»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by sorear on 4 February 2011.
sorear hi 00:00
shinobicl what can i do to solve this "compile time" issue?
nom: role Counter [**@ranges] { has @.ranges; has @.values; method new(*@new_values) { die if @new_values.elems != @ranges.elems; self.bless(*, :@new_values);} }; class TimeofDay does Counter[0..23, 0..59];
p6eval nom: OUTPUT«Cannot use '0..23, 0..59' as an argument to a parametric role as its value is not known at compile time at line 1, near ";"␤current instr.: 'nqp;HLL;Grammar;panic' pc 23510 (src/stage2/gen/NQPHLL.pir:6316) (src/stage2/gen/NQPHLL.pm:329)␤»
sorear shinobicl: not use parametric roles like that 00:01
shinobicl how should i use them?
or is NYI?
sorear well, what you're seeing is a NYI error 00:03
but that code is IMO very bad form 00:04
so it's not likely to be made to work soon
the normal way to get values into an object is by using constructor arguments 00:05
not by constructing a special class with defaults
shinobicl i want to implement the "counter" behavoir in a role, that's why i use a class to specify the ranges. 00:07
but the thing is, besides the class, how should i use the parametric roles?
00:07 ab5tract left
sorear why? 00:08
00:08 ascent_ joined 00:09 molaf joined
shinobicl i saw this in a example here in this irc channel... i think the idea was to transform a "TimeOfDay" class into something more generic... i thought it was a good idea. The TimeOfDay class had just 3 'ranges', for hours, minutes and seconds. 00:10
sorear **@x was only added like last week 00:11
if excessively abstracting a system causes problems, I think the sensible thing to do is to not excessively abstract it 00:12
shinobicl just in time to try to implement it, then :). The example i saw only had an array of upper limits, like this: role Counter[**@maxvalues] 00:13
sorear parametric roles are a dark corner of Perl 6 semantics. don't go there unless you have a legitimate reason to be creating families of classes
shinobicl oh yes, i know :) I'm not planning to use it on a system of library. I just want to see what can i do with perl6 roles :)
mmm i see,.,. i'll wait some months then untile they are more mature 00:14
sorear parametric roles are quite mature - for arguments that are classes 00:17
that's what they're most designed for
generic collections
tbh I think someone just added "oh and they can do values too!" as an afterthought
also, by bindng 0..23, 0..59 to a **@ranges, you're shooting yourself in the foot in the long term 00:18
** coerces elements to List, I think
whereas you need the values to stay Range so that you can take advantage of the Range structure 00:19
shinobicl The idea was not only to use Ranges, but any list for specifying a counter range 00:24
example: TimeOfDay does Counter[0..11,0..59,0..59,<AM PM>]; 00:25
but was only an experiment
00:28 lichtkind left
sorear ok. 00:29
well for now I suggest using master
this will be a nom regression 00:30
shinobicl master?
sorear don't use the nom branch
just checkout rakudo
shinobicl ahh ok, you mean the "**" ?
sorear no, I mean the using computed values in roles
if you don't know what master is, you shouldn't be using nom
do you have an installed copy of rakudo? 00:31
shinobicl ok. i always use rakudo, i just wanted to test **
yes, the april version
sorear nom needs the arguments to roles to be literals
this will be fixed in ctmo
but that's internal development stuff 00:32
00:32 Chillance left
shinobicl by chance... do you know if '**' will be in the next release of rakudo? 00:32
sorear it won't be
very unlikely 00:33
nom's still at least a month away from being releasable
00:39 bluescreen10 left 01:00 noganex joined 01:01 wolfman2000 joined 01:02 noganex_ left 01:09 envi joined 01:29 donri left 01:46 whiteknight left 01:58 f00li5h is now known as f00li5h[HireMe] 02:00 f00li5h[HireMe] is now known as f00li5h 02:12 woosley joined 02:43 daniel-s left 02:44 daniel-s joined
shinobicl hi... how to show '1' as '01' ? 02:45
sorear .fmt('%02d') 02:53
shinobicl thanks :) 02:54
02:54 jaldhar left 02:55 jaldhar joined 03:27 Su-Shee left 03:28 Su-Shee joined 03:36 buubot_backup left 03:51 kboga left 03:53 buubot_backup joined 04:44 birdwindupbird joined
dalek kudo/nom: ed26e45 | pmichaud++ | / (4 files):
An initial implementation of $*ARGFILES and lines().
05:10
05:11 eternaleye_ is now known as eternaleye 05:13 shinobicl left 05:25 ZaphrodZenovka left, ZaphrodZenovka joined 05:33 _jaldhar joined 05:34 jaldhar left 05:38 molaf left 06:00 koban joined, koban left 06:04 ggoebel left
dalek p: b5f2c5d | pmichaud++ | src/HLL/Compiler.pm:
-e option to HLL::Compiler needs to pass an arg0.
06:14
06:18 cognominal_ left 06:20 a joined, amkrankruleuen left, a is now known as Guest99070 06:25 Mowah joined 06:27 wtw joined
moritz \o 06:31
pmichaud: did you see the regex bug in the backlog?
pmichaud which regex bug? the one for arrays?
phenny pmichaud: 24 Jul 12:17Z <moritz> tell pmichaud that m:s// now works in nom, but not :sigspace -- is that something you want to fix on the qregex level, or should I do it in nom (as best as I can)?
moritz nom: 'a' ~~ /<.upper>/; say $/.Bool; say $/.to; say $/.from 06:32
p6eval nom: OUTPUT«Bool::True␤0␤0␤»
moritz nom: 'A' ~~ /<.upper>/; say $/.Bool; say $/.to; say $/.from
p6eval nom: OUTPUT«Bool::True␤1␤0␤»
pmichaud there will be a fix of some sort at the qregex level.
dalek kudo/nom: ad0bb3f | pmichaud++ | tools/build/NQP_REVISION:
Bump NQP_REVISION to get -e argument handling fix.
kudo/nom: 953c714 | pmichaud++ | src/ (3 files):
Clean up $*ARGFILES, .lines, .eof, and .get.
