pugscode.org/ | nopaste: sial.org/pbot/perl6 | pugs: [~] <m oo se> (or rakudo:, kp6:, smop: etc.) || We do Haskell, too | > reverse . show $ foldl1 (*) [1..4] | irclog: irc.pugscode.org/
Set by TimToady on 25 January 2008.
00:04 jeffreykegler joined
namelessuser Heh, I agree about search.cpan.org. 00:05
00:24 evarlast left 00:48 ebassi left 00:56 cmarcelo left 01:00 Auzon joined 01:04 lyokato_ joined
meppl good night 01:11
01:12 pen1 joined, penk left 01:17 jeffreykegler left 01:19 namelessuser left 01:21 meppl left, thoughtpolice joined 01:29 simcop2387 left 01:30 devogon left 01:42 mncharity joined 01:46 jferrero left 01:47 __Ace__ left 01:48 CoreEvil joined 01:50 [particle1 joined 02:00 Limbic_Region left 02:04 [particle] left 02:05 [particle] joined 02:06 Jamtech left 02:07 kanru left 02:11 tobeya left 02:14 Lorn joined 02:18 [particle1 left 02:32 CoreEvil left 02:49 simcop2387 joined 02:51 Schwern left 02:56 sri_work left 02:59 avar left 03:03 alester joined 03:06 japhb left 03:10 jjore joined 03:12 kanru joined 03:14 devogon joined 03:18 REPLeffect joined 03:20 wknight8111 left 03:24 mj41____ joined, mj41____ is now known as mj41 03:33 Torment left
mncharity TimToady, when you say in STD svn.perl.org/perl6/pugs/trunk/src/p...0.0-STD.pm , "# A fairly complete (but almost certainly buggy) operator precedence parser", is this something someone should/could pull out and shake down, so if/when fglock picks it up, it is _less_ certainly buggy? 03:33
lambdabot mncharity: You have 1 new message. '/msg lambdabot @messages' to read it.
03:34 Lorn left
mncharity ponders "an instance of a grammar is the current parse state"... nifty... 03:35
hmm. so all rules, if called from outside the grammar (GrammarA.randome_rule(astring)) are instance constructors? /me squints... 03:38
03:41 mj41__ left 03:43 mncharity left 04:08 thoughtpolice left 04:19 araujo joined 04:20 kupopo joined 04:34 devogon left 04:57 icezinhu joined 05:06 nothingmuch left 05:34 monomorph_ joined 05:43 monomorph left 05:47 PerlJam left, PerlJam joined 05:49 lyokato_ left 05:55 ryan_ joined 05:56 ryan_ left 06:01 Alias__ joined
Alias__ \o 06:01
So, thoughts on the idea of identifying one particle grammar as "Standard Perl 6" and finalizing the grammar class? :) 06:05
Auzon I assume you mean the current spec as Standard Perl 6? 06:07
06:17 justatheory left
allbery_b Perl-6.0.0-STD is being firmed up, yes 06:18
you'd have to ask TimToady its current status
06:48 alester left, Jedai joined
jjore Alias_, it occurs to me that your PPI6 should mimic the reflection API available in Perl 6 (whatever that is). Then it is just a VM-less reflection API and can be swapped for a VM doing direct reflection. 07:21
At some level anyway.
