperation to build a simple regex
05:40 svnbot6 r7455 | qtplatypus++ | implementation as a work around for "Noting having PGE" installed problems.
05:41 evalbot_7454 has quit IRC (Remote closed the connection)
05:42 evalbot_7455 has joined #perl6
05:47 QtPlatypus ?eval my %hash = {foo => "bar",baz => "buz"};%hash{"foo"|"baz"}
05:47 evalbot_7455 [undef]
05:48 QtPlatypus ?eval my %hash = {foo => "bar",baz => "buz"};$temp="foo"|"baz";say %hash{$temp}
05:48 evalbot_7455 Error: Undeclared variable: "$temp"
05:48 QtPlatypus ?eval my %hash = {foo => "bar",baz => "buz"};my $temp="foo"|"baz";say perl %hash{$temp}
05:48 evalbot_7455 \undef bool::true
05:49 QtPlatypus Damn.
05:49 b6s has joined #perl6
05:56 iblechbot has joined #perl6
05:59 gaal morning
05:59 QtPlatypus gaal: Good morning.
06:00 gaal we've moved to winter time :(
06:01 QtPlatypus Summer time for me in a few weeks.
06:01 gaal and the livin' is easy
06:09 dduncan I'm back
06:17 gaal wb
06:17 iblechbot has quit IRC (Read error: 110 (Connection timed out))
06:35 QtPlatypus ?eval Bool::True
06:35 evalbot_7455 Error: No compatible subroutine found: "&Bool::True"
06:36 QtPlatypus ?eval bool::true
06:36 evalbot_7455 bool::true
06:43 QtPlatypus Is there somewhay to tell perl to recurse deeper?
06:44 scook0 QtPlatypus: recurse deeper?
06:45 QtPlatypus There seems to be a limmit to the depth of structures that .perl will give you
06:46 QtPlatypus ?eval ([=>] (1,2,3,4,5,6,7,8,9,10,11,12,13,14,15)).perl
06:46 evalbot_7455 '(1 => (2 => (3 => (4 => (5 => (6 => (7 => (8 => (9 => (10 => (11 => (12 => (13 => (14 => 15))))))))))))))'
06:46 * QtPlatypus mmmsss.
06:50 QtPlatypus Its spitting out (... => ...) for me in some situations
06:55 gaal QtPlatypus: backlog a few days, this was discussed here
06:55 gaal i think autrijus changed the limit to 1024 or something
06:56 svnbot6 r7456 | gaal++ | use parametric $(MAKE) instead of assuming `make'. Should fix build for some systems.
06:57 evalbot_7455 has quit IRC (Remote closed the connection)
06:58 evalbot_7456 has joined #perl6
06:59 gaal here's a musing. now that we have auto smoke uploads, i want to tag each smoke matrix with "clean source" or "has local mods" (and, for bonus points, a list of the locally changed files).
06:59 gaal not sure where to get that information, though. the easiest way is to check mtimes on source files
07:00 gaal but that's obviously not correct.
07:00 gaal also, we probably don't want to maintain a list of interesting files manually
07:01 gaal and finally, i don't want to couple this too tightly with any source control system :)
07:01 * QtPlatypus makes a request for the lambda camels to build some debugging tools.
07:01 gaal there's alread a debugger in perl6, QtPlatypus
07:02 gaal only it's not for perl6 :)
07:02 gaal examples/obfu/l33t.p6 :-p
07:02 * QtPlatypus looks puzzled.
07:02 gaal i'm just being silly :)
07:04 jp-autark Hm, I'm still getting build errors with a fresh checkout of 7455
07:05 jp-autark *** Wasn't able to find 'Syck_stub.o', aborting...
07:05 nothingmuch has joined #perl6
07:05 jp-autark I have Syck_stub.o in ./dist/build/src/Data/Yaml/Syck_stub.o
07:06 nothingmuch morning
07:07 gaal hola and aloha nm
07:07 * nothingmuch discovered, to his great shock, that he is actually neat and tidy
07:07 scook0 howdy nothingmuch
07:07 nothingmuch my family went away for holiday
07:07 gaal alek
07:07 nothingmuch and I stayed due to lack of vacation days from work
07:07 nothingmuch the house is clean
07:07 nothingmuch the kitchen is spotless
07:07 nothingmuch i ironed my shirt
07:07 nothingmuch s
07:08 gaal really? you should come visit more often.
07:08 nothingmuch there is no hairbrush hair or toothpaste around the sink in the bathroom
07:08 nothingmuch no mess has been made of the living room
07:08 nothingmuch and the real issue is that I'm usually blamed for this stuff when they're home1
07:08 * nothingmuch has shift key sync problems
07:11 nothingmuch mmm... someone brought nice chocolate to work
07:13 gaal you're already at work? doing good with the clock change i see.
07:15 nothingmuch didn't I gain an hour?
07:15 Aankhen`` has joined #perl6
07:15 nothingmuch 24h ago it was 10:15, right??
07:24 * nothingmuch considers porting Path::Class instead of File::Spec
07:24 nothingmuch and assimilating File::Spec into Path::Class
07:24 gaal nothingmuch: yes.
07:24 gaal off to work &
07:25 nothingmuch ciao
07:26 nothingmuch it should be delightful with $path.=subdir("foo");
07:28 dduncan has quit IRC ()
07:45 Aankh|Clone has joined #perl6
07:54 ilogger starts logging #perl6 at Sun Oct 09 07:54:34 2005
07:54 ilogger has joined #perl6
08:04 Aankhen`` has quit IRC (Read error: 110 (Connection timed out))
08:08 Aankh|Clone has quit IRC (Read error: 110 (Connection timed out))
08:10 * autrijus wakes up to drink something and sleep again
08:10 autrijus scook0: please don't hold off patches... you can do a "make test" :)
08:11 autrijus if you see parsefails, then you know that your Parser.hs refactoring breaks things
08:11 autrijus so you'd know where to look; otherwise it's as safe as it can be and please commit ahead.
08:12 autrijus bbiab&
08:12 scook0 autrijus: "make test" would take my box a few weeks :)
08:13 autrijus you don't have a feather account? :)
08:13 nothingmuch scook0: if you'd like you can test on my box
08:13 nothingmuch or probably on feather
08:13 autrijus ah, there you go
08:13 * autrijus goes back to sleep
08:14 scook0 well, I'm currently hunting down a subtle bug in it
08:14 scook0 there's a temporary workaround, but I'd rather do a proper fix
08:14 autrijus *nod* take your time :) scook0++
08:14 nothingmuch scook0: /msg me if I can help CPU power wise... irssi is in the BG
08:15 scook0 ooh, I think I found the bug!
08:16 scook0 apparently `foo($bar, $baz,)` is legal
08:17 scook0 building...
08:21 meppl guten morgen
08:21 joepurl has joined #perl6
08:22 scook0 ooh, it works!
08:24 * nothingmuch hates the fact that in order to write multitasking code in perl 5 you have to think too much up front
08:24 nothingmuch i wish we had lightweight threads that behaved like closures
08:25 rep well, concurrency is not easy
08:25 rep (in any language)
08:25 nothingmuch right
08:25 nothingmuch but this stuff is really not complex concurrency
08:25 nothingmuch it's separate parts
08:25 nothingmuch that must run at the same time
08:25 nothingmuch and report to the same parent
08:26 nothingmuch threads introduce too much of a headache (all data must serialize properly, can't rely on e.g. Tie::RefHash for proper identity keying, etc)
08:26 nothingmuch the issue is that you not only have to design your app
08:26 nothingmuch but you also haave to figure out how you are going to implement everything at the lowlevel
08:27 nothingmuch like what tradeoffs you're willing to make to work with forks
08:27 nothingmuch or whether you can fake concurrency by using an event loop
08:27 nothingmuch if your code has good enough latency
08:28 scook0 nothingmuch: care to test my patch for me?
08:28 nothingmuch scook0: sure... nopaste a unidiff
08:30 pasteling "scook0" at 220.253.96.162 pasted "Parser patch" (109 lines, 4.5K) at http://sial.org/pbot/13573
08:30 nothingmuch full test suite?
08:31 scook0 I don't know
08:31 scook0 whatever you're happy with
08:32 nothingmuch okay
08:32 K has quit IRC ("Leaving")
08:33 scook0 it shouldn't be changing any behaviour
08:33 nothingmuch just see that it compiles?
08:33 scook0 (at least any behaviour that was already working)
08:34 scook0 oh, it compiles
08:34 scook0 but last time I committed it, it caused parsefails in Prelude
08:34 scook0 it runs Prelude fine now
08:34 scook0 I'm just not sure if it will break anything else
08:38 nothingmuch has quit IRC (Read error: 104 (Connection reset by peer))
08:39 nothingmuch has joined #perl6
08:39 nothingmuch fsck
08:39 * nothingmuch 's puter made a weird sort of freeze-ish thing
08:39 scook0 eep
08:39 nothingmuch i wish osx had virtual consoles
08:39 nothingmuch the player kept making noise
08:39 nothingmuch and the UI sort of responded
08:39 nothingmuch i think the UI server crashed or something
08:40 nothingmuch you could move the mouse, but not click anything
08:40 nothingmuch and the cmd+tab app switcher thingy was frozen in place
08:40 nothingmuch and now i forgot if I applied your patch already or not ;-)
08:41 nothingmuch seems like I did
08:42 * scook0 's computer semi-freezes due to thrashing every time he builds pugs
08:42 nothingmuch ah, that I don't get... =)
08:42 scook0 which sometimes crashes nautilus :(
08:42 nothingmuch that's not nice
08:43 * nothingmuch bought plenty of ram
08:43 nothingmuch src/Pugs/Parser.hs:1734:0: Warning: Defined but not used: `parseParamList'
08:43 scook0 the fact that I'm running GHC + Gnome + Firefox with 256m probably doesn't help
08:43 nothingmuch is that supposed to be there?
08:43 nothingmuch hehe
08:43 scook0 oh, I'll take it out when I commit
08:44 scook0 the code that used to call that now calls my new code
08:44 nothingmuch ah
08:45 scook0 thanks for telling me though--I'd forgotten to take it out
08:45 nothingmuch dum dee dum... still compiling
08:53 * nothingmuch likes Schwern even more after the recent Test::Builder::Tester issues =)
08:53 nothingmuch http://use.perl.org/~schwern/journal/27075
08:56 scook0 so he made an update to a test module
08:56 scook0 which then broke other test modules
08:56 nothingmuch right
08:56 scook0 which broke large parts of CPAN
08:56 scook0 wow
08:56 nothingmuch Mark Fowler was sick so he took over T::B::T
08:56 nothingmuch now it's being fixed
08:56 nothingmuch overall it was handled quite well
09:00 nothingmuch dum deee dum
09:00 nothingmuch i think it's been compiling Pugs.Run for the past few minutes now =)
09:00 nothingmuch perhaps I shouldn't have built Test.pm into it ;-)
09:02 scook0 I remember once my compiles were taking heaps longer then usual
09:03 scook0 so I looked in Makefile.PL
09:03 scook0 and found that somebody had accidentally made `make unoptimised` and alias for `make optimized`
09:03 nothingmuch i'm making optimized =)
09:05 scook0 &
09:05 nothingmuch ciao
09:10 pdcawley Does pugs still need haskell plugins installed?
09:11 nothingmuch pdcawley: no, that's only for eval_haskell('...') last I heard
09:12 pdcawley Something I'm unlikely to do in the immediate future.
09:12 pdcawley By gum, it's been a while since I brought pugs up on this box.
09:12 nothingmuch did you get the compilation stuff fixed?
09:12 pdcawley I hope so; got the ghc binary package installed
09:13 pdcawley I'm about to find out :)
09:13 nothingmuch =)
09:15 csgeek has joined #perl6
09:15 pdcawley It's even been a while since I last did an update of parrot.
09:16 nothingmuch how can you live with yourself?
09:16 pdcawley Easily enough.
09:16 nothingmuch okay =)
09:16 pdcawley I've been doing Ruby On Rails stuff in my programming life.
09:17 nothingmuch if I had more time I would probably learn that
09:17 pdcawley Ruby's not as nice as Perl 6, but it has the massive advantage of being here now.
09:17 pdcawley And, as I don't know either C or Haskell...
09:17 pdcawley Well, I *read* C. But I don't write it.
09:17 nothingmuch as you don't know them... ?
09:18 pdcawley I can't really go about implementing much of either Parrot or Pugs.
09:18 nothingmuch pugs can use lots of non c knowlege
09:19 nothingmuch what about a PIL backend on ruby to help both pugs and your ruby learning?
09:19 nothingmuch i think that would be uber cool
09:19 pdcawley Yeah. I might well come back to that; hence the getting pugs built again.
09:19 nothingmuch write with rails in perl 6
09:19 pdcawley :)
09:19 pdcawley Oh. Bollocks. Build failed.
09:19 pdcawley RTLD_NEXT undeclared.
09:20 nothingmuch it's been a while since I've seen that... let me see if I can remember
09:20 nothingmuch can you nopaste more of the braindump?
09:20 pdcawley Ah... I think I may have gcc_select 4.0 in power.
09:21 nothingmuch ghc 6.4 or 6.4.1?
09:21 pdcawley 6.4.1
09:21 nothingmuch i thought that was supposed to be fixed =(
09:21 pdcawley And I didn't do a make clean, so starting again.
09:21 nothingmuch scook0: tests seem to be passing
09:22 scook0 nothingmuch: cool
09:22 scook0 I just have to finish cleaning the kitchen &
09:22 nothingmuch t/builtins/caller..........​........................***
09:23 nothingmuch    unexpected "c"
09:23 nothingmuch    expecting ";", statements or "}"
09:23 nothingmuch    at t/builtins/caller.t line 12, column 19
09:23 nothingmuch t/builtins/caller............​......................dubious    
09:23 nothingmuch is that your fault?
09:25 nothingmuch scook0: ?>
09:25 scook0 hmm
09:25 scook0 maybe
09:25 scook0 I'll look into it
09:25 scook0 does it work in normal pugs?
09:25 nothingmuch let me see
09:26 scook0 ?eval sub b_sub { try { caller.subname } }
09:26 evalbot_7456 undef
09:26 scook0 must be the patch
09:26 pdcawley Oh bollocks, bollocks and bollocks some more. Still throwing the RTLD_NEXT error.
09:26 nothingmuch scook0: yes, on the other machine it works
09:26 scook0 &
09:26 nothingmuch pdcawley: did you relclean and regenerate the gcc
09:27 pdcawley Did a make clean. Isn't that enough?
09:27 nothingmuch regenerate the makefile
09:27 pdcawley Yup, did that.
09:27 * nothingmuch got confused (i was talking to someone else while typing)
09:27 nothingmuch realclean or distclean ought to fix it
09:27 pdcawley K
09:28 pdcawley Shame there's not a 'make realclean_but_keep_makefiles'
09:28 nothingmuch that wouldn't be really clean, would it? =)
09:28 pdcawley Well, it depends what you're trying to clean.
09:29 csgeek has quit IRC ()
09:29 pdcawley make clean_generated_c_files ?
09:29 nothingmuch i think what you really want is 'make optimized_but_really_quick​ly_without_making_a_fuss'
09:29 pdcawley Bollocks!
09:30 * nothingmuch buys some suspiciously cheap tea (http://cgi.ebay.com/ws/eBayISAPI.​dll?ViewItem&item=4400323366)
09:30 nothingmuch scook0: more parse barfs, please wait
09:31 nothingmuch t/oo/attributes/instance.t parses on regular pugs, but not with your patch
09:31 * pdcawley distcleans this time.
