Pugs t-shirts www.cafepress.com/pugscode | Pugs 6.2.9 released | pugscode.org | pugs.kwiki.org | paste: sial.org/pbot/perl6 | www.geeksunite.net
Set by stevan on 15 August 2005.
svnbot6 r6762 | Darren_Duncan++ | /perl5/org.perl6.metamodel : updated 10 files to replace tabs with spaces 00:02
r6763 | Darren_Duncan++ | /perl5/Perl6-Container : updated Changes file to replace tabs with spaces
r6764 | Darren_Duncan++ | /perl5/Perl6-MetaModel : in 7 files, replaced all tabs with spaces 00:08
r6765 | iblech++ | t/01-sanity/:
r6765 | iblech++ | * Renamed 07-test.t to 08-test.t.
r6765 | iblech++ | * Added new 07-tests testing for misc. other features needed by Test.pm and not
r6765 | iblech++ | already covered by 01-06 (suggested by putter++ IIRC):
r6765 | iblech++ | split, range, defined, binding, ref, isa, try, simple-multisubs, end-blocks,
r6765 | iblech++ | substr, tailcall, for.
r6766 | Darren_Duncan++ | /perl5/Perl6-MetaModel2.0 : replaced a single tab with 4 spaces in metamorph.pl
r6767 | Darren_Duncan++ | /perl5/Perl6-Value : replaced 3 tabs with spaces in the Changes file
Juerd ~/last juerd 3 00:11
argh
svnbot6 r6768 | Darren_Duncan++ | /perl5/Perl6.MetaModel : replaced 7 tabs with spaces in Builder.js 00:14
r6769 | Darren_Duncan++ | /perl5/PIL-Run : in 7 files, replaced all tabs with either 4 or 8 spaces each 00:20
dduncan and with r6771, all /perl5 files are detabbed ... and most of them did mix tabs and spaces 00:24
svnbot6 r6770 | Darren_Duncan++ | /perl5/PIL2JS : updated 3 files to replace tabs with spaces 00:26
r6771 | Darren_Duncan++ | /perl5/T2-Perl6 : replaced tabs with spaces in 3 files
geoffb iblech: Your "slow svnbot" FAQ answer did not explain why it is slow lagged that several minutes *AND* several messages can occur before it shows an update. For instance, when I whined earlier, svnbot didn't post until almost 3 minutes and several messages after evalbot had recompiled and reloaded.
Simple waiting for IRC to return doesn't sufficiently account for that
iblech: also, why did you number all of the new tests in 01-sanity 07-*? Are they all completely independent, with no ordering required? 00:30
iblech geoffb: There have to be two conditions fulfilled to trigger a check for new commits: (time - $last_check) >= $interval *and* out_of_blocking_readline 00:34
geoffb: Right, this is the very reasion. There is no clear ordering between the new 07-tests, i.e. 07-for.t is independent of 07-substr.t 00:35
svnbot6 r6772 | Darren_Duncan++ | updated ChangeLog file to account for my recent actions 00:38
r6773 | iblech++ | * New t/operators/binding/subs.t: Tests for binding the return value of 00:51
r6773 | iblech++ | subroutines (normal ones and "is rw" subs) as LHS and RHS.
r6773 | iblech++ | * New t/operators/binding/nested.t: Many tests for binding elements of complex
r6773 | iblech++ | structures as LHS and RHS (and this way creating abbreviations, infinite
r6773 | iblech++ | structures, etc.). (PIL2JS passes 43/43 :)).
iblech ok, really need to sleep now. Night all! :) 00:52
dduncan fyi, I'm starting on /modules now 01:03
svnbot6 r6774 | Darren_Duncan++ | /modules/Class-Events : updated Events.pm to replace 159 tabs with spaces 01:10
r6775 | Darren_Duncan++ | /modules/Date : replaced tabs with spaces in 3 files
dduncan as with r6778, all /modules are done 01:15
svnbot6 r6776 | Darren_Duncan++ | /modules/Grammars : replaced 35 tabs with spaces in MatchX.pm
r6777 | Darren_Duncan++ | /modules/JavaScript-FrontEnd : replaced 46 tabs with spaces in ARuntime.pm
r6778 | Darren_Duncan++ | /modules/Perl-MetaModel : replaced tabs with spaces in Class.pm, Trig.pm, Basic.pm
01:18 _SamB_ is now known as SamB
svnbot6 r6779 | Darren_Duncan++ | /modules/* : detabbed 2 more Date files and 1 more Perl-MetaModel file 01:21
01:54 sleepster is now known as typester
svnbot6 r6780 | Darren_Duncan++ | detabbed 3 root dir files: debian/rules, AUTHORS, Makefile.PL; but, Makefile.PL retains its literal tabs in its output to the generated Makefile 02:16
r6781 | Darren_Duncan++ | reversed the changes made to lib/Perl6/MakeMaker.pm in r6760, since it appears that 'make' actually needs those literal tabs in Makefiles; reverted 3 space-runs into tabs 02:22
r6782 | Darren_Duncan++ | updated ChangeLog re my recent updates 02:28
dduncan okay, that's all I'm going to do for the next few hours, or today 02:30
luqui wakawakawakawakawakawakawakawakawakawikiwakawakawakawakawakawakawakaspamwakawakawaka 02:58
?eval [76, 101, 97, 118, 105, 110, 103].pack("C*") 02:59
evalbot_6782 Error: No compatible subroutine found: "&pack"
luqui ?eval (76, 101, 97, 118, 105, 110, 103).pack("C*") 03:00
evalbot_6782 Error: No compatible subroutine found: "&pack"
luqui ?eval pack 'C*', 76, 101, 97, 118, 105, 110, 103
evalbot_6782 Error: No compatible subroutine found: "&pack"
luqui huh
QtPlatypus luqui: We don't have a pack yet, or a speficiation for pack. 03:38
luqui grepped pugs and noticed
I note that much of pack will be done through class declarations 03:39
but there should still probably be a lower level version
QtPlatypus wrote up a spec for pack posted to p6l "And when we get rawstrings I'll implement it" 03:40
luqui can you give me a messageid?
luqui found it 03:42
QtPlatypus [email@hidden.address]
luqui QtPlatypus, what are you blocking on in order to implement this? 03:47
QtPlatypus luqui: The ability to create raw strings. 03:48
pugs current "Everything is utf8 encoded unicode" would mean that most of pack's test would fail. 03:49
luqui ahh
bbiab
putter prolog p5 regex engine now 95% passes re_tests, with no skips. a few more tests I care about, a cleanup pass, and p6 starts getting added. 04:10
*weary smile*
luqui contemplated porting PGE to perl 6
after this prolog exercise, what kind of work do you think that would be?