jjore sleeps. 07:23
Alias__ jjore: Ya, having conceptual analogies would be great 07:30
Much like the way PDOM is similar to HTML DOM
In a ~~ way anyways
07:30 REPLeffect left
Alias__ bbl 07:30
07:30 Alias__ left 07:57 Aankhen`` joined 08:05 jisom joined 08:11 iblechbot joined 08:29 Alias_ joined 08:32 pen1 left, penk joined 08:35 masak joined 08:36 cognominal_ left 08:40 japhb joined 08:43 viklund joined 09:00 njbartlett left 09:07 kanru left 09:17 japhb left 09:25 zev left 09:35 Schwern joined 09:45 jisom left 09:59 iblechbot left 10:01 babbage joined 10:02 babbage_ joined, njbartlett joined 10:04 agentzh left 10:07 devogon joined 10:11 babbage left, babbage joined
hcchien use.perl.org/article.pl?sid=08/01/29/1545202 10:14
lambdabot Title: use Perl | OSDC.tw 2008 CFP will be end on 15th, February 10:15
10:25 ebassi joined 10:27 meppl joined 10:28 babbage_ left 10:32 orafu joined 10:44 jferrero joined 10:45 Lorn_ left 10:58 Schwern left 11:11 njbartlett left 11:12 kanru joined 11:18 babbage left
pugs_svn r19749 | ruoso++ | [smop] smop/sm0p/KP6Test.pm in better shape, it is already returing a more sane code, some values still missing. 11:24
11:29 meppl left 11:34 njbartlett joined 11:35 jiing left, cosimo joined 11:44 iblechbot joined 11:47 Lorn joined 12:11 nothingmuch joined 12:19 agentzh joined 12:26 njbartlett left 12:31 njbartlett joined 12:37 iblechbot left 12:39 cmarcelo joined
pugs_svn r19750 | ruoso++ | [smop] YAY!!!! sm0p preprocessing kp6 grammar seems to be working... 12:41
moritz_ ruoso++
ruoso script/kp6 ../smop/sm0p/KP6Test.pm from kp6 dir shows it :) 12:43
12:54 viklund left
moritz_ [particle], pmichaud: how stable is the rakudo grammar? 12:55
a few month ago I proposed a Perl 6 syntax explainer
and [particle] convinced me to implement it as a action class for STD.pm 12:56
so is it worth to prototype that for rakudo now, and then later move it to STD?
13:02 viklund joined 13:33 icezinhu left 13:35 pbuetow left 13:41 jferrero left 13:42 pbuetow joined 13:51 ebassi is now known as ebassi|lunch
[particle] moritz_: the rakudo grammar is growing and changing all the time 13:52
but it's based on std, and will eventually match std
so it seems a good place to experiment 13:53
13:54 njbartlett left
moritz_ uhm, that looks like a lot PIR - can I use nqp already for the actions? 14:00
14:00 avar joined
[particle] the actions for the compiler are in src/parser/actions.pm 14:01
moritz_ ah, just found it
that looks very p6ish *g*
[particle] indeed :)
i've just started working on perl6doc
i'll be creating a separate grammar for that, though
moritz_ what happens to the return value of such an action? is it somehow integrated into the $/ of the calling regex? 14:03
14:03 iblechbot joined
pugs_svn r19751 | moritz++ | perl6.vim: 'make' is a control flow keyword, just like return 14:04
[particle] use 'make' to build up the result object
moritz_ let's say I have a rule { foo {*} bar {*} }
now the action will be called twice, right?
so there will be two return values 14:05
what happens to them?
is $/ of the orignal match now a list of these two return values?
[particle] well, not two return values.
two places the result object is modified
make = modify result object
return = specify return values 14:06
moritz_ ok
[particle] see S05 for something more clear if you need it
if you look at the rakudo grammar (or STD) you'll see #= foo
moritz_ S05 doesn't seem to mention {*} at all 14:07
at least not the last time I checked
[particle] right, we're ahead of the spec there.
moritz_ so are actions called fromn {*} treated exactly as inline closures?
[particle] whatever is written after '#= ' is passed to the action as a second argument
moritz_ I know ;) 14:08
but apart from that, I mean
[particle] yes, actions are inline closures
the parse and past are generated simultaneously
moritz_ past? psomething abstract syntax tree? 14:09
[particle] parrot
docs/pdds/pdd26_ast.pod
it'll be important to understand that if you want to use nqp effectively 14:10
moritz_ ok, I'll read it
but in the end I don't want to generate a "normal" syntax tree
[particle] right. in fact, you may want to work solely with the parse tree. 14:11
moritz_ just something of the form DocTree.new(text => '*', pos => 4, description => 'numerlical mulitplication')
14:21 lisppaste3 left
moritz_ when I want a clean parse I have to execute BEGIN blocks... so that means that I have to built a "normal" parse tree anyway 14:24
[particle] rakudo doesn't handle BEGIN {...} yet, so you're good for now :) 14:26
moritz_ but it's something I have to take into account
[particle] i'll talk to patrick today about adding a flag to rakudo's perl6 exe that doesn't run BEGIN blocks 14:27
it's a tiny patch
14:34 lisppaste3 joined
moritz_ I think I'lll write to p6c and ask for discussion/advice 14:34
14:39 sri_work joined 14:42 ebassi|lunch is now known as ebassi, pbuetow left 14:48 apple-gunkies left 14:49 cognominal_ joined, sri_work_ joined 14:54 macae joined, AndyAway is now known as alester 14:56 IRSeekBot left 14:57 njbartlett joined 15:01 hareldvd joined 15:04 sri_work left 15:18 pbuetow joined 15:19 ingrediant joined
ingrediant PRIVAT-GEKNIPST.dnsdojo.net/?id=5420a7e0 15:22
15:22 ingrediant left, kanru left 15:23 jjore left
moritz_ hates spammers 15:24
alester Might as well hate the rain.