pmichaud I'm still thinking about whether or not we can return the type object to indicate a failed match. I'm guessing that probably has a few holes in it. 06:33
06:33 ggoebel joined 06:35 perlhack joined
pmichaud anyway, time for sleep here -- will fix the regex bug after sleep 06:35
moritz good night
06:36 perlhack left 06:52 kfo joined 06:59 kjeldahlw joined 07:04 kjeldahlw left 07:05 mj41 joined, kjeldahlw joined 07:10 kjeldahlw left 07:11 kjeldahlw joined 07:26 xinming left, xinming joined 07:57 Jackneill joined, Jackneill left, Jackneill joined 08:05 daniel-s left 08:12 Trashlord joined 08:23 thou left 08:41 daxim joined 08:46 scottjl joined
scottjl Hi all. I'm trying out perl6... after a git clone, during perl congifure.pl --gen-parrot, I get a "Can't locate Test/Harness.pm in @INC" 08:48
08:50 dakkar joined
scottjl Does anyone have any suggestions? 08:50
moritz scottjl: then your perl installation is broken or incomplete. Test::Harness is a core module since perl 5.0
scottjl: what kind of system is that?
operating system, perl version
(for example on debian you need to install the perl-modules package to get all perl core modules) 08:51
scottjl fedore f14 upgraded to f15. perl --version is v5.12.4
moritz then please check if there's a 'perl-modules' package or something of a similar name 08:52
BinGOs it might be 'perl-dev'
moritz wouldn't that be header files? 08:53
BinGOs You'd think wouldn't you.
moritz I would :-)
scottjl Thank you moritz. Fixed with "yum install perl-Test-Harness". 08:55
BinGOs rpmfind.net//linux/RPM/mandriva/upd....i586.html 08:56
For example
moritz they really put core modules into separate packages and not install them by default? that's kinda... scary
BinGOs Oh it requires perl-Test-Harness 08:57
moritz seems that it's a meta package that pulls in the core modules
that factoring might make it easier to upgrade a core module
BinGOs It's easy to upgrade a 'core' module right now. 08:58
moritz just uses perlbrew with a sufficiently new perl :-)
BinGOs I've not been that familiar with Redhat-esque systems, but have had to get up to speed with them in the past week or so. 09:03
I've managed to just use Debian/Ubuntu etc until now. 09:04
09:06 Jackneill left 09:15 kjeldahlw left 09:17 Trashlord left
arnsholt moritz: Ubuntu doesn't come with perldoc, annoyingly 09:20
09:20 Jackneill joined
moritz same with debian, needs perl-doc installed 09:21
09:21 Trashlord joined
arnsholt Right. It's really strange. Especially since the documentation is in the source code anyways >.< 09:22
09:23 perlhack joined 09:26 wamba joined 09:27 wolfman2000 left 09:29 perlhack left 09:45 Jackneill left 09:54 scottjl left
moritz if I were to write a matrix of feature comparison for different Perl 6 compiler, what software could I use to generate them? 09:56
09:57 Jackneill joined
moritz nom: my @x = (1, 2, 3); @x[0] := my $x; $x = 8; say @x 09:58
p6eval nom: OUTPUT«Cannot use bind operator with this LHS at line 1, near " my $x; $x"␤current instr.: 'nqp;HLL;Grammar;panic' pc 23510 (src/stage2/gen/NQPHLL.pir:6316) (src/stage2/gen/NQPHLL.pm:329)␤»
09:58 ab5tract joined 09:59 woosley left 10:15 kaare_ joined 10:16 Jackneill left 10:19 Jackneill joined 10:27 Jackneill_ joined 10:29 Jackneill left 10:31 kjeldahlw joined, Jackneill_ left 10:40 domidumont left 10:44 Jackneill_ joined, domidumont joined 10:45 pernatiy left 10:54 wamba left 11:05 wamba joined 11:07 Jackneill_ left 11:08 Jackneill joined, Jackneill left, Jackneill joined, MayDaniel joined 11:18 cosimo left, cosimo joined 11:19 fhelmberger joined, fhelmberger left 11:20 fhelmberger joined, pernatiy joined 11:29 fhelmberger left 11:30 fhelmberger joined 11:36 kaare_ left, mj41 left, Instil left, mj41_nb joined, kaare_ joined
dalek kudo/nom: 512670d | moritz++ | / (3 files):
enable ro-accessors for natively typed attributes
11:43
11:45 icwiener joined 11:48 Instil joined 11:49 MayDaniel left 11:54 ab5tract left 11:59 icwiener left 12:01 icwiener joined 12:10 icwiener_ joined, icwiener left 12:16 bluescreen10 joined 12:33 ab5tract joined
moritz nom: say (* + 2).(5) 12:37
p6eval nom: OUTPUT«7␤»
moritz nom: say (* + 2).WHAT
p6eval nom: OUTPUT«_block1014␤»
daxim .WHAT # .oO(this value is unacceptable, it's an outrage!) 12:43
12:50 daniel-s joined
moritz github.com/perl6/features 12:52
daxim rm Rakudo (branch master) 12:53
rm Rakudo (branch nom)
↑ typo 12:54
sbp moritz: + means has feature, - means doesn't have feature, and missing entry indicates not tested?
moritz sbp: missing entry = no data available
and rm+- means "rakudo master partially implements it"
optionally with notes in parenthesis
12:54 daniel-s left 12:55 daniel-s joined
moritz Zip: rm+- (only for two lists) 12:55
daxim: thanks, fixing
eventually somebody could automatically derive most of that data by associating features with test files, and smoking the compilers 12:56
but that's so much work that I don't see that happening in the next few weeks 12:57
dalek atures: 7eee291 | moritz++ | features.txt:
typo noticed by daxim++
12:58
12:58 wamba left
moritz is anybody interested in writing a script that turns that file into a HTML table? 13:01
13:02 Holy_Cow joined 13:03 Holy_Cow left
moritz nom: sub f(::T $x) { say $x.WHAT }; f 12 13:04
p6eval nom: OUTPUT«Int()␤»
moritz perl6: sub f(::T $x) { say $x.WHAT }; f 12
p6eval niecza v7-44-g54b5188: OUTPUT«Unhandled exception: Name component T not found in ::GLOBAL␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 570 (CORE die @ 2) ␤ at /home/p6eval/niecza/src/Metamodel.pm6 line 146 (Metamodel Namespace.get_pkg @ 22) ␤ at /home/p6eval/niecza/src/Metamode…
..pugs: OUTPUT«Int␤»
..rakudo 922500: OUTPUT«Int()␤»
moritz perl6: sub f($x as Int) { say $x.WHAT }; f "12" 13:05
p6eval pugs: OUTPUT«*** ␤ Unexpected "as"␤ expecting "?", "!", trait, "=", default value, "-->" or ")"␤ at /tmp/5Tr5lfJcqc line 1, column 10␤»
..niecza v7-44-g54b5188: OUTPUT«===SORRY!===␤␤Action method trait_mod:as not yet implemented at /tmp/gfPOJnUoLz line 1:␤------> sub f($x as Int⏏) { say $x.WHAT }; f "12"␤␤Unhandled exception: Cannot use hash access on an object of type Str␤ at /home/p…
..rakudo 922500: OUTPUT«Int()␤»
moritz niecza: sub f(**@x) { say @x.perl }; my @a = 1, 2; my @b = 4, 5; f @a, @b 13:06
p6eval niecza v7-44-g54b5188: OUTPUT«===SORRY!===␤␤Slice parameters NYI at /tmp/YfF4mADPTU line 1:␤------> sub f(**@x⏏) { say @x.perl }; my @a = 1, 2; my @b =␤␤Unhandled exception: Check failed␤␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 570 (CORE …
dalek atures: 795a430 | moritz++ | features.txt:
add more data points
13:09
13:16 donri joined 13:34 tokuhirom joined 13:41 drbean left 13:56 daniel-s left
dalek atures: 569fb82 | moritz++ | features.txt:
fix a few typos
13:59
kthakore_ hello 14:05
moritz hi kthakore_ 14:06
kthakore_ Su-Shee: ping? Before I left for India you mentioned there are some resources for making perl6 modules, what are thouse again?