09:32 pasteling "nothingmuch" at 212.143.92.226 pasted "./pugs -Iext/Test/lib t/oo/attributes/instance.t output" (4 lines, 411B) at http://sial.org/pbot/13574
09:36 pdcawley And more bollocks.
09:36 nothingmuch pdcawley: =(
09:36 nothingmuch do you have an account on feather?
09:38 pasteling "nothingmuch" at 212.143.92.226 pasted "more parse fails for scook0: t/operators/operator_overloading.t" (4 lines, 200B) at http://sial.org/pbot/13575
09:42 scook0 I see a common theme
09:43 scook0 it doesn't like `foo.bar`, `Class.bar` or `Class.bar()`
09:43 nothingmuch i'm glad testing is helping =)
09:44 scook0 nothingmuch: thanks
09:44 scook0 I can see the general shape of the problem
09:44 scook0 but I'll need to do some more investigating
09:45 pdcawley Yes. But I'm going to try a build on the laptop first.
09:45 pdcawley Which seems to have a clean ghc installation.
09:45 pdcawley No, I don't know why.
09:45 nothingmuch pdcawley: okay... if you need an OSX that works I can give you ssh to $home_box
09:45 nothingmuch scook0: it still hasn't finished the test suite
09:45 pdcawley It's a local box that works I want more than anything.
09:46 pdcawley But thanks for the offer.
09:49 nothingmuch scook0: out to lunch - will report more later
09:49 scook0 ok
09:52 scook0 ?eval class Foo {}; Foo.new
09:52 evalbot_7456 Foo.new();
09:53 scook0 ?eval foo.bar
09:53 evalbot_7456 Error: No compatible subroutine found: "&foo"
09:53 scook0 ?eval sub foo {}; foo.bar
09:53 evalbot_7456 Error: No compatible subroutine found: "&bar"
09:53 scook0 ?eval sub foo {}; sub bar {}; foo.bar
09:53 evalbot_7456 undef
09:53 scook0 ?eval sub foo {}; sub bar { say $_ }; foo.bar
09:53 evalbot_7456 Error:  unexpected "b" expecting block construct, ":", term, term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
09:54 scook0 ?eval sub foo {}; sub bar { say @_ }; foo.bar
09:54 evalbot_7456 undef
09:54 scook0 ?eval sub foo {3}; sub bar { say @_ }; foo.bar
09:54 evalbot_7456 3
09:54 scook0 ?eval sub foo {3}; sub bar { "A: @_[]" }; foo.bar
09:54 evalbot_7456 3
09:56 joepurl has quit IRC ("Lost terminal")
10:04 pdcawley How is foo.bar legal in those contexts?
10:06 rep win 13
10:06 rep oops
10:12 svnbot6 r7457 | autark++ | * Don't abort when Syck_stub.o is in the correct directory.
10:13 jp-autark is linking with parrot-0.3.0 supposed to work at this time?
10:14 evalbot_7456 has quit IRC (Remote closed the connection)
10:14 evalbot_7457 has joined #perl6
10:18 scook0 pdcawley: I don't think any of those is *supposed* to work
10:19 scook0 `foo.bar` should mean `foo().bar()`
10:20 scook0 and `MyClass.new` should mean something like `(::MyClass).new()`
10:20 scook0 but I'm not yet sure how pugs currently makes those work
10:22 scook0 nothingmuch: when the test suite finishes, could you email me the results as a file? (@gmail)
10:30 nothingmuch has quit IRC (Read error: 110 (Connection timed out))
10:33 QtPlatypus ?eval multi sub foo(bool $test) { say "Match" };foo(bool::true);
10:33 evalbot_7457 Match bool::true
10:34 QtPlatypus ?eval multi sub foo() {say "MissMatch"};multi sub foo(bool $test) { say "Match" };foo(bool::true);
10:34 evalbot_7457 Match bool::true
10:36 larsen has joined #perl6
10:40 evalbot_7457 has quit IRC (Remote closed the connection)
10:40 evalbot_7458 has joined #perl6
10:42 svnbot6 r7458 | qtplatypus++ | Can now mark some state as "Final".
10:47 QtPlatypus gaal: Can you check if Mine::Base64 still hangs?  I think I've resolved the issue, but since my copy never hanged I don't know if that is the case.
10:48 svnbot6 r7459 | qtplatypus++ | Resolving conflicts with the ruleless version of Base64.  Test still
10:48 svnbot6 r7459 | qtplatypus++ | diked out untill I can confirm there non hanging state.
10:48 lao has joined #perl6
10:50 evalbot_7458 has quit IRC (Remote closed the connection)
10:50 evalbot_7459 has joined #perl6
10:51 rafl_ autrijus: What are gmakeisms? :-)
10:54 autrijus rafl_: for foo in perl5 :)
10:56 Lopo_ has joined #perl6
10:56 rafl_ Oooch.. why? :-)
10:56 Lopo has quit IRC (Read error: 145 (Connection timed out))
10:56 Lopo_ is now known as Lopo
10:56 autrijus rafl_: because nmake does not understand that.
10:57 autrijus rafl_: so, you going to check in the privlib changes?
10:57 rafl_ Yes, but I'm going to write them first.
10:57 autrijus cool. I'll adjust the hs side now
10:58 rafl_ How will the hs side know where possible locations are? Do we set something in PugsConfig?
11:00 ron_ has joined #perl6
11:01 rafl_ Whops:
11:01 rafl_  Interface file inconsistency:
11:01 rafl_    home-package module `Pugs.AST.SIO' is mentioned,
11:01 rafl_    but does not appear in the dependencies of the interface
11:01 rafl_ ghc-6.4: panic! (the `impossible' happened, GHC version 6.4):
11:01 rafl_        forkM Declaration for retError{v}
11:02 wolverian I just love errors with "panic" and "the impossible happened" in them
11:02 rafl_ I don't for some reason.
11:02 adamc00 has joined #perl6
11:03 autrijus rafl_: we already have access to privlib and sitelib so I'll check both
11:03 rafl_ autrijus: OK.
11:03 autrijus rafl_: hm, did you restart the build?
11:03 autrijus incremental nonclean builds in ghc 6.4 is known to be troublesome
11:04 rafl_ Hrm, I hate waiting for a full build..
11:04 rafl_ OK, I'll be offline until the build finished now.
11:07 autrijus k.
11:19 meppl *g
11:19 vladtz has joined #perl6
11:19 rafl_ autrijus: You said you reverted my install vs. copy changes. But I can't find a $setup install anywhere..
11:20 gaal QtPlatypus: at $work now for at least five more hours, better get someone else on win32 to check it
11:21 rafl_ It get's install though. Dumb rafl..
11:21 nothingmuch has joined #perl6
11:21 nothingmuch beer++
11:22 autrijus rafl_: right... see util/ghc_setup.pl
11:22 rafl_ nothingmuch: Indeed!
11:22 nothingmuch scook0: no more errors in the test run
11:22 autrijus (r7460)
11:22 nothingmuch beer at lunch++ for 1 .. 5
11:22 * nothingmuch is underslept (also works when oversleeping)
11:22 nothingmuch and it really helps me calm down
11:22 nothingmuch big calorie boost
11:22 nothingmuch the cultural aspect is also significant
11:22 rafl_ autrijus: Ah, works well.
11:22 nothingmuch and ofcourse a small amount of alcohol
11:23 gaal beer and peanuts. you're hitchhiking?
11:23 nothingmuch gaal: nope, just went out to lunch
11:24 svnbot6 r7460 | autrijus++ | * do install instead of copy when there's no DESTDIR...
11:26 evalbot_7459 has quit IRC (Remote closed the connection)
11:26 ron_ has quit IRC ("Leaving")
11:26 evalbot_7460 has joined #perl6
11:26 gaal nothingmuch: you just reminded me of ford prefect, is all.
11:26 nothingmuch ah
11:26 nothingmuch =)
11:26 * nothingmuch is rereading the guide right now
11:26 nothingmuch with the 5th part
11:27 nothingmuch it's going a bit slow due to too many nights spent doing other stuff, but i'm really having fun
11:30 svnbot6 r7461 | rafl++ | * Install perl5 into perl6 {sitelib,privlib}/auto/pugs instead of pugslibdir.
11:30 svnbot6 r7461 | rafl++ | * Removed pugslibdir from util/PugsConfig.pm.
11:30 svnbot6 r7461 | rafl++ | * Make clean and realclean for perl5/ not fail if there were errors.
11:33 scook0 erp
11:33 scook0 I got `foo.bar` to parse again...as `foo ($_.bar)` :(
11:34 iblechbot has joined #perl6
11:34 evalbot_7460 has quit IRC (Remote closed the connection)
11:34 scook0 it might be time to cheat a little...
11:35 evalbot_7461 has joined #perl6
11:35 rafl_ OK, the HS, PIL-Run and PIL2JS site needs to be fixed WRT r7461
11:36 autrijus already fixed. testing
11:36 nothingmuch 3http://okmij.org/ftp/papers/zfs-talk.pdf
11:36 nothingmuch cool!
11:36 nothingmuch i wonder how it performs
11:37 autrijus nothingmuch: not terribly amazingly; the memory is never freed.
11:37 autrijus and also GHC did not have native support for the thousands of delimited continuations oleg uses.
11:37 autrijus hence, his request to me that GHC should get native support for resume.
11:38 * nothingmuch starts backing up 30GB of music
11:38 integral does it even have exposed support for any kind of continuation?
11:39 autrijus integral: Control.Exception.throw/catch is escape continuations
11:39 nothingmuch what exactly are delimited continuations?
11:39 autrijus nothingmuch: see ccshan's "shift to control" slides
11:40 integral oooh.
11:40 autrijus http://www.eecs.harvard.ed​u/~ccshan/recur/talk.ps.gz
11:40 nothingmuch what about this? http://www.eecs.harvard.ed​u/~ccshan/recur/recur.pdf
11:41 autrijus same, but in paper form
11:41 svnbot6 r7462 | autrijus++ | * adjust the helper launchers to find auto/pugs/ in either
11:41 svnbot6 r7462 | autrijus++ |   privlib or sitelib, now pugslibdir is gone.
11:41 scook0 nothingmuch: there's also a good explanation of delimited continuations way back in the #perl6 logs
11:41 scook0 from when autrijus explained them to me
11:41 nothingmuch scook0: how way back?
11:42 * nothingmuch gimbles over to the chatlogsa
11:42 scook0 nothingmuch: a few months
11:42 nothingmuch oh my, that won't be easy
11:43 evalbot_7461 has quit IRC (Remote closed the connection)
11:43 evalbot_7462 has joined #perl6
11:43 * scook0 wishes he'd started keeping his pugs notebook earlier
11:43 scook0 then I'd have the date
11:44 adamc00 has quit IRC ("Chatzilla 0.9.68.5.1 [Firefox 1.4.1/undefined]")
11:44 nothingmuch scook0: could you share that notebook?
11:45 scook0 nothingmuch: it's made of dead trees
11:45 nothingmuch ah
11:45 scook0 so sharing it wouldn't really be feasible
11:46 autrijus nothingmuch:
11:46 autrijus "reset introduces a new scope, in which you can call shift to get a partial continuation. Calling this partial continuation will result in starting at reset again. At the end of the shift, shift will return from the reset."
11:46 rafl_ autrijus: Does it work now? Because we installed into perl6 privlib/sitelib, not the perl5 equivalents.
11:47 nothingmuch what does reset do?
11:47 nothingmuch postpones execution and returns a continuation instead of a result?
11:47 autrijus nothingmuch: no, reset on itself does nothing.
11:47 autrijus it's jsut a mark.
11:47 autrijus rafl_: oh, I totally misunderstood you.
11:47 autrijus rafl_: I thought we were installing into perl5 sitelib/privlib.
11:48 nothingmuch so shift does what I thought reset does up to the parent reset?
11:48 autrijus rafl_: fixing
11:49 autrijus nothingmuch: er no, shift is just like call/cc
11:49 autrijus it takes a block, then runs that block with the cc as the argument.
11:49 autrijus the only difference is that the cc is not until end of program
11:49 autrijus but only until the reset
11:49 nothingmuch oh, i got it backwards
11:49 scook0 nothingmuch: there's a conversation starting at http://colabti.de/irclogger/irclogge​r_log/perl6?date=2005-06-04,Sat#l283
11:49 nothingmuch scook0++
11:49 autrijus nothingmuch: so unlike normal call/cc which never returns
11:49 nothingmuch ... it returns via it's caller
11:49 integral I think it looks really confusing in scheme where the future is to the *left* of the reset/shift not the right :-/
11:49 autrijus shift can return just fine
11:49 scook0 nothingmuch: also, look at the haddocks for shiftT/resetT in pugs
11:50 scook0 integral: shift/reset is SO much easier to understand in Haskell :)
11:50 nothingmuch callCC is essentially a clean goto
11:50 nothingmuch right?
11:50 integral scook0: I agree :-)
11:51 scook0 eep, it's late
11:52 scook0 hopefully I can finish my patch tommorow evening
11:52 scook0 otherwise it'll be at least a week :(
11:52 scook0 bye all
11:52 nothingmuch ciao!
11:52 scook0 has quit IRC ("Leaving")
11:52 autrijus nothingmuch: it's goto that passes values, but is
11:53 autrijus otherwise comparable
11:53 nothingmuch and how does shift/reset compare to goto?
11:53 autrijus the good thing about delimited continuations is that they act more like normal functions
11:53 autrijus so they are like... gosub!
11:53 nothingmuch gosub?
11:53 autrijus (i.e. they can be reasoned about like normal functions.)
11:54 autrijus never mind, crude BASIC joke.
11:54 nothingmuch ah
11:54 nothingmuch which are closer to coroutines?
11:54 integral they also label more than a goto label does;  you typically can't goto out of the current function for example
11:54 autrijus true
11:55 nothingmuch integral: assume goto is sub goto { $*INSTRUCTION_POINTER = $_[0] }
11:55 nothingmuch and the rest is just compile time mishmash
11:55 integral nothingmuch: yep.  a continuation OTOH sets the IP, and fiddles the return stack
11:55 * nothingmuch is very proud that in his p5orth that's how jumps are done
11:56 nothingmuch EXECUTE keeps it's stack frame but overwrites the instruction pointer such that the passed in address looks like it was really the caller
11:56 nothingmuch and it's all pure forth
11:57 nothingmuch shift reset is basically call with caller continuation, but also give it the current continuation on the side as if we called it regularly?
11:59 autrijus nothingmuch: reset marks the boundary of shift's return
12:00 * nothingmuch gives up on the distinction
12:01 rafl_ autrijus: Seems to work now. Thanks.
12:01 rafl_ Anyone who want's to fix PIL-Run and PIL2JS?
12:02 autrijus rafl_: they are implicitly fixed.
12:02 integral nothingmuch: reset puts a label in the return stack at that point.  shift as opposed to call/CC only saves the stack upto the last reset marked into the continuation it makes
12:02 autrijus rafl_: try "pugs -BPerl5 -e 'say 123'" after moving the srcdir aside.
12:02 nothingmuch integral: the problem is that I'm reading too fast to really grok now... @work. I'll read carefully later
12:02 nothingmuch but that is a good explanation
12:02 autrijus iblechbot: PIL2JS will die with "permission denied" if run with nonroot for precomp prelude/Test
12:02 svnbot6 r7463 | autrijus++ | * look into perl6 sitelib/privlib, not perl5 one...