Auzon How come lambdabot didn't announce the title of it?
moritz_ dunno
[particle] lambdabot hates spammers too 15:25
moritz_ test: pugscode.org/
lambdabot Title: Pugs - pugscode
[particle] maybe there's no title?
moritz_ perhaps lambdabot has an adult filter
anyway, I cleand the logs 15:26
alester The temperature here is -0. 15:27
According to Google.
Auzon as opposed to regular 0? 15:28
alester Apparently.
Auzon Is it in Celsius?
[particle] it's a valid temp. weather forcasters hate ieee.
15:28 meteorja1 joined
[particle] no, it's Kelvin :) 15:28
moritz_ wow
Auzon It could be Fahrenheit 15:29
[particle] and likely it's fahrenheit. chicago's been pretty cold
moritz_ then your computer must work _really_ fast
alester It's F
15:29 meteorjay left
alester It's -16 in Athens, WI 15:30
[particle] i was in chicago two weekends ago. 1F, with wind chill of -22F
Auzon 8 here. and I have to walk somewhere :P
15:33 Auzon left, chris2 joined
alester It was a high of 52 yesterday. 15:35
15:49 cosimo left 16:16 kanru joined 16:21 blindfish joined 16:22 Psyche^ joined
pugs_svn r19752 | ruoso++ | [smop] YAY YAY YAY!!! It really is working with a complex node !!! 16:25
moritz_ ;-)
Tene ruoso++ # yay 16:30
16:32 pen1 joined
[particle] ruoso: how can i play? 16:33
oh, i can't :(
i'm on windows 16:34
16:34 Patterner left, Psyche^ is now known as Patterner, hareldvd left 16:37 JarJarBinks left
moritz_ perlpunks.de/paste/show/47a0a7f3.54ec.290 # that's how the output looks 16:39
[particle]: the last commit only modified a file that is compiled with kp6
doesn't kp6 run on windows?
ruoso [particle], in theory you can, why couldn't you? 16:40
autotools may not be completely windows friendly... but it does work on it
in theory it's easier to make it work in windows by using autotools than writing my own makefile 16:41
moritz_ ruoso: perlpunks.de/paste/show/47a0a899.5d52.231 # autotool build failure
pugs_svn r19753 | ruoso++ | [smop] first shot on the KP6sm0p preprocessor. 16:42
[particle] autotools 16:43
ok, i'll look for a windows download. i thought they didn't work there
ruoso you probably need cygwin
moritz_, try changing the call order in Makefile.cvs to call autoconf before automake 16:44
moritz_, which os are you on?
moritz_ ruoso: debian
ruoso moritz_, which version? 16:45
moritz_ Etch
automake 1.10
autoconf 2.61-4
[particle] cygwin != windows.
moritz_ when I change the call order I get 16:46
autoconf
configure.ac:6: error: possibly undefined macro: AC_PROG_LIBTOOL If this token and others are legitimate, please use m4_pattern_allow. See the Autoconf documentation.
make: *** [all] Error 1
ruoso moritz_, do you have libtool installed?
you need automake autoconf and libtool
moritz_ no
16:47 manfred joined
moritz_ ok, the Makefile is built correctly 16:48
now sm0p.pl dies because gensym isn't defined
Bareword "gensym" not allowed while "strict subs" in use at ../sm0p.pl line 31. 16:49
Execution of ../sm0p.pl aborted due to compilation errors.
ruoso moritz_, yeah... I'm working on it right now 16:52
moritz_ ok
pugs_svn r19754 | ruoso++ | [smop] sm0p.pl now works... but the code in smop_lowlevel.c is not... 16:53
16:54 IRSeekBot joined 16:56 meppl joined
ruoso [particle], gnuwin32.sourceforge.net/summary.html this should do the trick for you 17:02
lambdabot Title: GnuWin32 Summary
ruoso it's not cygwin... they are native win32 binaries...