hi moritz
those*
Su-Shee kthakore_: I have no idea? Sure it was me? 14:07
kthakore_ ... yeah
I have logs
I was talking about how I failed at Parrot dev and suck at language/compiler coding 14:08
moritz kthakore_: all the links are on modules.perl6.org, iirc
Su-Shee then I don't remember what I answered, I don't recall.
kthakore_ moritz: oh Su-Shee mentioned some other stuff .. I guess I can just peruse logs again
I prolly should have done that from the start. Sorry for wasting time. 14:09
moritz: The link you gave me doesn't help me with handling C deps and installation 14:11
I have to check for SDL install and/or install it
moritz I don't think there's much support for that in Perl 6 land 14:12
qx// and parsing the output maybe
kthakore_ ah ... ok 14:13
[Coke] wonders why facebook shows him as ... The Perl Foundation
kthakore_ is stuck
[Coke]: Cause you are TPF ... all of it!
moritz: how can I do this? In perl5 we have modules that search PATH and stuff for this. Also XSLoader allows us to preload libraries from sharedirs. 14:14
Maybe for now I can keep the SDL 1.3 code with the module, but how to compile that stuff on varying machines? 14:15
moritz well, you can search %*ENV<PATH> too
rakudo: say %*ENV<PATH>
p6eval rakudo 922500: OUTPUT«/home/p6eval/rakudo/parrot:/home/p6eval/perl5/perlbrew/bin:/home/p6eval/rakudo/parrot:/usr/local/bin:/usr/bin:/bin:/usr/local/games:/usr/games␤»
moritz kthakore_: and there's zavolaj/NativeCall.pm for calling C functions 14:16
kthakore_ moritz: yeah those are clashing with the differing versions of parrot 14:17
it didn't work with parrot 3.5 for me.
moritz I know :(
kthakore_ oh ...
Hmm maybe if I start making a list of issues and just keep buggin for help 14:18
moritz they need fixing, or you need to use older versions
14:18 mkramer joined
kthakore_ right ... that just adds way too much on my plate 14:18
14:19 mkramer left
kthakore_ that means I would have to kickstart NativeCall, Alien::SDL and SDL6. Then testing ... and ... phew 14:19
moritz well, you're a pioneer
kthakore_ I am gonna start with making a list 14:20
moritz: yeah I have been here before ( SDL in p5p )
14:20 icwiener_ left
kthakore_ having NativeCall would make my life alot easier tough 14:20
a parser for C header files to Native Call might be a good idea 14:21
14:21 masak joined
masak popping in because I just realized something. 14:21
moritz waits for a big revelation 14:23
masak (sorry, mandatory keyboard break software kicked in) :) 14:25
if github.com/perl6/specs/commit/e84b...361054ea05 is now the case...
...then the subs at github.com/rakudo/rakudo/blob/mast...st.pm#L176 are wrongly named.
kthakore_ moritz: this is what I thinking of doing github.com/PerlGameDev/SDL6/wiki
moritz masak: no, actually they are named correctly now 14:26
masak: before, and eval() would never die... now it can
masak moritz: huh? the mechanism is clearly meant to wrap around things to catch exceptions. hence, it should be named 'try_', not 'eval_' 14:27
flussence does that mean there's a way to test for the presence of a warn() now?
moritz masak: eval_dies_ok actually does an eval
masak I know, and that's wrong now. 14:28
moritz no.
masak it should do a try.
moritz it should do a try and an eval.
masak oh, ok.
yes, I see now.
moritz but the try{} is really just an internal detail
like it is in lives_ok
14:29 wtw left 14:30 birdwindupbird left
masak right. 14:31
14:31 ymasory joined
masak ok, so false alarm. :) no renaming necessary. 14:31
see y'all tonight. o/
14:31 masak left
moritz \o 14:31
14:32 _jaldhar left
kthakore_ I guess I am wasting time just yapping. thanks for the help. 14:32
pmichaud good morning #perl6 14:39
colomon o/
domidumont hi
14:40 daniel-s joined
moritz \o 14:41
14:41 wamba joined
domidumont BTW, this has been long overdue: Rakudo is now available on Debian/unstable (I've ironed out the last issues this week end) 14:42
14:42 MayDaniel joined
moritz domidumont++ # that's great! 14:42
pmichaud domidumont: I tried looking through the debian sites to find the details and couldn'g find them
domidumont pmichaud: here's rakudo package page: packages.qa.debian.org/r/rakudo.html
more details are on: ddumont.wordpress.com/ 14:43
Debian pkg-rakudo team is on Alioth: alioth.debian.org/projects/pkg-rakudo/ 14:44
takadonet www.reddit.com/r/perl6/comments/iz7...on_debian/ :) 14:45
domidumont ok, this team is currently made of 2 people which is admitedly not much :-p
14:45 daniel-s_ joined, daniel-s left
pmichaud domidumont: is the package just the compiler, or is it packaging rakudo star? 14:46
14:46 tokuhirom left, pmurias joined
domidumont pmichaud: We packaged rakudo-star tarball 14:47
pmichaud excellent. there's a new tarball coming out tomorrow.
