pugscode.org/ | nopaste: sial.org/pbot/perl6 | ?eval [~] <m oo se> | We do Haskell, too | > reverse . show $ foldl1 (*) [1..4] | irclog: irc.pugscode.org/
Set by diakopter on 11 July 2007.
00:00 REPLeffect left
TimToady there are two competing subcultures here that come up with different answers to the question, "Is an unknown contributor more likely to help us or hurt us?" 00:01
also, it's not clear whether the "right" answer to that changes over time, and if so, whether the answer has changed yet.
the answer also depends on how you view licensing and copyright control 00:03
mncharity the "is there a wiki?" question was triggered by an old item on my infinite todo list, triggered by the Container paragraph in S10. basically, the Snn have a shape that doesn't fit well inside my head. so I've long considered spending a few days stuffing the Snn into my head, and holding them long enough to dump them in a shape which fits my head better. 00:05
re wiki, there's another factor,
00:05 Lorn joined 00:06 theorb joined
mncharity wikis are under active spam threat. you need to have enough of a critical mass community to keep it under control. if you don't, the wiki melts down. p.n.a is critical mass, pf.o isnt yet. 00:06
re Container, I'd love if someone did a more "Perl 6 - a reference manual" style presentation. Eg, Namespaces, Names, Containers, Variables, Types, etc. I find Snn ok as reference when I already understand something, but as having a very high barrier to entry when I don't. 00:10
00:10 stevan_ joined
mncharity as there are core ideas, with scattered incremental descriptions (without crossrefs). eg, Containers.] 00:11
hi stevan_
Sigh. Speaking about subcritical mass wikis, one which was created because the oz/mozart language had similar documentation issues, and thus need a single integrated "virtual reference manual", implemented as a table of contents to the scattered parts,... I guess I know what I'm doing this evening... fixing a smashed wiki. :/ 00:15
00:19 theorbtwo left 00:39 thoughtp1lice joined 00:40 lidden joined 00:46 nipotaway is now known as nipotan
mncharity eh, not so bad. back up. www.vendian.org/oz/wiki/index.cgi?VirtualManual 00:51
lambdabot Title: Unofficial Mozart/Oz Wiki: VirtualManual
00:55 thoughtpolice left
jjore-m Neat. I like Mozart/Oz and would like to use it in my perl one day. 00:57
Or use Perl in my Mozart/Oz.
Whatever.
mncharity :) 01:00
01:06 theorb left 01:07 theorbtwo joined 01:16 ispy_ joined 01:18 jjore-m left 01:19 Notrix left 01:38 hkb_zz is now known as hakobe 01:45 kanru left 01:54 amnesiac left 01:55 cnhackTNT joined 01:56 ferreira joined, ispy_ left 02:01 araujo joined 02:05 Jamtech joined 02:13 thoughtp1lice is now known as thoughtpolice 02:40 ferreira left 02:42 kanru joined
jmacclure quick question: how does one become a known contributer? 02:45
besides the obvious, of course. I'm referring to TimToady-sama's original comment... 02:46
03:10 Jamtech left
mncharity jmacclure: re 'known contributor, original comment', you mean about wikis? 03:16
actually, have to run. bbl &
03:17 jjore-m joined
jmacclure I thought he was talking in general re: perl6 03:31
03:47 amnesiac joined 03:54 SubStack left
mncharity jmacclure: well, the discussion context was the different policies of perl.net.au/wiki/Perl_6 and www.perlfoundation.org/perl6/index.cgi towards anonymous edits. Pugs is certainly also more open than many other software projects, and the licensing/copyright issues apply there too, so it could have been a more general observation as well. 03:55
lambdabot Title: Perl 6 - PerlNet
03:56 SubStack joined 04:41 hakobe is now known as hkb_lab 05:23 Southen left 05:26 Southen joined 05:33 Southen left 05:49 hkb_lab left 06:04 Ashizawa joined 06:07 thoughtpolice left 06:11 drbean left, Southen joined
mncharity TimToady: Is there a dual of CANDO, to intercept bindings? "DO!()"? :) 06:17
06:19 agentz1 joined 06:20 drbean joined 06:25 drbean left, drbean joined
mncharity Hmm, and actually, CANDO only applies to packages/modules/classes and such. For intercepting lexical lookup, it doesn't help, no? So to MOPify lookup, we're back to a MOPmumble::mumble_lookup() macro (or usually-optimized-away multi I suppose), which calls CANDO for the package portion of resolution. yes? 06:26
06:30 ahmadz joined, ahmadz left, jisom joined, Southen left