17:06 REPLeffect joined
cj @tell TimToady how do you want to handle inclusion of CLI libraries when their namespace is defined in the CPAN for instance? 17:09
lambdabot Consider it noted.
17:17 viklund left
pugs_svn r19755 | ruoso++ | [smop] YAY!!! sm0p.pl already processes smop_lowlevel.sm0p and smop_lowlevel.c almost compiles... 17:36
TimToady cj: see S11:391 for syntax of importing from foreign namespaces 17:42
lambdabot TimToady: You have 1 new message. '/msg lambdabot @messages' to read it.
17:44 orafu left
cj TimToady: danke 17:48
moritz_ in case anybody hasn't noticed, the irclogs make strings like S11:391 to nice links ;-) 17:49
cj TimToady: where is the canonical textual representation of that synopsis located?
moritz_ the "canonical" HTML is at perlcabal.org/syn/ 17:50
lambdabot Title: Official Perl 6 Documentation
moritz_ the POD is at svn.pugscode.org/pugs/docs/Perl6/Spec and ... uhm... I forgot the second URL 17:51
lambdabot Title: Revision 19755: /docs/Perl6/Spec
cj thanks, moritz_
moritz_ svn.perl.org/perl6/doc/trunk/design/syn/ # that's the second
lambdabot Title: Revision 14496: /doc/trunk/design/syn
17:52 masak left
monomorph_ newbie question: is there a more elaborate discussion of tokens and the longest-token rule than S05? 17:57
maybe an old mailing list discussion or something
17:58 orafu joined
moritz_ monomorph_: what kind of token discussion would you like to read? 17:59
17:59 orafu left
moritz_ monomorph_: iirc the exact semantics of longest token matching aren't yet fully settled 17:59
17:59 orafu joined
monomorph_ moritz_: good to know 18:00
18:00 meppl left
Tene monomorph_: last I heard, timtoady was still trying to figure out how to make it work. 18:01
pugs_svn r19756 | ruoso++ | [smop] It compiles!!!!!!!! the test still fails... but it compiles... 18:02
allbery_b <linus> ship it! </linus> 18:04
ruoso heh
18:05 docj joined
cj Hurm... I haven't finished reading the spec yet... I'm about half way through, but it occurs to me that there seems to be the assumption that modules will be imported from a single external location, and there is probably no concept of libraries defined by the underlying VM, vs. libraries defined in .pm files in the library search path 18:09
[particle] cj: no, there's an :auth<> attribute, which allows you to specify the authority (from which the module was downloaded) 18:11
well, not 'downloaded' but i think you get the point
cj the specific question I am asking is this: when a module is defined by a library source known to the underlying VM, and there exists a library in the filesystem search path which duplicates the namespace / module name, a) how does perl6 decide which to use and b) how does the user over-ride the default decision? 18:12
[particle] as yet undefined
we'll need some kind of module registry
cj it seems to me that the perlish way to do it would be to default to the perl6 version, requiring intervention in order to use the VM-supplied thinger 18:13
monomorph_ about matching again: that STD.pm exists implies to me that people have at least an idea of it's supposed to work? 18:16
specifically, i don't understand if and when whitespace terminates a token
[particle] monomorph_: yes, but there are cases where it's still unspecced 18:17
you'll have to look at unspace and whitespace
moritz_ monomorph_: actually PGE implements much of the specs, and in some cases a bit more
[particle] like the ws, unsp, and unv tokens
18:18 tobeya joined, REPLeffect left
[particle] wonders how complete the whitespace tests are in the test suite 18:18
18:19 meppl joined
moritz_ [particle]: I think they aren't full up to date, and not very complete either 18:20
but I'm not sure
18:23 docj left
[particle] yeah, that's what i figured :( 18:23
18:24 IllvilJa left
rhr_ [particle]: I wrote some ws/unspace tests a long time ago: t/syntax/long_dot.t t/syntax/unicode-whitespace.t 18:30
[particle] rhr_: great. maybe they can help monomorph_ 18:31
monomorph_ looking right now
moritz_ it seems that the "long dot" term has disappeared from the synopsis 18:32
18:32 Jamtech joined
moritz_ at least mostly 18:32
allbery_b yeh, one of the revisions seems to have deprecated that term
pugs_svn r19757 | rhr++ | [t/syntax/long_dot.t] s/long dots?/unspace/gi 18:37
r19758 | rhr++ | t/syntax/long_dot.t is now t/syntax/unspace.t
rhr_ fixed :) 18:38
moritz_ rhr_: did you actually verify that they are still matching the specs?