putter after pugs oo works... no problem at all. the problem with doing "large"-scale things in p6 at the moment, is piljs doesnt have objects, and pugs is buggy. 04:12
so you architect something, and find yourself in a corner of design space where "if only simple thing foo was working, we'd be all set, but since not..." and unable to back out.
luqui can hack pugs now though, so maybe that would get some things fixed 04:13
putter hmm... well... since you mention it... pil p6 tree is lightly hung up on, let me see if I can dig it up... 04:14
class C{}; class C::D{method new($d){...}}; sub f($d,$cls){ ::C::($cls).new($d) } say f(3,"D"); 04:15
luqui "pil p6 tree is lightly hung up on" # parse error
putter for some ...
ext/Perl-Compiler/lib/Perl/Compiler/PIL/Node_pm_gen.pl 04:16
luqui uh, what is sub f() supposed to be doing?
luqui doens't think that is valid p6
putter pugs -CPerl5 ouputs bless( X, Y) and one wants to define bless(). 04:17
::C::($y).new(*$x)
luqui oh I get it 04:18
symbolic deref
?eval class C{}; class C::D{method new($d){...}}; sub f($d,$cls){ ::C::($cls).new($d) } say f(3,"D"); 04:19
evalbot_6782 Error: Must only use named arguments to new() constructor
putter the tripple of ("is leading :: present", "doing ($y)", "*$x works") doesnt play well together (my fuzzy recollection).
thats the *$x not working
luqui ?eval class C{}; class C::D{submethod BUILD($x) { say "$x" }}; sub f($d,$cls){ ::C::($cls).new( x=> $d) } say f(3,"D"); 04:20
evalbot_6782 3 <obj:C::D> bool::true
luqui I don't quite understand what you are asking... if anything
putter ?eval class C{}; class C::D{submethod BUILD($x) { say "$x" }}; sub f($d,$cls){ ::C::($cls).new(*$d) } say f([3],"D") 04:21
evalbot_6782 Error: Must only use named arguments to new() constructor
putter bless gets array and hash refs passed as X.
f here is bless 04:22
the first case was too simple a reduction
luqui o...kay. so what are you asking?
new() does not support positionals yet, that's the only thing about this that I can see 04:24
putter the ability to write a sub bless() which works for bless([3,...],"Foo") bless({mumble=>frotz,...},"Bar")
luqui and how is this not possible currently? 04:25
the examples above seem to be working how they should
putter ?eval class C{}; class C::D{method new($d){...}}; sub f($d,$cls){ ::C::D.new($d) } say f(3,"D"); 04:27
evalbot_6782 Error: Must only use named arguments to new() constructor
putter ?eval class C{}; class C::D{method new($d){...}}; sub f($d,$cls){ C::D.new($d) } say f(3,"D");
evalbot_6782 Error: *** ... - not yet implemented at <eval> line 1, column 38-41
luqui is there a reason you cannot pass named arguments to new? 04:28
putter that's why its only lightly stuck. the classes are all autogenerated. so mumble.new().now_init($d) and they do their own unpacking. 04:30
luqui what is wrong with: mumble.new(init => $d) 04:31
luqui is beginning to contemplate the correctness of the "named only" constraint on new, though 04:32
putter two thoughts
re new, note it actually worked above, without the leading ::. which are unfortunately needed for the $cls case. 04:33
luqui right...
putter re why, I guess because I had a very low "pugsbug workaround" threshold for that project. its only really useful if folks can then build big things on it. and having had other things lost in layered workaround space, I put a low threshold cut off on this one. 04:35
luqui fair enough 04:36
putter "if I run into problems just creating the objects, then that should be taken as a sign that I'm biting off more than the system is ready for again"
regards other things on pause, the regexp stuff in pugs is wating on objects returned by functions not being mis-typed. 04:38
luqui for example?
putter but that's also been a, "well if that gets fixed, I get to see what the next obsticle is". 04:39
luqui well you have to get past the obstacles sometime
you can't just magically make them all go away
putter notes a "putter is burned out" tone to this... sorry ;)
let's see...
?eval class C{has $.v} sub f(){C.new} f().v 04:40
evalbot_6782 Error: No compatible subroutine found: "&v"
luqui hmm 04:41
putter but not to worry about this stuff. p5 will soon have a p6 rules engine, and iblech is adding objects, and piljs runs fine linked with perl5, so there is now a path for the regexp stuff to progress... 04:42
luqui ?eval class C { has $.v } sub f () { C.new } my $x = f; $x.v 04:43
evalbot_6782 \undef
luqui hmm, i wonder what the trouble here is
that's cool
putter autrijus has described it as a "known problem" with the type inferencer 04:44
luqui there's a type inferencer?
putter looks guilty at having intended to do a more thourough look at pugs internals... for half a year now... 04:45
I think that's what autrijus said. something must be providing type info for the multimethods... 04:46
luqui hmm 04:48
well, I need to do homework
putter basically on the parse->compile-(pil)->emit->runtime parse and compile are all pugs, and emit and runtime are all p5/native. 04:49
end of day for me 04:50
err, no, pugs|p5/native 04:51
luqui yeah,
and the pugs runtime is completely kludgy
thus pil2, right?
putter yes
luqui make a new pil, and write a runtime from scratch?
putter my impression is the necessary changes have been described as highly localized 04:52
luqui hmm
like findSub
luqui did a little hacking on findSub the other day
what a mess
putter nada->p6 in 6 months. ;) 04:53
growing pains inevitable
luqui true
it's nice to step back and notice the amazing progress of this project every once in a while :-)
putter indeed :) 04:54
neat mix of cultures too
luqui pugs forced me to learn haskell
putter :)
luqui and learning haskell has really chaned the way i think about a lot of problems
putter yeah. that's the argument for learning languages with unfamiliar approaches - leaves your brain with that fresh scrubbed feeling 04:56
changes the way you think about hings
smith100 hings think about you 04:57
mugwump leaves you more prepared for questions like, "yeah, but aren't all turing complete languages equivalent anyway? what's the point in learning a langauge like that?"
luqui has thought doing a "real project" in prolog
but prolog is so yucky
so I have been considering curry
putter awww
luqui except I can't find a curry compiler!