12:03 autrijus iblechbot: you think it's wise to pregen the Prelude at build time for JS?
12:03 nothingmuch phew! all installed CPAN modules were updated
12:04 nothingmuch integral++; # you made it real clear as I was formulating my "thanks but no thanks" reply =)
12:04 integral np ;)
12:05 autrijus nothingmuch: I think you'll enjoy this too
12:05 autrijus http://www.eecs.harvard.ed​u/~ccshan/nepls8/talk.pdf
12:05 autrijus it's ccshan's dissertation -- using delimited continuations to solve linguistic problems
12:05 nothingmuch the harvard server is responding very very slow from here
12:06 evalbot_7462 has quit IRC (Remote closed the connection)
12:06 evalbot_7463 has joined #perl6
12:12 rafl_ Hrm.. how would you explain this: /usr/bin/make install DESTDIR=/home/rafl/projects/​perl6/pugs_local/debian/tmp
12:12 rafl_ make[1]: Entering directory `/home/rafl/projects/perl6/pugs_local'
12:12 rafl_ make[1]: *** No rule to make target `tmp-Prelude.pm', needed by `pm_to_blib'.  Stop.
12:13 autrijus rafl_: I think you ran Makefile.PL during a strange time that tmp-Prelude.pm is present.
12:15 rafl_ autrijus: OK. I'll just clean it.
12:15 autrijus rafl_: I fixed that.
12:16 rafl_ autrijus: btw, ./Setup copy wants --copy-prefix, not --prefix.
12:17 evalbot_7463 has quit IRC (Remote closed the connection)
12:17 evalbot_7464 has joined #perl6
12:17 xinming_Beijing has joined #perl6
12:18 iblech has joined #perl6
12:19 whiteg has joined #perl6
12:20 * autrijus is apalled that gobby is not in freebsd ports.
12:20 svnbot6 r7464 | autrijus++ | * s/Prelude-temp.pm/Prelude/
12:20 autrijus rafl_: noted
12:21 rafl_ autrijus: What also would be much easier is to do ./Setup copy --copy-prefix=/\$(DESTDIR). Then the default copy prefix is / which is what ./Setup install does, AFAIK.
12:21 autrijus rafl_: but register takes prefix as DISTDIR?
12:21 autrijus rafl_: er no, it's not / it's /usr/local
12:21 autrijus hence my problem
12:22 rafl_ Oh, of course.
12:22 autrijus rafl_: try r7465.
12:22 evalbot_7464 has quit IRC (Remote closed the connection)
12:22 evalbot_7465 has joined #perl6
12:22 rafl_ autrijus: I'm not sure about the register stuff. I don't use that target. It's only important for me that it is not executet implicitly with install.
12:23 xinming_Beijing has quit IRC (Success)
12:23 rafl_ autrijus: I take .installed-pkg-config and register manually with ghc-pkg  in a script that is executed when I install the Debian package (postinst).
12:23 whiteg has quit IRC (Client Quit)
12:23 autrijus ok, I'll keep it that way.
12:25 rafl_ autrijus++ # Works
12:25 svnbot6 r7465 | autrijus++ | * copy wants --copy-prefix according to rafl.
12:25 autrijus eggcellent.
12:25 elmex has joined #perl6
12:25 autrijus so... do we please the Great Hierarchy sufficiently for this release now? :)
12:28 rafl_ I'm not sure your sentence makes sense, but from what I can guess from it: Yes, I think the build system is perfekt for this release. :-)
12:29 autrijus excellent!
12:29 * autrijus goes back to tidy up changelog during the lunch.
12:32 rafl_ Only the JS Prelude should be generated during the build and also get installed.
12:33 autrijus I was afraid you'd say that.
12:34 autrijus if you or iblech can add that in, sure
12:34 rafl_ It tries to write to /usr/share/perl6/5.8/auto/​pugs/perl5/lib/Prelude.js which it can't otherwise.
12:34 autrijus but I really need to find food :)
12:34 autrijus nod.
12:34 rafl_ OK, I'll try to do that.
12:48 iblech rafl_: ./runjs.pl --compile-only -we 1 >/dev/null will generate Prelude.js and Test.js
12:49 iblech rafl_: (This command does not need an installed /bin/js)
12:49 dolmen has joined #perl6
12:52 svnbot6 r7466 | iblech++ | * Usual svn props.
12:52 svnbot6 r7466 | iblech++ | * PIL2JS: Makefile.PL: Clean Prelude.js, Test.js, output.html.
12:53 evalbot_7465 has quit IRC (Remote closed the connection)
12:53 evalbot_7466 has joined #perl6
12:53 rafl_ iblech: Oh, I just hacked it in in another way. But it doesn't find the prelude yet. Maybe you can hack that in?
12:55 iblech rafl_: ./runjs --p6preludepc=/path/to/where/th​e/Prelude.js/should/be/created --testpc=/path/to/where/Test.js/should/be/created --compile-only -we 1 >/dev/null
12:58 rafl_ iblech: Could you please hack that into Pugs.hs? The prelude will be in sitelib/auto/pugs/js/lib/PIL2JS/Prelude.js or in the same path in privlib.
12:59 iblech rafl_: Hm? Pugs.hs? Not into some Makefile.PL?
12:59 rafl_ iblech: Pugs.hs calls runjs in doHelperRun
13:00 iblech rafl_: Yep. And runjs will precompile the Prelude and Test.pm if not already done. But I thought we wanted to force precompilation at build-time?
13:01 iblech rafl_: (pugs -CJS --p6preludepc=... --testpc=... -we 1 >/dev/null works too)
13:01 rafl_ iblech: I think the options should go there.
13:01 rafl_ iblech: The PIL2JS Makefile already generates the prelude and get it installed.
13:01 rafl_ Hey, why aren't my changes pushed to the repository? Stupid me..
13:01 rafl_ Ah, hasn't been merged because you changed perl5/PIL2JS before.
13:02 svnbot6 r7467 | iblech++ | PIL2JS: jspugs.pl, runjs.pl: Document the --testpc option and added --metamodel-base.
13:03 svnbot6 r7468 | rafl++ |  r16904@ata:  rafl | 2005-10-09 14:48:46 +0200
13:03 svnbot6 r7468 | rafl++ |  * Generate precompiled JS Prelude and install it.
13:03 rafl_ iblech: Yeah, it's precompiled at build time, but runjs needs to know where it is. Therefor it needs to be called with the right --p6preludepc args.
13:03 rafl_ iblech: Does runjs really need a precompiled Test.js?
13:04 evalbot_7466 has quit IRC (Remote closed the connection)
13:04 evalbot_7468 has joined #perl6
13:05 iblech rafl_: Yep, Test.js is unconditionally linked in. (This is because there's no real reliable way for runjs.pl to read the sourcecode and search for /use Test/ (think -e, -M, etc.))
13:05 rafl_ Ah, OK.
13:07 iblech rafl_: Is pugs already installed when PIL2JS's Makefile.PL:pure_all runs?
13:08 rafl_ iblech: No.
13:08 iblech rafl_: (Because it'll either need a "pugs" in $PATH or we've to give it a --pugs=... option)
13:08 iblech rafl_: Hm.
13:08 rafl_ iblech: But there are only two paths where the precompiled stuff can be.
13:08 rafl_ p6{sitelib,privlib}/auto/pu​gs/js/lib/PIL2JS/Prelude.js
13:10 rafl_ iblech: pugs is already built at this time. ../../pugs should work.
13:10 evalbot_7468 has quit IRC (Remote closed the connection)
13:10 evalbot_7469 has joined #perl6
13:12 iblech rafl_: Hm. Currently, PIL2JS.pm expects Prelude.js and Test.js in ./, where . is the dir runjs.pl/jspugs.pl is in.
13:13 iblech rafl_: But if Prelude.js should be in js/lib/PIL2JS/, then we'd have to update PIL2JS.pm accordingly, resulting in runjs.pl/jspugs.pl not working when PIL2JS is not installed...
13:13 svnbot6 r7469 | rafl++ |  r16910@ata:  rafl | 2005-10-09 15:08:44 +0200
13:13 svnbot6 r7469 | rafl++ |  * Also generate Test.js as build time.
13:13 rafl_ iblech: Yes, therefor Pugs should pass a proper --p6preludepc to runjs.
13:14 rafl_ iblech: Shouldn't be too hard to let it look into two places.
13:15 iblech rafl_: Right, but the definition of "proper" varies: if PIL2JS is installed, --p6preludepc should point to p6{sitelib,privlib}/auto/pug​s/js/lib/PIL2JS/Prelude.js, but when developing, it should be $srcdir/perl5/PIL2JS
13:15 rafl_ Oh, right.
13:15 rafl_ iblech: Breaking it when it's not installed it maybe acceptable for this relase only if it's too much work to get both working..
13:16 iblech rafl_: Actually, it is, because we can't simply check for the existance of Prelude.js/Test.js (as we can do with other files, as PIL2JS.js or the metamodel)
13:16 iblech rafl_: Hm. What about if install Prelude.js and Test.js unconditionally in the dir jspugs.pl/runjs.pl get installed into?
13:16 iblech rafl_: This would work, I think.
13:17 iblech We'd only have to cp Prelude.js Test.js blib/lib/
13:17 rafl_ And not that hard to do. commited.
13:18 svnbot6 r7470 | rafl++ |  r16912@ata:  rafl | 2005-10-09 15:18:35 +0200
13:18 svnbot6 r7470 | rafl++ |  * Install precompiled js stuff next to runjs.pl.
13:18 iblech rafl_++ # great! :)
13:19 iblech Only thing left to do is setting $PIL2JS::cfg{pugs} appropriately
13:19 rafl_ -v please.
13:20 iblech Never mind, autrijus had already hacked that in into PIL2JS.pm
13:20 iblech So we're set, I think :)
13:20 rafl_ Yay!
13:20 rafl_ Generating some .debs to test it.
13:21 rafl_ * Error: Your precompiled Prelude is outdated.
13:21 evalbot_7469 has quit IRC (Remote closed the connection)
13:21 iblech This means that Prelude.js is older than Prelude/JS.pm.
13:21 evalbot_7470 has joined #perl6
13:21 rafl_ Possible. The order in which they get installed isn't predictable.
13:22 rafl_ (in the debian package)
13:22 iblech Hm, Prelude.js/Test.js get re-compiled each time one issues make... Fixing that.
13:23 rafl_ OK.
13:23 rafl_ Could you also disable that mtime check?
13:24 iblech Is this a one-time error only? If so, I would s/die/warn/
13:25 rafl_ No, it happens every time when I run pugs -BJS.
13:25 rafl_ For the installed pugs only.
13:26 iblech Ok, removing the check.
13:26 rafl_ Thanks.
13:32 svnbot6 r7471 | iblech++ | PIL2JS:
13:32 svnbot6 r7471 | iblech++ | * PIL2JS.pm: Removed the mtime check for the Prelude, as it breaks Debian.
13:32 svnbot6 r7471 | iblech++ | * Makefile.PL: Only precompile the Prelude and Test.pm if not already done.
13:32 svnbot6 r7471 | iblech++ |   Also make use of autrijus's $PIL2JS::cfg{pugs}, so Makefile.PL will find
13:32 svnbot6 r7471 | iblech++ |   ../../pugs or ..\..\pugs.exe.
13:33 rafl_ Generating new packages..
13:36 rafl_ Works. Gereat!
13:36 rafl_ iblech++
13:36 rafl_ iblech: pugs -BJS -e'"Foo".say'
13:36 rafl_ iblech: Why is '1..0' printed at the end?
13:37 evalbot_7470 has quit IRC (Remote closed the connection)
13:37 iblech FYI, just make cleaned, rm -rf /usr/lib/perl6/site_perl/auto/pugs/, make install'ed, mved $srcdir, typed pugs -BJS, and it worked! :)
13:37 evalbot_7471 has joined #perl6
13:37 iblech Because Test.pm is unconditionally linked in, and Test.pm always installs an END {...} block
13:38 iblech (Again, runjs.pl can't detect reliably whether Test.pm is really needed.)
13:39 * QtPlatypus needs someone to test his MIME::Base64 in Win32
13:40 iblech Look at runjs.pl:71 -- I tried to use some magic which'd make pugs-smokejs unnecessary, but failed because pugs wasn't able to parse the combination of -CPerl5, -M and -e I used...
14:00 rantanplan_ has joined #perl6
14:04 rafl_ is now known as rafl
14:04 svnbot6 r7472 | rafl++ |  r16916@ata:  rafl | 2005-10-09 15:56:13 +0200
14:04 svnbot6 r7472 | rafl++ |  Debian:
14:04 svnbot6 r7472 | rafl++ |   * Let the pugs package suggest perl and spidermonkey-bin.
14:04 svnbot6 r7472 | rafl++ |   * Removed usr/share/pugs from pugs-modules.install. Those stuff is now
14:04 svnbot6 r7472 | rafl++ |     installed in usr/share/perl6.
14:04 svnbot6 r7473 | rafl++ |  r16917@ata:  rafl | 2005-10-09 15:59:39 +0200
14:04 svnbot6 r7473 | rafl++ |  * Added perl5/*/t/*.t to test files.
14:04 svnbot6 r7474 | rafl++ |  r16918@ata:  rafl | 2005-10-09 16:02:14 +0200
14:05 svnbot6 r7474 | rafl++ |  * Reverted accidently changed util/yaml_harness.pl. I forgot to remove the
14:05 svnbot6 r7474 | rafl++ |    patches applied for the Debian package before commiting.
14:06 rafl iblech: Why is Test.pm handled specially at all?
14:06 iblech rafl: Performance of the testsuite
14:07 rafl :-(
14:07 rafl iblech: You told me a cool way do disable smoking/sending smoke results once. I forgot it. Can you tell me again how it worked.
14:07 rafl s/.$/?/
14:08 iblech rafl: Sure, PUGS_BUILD_OPTS="smoke_upload=0" (but untested)
14:08 evalbot_7471 has quit IRC (Remote closed the connection)
14:08 evalbot_7474 has joined #perl6
14:08 rafl Great, thanks.
14:09 iblech rafl: I quite like seeing JS smoke results after a hour (instead of at least three) :)
14:09 rafl iblech: Buy better hardware!
14:10 iblech In the future, we'll probably need /var/cache/pugs/js/Prelude.js, with appropriate measures that no one can write precompile an evil modul and save it as Prelude.js
14:10 iblech rafl: Do you want so sponsor me? :)
14:10 iblech s/so/to/
14:11 rafl iblech: Yes, nice feature.
14:11 rafl iblech: Well, actually... no.
14:12 rafl iblech: It would be nice to include the name of the smoker in the overview of smoke results on smokeserv.
14:13 rafl And maybe also display the architecture.
14:15 rafl pugs doesn't work on mipsel, arm, mips, ia64, etc. yet.
14:16 rafl It kind of works on hppa, but lots of tests seem to fail.
14:17 iblech Should both be relatively simple to implement. I'll have a look at it, probably tomorrow
14:17 rafl On s390 as well.
14:17 rafl Thanks.