TimToady CANDO is for emulating missing declarations. there's no reason a lexical scope can't just use a real declaration. and grammar rules reductions can call out to whereever they like in any case. 06:34
06:34 Southen joined 06:41 Ashizawa left 06:52 drbean left 07:01 luqui joined 07:02 drbean joined 07:04 mr_ank joined, mncharity left 07:06 Ashizawa joined 07:19 justatheory left 07:28 jmacclure left 07:31 mncharity joined
moritz_ CANDO sounds like a nuclear reactor to me ;) (close to CANDU) 07:36
07:47 amnesiac left 07:48 iblechbot joined 07:49 cnhackTNT left 07:50 Aankhen`` joined 07:52 cnhackTNT joined 08:03 mr_ank_ joined 08:07 mr_ank_ left, mr_ank_ joined, mr_ank_ left 08:08 penk left, franck___ joined 08:15 Southen left 08:16 mr_ank left 08:36 Southen joined
agentz1 one of my friends has hacked out a miniSQL parser using PCR: svn.openfoundry.org/openapi/trunk/m...miniSQL.pm 08:50
lambdabot tinyurl.com/yw8dyu
agentz1 we'll use it as part of our yahoo cn's web service API :)
moritz_ cool ;) 08:51
08:51 alc joined
agentz1 moritz_: :) 08:51
08:57 cnhackTNT left 09:24 IllvilJa left 09:30 ting_ joined, ting left 09:32 masak joined 09:37 pbuetow left 09:41 ludan joined 09:53 IllvilJa joined, Southen left 10:01 Southen joined 10:06 pbuetow joined, IllvilJa left 10:10 jjore-m left 10:12 fglock joined 10:37 ruoso left 10:44 Southen left 10:46 Southen joined 10:49 Lorn left
pugs_svn r19024 | fglock++ | [kp6-kp6] t-bootstrap/02-grammar.p6 fixed semicolons 10:51
10:52 ebassi joined 10:53 ruoso joined 10:54 Southen left 11:14 Southen joined 11:15 jisom left 11:28 Torment joined 11:32 GeJ left, GeJ joined 11:34 pravus_ joined 11:36 yahooooo joined 11:37 mj41__ joined 11:43 Patterner left 11:44 Psyche^ joined, Psyche^ is now known as Patterner
pugs_svn r19025 | fglock++ | [kp6-perl5rx] add some indentation when emitting 11:45
11:45 pravus left 11:46 Jedai left 11:51 Ashizawa left 11:54 mj41_ left
pugs_svn r19026 | fglock++ | [kp6-perl5rx] fixed last commit 11:57
fglock agentz1.friend++ 12:03
12:04 Patterner left
agentz1 fglock: :) 12:04
fglock: maybe the p5rx emitter is even more production friendly?
fglock agentz1: I've been using v6.pm for production, 12:05
it actually parses the code inside code blocks, instead of using a source filter
agentz1 fglock: heh. our environment is *very* busy ;)
fglock: v6.pm's code quality makes me a bit worried ;) 12:06
fglock the .pmc files help a lot
yes, I know :P
we are using Perl 6 for the grammars, and Perl 5 for the remaining code 12:07
agentz1 *nod*
fglock I'm now working on an MDX parser, which is a variation of SQL
agentz1 sorry, gotta run for supper...already extremely hungry atm ;)
MDX?
interesting
fglock multi dimensional database
agentz1 ah 12:08
fglock ok :)
agentz1 okay, later...gotta run...&
12:08 cognominal_ left, Southen left, Southen joined 12:09 Psyche^ joined, Psyche^ is now known as Patterner 12:10 Lorn joined 12:14 ferreira joined 12:18 Southen left 12:19 shachaf left, shachaf joined 12:20 Patterner left 12:23 Psyche^ joined, Psyche^ is now known as Patterner 12:24 iblechbot left, Lorn left
pugs_svn r19027 | fglock++ | [kp6-perl5rx] improved Match.perl5 12:26
12:34 Southen joined 12:43 Notrix joined
pugs_svn r19028 | fglock++ | [kp6-perl5rx] deep Match.perl 12:47
12:55 IllvilJa joined 13:06 luqui left, xinming joined, xinming__ left 13:20 cmarcelo joined 13:27 pravus_ is now known as pravus 13:32 iblechbot joined 13:54 iblechbot left 13:57 fglock left 14:01 duncanh joined 14:09 iblechbot joined 14:23 nothingmuch left 14:34 fglock joined 14:45 rindolf joined
pugs_svn r19029 | fglock++ | [kp6-perl5rx] Visitor::RegexCapture is buggy - moved away for now 14:48
r19030 | fglock++ | [kp6-kp6] t-bootstrap/02-grammar.p6 passes test #6 14:59
15:04 timbunce joined 15:06 timbunce left, timbunce joined 15:10 nothingmuch joined 15:13 timbunce left
nothingmuch [particle]: moose? 15:18
[particle] hola! 15:20
pugs_svn r19031 | fglock++ | [kp6-perl5] most library loading moved to Runtime::Perl5::Runtime 15:33
15:36 pmurias_ joined, rindolf left 15:43 Lorn joined
pugs_svn r19032 | fglock++ | [kp6-kp6] fixed uninitialized Code 15:48
r19033 | fglock++ | [kp6-kp6] more mp6/kp6/perl5 bridge code 15:56
ferreira I have a minor patch for pugs Makefile.PL. Where should I submit it? 16:02
pmurias_ ferreira: do you have a commit bit? 16:03
ferreira no
pmurias_ ferreira: you could send it to perl6-compiler and wait for someone to apply it 16:06
or get a commit bit and add it yourself
ferreira ok. how do I get a commit bit?