maybe it's just because I don't understand
but I find I need to know too much about prolog's execution strategy to write code that doesn't infinite loop 04:58
which then just seems to be a convoluted procedural code that I'm writing
putter yes to the first part, no to the second 04:59
autrijus luqui: cyc is a nice curry compiler danae.uni-muenster.de/~lux/curry/user.html
and my experience is that cyc feels more modern than prolog by far
but maybe that's because I'm brainwashed by haskell first :) 05:00
putter you do have to be up close and personal with engine. the "usual logic programming intro" claim of write blind and it works is, well, if not a lie then quite misleading.
but the fun of prolog is intimite with the engine, and collaborating to write the engine you need for your problem (because prolog's is invariably not what you need;). for that, prolog has historically been one of the best. 05:01
s/is/is being/
autrijus mm the MRT car starts moving. ciao :) 05:02
putter &
autrijus public wireless at high velocity is still an unsolved problem
putter lol
luqui :-p
putter planes seem to work...
...sometimes
arguably not wireless in the sense of original comment 05:03
some of the modern prolog like things leave me with a "baby out with bathwater" feeling... they may have a nice corner of language design space, but its one a bit divorced from the strengths and mindspace of prolog. 05:05
luqui though some people say the same thing about perl after they come from C++ 05:07
"sure, it's good at text processing"
putter perl _is_ rather distant from C++. its just that C++ is used for sooooo many things for which it is technically (rather than socially) ill suited... 05:09
luqui yep
but a lot of times C++ers don't like perl because they don't understand it yet
even though perl has a much larger problem space that C++
s/has/covers 05:10
well, I'm going to log off irc new
now
because I'm being distracted
:-)
adios
putter me too. its late
bye &
brentdax I wrote a new module for Perl 6. (Basically it's `trace "message", :tag`, plus the infrastructure to make it work.) Should it be called Trace.pm? Devel::Trace? Debug::Trace? Something else entirely? 06:21
dduncan now working on /examples 06:53
svnbot6 r6783 | Darren_Duncan++ | /examples/* : converted tabs to spaces in these files under [advocacy, algorithms]: motd-i.p6, lambda-calculus.p6, Newton.pm, shuffle.p6 07:12
r6784 | Darren_Duncan++ | /examples/* : replaced tabs with spaces for these files: network/hangmanbot.p6, network/seenbot.p6, network/bot_irc.p6, network/http-server.p6, games/wizard.p6, irclog2html.p6, graphics/make_swatch.p6, functional/length.p6, functional/monads.p6, functional/fp.p6, life.p6 07:17
r6785 | Darren_Duncan++ | /examples/* : replaced tabs with spaces for these files: cookbook/test.pl, p6explain/p6explain, sendmoremoney.p6, ppt/cat.p6, perl5/cpan-upload.p6 07:23
dduncan and with r6786, /examples is done 07:35
svnbot6 r6786 | Darren_Duncan++ | /examples/* : replaced tabs with spaces in 8 files under 'cookbook'; updated changelog
dduncan that's all I'm doing today ... good night 07:39
moreover, thats probably all the mass detabbing I'm going to do in pugs at all ... if any of you find more that can be safely removed, handle them as you will 07:42
07:51 GeJ is now known as Demesmaeker, Demesmaeker is now known as GeJ
dduncan okay, I just did /util too 07:54
at this point, it is just the /src and /docs directories that I haven't done; with /src, I don't know if I might break something, so I'll leave those; with /docs, any changes are also dubious 07:55
so as with my previous comment, the detabbing is basically done ... 07:56
try not to add any more tabs in your future work, unless they are essential as literal tabs
good night
svnbot6 r6787 | Darren_Duncan++ | /util/* : replaced tabs with spaces in 14 files 07:58
r6788 | renormalist++ | * An (X)Emacs mode for editing Perl6 files. 08:04
r6788 | renormalist++ | * Forked from cperl-mode and extended with Perl6 syntax situations
r6788 | renormalist++ | primarily for correct indenting and highlighting.
r6788 | renormalist++ | * Current features:
r6788 | renormalist++ | - font-lock highlighting for
r6788 | renormalist++ | class, has, say, returns, multi sub(method), gather, take, try
r6788 | renormalist++ | - indentation recognizes extended sub syntax, especially traits
r6788 | renormalist++ | - regexes via rx// as quote operator
r6788 | renormalist++ | - m// and s// similiar to new rx//
r6788 | renormalist++ | - recognize if/elsif/unless-conditions without parens
r6788 | renormalist++ | * documentation inside file
r6788 | renormalist++ | * Call for help: feel free to help if you program Emacs Lisp,
r6788 | renormalist++ | some things in this mode and/or Perl6 syntax are non-trivial
r6789 | renormalist++ | added myself as requested 08:09
rafl_ How to revert a single file to a given version? 10:22
svnbot6 r6790 | rafl++ | * Reverted dducan's tab replacement for debian/rules. It's a Makefile. It needs 10:52
r6790 | rafl++ | tabs!
r6791 | rafl++ | * Added tailing slashes for debian/*.dirs for consistency. 10:58
r6792 | Stevan++ | Perl6::MetaModel 2.0 - 11:10
r6792 | Stevan++ | * added $::Module into the hierarchy, now:
r6792 | Stevan++ | $::Class->isa($::Module) && $::Class->isa($::Object)
r6792 | Stevan++ | * moved name, version, authority and identifier methods
r6792 | Stevan++ | to $::Module
r6792 | Stevan++ | - adjusted tests and files appropriately
r6793 | autrijus++ | * massive retab for src/, expanding all tabs into spaces, 12:17
r6793 | autrijus++ | except for the generated PIL1.hs.