18:39 barney joined
[particle] you could migrate them to t/spec if you like, too 18:40
rhr_ moritz_: at a quick glance it still looks right, I think
TimToady monomorph_: what part of S05:1929 "Longest-token matching" is unclear?
pugs_svn r19759 | rhr++ | [t/syntax/unspace.t] delete comment about renaming it to unspace 18:42
TimToady the <ws> rule is considered procedural whether called explicitly or implicitly
monomorph_ mostly the 4th bullet: (things that terminate a token include...) "Any part of the regex or rule that might match whitespace, including whitespace implicitly matched via :sigspace. (However, token declarations are specifically allowed to recognize whitespace within a token.)"
TimToady you'll note that the definition of ws uses || at its top level
so it actually follows from the | vs || rule 18:43
but \h+ doesn't call <ws>, so it's declarative
monomorph_ so .* and similar things also don't terminate the token prefix?
18:44 IllvilJa joined
TimToady no, but anyone who wants to match the rest of the file as a token is nuts. :) 18:44
[particle] NVTS.
TimToady well, I can see some uses for being nuts, and we don't disallow it :)
basically, anything we can hand off to a non-backtracking NFA engine is fair game for longest-token 18:45
and that certainly includes .*
pugs_svn r19760 | rhr++ | [unspace.t, unicode-whitespace.t] move to t/spec/S02-whitespace_and_comments
monomorph_ sounds reasonable
TimToady we're trying to be reasonable all over the place, but sometimes it's hard to explain what "reasonable" means... :) 18:46
the fundamental underlying goal is to make explicit lexers obsolete
monomorph_ okay, so what i misunderstood was that "might match whitespace" refers specifically to <ws>. maybe that's explained elsewhere (i read the whole thing yesterday, but it's hard to keep so many things in my head :)
TimToady yeah...
I should probably explain it in terms of || 18:47
thanks 18:48
monomorph_ "thanks" is what i wanted to say :) 18:49
TimToady It is part of my literary theory that communication failure is as much the fault of the author as the reader, if not more so. :) 18:50
18:51 fridim_ joined 18:52 orafu left 19:08 pbuetow left 19:12 justatheory joined
pugs_svn r19761 | ruoso++ | [smop] some advances... fetching named arguments from capture is failing for some weirdness in my use of libc bsearch 19:12
19:13 orafu joined 19:34 meppl left, kane_ left 19:56 japhb joined 20:00 manfred left 20:02 jwest- joined 20:03 macae left 20:04 wtgee joined 20:07 jwest- left 20:08 barney left 20:10 jwest- joined 20:15 jwest- left 20:17 cathyal joined
cathyal what is with freenode 20:18
horrible lags
like way horrible
20:19 namelessuser joined, Aankhen`` left
mofino like yea, like 20:30
cathyal pretty nasty
better now
mofino actually it was ok for me 20:32
20:33 blindfish left, baest_ left
mofino i just wanted to say like twice in the same sentence 20:33
20:34 baest joined
cathyal you mean like like yea! 20:34
if so you fucked it up pretty bad
:P
mofino I see
but that's just your opinion, perhaps others will enjoy, like yea, like 20:35
you never know!
cathyal talkl to me dirty moritz_
oh mofino
OK moritz_
mofino
!:P
mofino haha
cathyal hehe 20:36
TimToady Dude, dude, Dude!
cathyal so how come you're awake and nobody else
mofino DUDE
cathyal oh my god
its larry
mofino because it's 3:34pm
cathyal larry talks
cathyal chuckles
TimToady and dudettes...
cathyal hehe
hello larry!
it is indeeeeeeeed one hell of a pleasure tyo meet you 20:37
mofino to meat him
cathyal tries her best to be very diplomatic
TimToady curtsies
mofino if you ingest him, do you become him?
shall we try?
cathyal think i freaked him out 20:38
TimToady it would be difficult to freak me out more than normal...