Lorn its more easy to get a commit bit
:)
16:07 nipotan is now known as nipotaway
[particle] i'll give you one if you give me your email address and preferred username 16:08
16:09 pohyg joined 16:11 barney joined
pugs_svn r19034 | fglock++ | [kp6-kp6] t-bootstrap/02-grammar.p6 - all tests pass 16:13
Lorn [particle]: if possible, i want too user: lorn email: [email@hidden.address] 16:14
[particle] lorn: sent! 16:15
fglock pmurias_: see last commit :) 16:16
[particle] add your name to AUTHORS to test your commit bit
Lorn [particle]: thanks
fglock ferreira, Lorn, ola!
ferreira fglock: oi 16:17
[particle]: email is a.r.ferreira at gmail.com and preferred username is ferreira
Lorn fglock: :)
16:17 masak left
Lorn fglock: d'oh, i forget the irc logs :/ 16:17
pmurias_ fglock: nice 16:18
[particle] ferreira: sent!
Lorn and dont remember if have email protect
ferreira Lorn: just more a small pile of spam to collect from :)
[particle] don't worry, the irc logs obfuscate the email addrs 16:19
fglock kp6 is near to bootstrap... compiled to perl5 regex :P
[particle] @ is replaced with an image
Lorn [particle]: nice
ferreira [particle]: thanks - I could not answer you in private because I registered the nick and I am having trouble to authenticate :(
[particle] ferreira: i think by default the password is your username
(for nick auth)
fglock: pmichaud and i are making good progress converting perl6 to use nqp rather than pir 16:20
16:20 Psyche^ joined
ferreira if I use svk, there is some guideline to avoid messing up commit messages while pushing ? 16:21
[particle] many sanity tests are passing now
fglock [particle]: that's cool
I still wonder how we could work together
[particle] could write kp6 in nqp, i guess :) 16:22
Lorn what is nqp?
[particle] not-quite-perl
it's a subset of perl 6 used to implement languages on parrot 16:23
fglock kp6 is just plain Perl 6
simpler, even
pmurias_ fglock: more segfaults remain see func(1,2,3)
fglock pmurias_: I didn't convert the whole Pad.pm, just as-needed 16:24
16:24 amnesiac joined
fglock [particle]: does nqp has string-eval? 16:25
16:25 pmurias_ is now known as pmurias
[particle] fglock: it evaluates some string stuff, but not full p6 16:25
fglock string-eval, hash, array, and closures is all kp6 needs
pmurias fglock: i mean $_ = 'func(1,2,3)';KindaPerl6::Grammar.term() 16:26
fglock pmurias: looking 16:27
[particle] fglock: svn.perl.org/parrot/trunk/compilers...README.pod
fglock [particle]: that's all kp6 needs 16:28
converting the Perl 5 runtime to nqp should be straightforward, 16:29
but it depends on how you plan to implement the MOP
a straightforward conversion would implement the MOP in nqp
but you might want it implemented in C 16:30
or, s/C/Parrot/
[particle] parrot's oo support is quite nice
fglock even better :)
16:31 neuromu joined
rgs neuroĀµ! 16:31
neuromu rgs!
fglock [particle]: we can start with KindaPerl6::Visitor::Emit::Perl6, maybe it just works 16:33
pmurias fglock: you are considering having kp6 emit nqp? 16:35
fglock perl script/kp6 -Cperl6 -e ' say "hello, World" ' # untested
pmurias: sure
[particle] that'd be cool, but nqp requires parens on functions
fglock no problem, we can tweak the emitter 16:36
[particle] yep, should be straightforward
16:36 Patterner left, Psyche^ is now known as Patterner
[particle] i'm concentrating on getting changes in before our weekly status in a few hours atm 16:36
fglock currently compiles to something like: ($GLOBAL::Code_say)('hello, World') 16:37
[particle] kp6-on-nqp would be quite nice... i hope we can make something like that happen soon
fglock kp6 has several desugaring layers, you can choose how basic the target language is
16:38 xinming left, xinming joined
fglock [particle]: I would need some of your time for starting up 16:39
[particle] fglock: that can likely be arranged later today or tomorrow
fglock ok :)
hmm - string eval is only needed for bootstrapping 16:40
16:42 Onyxyte left
[particle] even easier :) 16:42
fglock bootstrapping in parrot would be cool
16:43 TJCRI joined
[particle] since i've been away, i haven't caught up on the latest kp6 progress. anything big happen since 1 nov? 16:46
16:53 penk joined 16:57 dlocaus joined
dlocaus ferreira: hi! this is daniel 16:57
16:58 neuromu_ joined
ferreira dlocaus: hi. you seem quite busy around here (perl6 land) - hoping you're enjoying your time :-) 16:59
dlocaus ferreira: I am... :)
ferreira: accept my hd drive died. I need -just- one file from it, I'm hoping that someone can restore it. I think the logical sectors 0-7? are dead.