12:30 locksy- is now known as locksy
svnbot6 r6794 | rafl++ | Debian: 13:04
r6794 | rafl++ | * Added smoker.yml to be able to identify smoke results from the buildds.
r6794 | rafl++ | * Added dpatch build-dependency.
r6794 | rafl++ | * Adjusted rules file to work with dpatch.
r6794 | rafl++ | * Added patches/10smoker.dpatch to change the path to smoker.yml for the build.
r6795 | rafl++ | Debian:
r6795 | rafl++ | * Removed patches/10smoker.diff which was a temp file.
stevan good morning boys and girls 13:06
rafl_ Hello stevan
fglock hi all
stevan $work project has now launched, so it's back to Pugs hacking :)
hey rafl_, fglock
rafl_ Do you guys know if I can make ghc use ccache? 13:08
stevan rafl_: sorry, no idea 13:09
fglock: have you done any work on Package or Module implementations
fglock no, all 'namespace' code is in PIL-Run, and it reuses p5 13:10
stevan hmm 13:11
I just added Module to the metamodel 2.0 this morning
cause I was reading Apoc 12
which stated that Role and Class will likley inherit from Module
we will have to address namespace in PIL-Run thouh 13:12
in order to account for names like "Foo-0.0.1-cpan:JRANDOM"
the package management for Perl 6 is going to me much more complex then in Perl 5
fglock do you think we should have p6 pads instead of reusing p5?
stevan I would expect we will need too eventually 13:13
fglock I currently have a problem, I need to clone a closure and I don't have easy access to the lexicals
being able to manage the pads would make that easier 13:14
stevan I imagine that p6 pads are more sophisticated then p5 pads
fglock I think a tree of hashes would do - with some extra accessors for short/long names 13:18
stevan for pads? 13:19
fglock yes
stevan sounds sane to me 13:20
fglock then, you may want to add support for continuations, coroutines, etc - a lot of work ahead :) 13:21
svnbot6 r6796 | iblech++ | t/var/constant.t: luqui confirmed the existance of "my constant" and "our 13:34
r6796 | iblech++ | constant", so I removed =for discussion and added some more tests (see
r6796 | iblech++ | www.nntp.perl.org/group/perl.perl6....ge/23003).
r6797 | fglock++ | * PIL-Run - commented out grep() and uniq(), because they break PrimP6 compilation
fglock stevan: I think we can think of pads the same way we do 'super' methods 13:35
and 'caller' 13:36
svnbot6 r6798 | fglock++ | * PIL-Run - logic value of 'undef' is false 13:58
r6799 | fglock++ | * PIL-Run - uniq() works - but only once, because the %seen is global
nothingmuch hah! I knew it! rafl++ 14:07
for 6790
(massive mindless changes)--
what I don't know is why it's been massively revertec
rafl_ nothingmuch: Err? 14:08
nothingmuch wait, re 6793 - what was actually done?
rafl_: well, i thought that making everything spaces was very stupid
because there is no real point in fixing what isn't broken 14:09
especially when in some people's opinion the fixing is actually breaking
and the makefile issue is a good example of why it was a bad idea
so thanks for fixing it =)
rafl_ nothingmuch: Well, I also like tabs more than spaces, but being consistent is good imho. Use tabs or spaces, not both. 14:10
masak rafl_: there are also ways of being consistent using both
nothingmuch i think this project is too spaced out to care about that
imho it should be 'use whatever this file is in' 14:11
rafl_ nothingmuch: That's probably a good point.
nothingmuch for new files I prefer tabs for block indenting
and spaces for arbitrary indenting
like my $foo = $moose;\nmy $gorch = $bing;
so that tab stops never make a difference
i wouldn't mind spaces except that to delete them you need to press the delete key N times 14:12
and inserting them "manually" is sometimes annoying
rafl_ Well, I actually don't care that much about what's used for indenting as long as it doesn't break something, of course. :-)
nothingmuch right
anyway, my opinion remains that the conversion did more harm than good, but whatever. too late. 14:13
food time
svnbot6 r6800 | fglock++ | * PIL-Run - grep() compiles - it still doesn't work, but this works: 14:22
r6800 | fglock++ | (1,2,3).map:{ if $^a != 2 { $^a } else { () } }
r6800 | fglock++ | The error is related to calling the grep sub inside map()
r6801 | iblech++ | * Usual svn props. 15:04
r6801 | iblech++ | * t/var/constant.t: Minor fixes.
r6801 | iblech++ | * Pugs.PIL1, Pugs.Compile: PSub and PCode contain a p[Sub]?IsMulti field now,
r6801 | iblech++ | so PIL2JS and PIL-Run can know which subs are multis and which subs are not.
r6801 | iblech++ | * Pugs.CodeGen.PIR: Accomodate for the above change.
r6801 | iblech++ | * PIL2JS: (Simple) multi subs!
r6801 | iblech++ | * PIL2JS.js: If necessary, try to call all .variants of a multi to find out
r6801 | iblech++ | whether the parameters match the sub signature. (Hack)
r6801 | iblech++ | * PIL::Subs: Emit appropriate code for multis --
r6801 | iblech++ | multi foo {...} # is compiled as
r6801 | iblech++ | our &foo ::= PIL2JS.new_multi(); &foo.add_variant({...});
r6801 | iblech++ | * Test.pm: Removed the skip_rest_anti_multi-hack which was previously needed
r6801 | iblech++ | for PIL2JS. :) 15:05
nothingmuch luqui: ping 15:10
rafl_ The MIME-Base64 tests block infinitely for me if parrot is installed. Can someone reproduce that? 15:14
dada rafl_: IIRC, QtPlatypus was working on that 15:18
nothingmuch rafl_: i can 15:22
as do Net-IRC/t/basic.t, and something in Set
caomu_ ?eval $PUGS_VERSION 15:23
evalbot_6801 Error: Undeclared variable: "$PUGS_VERSION"
integral ?eval $?PUGS_VERSION
evalbot_6801 \'Perl6 User\'s Golfing System, version 6.2.9, August 3, 2005 (r6801)'
integral of course, the nick is a bit of a giveaway ;-)
caomu_ hmm, I need to buy a notebook, so, Anyone here can give me a suggestion? As It seems It's not easy to choose one fit my needs. :-S
oops. I am xinming, I was now in Beijing 15:24
rafl_ caomu_: I can recommend the IBM X-Series.
15:24 caomu_ is now known as xinming_BeiJing
xinming_BeiJing that xinming is the one which I sshed to my home computer. :-P 15:24
rafl_, hmm, Which one, Could you please give me a url?
ping autrijus ~~~ :-) 15:25
rafl_ I have an X41. Boguth at nofost.de
integral likes spiralbound, good for laying flat on a desk 15:26
wolverian ?eval class Foo { method bar ($x) { 1 } } Foo.new.bar (1) 15:31
evalbot_6801 Error: unexpected "(" expecting operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
wolverian is method (args) illegal? (compare to method(args))
xinming_BeiJing wolverian, Foo.newbar(1)
?eval class Foo { method bar ($x) { 1 } } Foo.new.bar (1)
evalbot_6801 Error: unexpected "(" expecting operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
xinming_BeiJing ?eval class Foo { method bar ($x) { 1 }; } Foo.new.bar ( 1 )
evalbot_6801 Error: unexpected "(" expecting operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
xinming_BeiJing oops...