[particle] sweet!
cathyal i would sure try mofino
TimToady dude!
Hey, dude, where's my perl6? 20:39
mofino haha
haha man 20:42
this topic is pretty intense
pugs: [~] <m oo se> (or rakudo:, kp6:, smop: etc.) ??
exp_evalbot OUTPUT[*** ā¤ Unexpected "("ā¤ expecting operator, ":" or ","ā¤ at /tmp/GCvlp9A3o3 line 1, column 15ā¤]
mofino !!
pugs: [~] <m oo se> 20:43
exp_evalbot RESULT["moose"]
mofino madness
moritz_ no madness - very cool 20:47
Gothmog_ rakudo: [~] <m oo se> 20:48
exp_evalbot OUTPUT[Parrot VM: Can't stat languages/perl6/perl6.pbc, code 2.ā¤main: Packfile loading failedā¤]
Gothmog_ kp3: [~] <m oo se>
kp6: [~] <m oo se>
exp_evalbot r19761: OUTPUT[syntax error at position 0, line 1 column 0:ā¤[~] <m oo seā¤^ HEREā¤]
Gothmog_ smop: [~] <m oo se>
mofino hmmm
Gothmog_ Dumdeedum.
mofino interesting
so what exactly are you doing?
running that expression against different engines? 20:49
Gothmog_ Mhmyup.
moritz_ aye
mofino nifty
what's kp6?
moritz_ KindaPerl6
mofino haha awesome
[particle] perl: say("hi");
eval: say("hi");
darn, which one runs all impls?
moritz_ a (nearly bootstrapped) Perl 6 compiler written in a subset of Perl 6
perl6: say("hi"); 20:50
exp_evalbot kp6: OUTPUT[hiā¤]
..pugs: OUTPUT[hiā¤]
..rakudo: No output (you need to produce output to STDOUT)
mofino written in a subset of perl6?
moritz_ rakudo: say "hi";
exp_evalbot OUTPUT[Parrot VM: Can't stat languages/perl6/perl6.pbc, code 2.ā¤main: Packfile loading failedā¤]
[particle] seems rakudo eval is broken
moritz_ ouch
mofino: yes
mofino but ... that... madness
[particle] change that to 'make perl6' and you can run the exe rather than ../../parrot ...
mofino well maybe not 20:51
just hard to understand.
moritz_ mofino: aye, it's a long bootstrapping chain
TimToady you don't have to understand English to use it...
mofino for sure, it can be confusing in any language 20:52
:)
moritz_ and who bootstrapped English?
araujo gives TimToady the prize "philosopher of the year" 20:53
mofino so why do many engines?
moritz_ mofino: there are a number of reasons 20:54
-Ofun being one of them
mofino heh ok
moritz_ differnt approaches (top->down vs. buttom->up) being another
TimToady that would be for the year containing Christmas, I presume...
moritz_ differnt tools (p5 vs. C vs. haskell) being the thirds 20:55
mofino moritz_, yeah, ok
[particle] the fourth reason is that one of them is bound to work
moritz_ rebuilds parrot on the evalbot server 20:56
lol
mofino ok
allbery_b ..you hope
moritz_ allbery_b: not only hope. Especially [particle] works for it
I do as well, to a lesser extend 20:57
and mostly in less usefull ways :/
TimToady me, I'm just all talk :)
mofino I better read about this again
moritz_ TimToady: that's what we pay you for... wait. never mind ;)
20:57 jferrero joined
mofino Ok, these engines are independant of parrot? 20:58
moritz_ rakudo: say "hi";
exp_evalbot OUTPUT[hiā¤]
TimToady not that one
moritz_ mofino: rakudo is parrot based
mofino ok!
cool
moritz_ mofino: kp6 is ... well.. kp6 and mp6 based
pugs is built with haskell
mofino so kp6 would be rakudo? 20:59
based from?