ferreira: stupid harddrive. I should have gone with a SCSI raid. 17:00
ferreira :( failing hardware sucks as most things that fail when they should not :P 17:01
dlocaus I know... the only hardware that EVER fails is the hardware that doesn't have a backup... Things like this make me think the devil really does exist...
allbery_b s/the devil/murphy/ 17:02
ferreira oh, not the devil, only chance: chance to win or to lose - Murphy's law asserts a high probability for failure however 17:03
moritz_ re
dlocaus ferreira: my scsi system has been up for years... at my former work place, once issues were streighted out, harddrive/hardware failures are rare. 17:04
asssuming I heard about them...
fglock [particle]: re happen, several bugs were fixed (infinite loops in regexes, Scalar/Array/Hash autovivification, Array x List), some documentation added, and the bootstrapping process is going on 17:07
and the Lisp backend got unmaintained :( 17:08
[particle] lisp stale already? 17:09
i guess it's really a dead langauge :/
pugs_svn r19035 | fglock++ | [kp6-kp6] simple expression parsing works 17:11
fglock [particle]: kp6-kp6 is the bootstrapped compiler - it is just starting to work 17:12
pmurias [particle]: and some things got cleaned up 17:13
17:13 dlocaus left
[particle] o frabjous day! callooh! callay! 17:13
pmurias [particle]: avar is waiting for the bootstrap
[particle] ah, for the lisp backend? gotcha. 17:14
pugs_svn r19036 | ferreira++ | adding myself to AUTHORS: testing my commit bit
pmurias [particle]: i didn't refer that the lisp backend going unmaintained as cleanup, just a bit of nastines in kp6 got cleaned up 17:16
on the other hand kp6 is no fully 'no strict' by default
s/no/now
[particle] i mean, avar waiting for bootstrap to do lisp backend?
pmurias [particle]: yes 17:17
avar Well it's a mixed running around chasing other stuff / waiting for bootstrap
It's a bit harder if I do it right now, but it'll be easier once it's bootstrapped
pugs_svn r19037 | ferreira++ | r16@dracma: perl | 2007-12-04 15:18:33 -0200 17:19
r19037 | ferreira++ | try Haskell package time as well (contains Data.Time.*) - this is coherent with the dependencies in the Debian package
avar Mainly because the main problems I was having with the lisp backend were Aankhen`` rewriting everything to be unlike the p5 one:) 17:20
Which will go away once all the .pm stuff is in p5 and I only have to emit more primitive stuff and not care about implementing array or pads and stuff like that
17:20 neuromu left
avar so my all means tell me when this is bootstrapping so I can haaack 17:21
fglock avar: sure 17:23
17:24 neuromu_ is now known as jmacclure
pmurias avar: what do you mean by having all the .pm stuff in p5? 17:24
pugs_svn r19038 | fglock++ | [kp6-kp6] more tests
17:26 ruoso left
avar uh, p6 I mean 17:26
pmurias avar: arrays are primitive stuff in p6, you don't generally get to play around with pointers in p6 17:32
avar Well pads then 17:33
Anyway more stuff in p6 .pm makes writing backends easier
17:33 ruoso joined
pmurias yes, i don't doubt it 17:34
avar pypy looks like a very interesting backend:)
p6 -> rpython then you have jit and whatnot 17:35
they're already running scheme and js
moritz_ parrot might be an intersting target as well ;-) 17:36
17:36 chris2 joined
moritz_ they already support most of p6's object model (what a coincidence... ;-) 17:36
pugs_svn r19039 | dlo++ | Updated documentation, no code changes. 17:37
avar Yeah sure, I still like multi-backend targets like lisp/pypy better if I had to choose:)
pypy does stand-alone C executables already, and they'd be doing optimization too
anyway I thought I'd mention it since people haven't been talking about it here that I've seen 17:38
w6
moritz_ avar: right. Of course you're welcome to write any backend you like ;)
17:39 dlocaus joined
dlocaus fglock: I remember that you said a while ago, that all of kp6 would be "obsolete" in favor of onion (successor to kp6) is that still the case? 17:40
pmurias dlocaus: i hope so 17:41
dlocaus: it's more replaced the obsoleted
fglock the current plan is that the first "onion" will be just a copy, with the remaining MiniPerl6-isms removed 17:42
moritz_ kp6: print "string\ninterpolation"
exp_evalbot r19039: OUTPUT[stringā¤interpolation]
dlocaus is that "obsolete" in terms of "we're just going to delete it, because we don't need it" (meaning there is no point in more that documetnation for other programmers?) or "We will be using it, like we use miniperl6, so there is some worth to documenting it?" 17:43
fglock it will be incrementally obsoleted, most of the runtime will survive
which is the documented part 17:44
dlocaus KindaPerl6\Runtime\Perl5 correct?