domidumont Does it depend on new parrot (3.6.0) or is 3.3.0 good enough ?
pmichaud it uses the new parrot
domidumont ok, then a parrot package update is required. 14:48
pmichaud 3.3.0 had... issues (mainly with speed and gc)
domidumont Updating parrot may take some time ... 14:49
14:49 daniel-s__ joined
pmichaud okay. I'd probably recommend using rakudo star 2011.01 instead, then, it's a fair bit faster. 14:49
14:49 daniel-s_ left
pmichaud 2011.04 isn't really the optimal release to be using. 14:50
moritz wouldn't mind if it took a bit
domidumont Hmm, we don't have a package for 2011.01 and releasing an older version of rakudo may raise some eyebrows 14:51
pmichaud I agree, I'd be in favor of waiting for parrot-3.6.0 to be packaged.
domidumont Well, I can work on the package, but there are some organisational issues with debian-parrot-team to work out... 14:52
I mean I can work on parrot package. (I did most of the packaging work for parrot 3.3.0)
14:53 daniel-s__ left, daniel-s__ joined
pmichaud the new versions of star coming out over the next few months all will depend on 3.6.0 also. 14:54
domidumont I'm all for updating parrot 14:55
pmichaud should I join the packaging team, ooc? I've been meaning to get involved in debian packaging
domidumont pmichaud: no problem. The more, the merrier :-) 14:57
BTW, since you 14:58
14:58 _daniel-s__ joined, daniel-s__ left
domidumont pmichaud: BTW, since you're also admin for parrot team, could you approve my request to join parrot team with my new DD account (dod) and remove my old account (ddumont-guest) ? 14:59
pmichaud parrot team.... where?
moritz github?
pmichaud on github?
domidumont on Alioth
15:00 thou joined
[Coke] O_o 15:00
domidumont Hmm, sorry, I got confused, you're not admin, just member: alioth.debian.org/projects/pkg-parrot/
pmichaud right... didn't think I was an admin there
domidumont and I've just seen that my account was updated. Sorry about the noise 15:01
15:02 daniel-s joined 15:03 _daniel-s__ left
domidumont ok. I'll work on updating parrot. OTOH, I'll be on vacation soon, so it may take a while 15:04
pmichaud domidumont: okay. maybe I can get some of the other parrot team members to do an update. 15:06
I just requested to join pkg-rakudo
15:07 daniel-s left
domidumont what role do you want ? 15:07
moritz
.oO( world domination )
pmichaud admin seems appropriate, but I'm a newb on debian packaging so whatever role you think is appropriate is fine for now :) 15:08
domidumont :-)
pmichaud: welcome aboard, you're admin.
For me, admin role does not require packaging skills. Just common sense and good communication and some availability to answer questions :-p 15:11
pmichaud domidumont++ # thanks
domidumont pmichaud: you're welcome :-)
eiro domidumont++ 15:12
hello world
pmichaud and yes, domidumont++ for the packaging. I was just at FOSSCON this past weekend and my talk lamented that we didn't have a rakudo debian package yet... now I can say that we do. :-)
domidumont Excellent :-D 15:13
We also have to thanks ghedo (Alessandro Ghedini) for rakudo packaging, he did a lot of work too 15:14
15:15 MayDaniel left, shinobicl joined
moritz ghedo++ then :-) 15:17
domidumont Indeed :-) 15:18
dalek atures: 7701afe | moritz++ | / (3 files):
basic processing script
15:19
moritz moritz.faui2k3.org/tmp/features.html
15:20 Trashlord left 15:21 Trashlord joined
moritz it's really ugly right now, but I hope somebody with CSS skills can fix that for us :-) 15:22
pmichaud moritz++ 15:24
do we have a canonical place for the page itself to live? 15:25
or is that still being worked out?
moritz I want to put it somewhere on perl6.org, if nobody objects
pmichaud +1
forgiveness > permission
moritz perl6.org/compilers/features or so 15:26
pmichaud would love to have a way to add footnotes to that page (Yes, I know, "patches welcome!" :-) 15:27
moritz pmichaud: footnotes are already in the source file
pmichaud: please reload, just reduced the amount of text a bit :-)
pmichaud I see them in the source file... but now features.html isn't displaying correctly :) 15:28
moritz pmichaud: what's wrong? 15:29
pmichaud all I see are '+' and '+-' in the table cells
moritz and colors, hopefully? 15:30
pmichaud yes, and colors
oh, I see, that's the new label.
hmmmm.
moritz that's intended
pmichaud they're a little hard to see
moritz maybe I should put the legend box near the top
and with colors
pmichaud might help if the columns were equal-width, too.
and maybe center the +/-
dalek p: dc942fb | pmichaud++ | src/QRegex/Cursor.nqp:
Have !cursor_start create failing cursors by default. Fixes problem with default alpha/alnum/upper/etc. rules not failing properly (moritz++).
15:34
[Coke] (css skills) which now? 15:35
oh, features.html? Sure, I can take a look.
Is there a repo it's in? 15:36
pmichaud perl6/features
moritz there it's generated from template.html and process.pl (needs HTML::Template::Compiled) 15:37
dalek atures: 9e8d0db | moritz++ | / (3 files):
some small tweaks; center-align +/-, pmichaud++
15:38
moritz ok, I'll stop touching it, [Coke]++'s turn :-)
dalek atures: 46bde8a | pmichaud++ | features.txt:
Some features updates, corrections.