14:17 iblech IIRC putter runs Pugs on ia64, not sure
14:18 rafl OK, I'm off till this evening. I hope a have a new release to put in Debian then. :-)
14:18 * nothingmuch wonders how much effort it'll be to make smokeserv.pl into a catalyst based app
14:18 nothingmuch with slightly more abstraction WRT the naming
14:18 nothingmuch and have it submit storable dumps instead of htmls
14:18 rafl iblech: Well, it doesn't work with Debian's ghc6 on ia64. Haven't tested other compilers.
14:18 iblech rafl: Ah, ok.
14:18 rafl nothingmuch: Hm. I'd volunteer to do that if iblech agrees.
14:19 nothingmuch it's opensource, iblech doesn't have to agree ;-)
14:19 rafl Well, I don't want to crack the machine it runs on to install it. :-)
14:19 nothingmuch if that becomes a problem i'll host it
14:19 rafl Beside iblech agrees ;-)
14:20 iblech I don't have any knowledge of Catalyst, but if rafl wants to do it, I'm fine :)
14:20 nothingmuch rafl: what I would really like is that once it gets storables and slightly more structure, we use Test::TAP::Model::Consolidated to display diffs
14:20 rafl OK, I'll start on that this evening if the build of the Debian packages I currently do is finished and fine.
14:20 nothingmuch rafl++
14:20 iblech Installing Catalyst on the smokeserver is absolutely no problem :)
14:20 iblech rafl++ indeed
14:20 nothingmuch iblech: what about mod_perl?
14:21 rafl OK, bye. (really now) :-)
14:21 nothingmuch ciao!
14:21 iblech nothingmuch: I'm root on the server, so that'd be no problem either :)
14:21 integral why storable rather than something more inter-language like XML, YAML, or JSON?
14:21 iblech bye rafl
14:21 nothingmuch yummy
14:21 nothingmuch integral: performance
14:21 nothingmuch YAML takes 40 seconds to load
14:21 nothingmuch XML is probably not as bad
14:22 nothingmuch but Storable is optimized for speed
14:22 nothingmuch and the structure is very big, at 10000 unit tests
14:22 integral that's irritiating about YAML
14:22 nothingmuch very =(
14:22 nothingmuch rafl: if you backlog, use Data::Serilizer so that the serialization backend is pluggable ;-)
14:24 elmex has quit IRC ("essen 5*")
15:09 gaal autrijus: re r7464, sorry for the bikeshed paintjob suggestion, but imo "Prelude.pm" is better than having no extension at all...
15:12 autrijus gaal: problem was if it failed it left a .pm around for MakeMaker to pick up
15:12 autrijus gaal: maybe a END{} block to clean that away?
15:18 iblech has quit IRC ("installing more RAM")
15:18 geoffb See, now *there's* a good reason to be offline
15:20 nothingmuch darnit
15:24 evalbot_7474 has quit IRC (Remote closed the connection)
15:25 evalbot_7475 has joined #perl6
15:25 svnbot6 r7475 | autrijus++ | * "make haddock" now complies with Cabal convention and
15:25 svnbot6 r7475 | autrijus++ |   put the HTML docs into dist/doc/html/.  This should keep
15:25 svnbot6 r7475 | autrijus++ |   the src/ tree cleaner.
15:29 wcting has joined #perl6
15:30 evalbot_7475 has quit IRC (Remote closed the connection)
15:31 svnbot6 r7476 | autrijus++ | * Expose all Pugs.*, Emit.* and PIL.* modules for Haskell programs.
15:31 svnbot6 r7477 | autrijus++ | * Inline::Pugs: revert to the previous, non-PIL-Run UI
15:31 svnbot6 r7477 | autrijus++ |   as the API has changed and I don't quite feel like to
15:31 svnbot6 r7477 | autrijus++ |   chase it right now...
15:31 evalbot_7476 has joined #perl6
15:33 joepurl has joined #perl6
15:35 gaal autrijus: or put in a tmp/ directory MM won't touch?
15:35 gaal (end block where, build_pugs.pl?)
15:35 autrijus gaal: end on genprelude
15:36 iblechbot has quit IRC (Connection timed out)
15:36 svnbot6 r7478 | autrijus++ | * Pugs.Run.Perl5: remove commented-out debug statements.
15:36 svnbot6 r7479 | autrijus++ | * Pugs.hs: refactor out noEnvDebug so external programs
15:36 svnbot6 r7479 | autrijus++ |   can elect to run without debug traces.
15:37 evalbot_7476 has quit IRC (Remote closed the connection)
15:37 evalbot_7479 has joined #perl6
15:38 autrijus gaal: done.
15:38 gaal too fast for me, as usual :)
15:38 autrijus :)
15:39 iblechbot has joined #perl6
15:40 notkevin has joined #perl6
15:40 notkevin has left
15:42 svnbot6 r7480 | autrijus++ | * adopt gaal's bikesheding correction to paint the tmp Prelude back
15:42 svnbot6 r7480 | autrijus++ |   as Prelude.pm, and unlink it at END of gen_prelude.pl.
15:42 svnbot6 r7481 | gaal++ | add --keep option for Prelude debugging
15:42 iblech has joined #perl6
15:42 gaal current music, incidentally: Paint It Black
15:43 * autrijus chuckles
15:43 evalbot_7479 has quit IRC (Remote closed the connection)
15:43 gaal coincidence, I swear :)
15:43 evalbot_7480 has joined #perl6
15:44 zuulvin_ has joined #perl6
15:45 * gaal imagines a PHB giving a speech: "... now this, our bikeshed feature for which we had worked so hard in design meetings..."
15:45 gaal (corporate malapropisms)++
15:47 autrijus civilization is wonderful: it replaces bloodshedding with bikeshedding
15:48 gaal cats are perfection of this process, just shedding.
15:48 autrijus lol
15:49 evalbot_7480 has quit IRC (Remote closed the connection)
15:49 evalbot_7481 has joined #perl6
15:49 autrijus I think I'll be in OSDC.
15:49 autrijus (preferably if damian will also be there)
15:49 autrijus # http://www.osdc.com.au/
15:50 gaal monash!
15:50 gaal sounds close to home for him.
15:51 autrijus bleh, ::Pair also participated in magical pairs
15:51 autrijus I mean the ::Pair class object itself
15:51 autrijus simply because it .isa Pair
15:51 * autrijus feels like hlaghing
15:53 gaal Pair: Pairs are indeed recursive
15:55 evalbot_7481 has quit IRC (Remote closed the connection)
15:55 evalbot_7482 has joined #perl6
15:56 bsb has joined #perl6
15:58 autrijus down to two failures.
15:58 svnbot6 r7482 | autrijus++ | * even more giving-up on magical pairs: stringify ::Pair in is()
15:58 svnbot6 r7482 | autrijus++ |   calls as otherwise it would participate in Pair magical binding
15:58 svnbot6 r7482 | autrijus++ |   because it .isa Pair.  Special casing for that feel wrong, so
15:58 svnbot6 r7482 | autrijus++ |   let's also write it off and wait for the sane-magic-pairs revolution...
15:58 svnbot6 r7483 | autrijus++ | * ditto for Perl6-Value-List/t/.
15:58 * autrijus relooks arcfour
16:01 evalbot_7482 has quit IRC (Remote closed the connection)
16:01 evalbot_7483 has joined #perl6
16:01 Blicero has quit IRC (Read error: 110 (Connection timed out))
16:02 * gaal $works :(
16:02 autrijus enjoy :)
16:02 gaal :)
16:03 gaal I like your attitude!
16:03 stevan morning autrijus, gaal
16:03 gaal yoyoyo
16:03 * stevan thinks that p6 will be the new LISP
16:03 stevan we just need to get nothingmuch's exceptuations accepted
16:04 leo__ autrijus: are you planning to write some official messages about Pair's insaneness?
16:04 autrijus leo__: it's been extensively documented on p6l by luqui et al
16:04 autrijus with a working proposal even
16:04 autrijus s/proposal/counterproposal/
16:05 autrijus so I'm not sure what I can add to this, other than outlining what I'd like to support instead in pugs...
16:05 leo__ ok - rather warnocked iirc
16:07 autrijus nod.
16:10 efilnikufesin has joined #perl6
16:13 autrijus hm, arcfour seems to have actual problems
16:13 autrijus as the ciphertext did not match and the garbled chars killed my terminal
16:13 gaal unicode problems probably
16:14 gaal our io assumes utf8
16:14 gaal getc etc.
16:14 autrijus oh hm. yeah.
16:14 autrijus though we have binmode
16:15 gaal but that doesn't help if the prim reads more than one byte and assumes an encoding :)
16:15 autrijus true.
16:15 gaal binmode is Haskell's binmode, i think it just does eol/eof?
16:16 autrijus you are correct
16:16 autrijus okay, we need to integrate the Data.FastPackedString based IO.
16:16 autrijus (it's already in src/Data/)
16:16 gaal we can add some metadata to VHandle
16:17 autrijus *nod*
16:17 gaal there's a dummy $layer arg to open :)
16:17 autrijus I saw that :)
16:18 gaal P.hGet assumes latin-1
16:19 gaal also, i don't see a character based read
16:19 gaal but it's very nice that they have gz io.
16:20 gaal can Haskell strings contain NULs?
16:20 svnbot6 r7484 | autrijus++ | * todo-fail arcfour.t again as its test failures garbles my
16:20 svnbot6 r7484 | autrijus++ |   terminal and there is no obvious piece at fault; gaal
16:20 svnbot6 r7484 | autrijus++ |   suggests utf8ness, which is plausible.
16:21 autrijus gaal: of course...
16:21 autrijus gaal: CStringLen too
16:21 gaal (i mean other than at the last byte ;-)
16:21 gaal cool
16:21 gaal @t Char
16:21 gaal er, not here
16:22 evalbot_7483 has quit IRC (Remote closed the connection)
16:22 evalbot_7484 has joined #perl6
16:26 GeJ has joined #perl6
16:26 bsb has left
16:28 pasteling "iblech" at 62.216.196.182 pasted "luqui's pair proposal" (10 lines, 295B) at http://sial.org/pbot/13579
16:31 autrijus iblech: that, and also more importantly
16:31 autrijus $pair = :foo(1)
16:31 autrijus blah($pair) # positional
16:31 autrijus blah(*$pair) # named
16:32 iblech ok. I like the proposal very much :)
16:33 iblech How will we implement the normal pair/magical pair distinction? Two classes (Pair and Pugs::Internals::PairUsedForNamedParameters)?
16:34 autrijus no, the named for parser is a parser form
16:34 autrijus it's not really a pair in any sense
16:34 nothingmuch has quit IRC (Read error: 110 (Connection timed out))
16:34 autrijus it's syntactic.
16:35 iblech Ah, makes sense. But then how'll the PIL for foo(a => 42) look like? Currently, it's simply serialized as foo(infix:«=>»("a", 42))
16:35 autrijus yeah, in vanilla core we use Syn=> to distinguish
16:35 autrijus I think the App form needs extension to support all annotations
16:36 autrijus so instead of using substructural forms, it's all represented at App level
16:36 iblech example please?
16:37 iblech PApp { pFun = ..., pNamedParams = {...}, pPositionalParams = [...] }?
16:40 autrijus yeah, have a structure for each zones sounds prudent
16:41 iblech ok. And *$pair will simply add $pair to pNamedParams then. Makes sense and is easy to implement in PIL2JS :)
16:42 autrijus which is always good to hear :)
16:44 * iblech goes to write tests for it
16:45 autrijus :todo<revolution>
16:45 justatheory has joined #perl6
16:45 iblech :)
16:46 reffie has joined #perl6
16:47 autrijus seriously, <unspecced> would do
16:47 autrijus would be good to have it in the release and then posted to p6l.
16:47 svnbot6 r7485 | iblech++ | * util/gen_prelude.pl: Removed now unnesessary unlink() call.
16:47 svnbot6 r7485 | iblech++ | * PIL2JS:
16:47 svnbot6 r7485 | iblech++ |   * jspugs.pl, runjs.pl: Better handling/ignoring of -BJS.
16:47 svnbot6 r7485 | iblech++ |     This means make test-js works again.
16:47 svnbot6 r7485 | iblech++ |   * runjs.pl: New option --precompile-only (only precompile the Prelude and
16:47 svnbot6 r7485 | iblech++ |     Test.pm, do not run or compile any user code).
16:47 svnbot6 r7485 | iblech++ |   * Makefile.PL: s/compile-only/precompile-only/
16:48 evalbot_7484 has quit IRC (Remote closed the connection)
16:48 evalbot_7485 has joined #perl6
16:50 rep has quit IRC ("leaving")
16:51 reffie is now known as rep
16:56 nothingmuch has joined #perl6
16:56 justatheory has quit IRC ()
16:58 leo__ autrijus: what do you think of ~lt/named-args.txt
16:58 leo__ (at feather)
16:59 pasteling "autrijus" at 194.145.200.126 pasted "lt's proposal" (20 lines, 647B) at http://sial.org/pbot/13580
17:00 autrijus leo__: sure, that is the same with luqui/iblech/me/etc
17:00 autrijus the only addition is a *$pair form
17:00 autrijus akin to *%hash but only for one pair
17:01 nothingmuch dvd 7 is now burning.... *sigh*
17:01 leo__ not quite, luqui still had some magic in his proposal I think
17:01 * nothingmuch wonders what we're talking about
17:01 leo__ e.g. difference between a=>'b' and :a(b) or :a<b> - whatever
17:02 leo__ *$pair is *%hash with excatly one kv elem
17:03 nothingmuch has quit IRC ("leaving")
17:03 nothingmuch has joined #perl6
17:03 autrijus leo__: yeah, I think that :a<b> and a=>'b' should at all times be interchangeable
17:03 autrijus makes no sense otherwise
17:03 * iblech agrees
17:04 leo__ yep
17:04 dduncan has joined #perl6
17:04 autrijus good we are in vehement agreement.
17:04 leo__ great
17:05 autrijus so the entire "unexpected pairs jump into named zone" foo can go away.
17:06 leo__ for me - yes ;-)
17:06 autrijus good good.
17:06 leo__ a Pair arg is passed like a Foo arg - no special case at all
17:06 autrijus leo__: parrot of course has support for nonslurpy but still nameless arguments?
17:06 autrijus will it have named-only support?
17:07 nothingmuch you guys are discussing touples vs parrot calls?
17:07 autrijus nothingmuch: we guys are talking about demagicalizing the insane magic pair semantic :)
17:07 nothingmuch ah
17:08 nothingmuch has there been any attempt besides lexical sensitivity?
17:08 leo__ named support just depends on the 'saneness' of implementation possiblities - currently named args aren't supported
17:08 autrijus leo__: right, but iirc chip is working on a pythonesque support?
17:08 leo__ yep
17:08 autrijus nothingmuch: *$pair acts like *%hash with one pair
17:08 svnbot6 r7486 | stevan++ | Perl6::MetaModel -
17:08 svnbot6 r7486 | stevan++ | * new version of the Mini-MetaModel which has explicit EigenClasses for
17:08 svnbot6 r7486 | stevan++ |   all classes created. This is an implementation of the diagrams I did
17:08 svnbot6 r7486 | stevan++ |   yesterday. I still need to resolve how to deal with the class pointer
17:08 svnbot6 r7486 | stevan++ |   since it now points to the eigenclass instead of the real class. But
17:08 svnbot6 r7486 | stevan++ |   this is a somewhat minor issue really.
17:08 nothingmuch versus?
17:09 dduncan after back-logging, I agree with the idea of *$pair being like *%hash, making it bind to a named arg, whereas no * means it is positional
17:09 evalbot_7485 has quit IRC (Remote closed the connection)
17:09 leo__ ther is really no need for a special *$pair
17:09 autrijus nothingmuch: versus no such syntax.