?eval class Foo { method bar ($x) { 1 }; } Foo.new.bar( 1 ) 15:32
evalbot_6801 1
wolverian right, I know it works that way
dada but why?
xinming_BeiJing wolverian, In synopsis, It told you.
dada, hold on please.
wolverian xinming_BeiJing, it told me what? I don't remember reading about _method_ parentheses whitespace rules, but maybe I missed it 15:33
dada you mean that you can _never_ have spaces before ( in function calls?
xinming_BeiJing dada, hmm, In fact, I think It's for the problem of the ignorance of the parenthesis.
just like, you can use `sub fun( $x ) { ... }; fun 1;` 15:34
ou don't need the parenthesis to do the sub call.
dada but you can't do "fun (1);"?
dada sees whitespaces mattering all around and flees in panic 15:35
xinming_BeiJing But There is a special case, which is, If you need to pass an "ARRAY" to the function, You have to distinguish the "WHOLE" parameters and the parameter which with "ONLY" an "ARRAY"
wolverian dada, sure you can. it's just in methods, where parens are _required_ that it is different.
at least pugs thinks so 15:36
dada this is not encouraging
xinming_BeiJing so, If there is space between `fun` and `(`
wolverian and it is consistent with '%foo {1}' being illegal
but that is because of 'if %foo { ... }', and parens have nothing to do with that
dada I _hate_ whitespace rules 15:37
xinming_BeiJing the later (1) will be come a array in fact, and pass the (1) to the function, But the function need a argument which is a "SCALAR"
dada boo
xinming_BeiJing dada, neither do I. :-) But In fact, you have no choice... Perl 6 is complex to confuse many people. :-) 15:38
dada, hmm, In fact, you can use `fun .(1)` format
pay attention to the "DOT"
wolverian that works with {} too 15:39
%foo .{1} # legal
xinming_BeiJing wolverian, yes. `.` becomes like a "grammar" connector. :-)
wolverian dada, note: print (1+2)*3; # this is extremely non-obvious in perl5 15:41
xinming_BeiJing Parentheses are required on the dot notation if there are any arguments (not counting adverbial arguments). There may be no space between the method name and the left parenthesis unless you use the dot form of parentheses: 15:42
.doit # okay, no arguments
.doit() # okay, no arguments
.doit () # ILLEGAL (two terms in a row)
.doit.() # okay, no arguments, same as .doit()
.doit .() # okay, no arguments, same as .doit()
read S12 carefully please. :-)
?eval print(1+2)*3; 15:43
evalbot_6802 33.0
xinming_BeiJing ?eval print (1+2)*3;
evalbot_6802 9bool::true
wolverian xinming_BeiJing, ah, of course. thanks :)
dbrock what about .doit . ()? 15:44
xinming_BeiJing dbrock, hmm, I think, the "DOT" must be strictly standing before the `(` 15:46
svnbot6 r6802 | rafl++ | Debian:
r6802 | rafl++ | * Added libtest-tap-model-perl and libtest-tap-htmlmatrix-perl to
r6802 | rafl++ | Build-Depends. They are needed for smoking. The packages are currently in NEW
r6802 | rafl++ | and will soon be in the archive.
r6803 | Stevan++ | Perl6::MetaModel 2.0 -
r6803 | Stevan++ | * is_a() now expects a class instance, and not a string
r6803 | Stevan++ | - fixed a number of tests to reflect
r6803 | Stevan++ | * cleaned up isa() handling between $::Class and $::Object
r6803 | Stevan++ | - added some tests for this
r6803 | Stevan++ | * more tests in relation to $::Module being added
r6803 | Stevan++ | * added docs/misc_drawings.txt for misc. ASCII art
xinming_BeiJing seen autrijus
dada .doit . . . . . . . . . . . . . . . . . . . . . . . .() 15:47
dbrock would that work? 15:48
wolverian no. 15:49
dada why not!?
wolverian why would it?
stevan well it all depends on what language you are writing in ;)
xinming_BeiJing If in a dialect of perl 6, It might work. ;-)
dada ?eval sub doit { 42 } doit();
wolverian . . is almost certainly a typo of ..
evalbot_6803 42
dada ?eval sub doit { 42 } doit .(); 15:50
evalbot_6803 42
dada ?eval sub doit { 42 } doit . .();
evalbot_6803 Error: cannot cast from VInt 42 to Pugs.AST.Internals.VCode (VCode)
wolverian hm
?eval sub doit { return { 42 } } doit . .()
evalbot_6803 42
wolverian haha. that is evil.
dada ?
xinming_BeiJing you can even write a rule to alter the default perl 6 pharser... :-P Just like... make "DOT" equal to fun DOOOOOOOOOOOOOOOOOT() :-P
wolverian I'm thinking that's illegal though 15:51
?eval sub doit { 42 } doit .
evalbot_6803 42
dada ?eval sub doit { 42 } doit. ();
evalbot_6803 Error: unexpected "(" expecting operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
svnbot6 r6804 | Stevan++ | Perl6::MetaModel 2.0 -
r6804 | Stevan++ | NOTE: is_a() now checks the $obj.id and not $obj.name
r6804 | Stevan++ | I forgot to mention that in the last checkin
r6804 | Stevan++ | * $::Class no longer has a $:name field until it is
r6804 | Stevan++ | inherited from $::Module (in genesis.pl)
r6804 | Stevan++ | - same with $::Object as well
dada woo, I like the phoenix effect :-) 15:52
wolverian hmm, someone should ask p6l if 'doit .' should be legal :) 15:53
xinming_BeiJing autrijus, you there please? :-/ 16:29
svnbot6 r6805 | Stevan++ | Perl6::MetaModel 2.0 - 16:51
r6805 | Stevan++ | * added $::Package, so now:
r6805 | Stevan++ | $::Class isa $::Module isa $::Package isa $::Object
r6805 | Stevan++ | see article.gmane.org/gmane.comp.lang.p...guage/4599
r6805 | Stevan++ | for more details.
r6805 | Stevan++ | - adjusted and added some test to accomidate this change
r6806 | fglock++ | * PIL-Run - fixed @a=() 16:57
r6807 | iblech++ | * Usual svn props. 17:16
r6807 | iblech++ | * t/junction/junction_functions.t: Substituted (1|2|3) by any(1,2,3) so
r6807 | iblech++ | PIL2JS and PIL-Run have a chance of passing this test.
r6807 | iblech++ | * New t/junction/chained_operators.t: New test testing that (1|2|3) is the same
r6807 | iblech++ | as any(1,2,3) (passes 7/7 in normal Pugs).