fglock yes
dlocaus ok
ty
17:44 ferreira left
fglock coffee & 17:45
pugs_svn r19040 | dlo++ | This file did not have a terminating "1;" to terminate the file. I have added it. I have also made minor documentation improvements. 17:46
17:47 ebassi left
pmurias dlocaus: lots of stuff will be changed to use new kp6 features 17:49
dlocaus pmurias: yeah, I saw some of that when I was looking at mp6.pl 17:51
one of the packages (I don't remember right off hand MATCH?) used kp6's match
pugs_svn r19041 | dlo++ | [kp6] Minor documentation improvements 18:01
r19042 | fglock++ | [kp6-kp6] added debug tracking info for current bug (arrow blocks) 18:02
fglock weird code: @(($$<arrow_sub_sig>).positional).>>key, 18:05
moritz_ aye 18:06
18:07 franck___ left, ludan left, xinming left, xinming joined 18:10 iblechbot left
pugs_svn r19043 | fglock++ | [kp6-kp6] minor fix in Pad.pm 18:20
r19044 | dlo++ | [kp6] Minor documentation improvements. I have also added a "1;" to the end of the file.
18:22 justatheory joined
pugs_svn r19045 | dlo++ | [kp6] Minor documentation improvements. 18:28
18:30 Aankhen`` left, Aankhen`` joined
dlocaus fglock: ping? 18:33
fglock: what does "XXX" mean in the code? I know its not saying this code is sexy...
pugs_svn r19046 | fglock++ | [kp6-kp6] documented possible bug, will fix later
fglock it mean "will fix this later"
moritz_ dlocaus: it means "that should be fixed"
dlocaus so it is === to "TODO"?
fglock like in last commit :P
yes 18:34
home &
dlocaus Do you mind if I change XXX to TODO?
fglock dlocaus: it's different, XXX doesn't prevent the code from working
moritz_ dlocaus: basically, XXX is more "this is probably a bug", whereas TODO is "this should be implemented"
fglock it's just "refactor this"
dlocaus Ah, ok.
pmurias "beware"
fglock :) 18:35
now &
moritz_ "unsexy code below"
18:40 alc left 18:51 lichtkind joined 18:55 fglock left
Lorn does anybody use svk to sync pugs? 19:01
moritz_ judging from the commit messages, ferreira does 19:02
Lorn moritz_: what commit message? 19:04
moritz_ Lorn: irclog.perlgeek.de/perl6/2007-12-04#i_156866 19:05
lambdabot Title: IRC log for #perl6, 2007-12-04
moritz_ that's the typical svk signature (if it's not configured to suppress the first line) 19:06
19:06 jisom joined
Lorn moritz_: thanks 19:06
dlocaus is Pad.pm still in use? 19:07
Lorn moritz_: do you know if for the svk its a diferent url?
moritz_ Lorn: you're welcome
dlocaus KindaPerl6::Runtime::Perl5/Pad.pm?
because the package declaration is "package Pad;"
moritz_ Lorn: I don't think so, but I'm not an svk expert 19:08
dlocaus I would have thought given its directory, it would be
KindaPerl6::Runtime::Perl5::Pad;
moritz_ Lorn: maybe look at feather.perl6.nl/svk.html
lambdabot Title: SVK on Feather
moritz_ ah wait, that's specific to feather, and I assume you don't have an account
19:17 justatheory left
pugs_svn r19047 | dlo++ | [kp6] Documentation added. 19:19
19:22 blindfish_ joined 19:24 blindfish_ left 19:26 justatheory joined, iblechbot joined 19:27 jhorwitz joined 19:34 TheNo1Yeti joined 19:43 Notrix left
[particle] idly wonders if he can put "five years of perl 6 experience" on his resume :) 19:56
19:56 marmic left, marmic joined
moritz_ ;) 19:58
Aankhen`` [particle]: Why not? :-)
20:06 marmic left 20:07 marmic joined 20:11 cognominal_ joined 20:25 jisom left
gaal O HAI 20:29
moritz_ moose!