17:09 evalbot_7486 has joined #perl6
17:09 * nothingmuch reads for more context
17:10 nothingmuch as for :a<b> and a => b interchangiable - me too
17:10 autrijus leo__: because it may be written as *{$pair} ?
17:10 dduncan likewise, I infer that one can also say foo( *$pair1, *$pair2, ... ) etc if they had several pairs to bind in the same method call
17:10 autrijus dduncan: aye
17:10 dolmen has quit IRC (Read error: 104 (Connection reset by peer))
17:11 nothingmuch when there is one pair tuple with one named and hash are indistinguishable?
17:11 leo__ may point is: if you want named args: do it in the call or pass a *% name hash - nothing else is needed IMHO
17:11 autrijus leo__: say I have
17:11 autrijus foo($name => $value)
17:12 autrijus and then I want to refactor it out:
17:12 autrijus my $pair = (get_name() => get_value()); foo(*$pair)
17:12 autrijus then
17:12 autrijus sub get_pair { ... } ; my $pair = get_pair(); foo(*$pair)
17:12 autrijus surely that is sensible?
17:13 leo__ my proposal would need my %pairs - ...   foo(*%pairs)
17:13 leo__ ther is no special *$pair at all
17:13 * nothingmuch thinks that 'sub foo (*$a) { other(*$a) }' alone is worth the complexity of the tuple type
17:14 svnbot6 r7487 | autrijus++ | * fix a genuine Test::Builder bug uncovered by correct class-method
17:14 svnbot6 r7487 | autrijus++ |   handling; Test::Builder.create should create a new builder object
17:14 svnbot6 r7487 | autrijus++ |   without falling back to .new for singletons.
17:14 autrijus leo__: what nothingmuch said
17:14 dduncan back in about 3 hours ...
17:14 leo__ ther is aready *@ and *% support - why just another level?
17:14 nothingmuch because *@ and *% aren't all encompassing
17:15 nothingmuch *$a is much more transparent
17:15 nothingmuch so it's suitable for delegation
17:15 evalbot_7486 has quit IRC (Remote closed the connection)
17:15 nothingmuch while the others are suitable for more "normal" slurpiness - accepting a variadic number of args
17:15 evalbot_7487 has joined #perl6
17:16 leo__ for me this looks like a premature 'pessimization' against 1 elem hashes
17:16 nothingmuch leo__: the tuple type can hold names and positionals
17:17 nothingmuch http://svn.openfoundry.org/​pugs/docs/notes/theory.pod
17:17 nothingmuch see =head2 Tuples
17:17 nothingmuch my $tuple = ("foo",);  
17:17 nothingmuch foo(*$tuple);
17:17 nothingmuch that passes in a positional
17:18 nothingmuch    my $a = (1,2,3);    # a tuple with positionals 1,2,3
17:18 nothingmuch    my $b = (0,*$a,4);  # a tuple with positionals 0,1,2,3,4
17:18 leo__ this is just a mix of zones imho and not needed at all
17:19 leo__ # when used in a call of course
17:19 nothingmuch then how do you do 'sub foo ($my_param, *$their_params) { $my_param + other(*$their_params) }' or something like that?
17:19 nothingmuch it's always been hard in perl
17:22 leo__ if *$their_params isn't modified at all (partially consumed e.g.) just pass an array or list
17:22 leo__ else use a named hash
17:22 iblech leo__: But then you lose type information etc.
17:22 nothingmuch no, because that doesn't capture the "magicness" of names, and no because you can't losslessly represent positionals in named hash without name mangling
17:22 nothingmuch also, type inferrencing will benefit from opaqueness
17:23 nothingmuch that way foo's signature can statically be other's signature plus one positional if static typing is enabled
17:23 leo__ I don't see how *$ can help here, or is better then *% or *@
17:23 nothingmuch because it makes no implication (even psychological) about whether the parameters are named or positional
17:24 autrijus *$ in this view is a zone-free device
17:24 autrijus it is indeed beyond the original spec of perl6 zones
17:24 autrijus (pugs had never really implemented zone separation either)
17:25 dolmen has joined #perl6
17:26 autrijus preflight.
17:27 leo__ ok, when I get that right Tuple ... ~lt/named-args.txt
17:27 nothingmuch are these zones in the metaphorical, syntactic, or implementation sense?
17:27 leo__ oops again
17:27 autrijus http://pugscode.org/tmp/Per​l6-Pugs-6.2.10-pre0.tar.gz
17:27 nothingmuch where is named-args.txt?
17:27 leo__ a Tuple ~lt/named-args.txt
17:27 leo__ shit
17:27 autrijus nothingmuch: read the spec... it's syntactic.
17:28 nothingmuch autrijus: that is the distinction between a => b and (a => b) in a call?
17:28 leo__ yeah, wanted to paste ;-)
17:28 leo__ anyway the tuple is the param list, it seems
17:29 leo__ what I'm talking about is the implementation of that thing
17:29 nothingmuch as I see it in it's most naive form it's structure with an array and a list, where most of the time it's structure is known in advance
17:29 nothingmuch array and hash, sorry
17:29 autrijus nothingmuch: er no, no
17:29 nothingmuch autrijus: what spec then? theory.pod?
17:30 autrijus nothingmuch: A06
17:30 nothingmuch ah
17:32 gaal oooh preflight! /me wants to go home now
17:32 nothingmuch gaal: how was your pm talk?
17:32 nothingmuch autrijus: can preflight precompile Test.pm etc? i see no config.yml
17:33 gaal it's generated the first time
17:33 nothingmuch ah
17:34 gaal the talk was nice, the turnout was low
17:34 gaal home is where the pugs is &
17:34 autrijus gaal++ #  home is where the pugs is
17:36 nothingmuch autrijus: make smoke started on preflight
17:36 autrijus excellent
17:36 * nothingmuch goes to make dinner
17:38 nothingmuch hah, airtunes is choking when I'm scping a big file =)
17:46 putter has joined #perl6
17:46 svnbot6 r7488 | autrijus++ | * we cannot add perl5/*/t/*.t to the test files because those are harnessed
17:46 svnbot6 r7488 | autrijus++ |   with pugs as the main executable, and we had not yet mastered the art of
17:46 svnbot6 r7488 | autrijus++ |   re-execing perl5 on non-perl6 files.
17:46 svnbot6 r7489 | autrijus++ | * add Setup* into MANIFEST.SKIP.
17:46 evalbot_7487 has quit IRC (Remote closed the connection)
17:46 evalbot_7489 has joined #perl6
17:49 putter fyi, a recent (hmm, how do I check version when pugs doesnt run...) (within a few hours) pugs on FC3 x86_64 w embedded perl, dies with
17:49 putter ./pugs: error while loading shared libraries: libperl.so: cannot open shared object file: No such file or directory
17:50 putter anyone creating /lib/ paths and forget to include /lib64/ ?
17:50 autrijus putter: hm, libperl.so is not in LD_LIBRARY_PATH?
17:50 putter checking...
17:51 putter my LD_LIBRARY_PATH is undefined.
17:51 autrijus where is libperl.so then
17:53 putter /usr/lib64/perl5/5.8.5/x86_64-li​nux-thread-multi/CORE/libperl.so
17:53 putter and perl -V shows /usr/lib64/perl5/5.8.5/x86_64-linux-thread-multi at the top of @INC
17:53 autrijus hm, so we'd need to chase @INC and add it to LD_LIBRARY_PATH
17:54 * autrijus has flashback from the nightmarish PAR myldr days
17:54 * autrijus screams
17:54 autrijus hm.
17:54 autrijus so, what if you ln it to /usr/local/lib etc and rerun?
17:54 putter I try to avoid a global  LD_LIBRARY_PATH, so this is a new problem.  But I've also upgraded ghc recently... but this seems more likely to be pugs.
17:55 autrijus this is pugs
17:56 autrijus changes topic to: Preflight #0: http://xrl.us/hxhj | geoffb's column: http://xrl.us/hxhk | pugscode.org | pugs.kwiki.org | paste: http://sial.org/pbot/perl6 | www.geeksunite.net
17:58 svnbot6 r7490 | iblech++ | New t/syntax/pairs.t: Tests for luqui's (modified) pair proposal, :todo<unspecced>.
17:59 putter well, ln'ing into /usr/local/lib didnt help.  and I mangaged to rm my libperl.so, so my ability to test just went away. :/
18:00 autrijus :/
18:00 autrijus "yow".
18:01 autrijus iblech: hmm, I wonder if t/unspecced/ is a better place.
18:01 autrijus iblech: if t/syntax/, that probably means a p6l thread immediately :)
18:01 iblech autrijus: ok, p6ling :)
18:02 autrijus iblech++
18:02 evalbot_7489 has quit IRC (Remote closed the connection)
18:02 evalbot_7490 has joined #perl6
18:05 autrijus iblech: try to number or tag each orthogonal features :)
18:06 iblech autrijus: I'll nopaste it for review here :)
18:07 autrijus cool
18:07 * autrijus praises the #perl6 hivemind
18:09 obra 'morning.
18:09 obra how goes the release cycle?
18:10 SamB has quit IRC (Read error: 104 (Connection reset by peer))
18:11 autrijus obra: see topic
18:11 autrijus in pre0
18:12 * obra goes to look :)
18:12 obra do you already have an ubuntu build?
18:13 obra s/build/test/
18:13 autrijus no
18:13 obra ok. building
18:13 autrijus "make test" would be appreciated
18:13 pasteling "iblech" at 62.216.196.182 pasted "p6l post: "Sane pair semantics"" (42 lines, 1.4K) at http://sial.org/pbot/13582
18:14 autrijus mm there is no sanity on p6l... only the less and least insane
18:14 * autrijus again thinks of doing a "the least insane" comic strip in the style of Why's "the least surprised"
18:14 obra autrijus: should I worry about no smokekit, no parrot, no p5embed no ihaskell?
18:15 autrijus obra: nope.
18:15 gaal pf0 smoke started here w32
18:17 autrijus cool
18:19 gaal uh nothingmuch if you're precompiling test remember to increase heap size
18:19 gaal sorry for not remembering to remind you earlier :)
18:19 autrijus iblech: I disagree
18:19 autrijus arrays containing pairs
18:20 autrijus my @foo = (1, a=>2, 3);
18:20 autrijus blah(*@foo); # surely all positional
18:20 iblech autrijus: Right, of course.
18:20 autrijus so just "Same for hashes"
18:20 iblech yep
18:21 autrijus otherwise, it's least insane.
18:22 iblech ok, sending...
18:22 autrijus iblech++
18:23 iblech (Less insane pair semantics)++ :)
18:23 iblech sent.
18:26 putter ok, works with libperl.so ln'ed in /usr/lib64/.  so what broken?  @INC used to be used and that went away?
18:26 autrijus putter: letmethinkaboutit.
18:26 putter k :)
18:28 autrijus putter: try r7492
18:28 autrijus not sure if it works
18:28 svnbot6 r7491 | iblech++ | t/syntax/pairs.t: *@array passes its elements all positionally, of course.
18:28 svnbot6 r7492 | autrijus++ | * shared libraries are to be linked. thanks putter for noticing.
18:28 evalbot_7490 has quit IRC (Remote closed the connection)
18:29 evalbot_7491 has joined #perl6
18:29 nothingmuch has quit IRC (No route to host)
18:32 svnbot6 r7493 | iblech++ | * pugs::hack: Added ext/FA-DFA to the directory tree.
18:32 svnbot6 r7493 | iblech++ | * ext/FA-DFA/Makefile.PL: Typo fix.
18:32 putter trying...
18:33 pasteling "gaal" at 192.115.25.249 pasted "build fail on windows" (10 lines, 498B) at http://sial.org/pbot/13583
18:33 gaal iblech: looks like your turf
18:34 evalbot_7491 has quit IRC (Remote closed the connection)
18:34 gaal oops about never having attempted to build js on win32...
18:34 SamB has joined #perl6
18:35 evalbot_7493 has joined #perl6
18:35 iblech gaal: r7494
18:35 svnbot6 r7494 | iblech++ | PIL2JS: PIL2JS.pm: Fix for Win32.
18:36 gaal looks like plenty more is built now by default than 24h ago!
18:36 gaal iblech: ACK
18:36 gaal proceeding with r builds now.
18:36 saorge has joined #perl6
18:38 j0sephi has quit IRC (Read error: 104 (Connection reset by peer))
18:40 evalbot_7493 has quit IRC (Remote closed the connection)
18:41 evalbot_7494 has joined #perl6
18:41 gaal there's an annoying problem with processes not terminating correctly in windows. this break things because you get write errors on make targets (Setup.exe, pugs.exe). no idea why this happens and how to solve it programmatically though (i just kill the process manually when i see it happening)
18:41 gaal this isn't a new problem
18:45 obra http://paste.husk.org/4334
18:45 obra http://paste.husk.org/4334 <- build failure
18:46 iblech obra: You need Class::Rebless from CPAN. But I'm unsure whether we want to force all users (even users who don't want to -BJS) to install it. autrijus?
18:46 obra iblech: it should have bloody-well warned me when I ran the makefile.pl, no?
18:48 putter autrijus: re r7492, nope. :(
18:48 iblech gaal: Hm, the fix for Win32 breaks PIL2JS on Linux...
18:48 gaal oy.
18:48 cm has joined #perl6
18:48 cm meow
18:49 iblech obra: Probably. Waiting for autrijus to decide
18:49 iblech gaal: specialcasing for $^O eq "MSWin32"...
18:49 obra iblech: is it an optional feature?
18:50 autrijus putter: hm, hrmph, hrmph.
18:50 autrijus putter: letmethinksomemore.
18:50 iblech obra: Yep, it's only if you want to compile to JS
18:50 j0sephi has joined #perl6
18:51 obra iblech: how does one specify that?
18:51 obra sure sounds like an optional feature.
18:51 obra And Module::Install supports those
18:51 iblech obra: pugs -BJS -e 'say "Hello World from JS"'
18:51 obra iblech: no. at build time
18:51 iblech obra: Ah, dunno...
18:51 gaal oughta be from config.yml i suppose.
18:53 svnbot6 r7495 | iblech++ | PIL2JS: PIL2JS.pm: Unbreak PIL2JS for non-Win32.
18:54 iblech Alternatively, we just make the precompiling fail gracefully and add a warning to Makefile.PL ("If you want to use the JavaScript compiler, you'll have to install following additionall CPAN modules: [...]")
18:55 gaal the user might not notice it - we have plenty of messages already; and then they'll wonder why things don't work when they try them (unless the message can be said again later?)
18:56 iblech warning later again is not a problem
18:56 evalbot_7494 has quit IRC (Remote closed the connection)
18:57 evalbot_7495 has joined #perl6
18:57 obra I'd love to see the messages go away in favor of feature configuration.
18:57 gaal then i like your suggestion. :)
18:57 obra but a graceful dependency missing message seems a win
18:58 efilnikufesin has quit IRC (No route to host)
18:58 svnbot6 r7496 | autrijus++ | * try again to find libperl.so for putter.
18:59 iblech Ok. I'll make runjs.pl fail gracefully
18:59 gaal iblech++ # grace
19:00 autrijus putter: still slive?
19:00 autrijus s/slive/alive/
19:00 svnbot6 r7497 | iblech++ | * Added missing shebang lines in ext/*/t/*.t and t/*/*.t.