17:24 PerlPilot is now known as PerlJam
svnbot6 r6808 | iblech++ | Test.pm: Put skip_rest_anti_multi-hack in again (I removed it in r6801). 17:28
r6808 | iblech++ | PIL2JS's handling of multisubs needs some more work until we can get rid of the hack.
r6809 | fglock++ | * perl5/Array - $array.store() preserves cell binding
xinming_BeiJing seen autrijus 17:33
stevan iblechbot: does PIL2JS handle packages at all? 17:37
svnbot6 r6810 | iblech++ | t/operators/repeat.t: Fixed the test so fglock++ can test his work :) 17:40
fglock checking repeat.t ...
svnbot6 r6811 | fglock++ | * perl5/Array - fixed autovivification in slices 17:51
r6812 | rafl++ | Debian: 18:15
r6812 | rafl++ | * Added libwww-perl to Build-Depends. Needed by smokeserv-client.
r6812 | rafl++ | * Clean config.yml properly.
rafl_ Grr. Would someone please stop that evalbot maddness? 18:26
Is there an other way than precompiling Test.pm to speed up smoking a bit? 18:27
stevan rafl_: not running the rules test will usually cut smoking time down some 18:28
geoffb Why is the evalbot bothering you, rafl_ ?
rafl_ stevan: Ah, OK.
geoffb: Because it keeps parting and joining. That's annoying. 18:29
stevan rafl_: it is updating itself
geoffb rafl_, causing your chat client to beep at you?
rafl_ An update for a change in the debian packaging? Hrm.. 18:30
geoffb: No, but it's line noise I don't like that much.
geoffb Someone mentioned that there is a way to make the bot be invisible WRT part/join messages
ah.
I'm used to P/J line noise because this channel has so many people going in and out anyway. 18:31
stevan maybe it should only re-compile when src/ changes
geoffb And my client colorizes those lines so they fade out of the mess
stevan, that way lie edge cases and other dragons.
Like, what about changes to Makefile.PL, or util/? 18:32
stevan geoffb: true dat
maybe evalbot could just change it's nick?
I am not really sure the details of what exactly is happening anyway
stevan pleeds IRC ignorance and goes back meta-model stuff 18:33
geoffb Well, since the helper respawns pugs children every time, theoretically just changing nick would work, but someone has to do that. Since luqui is running evalbot these days, I guess him, but who knows if he has tuits that he wants to spend on that. :-) 18:34
Which of course leads to: rafl_, you can always fix it and run it from you box. ;-)
stevan thinks it might be running on feather
but I am not sure of that 18:35
geoffb stevan++ # Devoting so many non-work cycles to the MM
stevan geoffb: who said it wasn't $work related :) 18:36
rafl_ geoffb: It's running on feather. I don't want to change evalbot. /ignore fixes the problem for me. 18:38
I hope /ignore -regexp -pattern 'evalbot_[:alnum:]+' JOINS will work.. 18:41
svnbot6 r6813 | iblech++ | PIL2JS: pugs-smokejs.pl: Skip all rules tests for now (they exhaust far too 18:51
r6813 | iblech++ | much RAM) -- smoking PIL2JS should be safe now.
r6814 | fglock++ | * PIL-Run - ($a,\undef,$b)=(1,2,3) and ($a,$b)=($b,$a) work
r6815 | autrijus++ | * Add space-loving settings for Emacs in pugs::hack, following
r6815 | autrijus++ | iblech++'s paragraph on Vim settings.
rindolf Hi all!
dduncan Hi one
rindolf autrijus: here?
dduncan seen autrijus
stevan hey rindolf, autrijus just commited, but he has not been around much lately
rindolf Actually it may be too late for him.
stevan rindolf: he regularly IRCs in his sleep, so you might be able to access his unconcious mind 18:52
rindolf perlbot: karma autrijust
perlbot autrijust doesn't have any karma
rindolf perlbot: karma autrijus
perlbot Karma for autrijus: 46
rindolf OK. 18:53
dduncan I notice the brand-new /util/cperl6-mode.el has almost 10,000 tabs in it, and they are mixed in with spaces
stevan dduncan: I wouldn't mess with that one
dduncan oh? 18:54
stevan well actually eLISP shouldnt be too whitespace sensitive
dduncan well, I know that some file types need tabs
stevan dduncan: yes, the makefiles do
dduncan the hint here is that the tabs and spaces are mixed
apparently the haskell files don't, since autrijus converted those
stevan dduncan: I think tab consistency is more important in p6 source files 18:55
dduncan yes
I was mainly going after the perl 5 and perl 6 files anyway 18:56
svnbot6 r6816 | autrijus++ | * Change all `is readonly` to either `my constant` or `is constant` .
r6817 | autrijus++ | * lambda-calculus.p6 - Port the $VERSION line to Perl6:
r6817 | autrijus++ | our $VERSION = sprintf "%d.%02d", q$Revision: 0.1 $ ~~ m:P5:g/(\d+)/;
r6817 | autrijus++ | Also made some minor fixes to make the program more perl6ish.
dduncan and c and java, since I understand those languages
stevan, have a look at cperl6-mode and see if you think those tabs are necessary, and if not then how many cols the author saw when using them 18:58
with my editor, that uses 4 cols, various sections don't seem to line up right now
or I suppose I could ignore the file and let the emacs folks worry about understanding it 18:59
stevan dduncan: my concern with that file is that it might have some code-reformatting stuff in it 19:01
geoffb stevan, MM is $work related for you? Wunderbar!
stevan which means it could be whitespace sensitive in select areas
dduncan: so my suggestion is let the emacs people deal with it ;)
dduncan so I'll leave it alone then
stevan geoffb: loosely $work related
integral ]/window 23 19:02
stevan geoffb: $work uses a lot of OSS, so any time we are without an active $project we contribute to OSS :)
my co-worker is over on #jsan doing the same as I am here :) 19:03
geoffb stevan, good for you guys! Who is $work, if you don't mind saying?