15:42
pmichaud why is rakudo listed with a '-' for conjunctions, ooc? 15:43
(in Regexes)
moritz rakudo: say 'abc' ~~ /\w+ & .+/
p6eval rakudo 922500: OUTPUT«abc␤»
moritz pmichaud: because I didn't know :-) 15:44
pmichaud nom: say 'abc' ~~ /\w+ & .+/
p6eval nom: OUTPUT«abc␤»
moritz \o/
pmichaud nom: say 'abc' ~~ /\w+ & .**2/
p6eval nom: OUTPUT«ab␤»
moritz niecza: say 'abc' ~~ /\w+ & .**2/
p6eval niecza v7-44-g54b5188: OUTPUT«Unhandled exception: Unable to resolve method oplift in class Any␤ at /home/p6eval/niecza/src/NieczaActions.pm6 line 274 (NieczaActions NieczaActions.op_for_regex @ 2) ␤ at /home/p6eval/niecza/src/NieczaActions.pm6 line 290 (NieczaActions NieczaActions.qu…
moritz niecza: say 'abc' ~~ /\w+ & .*/ 15:45
p6eval niecza v7-44-g54b5188: OUTPUT«Unhandled exception: Unable to resolve method oplift in class Any␤ at /home/p6eval/niecza/src/NieczaActions.pm6 line 274 (NieczaActions NieczaActions.op_for_regex @ 2) ␤ at /home/p6eval/niecza/src/NieczaActions.pm6 line 290 (NieczaActions NieczaActions.qu…
15:45 donri left
dalek atures: f255ab6 | moritz++ | features.txt:
update regex conjunction, pmichaud++
15:45
moritz counts that as a -
15:46 donri joined
pmichaud nom: say 'abc' ~~ / <[bc]>+ / 15:46
p6eval nom: OUTPUT«bc␤»
moritz nom: say so "a\tb" ~~ / <[\t]> / 15:47
p6eval nom: OUTPUT«Bool::False␤»
pmichaud correct, we still don't do \-escapes in <[...]> yet. 15:48
moritz I wonder if we should <:alpha> as char classes or unicode properties
pmichaud unicode properties. 15:50
15:51 REPLeffect joined
dalek kudo/nom: 3c8f0b6 | pmichaud++ | src/Perl6/Compiler.nqp:
Fix bug in interactive mode introduced by @*ARGS changes.
15:52
kudo/nom: d973a42 | pmichaud++ | / (2 files):
Bump NQP_REVISION to get alpha/alnum/digit/etc. fixes.
moritz niecza: say 'a' ~~ /<:Letter>/
p6eval niecza v7-44-g54b5188: OUTPUT«===SORRY!===␤␤Action method assertion:sym<:> not yet implemented at /tmp/Y3iLKMoGsX line 1:␤------> say 'a' ~~ /<:Letter⏏>/␤␤Unhandled exception: Unable to resolve method oplift in class Str␤ at /home/p6eval/niecza/src/…
dalek atures: c37acce | moritz++ | features.txt:
add Unicode properties
15:53
moritz nom: $_ = 'a'; say so m:c(10)/<?alpha>/ 15:54
p6eval nom: OUTPUT«Rebuild in progress␤»
moritz is that also fixed with that nqp patch? 15:55
pmichaud I don't know if it is or not.
> $_ = 'a'; say so m:c(10)/<?alpha>/ 15:56
Bool::True
looks like not
need lunch here... I'll fix it after lunch. 16:02
16:11 itz` joined 16:12 itz` left 16:13 kjeldahl joined 16:20 Trashlord left
dalek atures: 17f43e8 | Coke++ | / (2 files):
minor updates to xhtml/css.
16:21
[Coke] moritz: some updates.
sorear good * #perl6 16:22
16:23 jedai joined, mj41_nb left 16:24 shinobicl left 16:27 benabik joined
dalek atures: fa8c84a | Coke++ | template.html:
Add rounded corners.
16:32
atures: 8bc2ad0 | Coke++ | template.html:
pick slightly more legible colors.
16:32 wjguo joined
[Coke] feather.perl6.nl/~coke/features.html 16:33
16:33 cdarroch joined, cdarroch left, cdarroch joined 16:46 Trashlord joined
moritz [Coke]++ 16:52
16:54 wolfman2000 joined, mj41_nb joined 16:56 pmurias left 17:01 ymasory left
PerlJam [Coke]++ except for the color chosen for not implemented. :) 17:03
17:05 soh_cah_toa joined 17:07 masak joined
masak hi, zebry. 17:08
17:09 dakkar left, [particle]1 joined
dalek atures: 8a260b4 | pmichaud++ | features.txt:
Fix typo, add entry for operators on natives.
17:09
masak moritz++ # feature comparison table 17:10
17:11 [particle] left
masak [Coke]++ # improvements 17:13
17:13 daxim left 17:14 uvtc joined
[Coke] PerlJam: I just picked a slightly less jarring red. you have an alternate suggestion? 17:15
PerlJam I liked the jarring red actually. Yours is just a tad too close to pink for me. 17:16
But I'm just bikeshedding it, so ignore me.
dalek atures: 3767f94 | masak++ | template.html:
improved color scheme
17:19
masak I'm just JFDI-ing, so ignore me :P
feather.perl6.nl/~masak/features.html 17:21
[Coke] yah, the orange is fine.
masak yellow on white is never a good idea IMHO.
(and bright cyan on white is almost as bad) 17:22
[Coke] masak: there was no yellow when i was done.
I wouldn't call it bright cyan, but orange is fine. ;) 17:23
masak [Coke]: right. just giving my reasons for going with orange instead of yellow. :)
uvtc Wow. Had to pipe in -- that implemented features comparison table is great.
felher masak++ for blog post. I played and solved it. But not without look into the source. I just couldn't find the right words sometimes. :). But great adventure. And quite funny :D
masak felher: thank you. :) 17:24
felher: I'm very interested to hear which parts had you stumped, and which words you tried.
(they're probably fine, it's just that my imagination is limited)
pmichaud (features) I think the red is too strong/saturated. It tends to overpower the table. 17:28
felher masak: the hardest problem was guessing the names for the disks. What i tried was little, and big. After some movements the game suggested AC, AB and so on and i was glad to use that. :) 17:29
uvtc re. the feature comparison table: I think it would be easier to read if "unknown" cells were at least rendered light gray or something.
pmichaud it's fine in isolation ("Natively typed attributes"), but large blocks of red ("Regexes") are a bit jarring to my eyes
masak feel free to change it to #f44 or #f66. 17:30
felher: oh, nice!
felher: yes, it's a tricky problem to lead people into the right path there.
uvtc masak: re. the crypt, I had to look in the source to figure out how to move the disks around. It would be helpful if you gave a hint in the game to use the "ab", "ca"... commands.
masak uvtc: noted. 17:31
soh_cah_toa not that i have anything to do w/ this, but when i have to come up with a color scheme www.colorschemer.com/schemes/ has always been helpful 17:32
felher masak: Yeah. :) There was one more thing but i can't recall it at the moment. Maybe i get it later. 17:33
masak I agree with uvtc that missing cells should be pale gray.
uvtc At perl6.org/compilers/ the link to Rakudo's status rakudo.org/status is broken.