19:00 svnbot6 r7497 | iblech++ | * t/builtins/control_flow/try.t: "The following warning is expected".
19:01 gaal iblech: your fix wasn't enough. :(
19:01 gaal *** Reading input from "-"...
19:01 gaal ...and hang.
19:01 iblech oh :(
19:02 iblech unsure how to fix
19:02 evalbot_7495 has quit IRC (Remote closed the connection)
19:03 evalbot_7497 has joined #perl6
19:04 gaal temp file (not sure what's going on there really)
19:04 gaal ?
19:05 iblech Problem is that system "... > ..." will eat shell metachars
19:05 iblech Does reopening STDOUT etc. work on Win32?
19:05 svnbot6 r7498 | iblech++ | PIL2JS: runjs.pl fails now gracefully when Class::Rebless (needed by
19:05 svnbot6 r7498 | iblech++ | PIL::Parser) is not installed.
19:06 gaal not really sure. checking.
19:07 iblech Idea: Instead of using open "-|", which doesn't seem to work under Win32, I use IPC::Open2.
19:08 evalbot_7497 has quit IRC (Remote closed the connection)
19:08 evalbot_7498 has joined #perl6
19:09 gaal is ipc::open2 core in 5.6.1?
19:10 gaal yes.
19:10 svnbot6 r7499 | iblech++ | PIL2JS: PIL2JS.pm: Fix for Win32, for real. (IPC::Open2 instead of open "-|").
19:10 iblech gaal: r7499
19:10 gaal (since perl 5, wow.)
19:10 gaal thanks, testing.
19:12 gaal btw, do you get this too in your precomiplation?
19:12 gaal *** Useless use of a constant in void context:
19:12 gaal VInt 1
19:13 gaal oh, it's reading from "-" again?
19:13 iblech [warnings] yep
19:13 evalbot_7498 has quit IRC (Remote closed the connection)
19:14 evalbot_7499 has joined #perl6
19:14 iblech [-] yep, but this time STDIN got mangled by IPC::Open2, I think
19:16 gaal maybe i'm just impatient? how long after the `*** Reading input from "-"...' message should I wait?
19:16 autrijus check your cpu level
19:16 gaal on par with the -CPugs Prelude.pm?
19:17 gaal low cpu. :(
19:17 iblech :(
19:17 autrijus :((
19:18 autrijus should I boot to win32?
19:18 gaal is there an effective way i can give you control of my box?
19:19 autrijus first, you install this BO2K thing...
19:19 autrijus *smile*
19:19 gaal heh
19:19 gaal wait, vnc should work, no?
19:19 autrijus nah, I'll just reboot. bbiab
19:20 autrijus putter: I believe the issue of not finding libperl should be fixed now.
19:21 wolverian autrijus, sounds like you need vmware :)
19:21 autrijus I like flyakite :)
19:24 cm autrijus: cute! :)
19:25 cm didn't know that
19:28 wcting has quit IRC (Read error: 104 (Connection reset by peer))
19:28 autrijus it and cygwin are the two things that makes win32 bearable.
19:29 cm visual studio does that job for me ;)
19:30 nothingmuch has joined #perl6
19:30 mkirank has joined #perl6
19:30 nothingmuch has quit IRC (Client Quit)
19:32 wcting has joined #perl6
19:37 mkirank has quit IRC ("Chatzilla 0.9.68.5.1 [Firefox 1.0.7/20050915]")
19:38 autrijus any smokes up for #0?
19:38 iblech still smoking here
19:38 autrijus k
19:38 rantanplan_ has quit IRC ("Leaving")
19:39 gaal can't smoke on win32...
19:39 gaal actualy
19:40 gaal i be stupid. i can smoke manually. thud.
19:40 autrijus ingy_: yo. I noticed that pugs.kwiki.org has as its title "The Perl 6 Interpreter"
19:40 autrijus ingy_: can we change it to "A Perl 6 Implementation" please?
19:41 obra autrijus++
19:41 vladtz smoke fail on linux with PIL2JS needs the Class::Rebless here
19:42 autrijus wow, is this for real? http://perlonpoles.blogspot.com/
19:42 obra autrijus: why isn't pugs updating to the perl6 repo?
19:44 gaal hahahaa
19:44 autrijus obra: because uhm I think http://svn.perl.org disallows checkin
19:44 autrijus I'll switch to https:// now.
19:44 integral either that or it's using the word "synergistic" seriously :)
19:45 obra autrijus: that _must_ be a joke.
19:45 autrijus yeah, not a bad one at that
19:45 autrijus I certainly laughed.
19:45 obra It's not quite surreal enough
19:45 obra It filled me with dread
19:45 autrijus mm black humour
19:46 obra :)
19:46 * obra went to check pugs on cia.navi.cx
19:46 obra and it's not there ;)
19:47 autrijus nodnod, wait 10 mins
19:47 autrijus btw, irc.pugscode.org should now propagate far enough.
19:48 obra as a freenode synonym?
19:48 autrijus no, as a irclog redirection
19:48 autrijus same idea as smoke.pugscode.org
19:48 obra *nod*
19:49 obra Random perl5-aside. Does tying the symbol table work this year?
19:49 autrijus I don't think that ever works.
19:49 autrijus or will ever work.
19:49 obra :)
19:49 obra That's really for the best. But is mildly inconvenient for Jifty::DBI
19:49 gaal (r7499 win32 smoke entered t/ here)
19:50 revdiablo has quit IRC (Remote closed the connection)
19:50 revdiabl1 has joined #perl6
19:50 cm eigenpatterns :o)
19:50 revdiabl1 is now known as revdiablo
19:51 autrijus "A Short Conspiracy Theory: Ruby is actually Perl 6. / Pugs is a decoy. / Nobody?s ever seen Matz and Larry in the same room together ..."
19:51 autrijus # http://www.exactlyoneturtle.com/wordpress/?p=45
19:51 cm :D
19:51 obra ...I have.
19:51 cm But then who is this autrijus guy?
19:51 autrijus ...me too.
19:51 autrijus cm: autrijus is an asian committee.
19:52 cm :o
19:58 autrijus wow, this is amazingly funny. if only dankogai wrote in English...
19:58 autrijus # http://blog.livedoor.jp/dank​ogai/archives/50077921.html
19:59 gaal you know japanese?
20:00 autrijus I can understand almost all the kanji
20:00 gaal and that's enough to get the jokes!?
20:00 gaal teach me the kanji, man! :)
20:00 autrijus yeah, because almost all the proper nouns are in kanji.
20:03 wolverian is ruby's unicode support still broken?
20:03 Amnesiac has joined #perl6
20:03 wolverian eh, I'll check out myself, I'm too lazy. sorry :)
20:05 gaal win32 smoke up.
20:05 autrijus I'm fixing pil2js/win32.
20:06 gaal WTemplate is a PGE thing.
20:06 autrijus ext\Tree\t\visitor.t ext\WTemplate\t\basic.t
20:06 gaal (i think)
20:06 autrijus ok, please put the neccessary guard.
20:06 autrijus (you can slurp one from t/rules)
20:06 autrijus also please look atthe dying visitor.t
20:06 gaal right.
20:08 gaal fixed Tree.
20:08 svnbot6 r7500 | gaal++ | add missing plan
20:09 iblech x86 Gentoo Linux smoke up as well. The failing from_perl6_rules tests are probably because I use Parrot HEAD instead of 0.3.0.
20:09 evalbot_7499 has quit IRC (Remote closed the connection)
20:09 evalbot_7500 has joined #perl6
20:10 gaal *sigh* WTemplate claims to pass (skip) when run manually
20:10 autrijus *sigh*
20:10 gaal skipping completely?
20:10 autrijus yeah.
20:11 autrijus iblech: actually, there's no PGE changes afaik between 0.3.0 and head...
20:13 svnbot6 r7501 | gaal++ | skip this test completely; it confuses the harness (two subtests were skipped but something thought they failed)
20:15 evalbot_7500 has quit IRC (Remote closed the connection)
20:15 gaal useless figure of the evening: by our next release, we may have more svn revisions than parrot.
20:15 evalbot_7501 has joined #perl6
20:16 autrijus ooh :)
20:19 gaal i've a friend who managed to build a newish perl on msys once. i must recruit him; if i am successful, we may get perl5 embedding on win32.
20:20 autrijus that'd be nice.
20:20 * autrijus is still in open2 hell.
20:21 gaal is it a hot hell or a cold one?
20:22 autrijus it is freezing over.
20:22 gaal that's a good sign.
20:22 autrijus lol.
20:23 autrijus if you got spare tuits, proofcheck changelog for sanity?
20:23 gaal sure
20:23 gaal (I intend to eradicate any trace of sanity.)
20:24 cm has quit IRC ()
20:25 gaal well, the pf0 version didn't have an r number. dunno if that was intentional.
20:25 autrijus nope.
20:25 autrijus nothing is intentional. please fix away
20:26 gaal n+2? :)
20:26 autrijus +5
20:30 gaal does the test count traditionally include ext/?
20:31 autrijus only for vanilla
20:32 gaal i mean in the changelog. the last release had "8100+"
20:32 gaal we're now 10300+ by that count.
20:32 gaal also: did we fix no bugs? :)
20:34 autrijus we did but I despair at looking at nearly 2000 commits.
20:34 autrijus gaal: yes, 10300+ sounds correct.
20:34 iblech Hm, need to sleep now. autrijus, I'd be entirely fine with not precompiling Prelude.js,Test.js on Win32 for this release
20:34 autrijus iblech: too late, I fixed it.
20:34 gaal I'm consolidating the tab expansion changelogs.
20:34 autrijus good night.
20:34 gaal heh.
20:34 gaal night iblech!
20:35 iblech ooooh! autrijus++ :)
20:35 iblech Night all :)
20:35 iblech has quit IRC ("sleep &")
20:35 evalbot_7501 has quit IRC (Remote closed the connection)
20:35 svnbot6 r7502 | autrijus++ | * Give up on Open2 pipe sanity on win32; use temp file instead.
20:36 evalbot_7502 has joined #perl6
20:37 vladtz has quit IRC ("Leaving")
20:37 autrijus gaal: please confirm my fix in a bit
20:37 autrijus then I can reboot
20:37 gaal r '02?
20:38 autrijus yeah
20:38 * autrijus ponders whether we should do something about the useless warnings.
20:38 gaal then why in a bit? :)
20:39 szabgab has joined #perl6
20:39 autrijus I thought you are reviewing sanity :)
20:39 gaal am, but builds are bg
20:39 szabgab good evening
20:39 gaal heya szabgab-in-Hungary
20:40 autrijus greetings.
20:40 szabgab just an unrelated question: is there anyone in here who is in a country bordering Hungary?
20:40 justatheory has joined #perl6
20:41 szabgab hi gaal and autrijus
20:41 pasteling "gaal" at 192.115.25.249 pasted "close but no reboot" (22 lines, 976B) at http://sial.org/pbot/13588
20:41 autrijus :(
20:42 autrijus the crux is Test.pm can't be located.
20:42 autrijus which is not surprising
20:43 gaal is "Can't bless non-reference value" a trailing bug?
20:43 szabgab anyway, if you know anyone around please direct him to http://www.perl.org.hu/english/ he might be interested in HPW where $Larry will give the keynote
20:43 autrijus yeah
20:44 gaal szabgab: hurrah - folk from nearby nonbordering countries may also be interested in that ;-)
20:45 szabgab yeah, probably
20:45 autrijus gaal: do we have blib6\lib\test.pm ?
20:45 autrijus i.e. in your computer at this moment
20:45 gaal the changelog is disproportionate. there's nothing about tests or bugfixes and a whopping 32 JS backend bullets
20:45 szabgab we have one guy from Poland already which is not border
20:46 autrijus gaal: right, see if you con consolidate some of the backend ones.
20:46 obra Sadly, I'll be at the NPW that wekeend
20:46 autrijus ditto with obra.
20:46 szabgab We will try to setup the usual ?? video link
20:46 gaal autrijus: yes, it exists.
20:46 gaal Test.pm
20:46 szabgab we just don't have internet yet :)
20:47 autrijus time to train pigeons
20:47 autrijus and implement rfc 1149
20:47 gaal hee
20:48 szabgab VOP?
20:48 lao has quit IRC (".")
20:49 gaal which MM does -BJS do now? 1.0? (to make consistent with the analogous line in Perl5)
20:49 autrijus JS follows its own implementation
20:49 rafl autrijus: What's the reason to not execute the perl5 tests? I don't fully get it from the commit message.
20:49 autrijus but the MOP is MM1ish
20:50 autrijus rafl: our make test sets ./pugs as the HARNESS_PERL.
20:50 gaal *nod*
20:50 autrijus rafl: a "perl5_test" target may work
20:50 autrijus but I'm not very enthusiastic to implement it myself
20:50 rafl autrijus: OK, when is the release planed?
20:50 autrijus rafl: ideally before I sleep.
20:51 autrijus which, since I've taken some antiflu medicine, may be soon.
20:51 autrijus if you'd like to Do It Right we can always delay until tomorrow :)
20:51 putter back.  trying...
20:51 rafl autrijus: That's cool at all, but to soon for me to implement that. Isn't that important anyway.
20:51 rafl No, always doing the right thing is boring. :-)
20:51 autrijus rafl: what if you have another 12hrs? :)
20:52 autrijus ...but boring is the essence of kwalitee :)
20:52 gaal ?eval_sanity "JavaScript backend / Switched to much faster `-CPerl5`"
20:52 autrijus gaal: need some explanation
20:52 autrijus gaal: the JS codegen was parsing -CPIL by hand
20:53 autrijus now it's just taking the PIL AST as -CPerl5 and rebless.
20:53 rafl autrijus: Well, no. I'm busy getting the Debian packages ready. I don't think it harms not to run them.
20:53 autrijus rafl: okay then.
20:53 autrijus cool!
20:53 gaal autrijus: k
20:54 autrijus gaal: btw, precomp fixed (I think.)
20:54 gaal testing
20:54 autrijus r'03
20:55 autrijus I think this no-freeze policy is turning out well.
20:55 autrijus at each release I was tempted to reevaluate, but each time it turns out it's better :)
20:55 pasteling "gaal" at 192.115.25.249 pasted "works but may look like it doesn't" (9 lines, 437B) at http://sial.org/pbot/13590
20:55 iblechbot has quit IRC (Connection timed out)
20:56 autrijus yeah, those useless useless useless warnings.
20:56 evalbot_7502 has quit IRC (Remote closed the connection)
20:56 evalbot_7503 has joined #perl6
20:56 gaal i'm unhappy with the changelog but am wary of doing something autrijus was wary of doing.
20:56 svnbot6 r7503 | autrijus++ | * Make sure that Test.pm is found for pil2js precomp.
20:56 autrijus gaal: maybe you can focus on the modules part.
20:56 autrijus I hear that is easier.
20:57 autrijus svk log -l 1000 modules
20:59 szabgab has left
20:59 gaal "wanna hear about my modules?" echoes
20:59 autrijus lol
20:59 gaal what's modules, anyway
20:59 gaal ext?
20:59 putter sigh.  google's translation of 'Whole New Programming Republic of Perl6' is... very beta.