stevan geoffb: www.iinteractive.com
geoffb is a consultant and writer, so considers pugs as "continuing education" :-)
stevan we are a small consultancy,.. the website is not terribly up to date right now
geoffb nodnod 19:04
stevan mostly we do LAMP apps for various clients
I am hoping I can get the MM stable enough to use in prod 19:05
that way I can use p6 OO in p5 :) 19:06
stevan is not a fan of the current crop of p5 class-generation modules
dduncan fyi, following recent changes, over 90% of all the literal tab characters are gone from the distros, where about 2000 of 25000 or so remain; but that the new cperl6-mode file adds almost 10000 more, bringing it to 11,822
geoffb stevan, :-)
dduncan s/the distros/pubs/
s/pubs/pugs/
stevan geoffb: but the real reason I am doing this is because its fun ;) 19:07
geoffb Well, of course. You could have chosen *any* OSS project to throw cycles at. 19:08
Thanks for picking this one. ;-)
autrijus yeah, apparently our optimizer is pretty good on this one :)
geoffb autrijus, yep, best choice for optimization, if you ask me. 19:09
-Of
autrijus :D
autrijus starts thinking t-shirt ideas based on -Ofun 19:10
stevan :)
dduncan autrijus, I noticed you did the /src last night
geoffb Did you do the Just Fear | Nothing one?
stevan hey autrijus, got DSL?
autrijus geoffb: yup, did that
geoffb schweet
autrijus stevan: yeah, finally
stevan nice
autrijus but then begineth the long strectch of $work that will extend to at least thursday
geoffb (faster connection)++
autrijus :-/ 19:11
stevan autrijus: we need to find you some funding
autrijus stevan: MM2 is making more and more sense every day
stevan this $work thing is no good
autrijus: thanks :)
geoffb autrijus, are you ready to accept TPF funds, or still no? 19:12
autrijus geoffb: er it's said on TPF website that I am
geoffb sorry, hadn't checked recently
stevan autrijus: did you see the recent $Package and $Module additions to MM2? 19:13
geoffb gets lazy when multitasking
autrijus but I'm $work'ing now on the premise of no fundings -- at this rate I'll need to work another two weeks before ICFP, then for another two to three months, before I can cover a year's worth of erdosing
xinming_BeiJing autrijus, are you awake now? :-P
stevan I am debating whether to add $Package symbol tables accessing and such
autrijus xinming_BeiJing: barely. a long day today -- and need to up early tomorrow, which means I need to sleep soon 19:14
stevan: I think they obviously belong to ::Package
xinming_BeiJing autrijus, I need a suggestion on buying a laptop, I know It's not a good to ask here. But I think I willl buy an laptop in BeiJing, But I don't know much on Laptop, So, I wish some one could give me some suggestions.
autrijus since ultimately the interop with underlying OS -- I mean runtime -- will base on that API
stevan autrijus: yes, but do they belong in the MM? or should I assume it is there?
xinming_BeiJing autrijus, Or, I will ask for your laptop "version" So that I might get the one "close" as yours. :-P 19:15
stevan in the underlying runtime that is
autrijus stevan: I think it's best to settle on a API and have a p5 impl, as part of Perl6-Package or something, or just part of PILrun
xinming_BeiJing: okay... I'm using Asus M2C, on the ground that it's got excellent support and very cheap, so I can break it freely without feeling guilty. 19:16
xinming_BeiJing thanks.
svnbot6 r6818 | Darren_Duncan++ | updated ChangeLog re other peoples' detabbing work; added /src/* re autrijus++, removed /debian/rules re rafl++ as it has a net zero change
autrijus stevan: also, I wonder how the ChangeLog thing shoudl be handled for this release 19:17
xinming_BeiJing autrijus, maybe It's cheap to you, But not to me, hmm, I will consider it for following the talent hacker. ;-)
autrijus which I hope to start relenging this weekend
xinming_BeiJing: er it's cheap compared to Toshiba or IBM. 19:18
stevan autrijus: I was thinking about that too (Changelog)
perhaps PILRun and PIL2JS should have their own changelogs?
autrijus stevan: maybe have you, iblech and fglock summarize the major areas
I think a single ChangeLog is still good, but probably difference section layout than the previous ones 19:19
stevan autrijus: ok
autrijus but I'm too fried to think clearly
so maybe tomorrow :)
stevan I will slate some extra time this weekend to do that
xinming_BeiJing autrijus, ä½ ēš„é‚£äøŖäøœč„æ会äøä¼šę¼ē”µå•Šć€‚ć€‚ć€‚ę€Žä¹ˆē½‘ē»œäøŠęœ‰äøŖ궈ęÆčÆ“čæ™äøŖäŗ§å“ęœ‰äø€ęŠ«ä¼šę¼ē”µć€‚怂怂 19:20
autrijus xinming_BeiJing: my laptop doesn't leak electricity
rafl_ dduncan: I don't really understand what your commit message want's to say to me.
autrijus stevan: you rocketh
xinming_BeiJing autrijus, thanks. I will consider it. ;-)
autrijus geoffb: wrt tpf funding url: www.perlfoundation.org/news/2005/al...npugs.html
stevan autrijus: :)
autrijus alright, journal up, I'll zzz now 19:21
ciao. :)
autrijus waves &
geoffb ciao
stevan nite
dduncan rafl, I just acknowledged that you fixed something I broke; my changelog update concerned what files were detabbed and what weren't
rafl_ dduncan: Ah, OK. 19:22
xinming_BeiJing autrijus, Is your CPU 1.6 G 19:23
dduncan in your specific case, since the changelog is for the benefit of people switching up distros and don't care about the intermediate steps, I removed the mention of a changed file because it was reverted between the same 2 distros it was changed
autrijus xinming_BeiJing: yes, dothan 1.6g. but I'll sleep now -- please direct further questions to my mail or something :)
&
xinming_BeiJing autrijus, well, I won't trouble you more, I'd rather like to see the finish of pugs faster. ;-) 19:24
sleep well, night.
svnbot6 r6819 | Stevan++ | Perl6::MetaModel 2.0 - fixed up the ASCII art 19:41
r6820 | fglock++ | * PIL-Run - %ENV is tied 19:53
fglock autrijus: done :) 19:54
rafl_ Juerd: ping 19:56
I see there's already a changelog entry for 6.2.10. Is there also a release shedule yet? 20:03
Juerd I have no idea. 20:04
I know nothing.