tadzik pops out of nowhere 17:34
masak tadzik! \o/ 17:35
tadzik empty fields on the features page mean "unknown"? 17:36
masak: o/
I has a computer, but it's less than fast, I may give it some love today and maybe it'll get a bit better
laptop fixing mission failed, today. "Oh Lenovo? We don't fix those anymore, yes, the official site lies" 17:37
masak does anyone want to submit github.com/masak/crypt to HN and/or Reddit? :)
tadzik Will try tomorrow
uvtc masak: about how long does it take crypt to start up on your system? 17:38
soh_cah_toa masak: a text adventure game, nice. i remembering writing those in qbasic :)
i have to try this out
uvtc On my system (fairly modern laptop, R* 2011.04) it takes roughly 15 s. 17:39
[Coke] tadzik: yah. 17:40
ENODATA.
masak uvtc: takes 66 s here on my Ubuntu VM. 17:42
17:43 Chillance joined 17:44 pernatiy left
soh_cah_toa masak: digg.com/news/technology/a_text_adv..._in_perl_6 17:44
uvtc masak: that's a big twinkie. 17:45
masak soh_cah_toa++ # \o/
uvtc masak: Is most of that time spent compiling crypt.pl into a form that Parrot can run? 17:47
tadzik I suppose so 17:48
uvtc Is there a way to pre-compile the file?
tadzik I suppose so 17:49
try perl6 --target=pir file.pl > file.pir
then parrot file.pir
or parrot -o pbc file.pbc file.pir; parrot file.pbc 17:50
to go full hardcore^Wbytecode :)
17:50 impious joined, impious left
masak IIUC, the speed gain for running the .pbc rather than the .pir isn't that great. 17:50
but .pir rather than .pl will surely speed things up, circumventing the parsing which is the heavy step. 17:51
uvtc Running `perl6 crypt.pl` took 15 s to start up. Compiling it using `perl6 --target=pir crypt.pl > crypt.pir` took about the same amount of time. 17:52
tadzik and running
uvtc Running that file via `parrot crypt.pir` started the game fairly quickly (a couple of seconds) but then died.
pmichaud I suspect precompiling stuff still doesn't work very well... there's a lot of startup stuff that gets missed. 17:53
uvtc tadzik: Thanks, tadzik++
pmichaud: If crypt gets some attention, my guess is that people will howl at how long it takes to start up. Would be nice if precompiling worked smoothly. 17:54
pmichaud uvtc: I agree, it would be nice. :) 17:55
just because it would be nice doesn't make it easy. :)
it's very challenging to get precompiled stuff to work properly (more)
17:56 soh_cah_toa left
pmichaud in the case where we compile the source, the compiler is loaded and initialized and *then* we start compiling. 17:56
In the case where we start from a pre-compiled .pir file, the .pir file gets loaded and then *it* has to load the compiler and make sure any initializations get run properly.
pmichaud.com/sandbox/features.html # toned down the red slightly, added grab boxes for the empty spaces 17:58
*grey
uvtc pmichaud: thanks for the info. Also, feature table colors look nice. I like the orange. 17:59
masak pmichaud: the red looks nice. the grey is almost invisible on my screen, though. 18:00
(and the contrast with the white question mark inside of it is too small)
18:01 tokuhirom joined
pmichaud well, I wanted the grey to be very subdued. I can make it less so. 18:01
reload, now at #aaa 18:02
masak better. maybe the question mark should be #666 or something. 18:03
pmichaud I think that calls too much attention to the question mark, personally.
uvtc masak: the question-mark of the beast 18:04
pmichaud maybe there shouldn't be a question mark. :)
maybe I'll use #aaa for the question mark :)
uvtc pmichaud: the ± looks too dark 18:06
pmichaud uvtc: I agree -- now switched to #333 18:09
dalek atures: 79b0910 | pmichaud++ | / (2 files):
Add placeholders for unknown entries, tone down colors a bit.
18:13
18:16 dodathome joined 18:21 spq1 joined
[Coke] nom: my $notFish; $notFish.++; 18:26
p6eval nom: OUTPUT«Could not find sub &postfix:<.>␤current instr.: '_block1002' pc 77 ((file unknown):63250011) (/tmp/h0bN7qZP6W:1)␤»
18:26 pernatiy joined 18:28 pyrimidine joined 18:29 nothingmuch left
masak rakudo: my $a; $a.++; say "alive" 18:35
p6eval rakudo 922500: OUTPUT«Could not find sub &postfix:<.>␤ in main program body at line 22:/tmp/LYLdHHj5Nc␤»
18:37 dukeleto left 18:38 bluescreen10 left 18:39 jevin left
dalek p: 4fb5124 | pmichaud++ | src/QAST/Compiler.nqp:
Avoid vtable_bool for determining whether subrule passed/failed.
18:39
18:39 bluescreen10 joined, jevin joined, birdwindupbird joined 18:44 uvtc left 18:48 Mowah left
dalek kudo/nom-install: 6a45ccb | moritz++ | / (2 files):
try to make nom installable

This is... wrong. It installs rakudo's pbcs into the general parrot install location, because I can't figure out where the pbcs are actually loaded, and how. On the plus side it gives us an idea what files need to be installed.
It gets us so far that we can actually run the perl6 binary from outside the installation path, but using an Int literal results in "Could not locate compile-time value for symbol Int"
18:48
[Coke] moritz: if you want to know where to install PBC, there's a parrot-config setting that should tell you. (or so I think) 18:50
moritz [Coke]: I know where I *can* install PBC to, but that's not where it *should* be installed to 18:51
18:51 dukeleto joined
pmichaud where do the NQP .pbcs get installed to? 18:52
moritz into the same global location that nom-install installs its pbcs 18:53
ie install/lib/parrot/3.6.0-devel/library/
pmichaud I'm willing to live with that for now.