gaal hey moritz_ :)
moritz_ what's up gaal? 20:30
gaal work, work, work
but I miss pugs! 20:31
moritz_ well, it's not running away ;)
gaal :/ 20:32
[particle] pugs misses you too, gaal 20:38
gaal pugs: say '> "pugs: say <moose>"' 20:40
exp_evalbot OUTPUT[> "pugs: say <moose>"ā¤] 20:41
gaal oh right
> "pugs: say <moose>"
lambdabot "pugs: say <moose>"
gaal tsk, tsk
a polyglot bot command looks infeasible 20:42
[particle] perl6: say("moose")
exp_evalbot kp6: OUTPUT[mooseā¤]
..pugs: OUTPUT[mooseā¤]
..p6: OUTPUT[mooseā¤]
..nqp: OUTPUT[src/pmc_freeze.c:1254: failed assertion 'must_have_seen'ā¤Backtrace - Obtained 19 stack frames (max trace depth is 32).ā¤ Parrot_print_backtraceā¤ Parrot_confessā¤ (unknown)ā¤ (unknown)ā¤ (unknown)ā¤ (unknown)ā¤ Parrot_thawā¤
..PackFile_Constant_unpack_pmcā¤ PackFile_Con...
[particle] ouch 20:43
gaal whoa
perl6: say localtime
exp_evalbot kp6: OUTPUT[no method 'APPLY' in Class 'Undef'ā¤ at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 345ā¤ KindaPerl6::Runtime::Perl5::MOP::__ANON__('HASH(0x824b75c)', 'APPLY') called at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 169ā¤
..main::DISPATCH('HASH(0x824b75c)', 'APPLY') called at compiled/perl5-kp6-mp6/lib/Kin...
..pugs: OUTPUT[*** ā¤ Unexpected end of inputā¤ at /tmp/CA2dUL7q5e line 2, column 1ā¤]
..p6: OUTPUT[Null PMC access in invoke()ā¤current instr.: 'parrot;Perl6::Compiler;main' pc 138 (perl6.pir:98)ā¤]
..nqp: OUTPUT[src/pmc_freeze.c:1254: failed assertion 'must_have_seen'ā¤Backtrace - Obtained 19 stack frames (max trace depth is 32).ā¤ Parrot_print_backtraceā¤ Parrot_confessā¤ (unknown)ā¤ (unknown)ā¤ (unknown)ā¤ (unknown)ā¤ Parrot_thawā¤
..PackFile_Constant_unpack_pmcā¤ PackFile_Con...
gaal what you say? 20:44
or is it "time"
20:44 thoughtpolice joined
gaal pugs: say time 20:44
exp_evalbot OUTPUT[250116216.850201ā¤]
20:44 Lorn left
gaal perl6: say time 20:44
exp_evalbot kp6: OUTPUT[no method 'APPLY' in Class 'Undef'ā¤ at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 345ā¤ KindaPerl6::Runtime::Perl5::MOP::__ANON__('HASH(0x824b75c)', 'APPLY') called at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 169ā¤
..main::DISPATCH('HASH(0x824b75c)', 'APPLY') called at compiled/perl5-kp6-mp6/lib/Kin...
..pugs: OUTPUT[250116222.955503ā¤]
..p6: OUTPUT[Null PMC access in invoke()ā¤current instr.: 'parrot;Perl6::Compiler;main' pc 138 (perl6.pir:98)ā¤]
..nqp: OUTPUT[src/pmc_freeze.c:1254: failed assertion 'must_have_seen'ā¤Backtrace - Obtained 19 stack frames (max trace depth is 32).ā¤ Parrot_print_backtraceā¤ Parrot_confessā¤ (unknown)ā¤ (unknown)ā¤ (unknown)ā¤ (unknown)ā¤ Parrot_thawā¤
..PackFile_Constant_unpack_pmcā¤ PackFile_Con...
pmurias gaal: evals hurts ;) 20:45
gaal who's evaling what? I'm up to slow
pugs_svn r19048 | moritz++ | [evalbot] disable nqp evaluation
20:45 exp_evalbot left, exp_evalbot joined
pmurias gaal: sorry "perl6:" 20:46
[particle] moritz: nqp should be working, but it's moved from languages/nqp to compilers/nqp
moritz_ [particle]: ah, that explains a lot
[particle] anyway, need to update that parrot repo, it's probably an old rev
moritz_ gaal: ^(kp6|nqp|pugs|p6) evals with the respective compiler... 20:47
gaal: and ^perl6: evals with all of them
gaal does the kp6 error mean time's not implemented, or that the kp6 bot is broken?
moritz_ kp6: say "I'm fine" 20:48
exp_evalbot r19048: OUTPUT[I'm fineā¤]
moritz_ so it's NYI
gaal Not Yncredibly Invalid? 20:49
oh oh
yeah.