geoffb rafl_, autrijus mentioned something about releng an hour or so ago, sound like will occur this weekend, perhaps beginning of next week
but that's just reading the tea leaves
(and that autrijus has work through Thursday and stevan is planning to do changelog work this weekend) 20:05
rafl_ geoffb: Cool. I think that will be the first version to be in Debian officially.
geoffb w00t! 20:06
geoffb can't wait for 'apt-get pugs' everywhere he touches
rafl_ We only need to wait for ghc6 to be updated to 6.4-4.1 on all architectures (hppa ia64 m68k sparc missing) and for libtest-tap-model-perl and libtest-tap-htmlmatrix-perl to be approved by the ftp-masters. 20:07
Juerd 22:09 * geoffb can't wait for 'apt-get pugs' everywhere he touches 20:11
+install
rafl_ Juerd: How about moving the pugs installation on feather to the debian package? 20:13
dduncan I added a template changelog entry for 6.2.10 but left the revision and date XXd out ... that way, people can add to it while they are making changes and best understand the nature of those changes, rather than trying to figure it out later ... this is also standard practice with my own distros 20:14
that's in response to rafl_'s question of a half hour ago
svnbot6 r6821 | iblech++ | * t/builtins/math/: log.t, exp.t: Added abs(...) <= EPSILON guards. 20:17
r6821 | iblech++ | * PIL2JS:
r6821 | iblech++ | * Prelude::JS::Array: sum() returns undef now.
r6821 | iblech++ | * PIL2JS: Added $*OUT.
dduncan oh, and the pugs release schedule is about once per month, shortly after each monthly parrot release
svnbot6 r6822 | Stevan++ | Perl6::MetaModel 2.0 - 20:29
r6822 | Stevan++ | * added a number of tests from MM 1.0
r6822 | Stevan++ | * fixed DESTROY to work correctly
Juerd rafl_: If that can be done automatically, without too much breakage, sure, I'm all for it. 20:39
rafl_ Juerd: For sure, it can be automated. 20:43
Juerd: Who run's the automated updates at the moment?
Juerd /me, but if you want totake over, please so 20:44
I have very little time and health to do feather stuff
rafl_ How do you do it? Do you trigger a recompile with every commit? Or maybe via cron? 20:45
Juerd See ~juerd
It's all readable
It's done via cron
The script autobuild.sh
rafl_ Ok, I think I'll do that this night.
svnbot6 r6823 | Darren_Duncan++ | updated lib/pugs/hack.pod to specify how to configure BBEdit to auto-expand tab key-presses, complementing the Vi and Emacs instructions for the same 20:47
rafl_ wonders if this information is really pugs specific.. 20:48
Juerd Which information?
rafl_ Juerd: How to configure $your_favourite_editor. 20:49
Juerd rafl_: It is. 20:50
(not really.)
(but who really cares?)
rafl_ Juerd: Hm. I see that you also compile parrot yourself. I think I'll preparea Debian package for it before. 21:02
jakx hello 21:05
when is perl 6 coming out 21:06
wolverian jakx, when it's done. 21:07
dduncan or, you could say its out right now!
but just not all of its pieces
wolverian right :)
jakx, pugscode.org has a in-development perl6 implementation that we are collectively hacking on 21:08
jakx, you are more than welcome to join in :)
rantanplan Pugs is really a good thing. Just startet coding in perl 6 with pugs the last weekend. Before i was coding perl 5 like 10 years. But i like the changes of perl 6. :) 21:09
wolverian rantanplan, wow, that's oldschool. :) did you ever use perl4? 21:11
jakx i have little experience in perl
does anyone here write object oriented code in perl
integral sure. And functional code too :) 21:12
rantanplan not really, i was only using perl 5. i also had some perl breaks to do other languages as well. but in perl 4 i never coded.
jakx when i bless a reference what really happens?
buu jakx: Why are you asking about perl5.. in perl6? 21:13
rantanplan jakx: i dont even know if in perl 6 is blessing still available. 21:14
jakx how do i create a constructor in perl6
integral bless is still in perl6, but your "new" function is automatically created basically
jakx: bless just takes a wee sticky label and sticks in on your data structure
jakx ok 21:15
integral Things later on can then look bemusedly at your handwriting on the label and find the right code to run
rantanplan jakx: Look at nr. 12: dev.perl.org/perl6/docs/synopsis.html 21:17
jakx it says file not found
rantanplan Sorry, my bad. s/docs/doc/ 21:18
jakx wow
a class keyword is in perl6
buu YES 21:20
jakx rantanplan:does the pugs offer this functionality
buu All hail the mighty.
wolverian jakx, read all the synopses :)
?eval class Foo { method bar { 42 } } Foo.new.bar
evalbot_6823 42
wolverian jakx, there you go. evalbot runs pugs. :)
rantanplan I ve printed out the synopsis today :)
jakx nice 21:23
the python people are getting competion 21:24
*competition 21:25
perl has it all now ,OOP,speed and a big comunity 21:27
can the pugs be installed on windows using active perl 21:29
rantanplan jakx, look at the pugscode hp. 21:30
there is a windows installer. afaik.
integral speed? Pugs itself isn't that speedy, and the PIR isn't brilliant yet
Pugs is included in PXPerl which is a Windows Perl distribution.
dduncan jakx, I write all my non-trivial perl in object oriented form 21:32
jakx ok
is there a site to get PXPerl from 21:33
rantanplan www.pugscode.org -> Binary Builds
jakx the pxperl linkis saying the lik canot be found 21:36
is there another site 21:37
buu ?eval (1,2,3)[1] 21:47
evalbot_6823 \2
buu That's weird.
?eval sub z { (1,2,3) } z()
evalbot_6823 (1, 2, 3)
buu ?eval sub z { (1,2,3) } z()[0]
evalbot_6823 \1
buu ?eval sub z { (1,2,3) } z()[2]
evalbot_6823 \3
jakx is there another site to get pxperl
buu I bet google knows! 21:49
svnbot6 r6824 | fglock++ | * TODO updates 21:52
21:52 nnunley2 is now known as nnunley
svnbot6 r6825 | rafl++ | * Updated documentation for the Debian package. 22:16
r6826 | vkon++ | fix logical typo in Types.hs in cxtSlurpy definition 22:50
mosnene hello 23:40
what are you doig?
buu I am aborting small children. 23:47
23:48 buu is now known as buubot
lscd congradulations 23:48
23:48 buubot is now known as buu
buu Thank you. 23:49