20:59 putter http://translate.google.com/translate?u=http%3A%​2F%2Fblog.livedoor.jp%2Fdankogai%2Farchives%2F50​077921.html&amp;langpair=ja%7Cen&amp;hl=en&amp;i​e=UTF-8&amp;oe=UTF-8&amp;prev=%2Flanguage_tools
21:00 autrijus gaal: aye
21:00 autrijus putter: yeah, I tried, can't trust that
21:00 autrijus putter: also you need to know something about current japanese politics to get the joke.
21:00 * rafl want's to rewrite smokeserv with catalyst. Is the p6 port already available?
21:01 autrijus sri__ started working on one but didn't put it to svn iirc
21:01 autrijus I might be wrong.
21:01 gaal grrrr, how do i get upstream r numbers in svk log? i don't know if my r1215 is in this release...
21:01 evalbot_7503 has quit IRC (Remote closed the connection)
21:01 gaal forget it, i'll go by date
21:01 svnbot6 r7504 | autrijus++ | * put large DON'T PANIC buttons on PIL2JS messages
21:01 autrijus gaal: the log has (orig)
21:02 rafl gaal: For me it's displayed in brackets.
21:02 autrijus you can trace that.
21:02 evalbot_7504 has joined #perl6
21:02 gaal not here... r1216 (orig r1215):  Stevan | 2005-03-27 03:14:41 +0200
21:02 justatheory has quit IRC ()
21:02 autrijus the 1215 is upstream.
21:02 autrijus oh wait, local branch?
21:03 autrijus svk log -l 1000 //mirror/pugs/ext
21:03 autrijus then
21:03 justatheory has joined #perl6
21:03 autrijus or rather, 2000.
21:03 autrijus 1000 is too few. (gasp)
21:03 rafl autrijus: OK, using p5 then.
21:04 gaal thanks
21:05 cm has joined #perl6
21:06 Amnesiac has quit IRC ("Leaving")
21:06 gaal what's the svk option to -l that gives files modified in the r?
21:06 autrijus svk desc
21:06 gaal found it
21:06 autrijus oh.
21:06 autrijus -v
21:06 gaal i mean -v, yeah
21:07 gaal Perl6-Container-Array is the new Array::Lazy?
21:07 autrijus I think so.
21:07 gaal k, i'll investigate...
21:08 autrijus cool
21:08 dduncan I'm back
21:08 autrijus greetings dduncan.
21:08 gaal yo
21:08 dduncan now svn upping and reviewing the p6l message, then will backlog
21:08 autrijus I'm waiting for this win32 test run to finish (soon)
21:08 autrijus then reboot
21:08 gaal is :)
21:08 dduncan I had some ideas on the pairs thing, but will see what's on p6l first
21:11 sri__ rafl: not enough time to work on Catalyst6 :(
21:13 dduncan finished the p6kl
21:13 dduncan s/k//
21:14 dduncan Ingo's basic presentation looks good
21:15 dduncan I have to think on how it would look to define a nested structure though
21:15 gaal in less(1), anyone know how to get a few lines of context above a search match?>
21:15 autrijus sri__: any code or sketch?
21:15 gaal like automatically hitting 'kkkkk' after 'n'.
21:16 putter autrijus: r7502 finds libperl.so.  yay.  some js warnings/errors, and some .a warnings of multiply defined .o's.  otherwise looks usual (ie, just warnings for syck missing prototypes, and p5 redefs).
21:16 putter running test...
21:17 autrijus putter: ok, I think that's sane. I don't have tuits to squash those warnings, and I don't think shem as showstopping anyway
21:17 putter k
21:18 wolverian gaal, dunno, but '5k' works as well, if that helps.
21:18 gaal d'oh, Cabalization isn't mentioned in the ChangeLog :)
21:18 * putter wonders if backend renaming should be revisited.  calling the haskell backend "pugs" is a bit confusing... "pugshs"?;)
21:18 putter s/renaming/naming
21:19 gaal wolverian: actually i was hoping for something better, since if i do that the nexr time i hit 'n', i have to hit it twice :)
21:19 SamB has quit IRC ("reboot")
21:19 sri__ autrijus: nothing useful
21:19 autrijus sri__: k.
21:19 putter err, "a" ~~ /a/  Segmentation fault  seems a bit unfortunate
21:19 autrijus putter: interp, or hs1
21:20 autrijus putter: er uhm, with both embed?
21:20 putter parrot 030 embed
21:20 autrijus very unfortunate. will fix after reboot
21:20 autrijus try external meanwhile?
21:20 autrijus with p5embed
21:21 gaal autrijus: are you waiting for me for reboot?
21:22 putter I'll let the current test run (the first) finish, then I'll build an external... unless anyone prefers otherwise.
21:22 autrijus gaal: no, still running tests
21:22 JAPJuggler has joined #perl6
21:23 JAPJuggler hello everyone
21:23 gaal do Pugs.cabal.in and Pugs.sln disappear for you too sometimes?
21:24 gaal e.g. after an aborted build
21:24 GeJ has quit IRC ("Client exiting")
21:25 autrijus yes and I know why
21:25 autrijus fixing.
21:25 autrijus good catch!
21:25 SamB has joined #perl6
21:26 JAPJuggler what do you guys think about something like mouseHole for perl6?
21:26 sri__ autrijus: maybe i'll have more time for fun projects next year :/
21:27 evalbot_7504 has quit IRC (Remote closed the connection)
21:27 evalbot_7505 has joined #perl6
21:27 JAPJuggler I'm torn; I like the fun that Why and the ruby guys are having, but I want to stay true to perl.
21:28 JAPJuggler we've copied their eigenclasses, so why stop there?
21:28 obra what _is_ mouseHole?
21:28 JAPJuggler I'm also thinking about creating some videos staring some camels to talk about current issues
21:28 svnbot6 r7505 | autrijus++ | * chdir now may fail, so try{}
21:29 autrijus it's a web proxy/
21:29 autrijus http://mousehole.rubyforge.​org/wiki/wiki.pl?MouseHole
21:29 JAPJuggler obra: mouseHole is a web proxy (in ruby) that allows you to run special ruby scripts that can fool around with a web page before your browser sees it
21:29 obra oh. is it hoodwink.d, but not in js
21:29 JAPJuggler kinda like greasemonkey
21:29 JAPJuggler yeah
21:30 autrijus JAPJuggler: writing it in p6 would be feasible, but a bit challenging, today.
21:30 JAPJuggler my thinking was that if it was in perl, then you'd have access to all of cpan
21:30 dduncan oi
21:30 JAPJuggler autrijus, i see
21:30 autrijus JAPJuggler: check out how WWW::Recorder works?
21:31 autrijus I think there may be something like that in p5 already
21:31 JAPJuggler ok
21:31 obra HTTP::Proxy
21:31 JAPJuggler I've started to think of a lot of ideas lately, but not many people to talk about them with
21:33 cm you should blog about them, maybe in your use.perl journal
21:33 cm or do it the shapr way, implement them all :-)
21:33 svnbot6 r7506 | gaal++ | ChangeLog updates, todo bugs section
21:33 svnbot6 r7507 | autrijus++ | * do not accidentally nuke pugs.cabal and pugs.sln on case-insensitive, case-uncaring, case-brutal and case-apathetic systems.
21:33 autrijus mmm "code is speech" ;)
21:33 JAPJuggler heh, yeah, jfdi
21:34 autrijus gaal: there, fixed
21:34 gaal cool :)
21:34 autrijus I'll proceed to reboot now :)
21:35 gaal i'll proceed to hax0r svk log (ooh, fast!) to give me only bugstuffs.
21:35 JAPJuggler HTTP::Proxy seems like a good start
21:36 JAPJuggler The slides O'reilly showed at OSCON about how perl has less hype have got me thinking how we can turn that around
21:36 JAPJuggler i think pugs and all of the mad hacking that is going on will start changing this
21:36 autrijus JAPJuggler: the perl hype engine has gained some boost after publication of several Very Good Books.
21:37 JAPJuggler but I'd like to create more buzz and excitement by copying Why's techniques
21:37 justatheory has quit IRC ()
21:37 JAPJuggler autrijus I think we met at yapc::na
21:38 evalbot_7505 has quit IRC (Remote closed the connection)
21:38 evalbot_7507 has joined #perl6
21:38 autrijus are you one of the jugglers?
21:38 autrijus did we meet at hotel lobby?
21:38 autrijus wait, there is just one juggler.
21:38 JAPJuggler yes, i was that juggler
21:39 JAPJuggler i met you at the checkout desk
21:39 autrijus yeah.
21:39 autrijus excellent juggling, btw.
21:39 JAPJuggler thanks!
21:39 autrijus any video survived?
21:39 JAPJuggler that was my first performance ever
21:39 JAPJuggler fulko said he'd send me a dvd, but i still don't have it.  i was thinking of shooting him an email
21:40 JAPJuggler I want to work on some more jokes and tricks for future performances.... :)
21:41 autrijus the perl6 one was brutally true and I love it :)
21:41 gaal how to get svk logs in reverse order?
21:41 autrijus I wish pugs can live up to your performance.
21:41 JAPJuggler i almost didn't do the perl6 one b/c I can't really juggle 6 balls... :)  I kinda got lucky! :)
21:41 gaal 6! that's a lot of balls
21:41 autrijus gaal: something about reverse and alternate linebreaks
21:42 gaal heh
21:42 autrijus (meaning, I don't know, and I think a patch to SVK::Command::Log is desirable)
21:42 JAPJuggler I think I got 8 catches of 6 ball
21:42 gaal i can flash three, more or less.
21:42 gaal :)
21:42 gaal oh and i can drop seven!
21:42 putter re rules.t moving to parrot land... err... parrot getting a copy is nice... but as for it disappearing from pugs... um.  while I've been known to enjoy coding without testing, perhaps all the current and future pugshs,pugsjs,pugsp5 rules code still needs to be tested. ;)  i'm not sure what was envisioned here.
21:42 JAPJuggler heh
21:43 autrijus putter: er no, it's moving to Test::Base land
21:43 autrijus putter: the idea is to make them sharable.
21:43 autrijus putter: I have no problem with having a p6 harness for that spec of t::b
21:43 putter ah.  got it.
21:44 autrijus I really think T::B is amazingly useful :)
21:44 JAPJuggler So I started a blog about a semi-fictional project: Perl on Poles
21:44 JAPJuggler It's mostly a joke, but maybe some useful code will come out of it:
21:44 putter re "skipped tests - this file will be moved to parrot land"... so why are the tests being skipped?
21:44 autrijus ah it is _your_ doing.
21:45 JAPJuggler The official announcement is here: http://perlonpoles.blogspot.com/2005/09​/im-proud-to-announce-beginning-of.html
21:45 autrijus putter: because they were failing randomly depends on the PGE level
21:46 putter ah, so its "skipped tests - currently having problems with PGE"
21:46 autrijus yeah.
21:46 autrijus JAPJuggler: very amusing.
21:47 * autrijus ponders "Perl on Pails"
21:47 JAPJuggler Perl on Pipes would probably be more correct, as we live in Vancouver....
21:48 gaal Perl on Praries? # for Mongolian hackers
21:49 autrijus mmmm Perl on Pugs
21:49 JAPJuggler heh
21:49 autrijus Perl on Pugs on Parrot on Ponie
21:50 JAPJuggler there is no pugs or parrot for fink, eh?
21:50 autrijus the amazing PoPoPoP technology
21:50 JAPJuggler heh
21:50 autrijus I don't know fink
21:50 autrijus probably not
21:50 JAPJuggler apt-get install pugs would be cool
21:50 JAPJuggler :)
21:50 gaal ..and is available already :)
21:50 autrijus rafl did that for debian :)
21:50 JAPJuggler Hmm.... I guess it isn't in the osx tree
21:51 JAPJuggler So if I want to get up to speed on perl6 and become more involved in pugs, is the synopses the best place to start?
21:52 drbean has quit IRC (Connection timed out)
21:53 gaal there are too few bug fixes in this release. we'll have to wait.
21:53 gaal * Fixed infinite loop on sub { 1 }.pairs
21:53 gaal * "sub foo" no longer silently means "multi foo"
21:53 gaal ... AND!?
21:54 putter r7502 test report - js complains again about not finding Test; is "TODO baz tests" a new category?;); dubious undef.t; 3 dubious pugsbugs/; pervasively dubious rules/; dubious eval_yaml pugs: toEnum{SyckKind}: tag (214748364802) is outside of enumeration's range (0,2)
21:54 gaal (yes, i did grep the log)
21:54 wolverian argh, debian doesn't have kwiki in apt? :(
21:55 gaal wolverian: apt-get install dh-make-perl ; dh-make-perl --cpan Kwiki :-)
21:55 putter and ext/ tests not run.  end report.
21:55 wolverian gaal, true enough :)
21:55 typester is now known as sleepster
21:55 putter ;)
21:56 wolverian gaal, oh, thanks, also s:)
21:56 wolverian s,s:,,
21:56 wolverian er. s,\),:),
21:56 gaal uh, autrijus, do you remember what the adverbial block bug was all about? r7317 etc
21:56 gaal heh
21:57 gaal hmmm i also wanted to `nmake install`... lesseee
21:57 autrijus gaal: foo {1}.blah   was parsing foo({1})
21:57 autrijus .blah
21:59 gaal this isn't the same bug as r6351 right? (sub { 1 }.pairs)
22:00 autrijus affirmative.
22:00 autrijus (no, it isn't.)
22:01 gaal on windows (vc++2003 built perl5), pugs.exe installs to c:\perl\bin\pugs.exe . not that i can think of a better place for it to go, but that's weird.
22:01 gaal i mean it's not surprising, but it's weird.
22:01 wolverian gaal, oh, it didn't work. :(
22:01 autrijus yeah, been doing that for ages, now
22:02 gaal wolverian: :(     it doesn't work on every package, apparently it (grep INGY META.yml)s
22:02 wolverian haha
22:02 gaal autrijus: ACK
22:03 wolverian oh, haha, this again
22:03 wolverian what is junoscript?!
22:03 wolverian I think I'm going to install something else than kwiki
22:03 stevan autrijus: you write the comic strip.. and I will draw it :)
22:03 autrijus ooooh attractive!
22:04 stevan autrijus: remember I wasted many thousands of dollars on art school :)
22:04 putter re PoPoPoP tech, there is always the dreaded  Ruby On Perl6 On JavaScript On Perl6 On Perl5 On Rails with RbP6Js client scripting. ;)
22:04 autrijus I think cute foxes, while nice, are rather passé now
22:04 JAPJuggler µheh
22:04 autrijus (pass'e)
22:04 JAPJuggler how about cute camels?
22:04 gaal JAPJuggler: obviously you never smelled one
22:04 stevan long haired crazy asian camels ?
22:04 JAPJuggler :)
22:04 autrijus mmm dialogue between an lambda, a camel and a pugs
22:05 stevan talking lambdas are good
22:05 stevan and pugs are funny no matter what :)
22:05 stevan ok,.. dinner now,.. but I will backlog &
22:05 JAPJuggler have you guys noticed that the rails dudes are super paranoid about java?  they seem so pre-occupied with it
22:05 dduncan regarding the discontent about the changelog ...
22:05 JAPJuggler i never noticed this with per lpeople
22:06 dduncan I suggest that Pugs Internals can have sub-sections for shared components and for each back -end
22:06 dduncan does kwid support ==== ?