moritz ok, then the main problem is that it doesn't work when I use literals 18:54
[Coke] moritz: I don't see a single flag, but "parrot_config prefix versiondir" should get you really close. 18:55
pmichaud [Coke]: in my experience, the problem isn't so much figuring out where to put them as figuring out how to load them on the other side
[Coke] hokay. 18:56
pmichaud iwbni we could take all of the .pbc's and pbc_merge them into a single .pbc, though. 18:57
I don't think that PERL6_ML_PBC should appear in both LIBDIR and LIBDIR/blib 18:58
moritz yep, should probably only the the latter
pmichaud actually I don't think there should be a LIBDIR/blib
blib is where we put stuff while building... but after building, there shouldn't be a blib 18:59
(or more accurately, the "built" versions shouldn't need a blib) 19:00
moritz the problem is that Perl6::SymbolTable.load_setting hard-codes the path 19:02
to ModuleLoader.pbc
because.... the logic for module loading is in ModuleLoader
and we get away with that either in the setting or in the installed location 19:03
the problem is really that we have different layouts in install and build dirs 19:04
maybe we should build everything in blib/ in the source dir
pmichaud I think we run into circularity/bootstrap issues there. 19:05
moritz do we? we can add a library path at the command line, I think
pmichaud yesterday while working on @*ARGS it also bugged me that we call load_bytecode so many times for the same file. 19:06
maybe that was in nqp, though. 19:07
19:08 bluescreen10 left
moritz I thikn I want to experiment with "everything in blib/" a bit 19:08
19:12 wamba left 19:13 Jackneill left 19:14 Jackneill joined, Jackneill left, Jackneill joined
pmichaud seems like each ModuleLoader should be smart enough to know where to look for its modules 19:14
by default it should look in the appropriate parrot language install directory, unless directed (by command line option) to look elsewhere. 19:15
19:17 mike joined, mike is now known as Guest1080
pmichaud so, Perl6::ModuleLoader should look in install/lib/parrot/<version>/languages/perl6 19:18
oh, there's a bunch of ickiness there, though
because a lot of the modules are being loaded by nqp's ModuleLoader
moritz including Perl6::ModuleLoader 19:19
pmichaud only the core setting gets loaded by Perl6::ModuleLoader
moritz but Perl6::ModuleLoader gets loaded by nqp's ModuleLoader, no?
hm, not sure
pmichaud no
it's loaded by load_bytecode 19:20
load_bytecode "blib/Perl6/ModuleLoader.pbc"
moritz ah right, we've talked about that before :-)
19:21 Guest1080 left
moritz this is frustrating, now I get "Cannot fetch object from non-existant serialization context 1311615152.93596" 19:21
that's something I don't know how to debug
pmichaud I suspect this is all something that really wants jnthn++ presence to resolve (more) 19:23
moritz aye, I fear so :/
19:23 bluescreen10 joined
pmichaud I can certainly come up with something that "works", but I think some of the current structure is also due to portability concerns 19:23
and there are several startup-ish issues we need to resolve as well... such as dealing with @*ARGS in precompiled modules 19:24
er, precompiled programs.
19:26 Jackneill left
moritz phenny: tell sorear about moritz.faui2k3.org/tmp/features generated from github.com/perl6/features/ - please review it and add missing information about niecza before I put up that page on perl6.org somewhere 19:26
phenny moritz: I'll pass that on when sorear is around.
19:28 cdarroch left
pmichaud > $_ = 'a'; say so m:c(10)/<?alpha>/ 19:28
Bool::False
moritz \o/ 19:29
dalek kudo/nom: 91d7ab5 | pmichaud++ | tools/build/NQP_REVISION:
Bump NQP_REVISION to get cursor boolean fixes. (moritz++)
[Coke] pbc_merge should be ok to use these days, pmichaud. 19:32
masak so far, Niecza and Rakudo appear quite evenly matched in the features table. 19:33
19:34 soh_cah_toa joined 19:35 pmurias joined
flussence nom: infix:<½>($a, $b) { ($a + $b) / 2 }; say 4 ½ 3; 19:35
p6eval nom: OUTPUT«Confused at line 1, near "infix:<\x{bd}>("␤current instr.: 'nqp;HLL;Grammar;panic' pc 23510 (src/stage2/gen/NQPHLL.pir:6316) (src/stage2/gen/NQPHLL.pm:329)␤»
flussence looks like you can fill in 'Defining new operators' there...?
rakudo: infix:<½>($a, $b) { ($a + $b) / 2 }; say 4 ½ 3; 19:36
p6eval rakudo 922500: OUTPUT«===SORRY!===␤Confused at line 22, near "infix:<\x{bd}>("␤»
flussence oh
unless I'm being stupid somewhere again :)
rakudo: sub infix:<½>($a, $b) { ($a + $b) / 2 }; say 4 ½ 3;
p6eval rakudo 922500: OUTPUT«3.5␤»
flussence nom: sub infix:<½>($a, $b) { ($a + $b) / 2 }; say 4 ½ 3;
p6eval nom: OUTPUT«Confused at line 1, near "say 4 \x{bd} 3;"␤current instr.: 'nqp;HLL;Grammar;panic' pc 23510 (src/stage2/gen/NQPHLL.pir:6316) (src/stage2/gen/NQPHLL.pm:329)␤»
flussence (that page looks much nicer than it did a few hours ago, btw!) 19:37
masak crowdsourcing++ 19:43
flussence and (I'm not trying to take all the fun for myself, honest) "is rw", "is copy", "@a».say" and "proto sub" all work in niecza :) 19:44
19:48 silug_ left, silug_ joined 19:53 rperry joined
pmichaud niecza: sub xyz($a is copy) { $a++; say $a }; xyz(3); 19:56
p6eval niecza v7-44-g54b5188: OUTPUT«4␤» 19:57
pmichaud niecza: sub xyz($a is copy) { $a++; say $a }; my $b = 3; xyz($b); say $b;
p6eval niecza v7-44-g54b5188: OUTPUT«4␤3␤»
pmichaud niecza: sub xyz($a) { $a++; say $a }; my $b = 3; xyz($b); say $b;
p6eval niecza v7-44-g54b5188: OUTPUT«Unhandled exception: Writing to readonly scalar␤ at /tmp/FCQbLF6BZc line 0 (MAIN xyz @ 0) ␤ at /tmp/FCQbLF6BZc line 1 (MAIN mainline @ 1) ␤ at /home/p6eval/niecza/lib/CORE.setting line 1933 (CORE C906_ANON @ 2) ␤ at /home/p6eval/niecza/lib/CORE.setting …
pmichaud niecza: sub xyz($a is rw) { $a++; say $a }; my $b = 3; xyz($b); say $b;
masak \o/
p6eval niecza v7-44-g54b5188: OUTPUT«4␤4␤»
pmichaud looks good to me!
fwiw, I think that @a».say is a meme we need to kill. 19:59
niecza: my @a = <the quick brown fox>; say @a».flip
p6eval niecza v7-44-g54b5188: OUTPUT«ehtkciuqnworbxof␤»
20:00 birdwindupbird left
masak pmichaud: what made you think of @a».say ? 20:00
oh, flussence mentioned it.
pmichaud 19:44 <flussence> and (I'm not trying to take all the fun for myself, honest) "is rw", "is copy", "@a».say" and "proto sub" all work in niecza
masak aye.
flussence I know what he means, it's semantically wrong
pmichaud in nom I'm going to make sure that hyper (methods) do things out-of-sequence. 20:01
masak good.
pmichaud niecza: my @a = <the quick brown fox>; say @a »~» 'x';
p6eval niecza v7-44-g54b5188: OUTPUT«thexquickxbrownxfoxx␤»
dalek atures: 965b89e | pmichaud++ | features.txt:
More features as noted by flussence++ on #perl6.