pugs_svn r19049 | moritz++ | [evalbot] reenable nqp evaluation (but not as part of 'perl6:') 20:51
20:51 ispy_ joined 20:52 exp_evalbot left 20:53 exp_evalbot joined
pmurias gaal: kp6 is 'no strict', so the unimplemented &time is undef 20:53
moritz_ kp6: say time() 20:54
exp_evalbot r19049: OUTPUT[no method 'APPLY' in Class 'Undef'ā¤ at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 345ā¤ KindaPerl6::Runtime::Perl5::MOP::__ANON__('HASH(0x824b75c)', 'APPLY') called at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 169ā¤
..main::DISPATCH('HASH(0x824b75c)', 'APPLY') called at compiled/perl5-kp6-mp6/lib/Kin...
gaal ah, I see, and undef() dies.
moritz_ right
gaal that error message should probably be improved :) but I wonder how, unless the undef is somehow "but named<&time>" 20:55
moritz_ gaal: the checking should be made at compile time 20:56
pmurias gaal: use strict
Patterner who needs messages? just return a large negative numb3r and let the people google for it...
gaal even without use strict, I like to know what went wrong when it did 20:57
Patterner: there aren't enough negative numbers to describe the various things that can go wrong
moritz_ I think it could be done in the MMD dispatcher, somehow
Patterner And I thought card(|R) is pretty big... 20:58
gaal not if the dispatcher sees a pure undef? but don't listen to me
moritz_ gaal: the dispatcher should see a name, not a ref 20:59
gaal Patterner: I wouldn't get into a contest against Murphy and computers combined.
moritz_ gaal: otherwise it wouldn't have to dispatch at all
nqp: say("back again");
exp_evalbot OUTPUT[back againā¤]
gaal nqp: 42 21:00
exp_evalbot RESULT[Syntax error at line 1, near "( ( do { 4"ā¤current instr.: 'parrot;PGE::Util;die' pc 120 (runtime/parrot/library/PGE/Util.pir:82)ā¤called from Sub 'parrot;NQP::Grammar;TOP' pc 588 (src/Grammar_gen.pir:129)ā¤called from Sub 'parrot;PCT::HLLCompiler;parse' pc 444 (src/HLLCompiler.pir:291)ā¤called
..from Sub 'parrot;PCT::HLLCompiler;compile' pc 34...
gaal O RLY?
pmurias moritz_: i would guess the dispatcher sees the value
($foo.bar).baz
moritz_ pmurias: perhaps my understanding of the dispatcher is a bit wrong ;) 21:02
I always thought the dispatcher takes a sub/method name and a capture of arguments, and determins (and calls) the right method 21:03
pmurias it's done a bit differently in kp6 21:04
in case of sub calls the method name is APPLY
21:06 barney left
pmurias moritz_: if the dispatcher recived a sub name it would need at least the lexical pad too 21:13
21:13 Aankhen`` left 21:23 lidden left 21:32 TheNo1Yeti left 21:48 kyrbe joined, kyrbe left
pmurias @tell fglock while attempting to remove eval uncovered a bug 21:52
lambdabot Consider it noted.
pmurias @tell fglock look at line 1586 compiled/perl5-kp6-kp6/lib/KindaPerl6/Grammar/Regex.pm inside eval 21:54
lambdabot Consider it noted.
akreal is back.
21:54 jmacclure left
pmurias akreal: to the computer or from the dead :) 21:57
akreal pmurias: aren't this things the same?) 21:58
hello world :) 21:59
pmurias akreal: hello world :) 22:11
22:11 pmurias left 22:12 TJCRI left 22:14 ispy_ left 22:15 ispy_ joined 22:17 justatheory left 22:26 kst left 22:27 cmarcelo left, justatheory joined 22:28 kst joined 22:30 jjore-w left 22:32 lumi joined 22:42 IllvilJa left, Schwern joined 22:45 pohyg left 22:46 iblechbot left 22:55 marmic_ joined, marmic left, marmic_ is now known as marmic
ispy_ does p6 have any integrated xml parsing abilities? 23:00
moritz_ ispy_: no
ispy_ ok, just curious :0
moritz_ just very powerfull grammars
that should make plain perl6 parsing really easy 23:01
ispy_ I've yet to mess with grammers yet.
integral once someone writes a module called "XML::ParseIt", you just type "use XML::ParseIt;" and then functions are available to parse XML. Is that integrated enough?
moritz_ they're less messy and more powerfull than regexpes ;) 23:02
ispy_ that's not native, smart ass.
so no.