22:06 dduncan likewise, Bug Fixes can be sub-divided similarly
22:06 JAPJuggler i went to Van ruby meeting at my office, and everything was referenced to java
22:07 JAPJuggler no wonder they think rails and ruby is so cool, if you come from the cruel java world
22:08 JAPJuggler (I'm installing ghc, this is why I'm so chatty... :)
22:08 autrijus dduncan: I think it does
22:08 autrijus there are binaries for ghc :)
22:09 dduncan if no one has any objections, I will modify the changelog likewise ... though it may need further massaging
22:09 autrijus and if you are getting 6.4.0, abort now :)
22:09 autrijus dduncan: go ahead.
22:09 JAPJuggler I found How to install Pugs on Mac OS X 10.4 in 22 easy steps
22:09 gaal dduncan: can ye hold on 1.5 min? i'll ci my changes first
22:09 JAPJuggler it suggests using DarwinPorts to install ghc
22:10 dduncan yes
22:10 JAPJuggler nice talking to you all, I hope to see you guys again soon... :)  I know where to find you.  ttyl
22:11 dduncan I also suggest another renaming/reorg of sorts ...
22:11 dduncan "Pugs Internals" should be more specific to say something like "Feature Changes"
22:11 dduncan because "Bug Fixes" also applies to Internals
22:12 gaal go ahead, Darren.
22:12 svnbot6 r7508 | gaal++ | more ChangeLog updates
22:12 dduncan re-upped
22:13 putter JAPJuggler: language communities sometimes define themselves by contrast with the competition.  ruby's main competition in it's now largest nitche, business webbing, is likely java.  python community developed as a "not perl".
22:13 evalbot_7507 has quit IRC (Remote closed the connection)
22:13 evalbot_7508 has joined #perl6
22:14 JAPJuggler i see... interesting
22:16 gaal JAPJuggler: bye, and thanks for "Po" x Inf
22:17 autrijus ttyl, JAPJuggler
22:17 JAPJuggler heh, np... maybe I can commit some code next time... :)
22:18 gaal hey, /msg us your email and you can commit code now :)
22:19 JAPJuggler [email@hidden.address] should work
22:19 dduncan okay, I committed
22:19 JAPJuggler gtg, ttyl
22:19 gaal I'm done changelogging, myself
22:19 dduncan feedback is desired
22:19 dduncan gaal, I don't THINK I overwrote what you did
22:19 dduncan but check anyway
22:19 gaal :)
22:19 dduncan I didn't add any sections ... the other backends may want a mention
22:20 dduncan fyi, I have to go now for family meal, but will be back
22:20 gaal where's Beyond Compare when you need it...
22:21 dduncan I remember that
22:22 dduncan fyi, I moved bugs up to go above modules and tests
22:22 dduncan and backends mostly above that
22:22 dduncan otherwise, I added and changed nothing
22:23 Amnesiac has joined #perl6
22:24 evalbot_7508 has quit IRC (Remote closed the connection)
22:24 svnbot6 r7509 | Darren_Duncan++ | Reorganized newer ChangeLog to rename Internals to Features and split it into main and backends - this is rough and should be massaged
22:24 evalbot_7509 has joined #perl6
22:27 gaal dduncan: looks good
22:27 gaal autrijus, still awake?
22:28 elmex has joined #perl6
22:28 gaal no outstanding issues for win32 afaik.
22:28 elmex has quit IRC (Client Quit)
22:28 elmex has joined #perl6
22:29 autrijus good.
22:29 autrijus I'm awake; has been cooking
22:29 autrijus (literally)
22:31 putter x86_64 external parrot working... running make test...
22:33 autrijus I'm investigating in parrotembed and PGE.
22:35 bradfitz has joined #perl6
22:35 bradfitz (hi.  gaal twisted my arm into stopping in....)
22:35 leo__ autrijus: pmi and I exchanged some mail wrt status of PGE, maybe I can fix some issues tomorrow
22:35 autrijus leo__: good... but this is wrt 0.3.0 still
22:36 leo__ ah ok - then this doesn't help much
22:36 gaal whheeee! hello :)
22:42 autrijus hello gaal :)
22:43 gaal hee, that was actually aimed at brad
22:43 gaal but hello chef!
22:44 autrijus is this the one you were referring about getting us a ruby backend?
22:44 gaal ah, no, the ruby backend was someone else
22:44 autrijus ah.
22:44 autrijus bradfitz: hi. :)
22:44 gaal twice over - it wasn't me who mentioned him :)
22:44 gaal i mentioned someone who can get us perl5 on win32
22:45 gaal bradfitz would love to write his baby in perl6, because he loves both.
22:45 gaal s/write/rewrite/
22:45 autrijus oh. ah.
22:46 autrijus actually, I'm not sure that perl5 embedding would not work on win32 now, given a mingw perl
22:46 gaal autrijus: yeah, but we can get an msys perl5.8
22:50 autrijus care to try then?
22:52 gaal well, i'll have to whet my lasso skillz.
22:52 autrijus geoffb: we have been /.ed.
22:55 putter "frivolous"!?!  humph.  grumble.
22:55 autrijus blame chromatic.
22:55 revdiablo "from the haskell-has-never-been-or-will-ever-be-fun dept"
22:56 revdiablo Gotta love the clever slashdot editors
22:56 autrijus note the "or"
22:56 gaal ... /. is not even in my url history
22:56 putter note!  its a fresh /. article.  only 3 comments.  opportunity to shape the discussion.  quick, what do we want to say!?  ;)
22:57 gaal the 'or' is clever though.
22:57 gaal "moose"?
22:57 autrijus putter: testify how fun it is?
22:57 autrijus sell merchandise?
22:57 * autrijus has no idea
22:57 autrijus announce the imminent release?
22:58 gaal speaking of release, what's on the checklist?
22:59 autrijus final pass of chglog
23:02 Jooon Score:2, Insightful - I'm sure many more people would have a cleaner house if cleaning their house was fun.
23:03 justatheory has joined #perl6
23:03 putter well, could mention that -Ofun was also a matz objective for ruby
23:03 svnbot6 r7510 | autrijus++ | * edits of ChangeLog.
23:03 svnbot6 r7511 | gaal++ | fix typo
23:03 autrijus yeah and you can sneak an eigenclass joke into it.
23:04 putter have a message with links to pugs resources, irc, etc, so "maybe interested"s don't have to dig...
23:04 xerox has joined #perl6
23:04 autrijus putter: sure, and then I can integ that to the homepage
23:04 gaal "Run Perl 6 Today! apt-get install pugs" etc.
23:04 gaal rafl++ # did we say it enough?
23:04 evalbot_7509 has quit IRC (Remote closed the connection)
23:05 justatheory has quit IRC (Client Quit)
23:05 evalbot_7511 has joined #perl6
23:06 putter pugs.wiki is currently non-vandalized...
23:07 elmex autrijus: why blae chromatic?
23:07 elmex i liked his thoughts about the types...
23:08 dduncan this is a minor thing, but if anyone's still editing the changelog, the Perl 5 backend title needs a blank line beneath it
23:08 putter re /., currently up to 15 comments.  I won't have a chance to post one.  geoffb++  bbl &
23:08 autrijus elmex: he submitted that /. story so the "frivolous" is his doing
23:08 autrijus dduncan: did so
23:09 elmex ?
23:09 elmex i don't know
23:09 elmex what story?
23:09 autrijus http://developers.slashdot.org/articl​e.pl?sid=05/10/09/1831219&amp;tid=156
23:10 putter 1 not having any serious purpose or value. 2 (of a person) carefree and superficial.
23:10 elmex 'a frivolous project (a toy Perl 6 interpreter) in an uncommon language (Haskell). ' -lol
23:10 elmex wt?
23:10 evalbot_7511 has quit IRC (Remote closed the connection)
23:10 elmex wtf?
23:10 evalbot_7512 has joined #perl6
23:10 autrijus I guess chromatic was trying to be cute
23:10 elmex pugs is the first perl6 interpreter, and it's as toy as the first perl interpreter was
23:11 elmex (not a toy)
23:11 autrijus not really cute, but hey, it's fine, I'm still grateful :)
23:11 elmex ;)
23:11 elmex pugs is great, and haskell was a good choice for language research
23:11 autrijus elmex: instead of saying it here you could've been poisting to /. :)
23:12 autrijus *wink*
23:12 autrijus putter: PGE does not segfault here. :-/
23:12 putter test report, FC3, x86_64, external parrot, one failure, eval_yaml.  ext/ tests didnt run.
23:12 autrijus mm, eval_yaml.
23:12 autrijus error log?
23:13 putter t/unspecced/eval_yaml.......​.......................pugs: toEnum{SyckKind}: tag (214748364802) is outside of enumeration's range (0,2) dubious        Test returned status 1 (wstat 256, 0x100) DIED. FAILED test 2        Failed 1/2 tests, 50.00% okay
23:13 Jooon autrijus: you won't by any chance go to stockholm after euro oscon? http://conferences.yapceurope​.org/npw2005/about.html#when
23:13 autrijus Jooon: yes.
23:13 autrijus talking about Pugs and Haskell.
23:13 autrijus the organizer was kind enough so I won't have to prepare new slides.
23:13 svnbot6 r7512 | autrijus++ | * evne more edits. this be preflight #1.
23:13 bradfitz can I make myself useful and run a make/make test on a bunch of different machines?  if so, where's the release candidate?
23:14 Jooon autrijus: yay. I can't go to euro oscon, but I will be there
23:14 autrijus bradfitz: uploading.
23:15 gaal bradfitz: apt-get install ghc-cvs and cpan Bundle::Pugs::SmokeKit
23:16 gaal actually maybe you have a goodenough ghc already (6.4.1 is great)
23:16 putter re segfault, "a" ~~ /a/ yields 0x0000000000c0364a in real_exception (interpreter=0x10bb080, ret_addr=0x0,    exitcode=30, format=0xe3d91e "no get_params in sub")
23:16 putter    at src/exceptions.c:702
23:16 putter 702         the_exception->severity = EXCEPT_error;
23:16 autrijus changes topic to: Preflight #1: http://xrl.us/hxkk | geoffb's column (/. ed): http://xrl.us/hxhk | pugscode.org | pugs.kwiki.org | paste: http://sial.org/pbot/perl6 | www.geeksunite.net
23:16 autrijus bradfitz: http://xrl.us/hxkk
23:17 autrijus aka http://pugscode.org/tmp/Per​l6-Pugs-6.2.10-pre1.tar.gz
23:17 bradfitz got it.
23:17 gaal 'hxkk' is a nice name.
23:17 autrijus gaal: indeed
23:17 elmex autrijus: i don't post @ /. :)
23:17 autrijus ok :)
23:17 dduncan I just noticed the '=' stripping ... good idea since a single main file heading appears not to be coming
23:18 dduncan getting preflight ...
23:19 autrijus ooh the cuteness backfired.
23:19 autrijus # http://developers.slashdot.org/comm​ents.pl?sid=164792&amp;cid=13753083
23:21 dduncan the pugs distro has grown a lot btwn the last 2 releases ... a little under doubling in size, vs around 10% increments between previous releases
23:22 pasteling "putter" at 66.30.119.55 pasted "segfault bt" (66 lines, 2.9K) at http://sial.org/pbot/13591
23:22 gaal freepan's in a lull
23:23 dduncan that /. poster's another example of people not knowing about what they speak
23:23 putter bbl &
23:23 dduncan the first sentence particularly
23:24 mdmkolbe has joined #perl6
23:24 mdmkolbe eval? perl 1+2
23:24 mdmkolbe ?eval perl 1 + 2
23:24 evalbot_7512 '3'
23:24 mdmkolbe ?eval help
23:24 evalbot_7512 Error: No compatible subroutine found: "&help"
23:25 mdmkolbe ?eval
23:25 mdmkolbe ?eval ?
23:25 evalbot_7512 Error:  unexpected end of input expecting term
23:25 elmex has quit IRC ("leaving")
23:25 gaal ?eval "it does work"
23:25 evalbot_7512 'it does work'
23:25 larsen has quit IRC ("later")
23:25 gaal ?eval 6*9
23:25 evalbot_7512 54
23:25 autrijus putter: still looking into your bt.
23:25 gaal # base 13
23:25 mdmkolbe ?eval perl ((1=>2)&(3=>4)).value
23:25 evalbot_7512 '({ref:<Scalar::Proxy>})'
23:26 mdmkolbe How up to date is the evalbot usuallly?  Is it built nightly?
23:26 gaal r7512 is the latest.
23:26 gaal it's built all-the-timely.
23:28 gaal autrijus: bump the version number :)
23:30 autrijus gaal: where? :)
23:30 gaal pugs -V
23:30 gaal pugs_version: Perl6 User's Golfing System, version 6.2.9, August 3, 2005
23:30 svnbot6 r7513 | autrijus++ | * more changelog edits.
23:31 mdmkolbe has left
23:31 evalbot_7512 has quit IRC (Remote closed the connection)
23:31 evalbot_7513 has joined #perl6
23:33 autrijus k done
23:34 autrijus leo__: ping
23:35 gaal smoke started, ETA 25
23:35 svnbot6 r7514 | autrijus++ | * adding throw_away to MANIFEST.SKIP.
23:35 svnbot6 r7514 | autrijus++ | * bump version number.
23:36 evalbot_7513 has quit IRC (Remote closed the connection)
23:36 evalbot_7514 has joined #perl6
23:37 RedEar has joined #perl6
23:37 gaal ?eval $?PUGS_VERSION
23:37 evalbot_7514 \'Perl6 User\'s Golfing System, version 6.2.10, December 10, 2005 (r7514)'
23:38 gaal hey, that's funny
23:38 gaal why is there a leading \ there?
23:38 gaal ?eval $?PUGS_VERSION.perl
23:38 evalbot_7514 '\\\'Perl6 User\\\'s Golfing System, version 6.2.10, December 10, 2005 (r7514)\''
23:41 revdiablo That is kind of bizarre
23:41 autrijus ?eval say $?PUGS_VERSION
23:41 evalbot_7514 Perl6 User's Golfing System, version 6.2.10, December 10, 2005 (r7514) bool::true
23:41 revdiablo Is evalbot running in the future?
23:41 revdiablo December 10, 2005?
23:42 autrijus revdiablo: see my commit.
23:42 autrijus putter: reproduced; was triggered by unclean build.
23:42 autrijus I suspect Parrot_call_sub is not happy.
23:42 revdiablo Oh, so autrijus is from the future. That explains a lot!
23:42 svnbot6 r7515 | autrijus++ | * It being the 10th month this year does not make it December.
23:42 svnbot6 r7515 | autrijus++ |   I blame the Romans.   s/December 10/October 10/
23:43 revdiablo ;)
23:43 autrijus :)
23:43 gaal That doesn't explain the leading '
23:43 gaal leading \
23:44 gaal leading lambda!
23:44 gaal may be *that* explains it.
23:44 autrijus gaal: that signifies a rw container.
23:44 autrijus since it is dumping the container
23:44 gaal ah!
23:47 evalbot_7514 has quit IRC (Remote closed the connection)
23:47 dolmen has quit IRC ("Leaving")
23:47 evalbot_7515 has joined #perl6
23:51 justatheory has joined #perl6
23:55 gaal has there been any change to t\rules\rules.t in this pf?
23:55 gaal it takes a while before it's skipped here.
23:57 autrijus gaal: the rules are passed to macros.
23:57 autrijus I fixed that.
23:57 autrijus I mean, I made those not parse at all.
23:58 gaal will there be a pf2? i think i can't stay.
23:58 autrijus probably not, this will be release if nothing wrong happens
23:59 gaal smoke up.

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

Perl 6 | Reference Documentation | Rakudo