20:02
20:10 soh_cah_toa left 20:18 molaf joined
[Coke] is moritz.faui2k3.org/tmp/features.html being updated with the changes? 20:19
moritz [Coke]: I'm updating it right now
[Coke]: but it's not the final/permanent URL
[Coke] something off perlcabal, perhaps? 20:20
moritz perl6.org
20:30 wamba joined 20:33 rperry left 20:34 dodathome left
sorear good * #perl6 20:35
phenny sorear: 19:26Z <moritz> tell sorear about moritz.faui2k3.org/tmp/features generated from github.com/perl6/features/ - please review it and add missing information about niecza before I put up that page on perl6.org somewhere
masak evenin', sorear.
sorear o/ masak
well it's the last monday of the month again 20:36
masak yay
sorear I'll need to cut a niecza release, though playing with J has meant I didn't get to everything I planned
well there's always Aug
pmichaud only one more Monday before financial doomsday strikes the U.S. :)
masak sorear: Håkan told me to mention K and Kona to you ;) 20:37
sorear Is Håkan hakank? 20:39
masak aye.
he's not on IRC, so I have to speak for him.
20:39 thou left
sorear heh. Would he prefer me to join some forum of his choice? 20:41
masak he seems to like email. :)
sorear heh. well he has my address 20:42
->backlog 20:43
[Coke] pmichaud: I thought the debt ceiling didn't explode until 02AUG 20:44
sorear back 20:47
pmichaud: hmm. do you think the debt ceiling will adversely impact my ability to release niecza v9? 20:48
moritz: Should I just correct and push to github.com/perl6/features/ ?
diakopter heh
moritz sorear: sure 20:49
masak "Niecza v8" sounds like it would have something to do with JavaScript. :) 20:50
20:56 soh_cah_toa joined
masak blog post! strangelyconsistent.org/blog/what-i...e-blogging 20:57
moritz masak++ 20:59
20:59 bluescreen10 left
dalek atures: a3b3dc7 | moritz++ | / (2 files):
add output filename option to script; expand README
21:01
atures: dddcf99 | moritz++ | / (2 files):
implement footnote display - more work for [Coke]++ and other CSS gurus
21:13 benabik left 21:14 bluescreen10 joined
felher masak++, just again. 21:14
masak thanks. 21:18
21:24 jevin left 21:25 jevin joined, kaare_ left 21:30 kjeldahl left 21:36 Trashlord left
diakopter moritz: maybe the feature items could hyperlink to their corresponding Sxx descriptions 21:37
21:37 drbean joined
frettled masak++ :) 21:38
sorear hmm 21:39
.u 177 21:40
phenny sorear: Sorry, no results for '177'.
sorear .u U+B1
phenny sorear: Sorry, no results
sorear ah, ±
21:41 bluescreen10 left 21:43 Psyche^ joined 21:46 Patterner left, Psyche^ is now known as Patterner 21:48 La4ux joined, La4ux left, pmurias left
pmichaud 20:44 <[Coke]> pmichaud: I thought the debt ceiling didn't explode until 02AUG 21:54
right. 01Aug is the "one more monday" before the ceiling caves in :) 21:55
21:56 dorlamm joined
pmichaud reads the "blog post!" and enjoys it. 21:56
22:01 Guest99070 left, cognominal_ joined
masak ok, there's now a PDF. it's 83 pages long. feather.perl6.nl/~masak/book.pdf 22:10
more exactly, there's now a build system in the adventure-game-book repository that takes the 30 chapters, converts them from Markdown to LaTeX, and then renders that as a PDF. 22:12
22:12 mj41_nb left
dalek atures: 2540598 | sorear++ | features.txt:
Add/tweak some entries for niecza
22:13
masak 'night, #perl6. 22:20
22:20 masak left 22:22 franek joined 22:25 dorlamm left 22:30 soh_cah_toa left
tadzik Rakudo compilation takes about 30 minutes on this machine. No way I'll do any gsoc on this :/ 22:36
tadzik looks at cpu prices 22:38
22:45 whiteknight joined, Trashlord joined 23:00 donri left 23:06 supernovus joined 23:09 Trashlord left
supernovus This may sound like a strange question, but is there anyway to reload a package once it's been called with "use" or "require" a first time. Say a long-running process and you've made changes to a class but don't want to have to restart the entire service? 23:09
23:11 Trashlord joined
[Coke] you can remove it from %INC and re-use it, I think (perl 5) 23:11
(though I'd use 'do <file>' in that case.) 23:12
evil: github.com/coke/tigerlily/blob/mas...cj.pl#L657 (unloads) github.com/coke/tigerlily/blob/mas...cj.pl#L267 (loads) 23:13
... please don't make me rewrite my lily Bot in perl 6. 23:14
23:14 _jaldhar joined
supernovus [Coke], thanks. Is there an equivilant to %INC in perl6? I can find @*INC but %*INC at least on rakudo master gives a "Contextual %*INC not found" error. 23:15
Hmm, and I'm thinking the biggest problem is going to be the dreaded "Illegal redeclaration of symbol 'ClassYouTriedToReload'" error. I think I'll just give on this and admit that changes to classes means restarting the daemon. Not optimal for a MOO/MUD but a necessary evil I guess. Only insane people try to write MOO engines in Perl 6... I blame Carl, it was his "crypt" game that inspired my latest bout of madness. 23:20
[Coke] you could just write it in MOO. :) 23:22
(which, coincidentally, is what the aforementioned lily is written in)
I'm not sure that's fully baked in p6 yet.
there is a p5 module that does a better job of reloading rather than just stomping on already loaded code. (hurm.) 23:23
Something like Module::Reload 23:24
supernovus I'll look into those, thanks :-) 23:27
23:29 soh_cah_toa joined 23:30 orafu left, orafu joined 23:48 jferrero joined 23:51 _jaldhar left, _jaldhar joined 23:52 ab5tract left, mattp_ left, franek left, molaf_ joined 23:53 molaf left 23:58 jferrero left 23:59 soh_cah_toa left