:)
integral What does native mean if the perl6 compiler is written in perl6?
ispy_ integral: stop being pedantic.
dork.
integral No, I'm asking what you really mean
PerlJam ispy_: I don't know what you mean by "native" here either 23:03
ispy_ shakes his head...
integral as I understand it, we'll be able to extend the perl6 grammar in modules, so we'll be able to say things like qq:xml{ <xml> </xml> } and have the quoting behave somewhat intelligently, and yield an AST
ispy_ being embedded into the interpreter... a native ( not requiring a module but part of the language )
PerlJam ispy_: Perl 6 will come with nothing but Perl 6. From that people will construct "distributions" that have commonly used modules bundled with Perl 6. 23:04
ispy_ how could it mean anything other than that?
so, same as p5.
cool...
PerlJam ispy_: no, there will never be a "native" xml parser in perl 6 in that sense.
ispy_ ok, didn't think so... but was curious.
moritz_ ispy_: there will be no "core" perl6 with many useful modules as in p5
PerlJam moritz_: aigh! Now you've said the C-word and we're likely to rehash the p6l thread! ;) 23:05
ispy_ so you are saying that unlike p5 which comes with several handy cpan modules, p6 will come with it's language only?
integral One problem is that if you admit something to be available without saying "use ...", you end up with having to have that available everywhere, even an embedded environment; if an environment doesn't have it available, it's suddenly not Perl 6, rather than just being Perl 6 but without a module installed
ispy_ I agree.
moritz_ ispy_: see www.nntp.perl.org/group/perl.perl6....28016.html and the whole threads 23:06
lambdabot Title: xml and perl 6 - nntp.perl.org, tinyurl.com/3cg3xj
PerlJam ispy_: p6 will come with its language and whatever else is needed to get and install modules from CPAN (or cpan-like things)
moritz_: is that larry's message?
PerlJam is too lazy to look 23:07
23:07 jhorwitz left
moritz_ PerlJam: that's the root of the "xml and perl" threads 23:07
ispy_ so are some of you active in the p6 development?
I've been wanting to contribute, possibly to cpan6 or something.
PerlJam moritz_: you should just link to larry's response :)
moritz_ ispy_: so in which language do you want to code? 23:08
PerlJam ispy_: i've contributed in some small ways to pugs and parrot and I've written some articles on parrot and perl6. I guess that's "active"
moritz_ same for me if you s/parrot/kp6/ 23:09
ispy_ moritz_: honestly, my haskell isn't impressive. However I am and continue to be a professional p5 developer. And have really taken a liking to p6. So I was wondering if ways existed for me to contribute p6 code.
moritz_: I've really just gotten into haskell over the past month. 23:10
PerlJam: I plan on writing a p6 crash course for Detroit.pm
moritz_ ispy_: you could work on kp6, that's a perl6 compiler in a subset of p6, and emits p5
PerlJam ispy_: tests are always welcome. The pugs repo is the standard place for all perl6 tests.
ispy_ excellent...
moritz_: where could I find out more about getting involved with kp6? 23:11
moritz_ ispy_: here ;-)
ispy_ sweet :)
so basically this is a p6 interpreter that converts p6 code to p5?
moritz_ ispy_: first check out the pugs repo, which contains kp6 under v6/v6-KindaPerl6/
ispy_ ok
moritz_ ispy_: a compiler, do be more specific ;) 23:12
ispy_ moritz_: a compiler that doesn't compile? or am I missing something?
:)
moritz_ ispy_: it compiles to a p5 image
PerlJam ispy_: you're probably stuck with an archaic idea of "compiler" :)
ispy_ oh, ok.
I'm sure I am :)
But I'm willing to expand my mind.
gonna eat some dinner, I'll be back to talk some more shop with you guys... I'm really interested in getting involved. I love Perl. 23:13
PerlJam ispy_: be sure to read the synopses if you haven't already
moritz_ the right people to ask about the interna are fglock and pmurias.. both of which (and me too) live in europe, so you won't reach any of us in the next ~6hrs ;-) 23:14
ispy_ PerlJam: I have, thank you. A very important read.
moritz_ you can get a commmit bit to the repository here, just /msg your email and desired nick to somebody
I'm still not sure if I've read all of the synopsis entirely :/ 23:15
PerlJam moritz_: you have to periodically re-read them or you will miss changes.
moritz_ PerlJam: I read the diffs on p6l ;) 23:16
PerlJam (plus the re-reads will keep the static parts in your head :)
yeah, I read the diffs too, but occasionally you need to see things in contet.
er, context
(a lrager context than the diff :) 23:17
s/lra/lar/
moritz_ aye
PerlJam: I try to translate some of the synopsis periodically (and get stuck at 10%-50%) 23:18
that way I want to make sure I actually _understand_ them ;)
moritz_ -> bed 23:20
23:23 amnesiac left
ispy_ PerlJam: honestly, the synopsis is confusing to me :) 23:30
23:38 lidden joined

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!