moritz_ no
TimToady rakudo: say [~] <m oo se>
mofino ohj
exp_evalbot OUTPUT[Statement not terminated properly at line 1, near "[~] <m oo "ā¤current instr.: 'parrot;PGE::Util;die' pc 120 (runtime/parrot/library/PGE/Util.pir:82)ā¤called from Sub 'parrot;Perl6::Grammar;statementlist' pc 14841 (src/gen_grammar.pir:2402)ā¤called from Sub
..'parrot;Perl6::Grammar;statement_block' pc 12028 (src/gen_grammar.pir:1396)ā¤called f...
moritz_ two entirely differnt things
20:59 minge left
mofino so how does kp6 build itself? 20:59
maybe that's a loaded question
moritz_ with an earlier version of itself ;)
partly
mofino madness 21:00
moritz_ and partly with mp6
mofino mp6?
moritz_ and mp6 is yet another, smaller subset of perl6, and a compiler that implemetns that
mofino hahah this is so intense
I guess it's not that intense, gcc is C
just really cool
[particle] what's the smallest subset of perl 6 that's turing complete? 21:01
moritz_ gcc is written in C as well, and nobody seems to care
[particle] we can call it 'tp'
moritz_ [particle]: I think that lrep is turing complete
you just need arrays + while loops for turing completeness 21:02
and an if-statement, perhaps
[particle] yep
TimToady if and lambda gets you loops
21:02 riffraff joined
moritz_ uhm, but lambda... that's cheating ;) 21:02
monomorph_ even if is unnecessary with lambda :) 21:03
if you don't compare things that is
moritz_ uhm
then who do you abort rekursion?
turing completeness implies ability to compare, doesn't it? 21:04
monomorph_ you'd encode every value as a function 21:05
for example, true is a function so that true($x, $y) == $x, false($x, $y) == $y 21:06
(boolean_expression)(then_argument, else_argument) can encode if 21:07
moritz_ scary
mofino heh neat
monomorph_ it's lambdas all the way down 21:09
21:09 ggoebel joined 21:10 ggoebel left
moritz_ but... it's ugly! scary! evil! 21:10
actually it's rather pretty, if you think a bit more about it ;)
monomorph_ slow... pointless... theoretic... 21:11
allbery_b en.wikipedia.org/wiki/Turing_tarpit :) 21:12
lambdabot Title: Turing tarpit - Wikipedia, the free encyclopedia
mofino you're no fun
21:28 japhb left 21:37 OuLouFu joined 21:53 orafu left 22:09 orafu joined
cj are there contract definition bits in perl6? 22:12
moritz_ not that I know of
uhm, wait 22:13
du you mean in the juristic sense?
or in the "design by contract" sense?
cj moritz_: "design by contract"
moritz_ ah, yes
they are defined in terms of traits
cj cool. got a url or a S version?
er, number 22:14
moritz_ wait a sec
PRE/POST in Subroutine.pod 22:15
perlcabal.org/syn/S06.html 22:16
lambdabot Title: S06
22:17 japhb joined 22:19 smtms left 22:21 wtgee left
pugs_svn r19762 | cosimo++ | [t/spec] Test suite refactoring. Relevant bits for cmp tests have been moved to S29-any 22:24
22:26 cosimo_ joined 22:27 OuLouFu left 22:44 Limbic_Region joined
pugs_svn r19763 | cosimo++ | [t/spec] Test suite refactoring. Starting work on S16! 22:45
r19763 | cosimo++ | Moved io::dir and getc tests under S16-io and S16-filehandles
r19764 | cosimo++ | Fixed name of test script (was get.c)
22:48 Auzon joined 22:54 Psyche^ joined
pugs_svn r19765 | cosimo++ | [t] Test suite refactoring. 22:54
r19765 | cosimo++ | Moved other io/filehandles tests under t/spec/S16
r19765 | cosimo++ | Converted pod syntax to pod6.
r19766 | cosimo++ | [t/spec/S16] Fixed test script names (need italian coffee) 22:56
22:57 iblechbot left 23:01 jeffreykegler joined
cosimo_ good localtime() to all! 23:02
23:03 ilogger2 joined
cosimo_ What should I do with all the `if $*OS eq "browser" { skip_rest() }' things? 23:04
Makes sense to keep them?
I mean, for rakudo 23:05
moritz_ I don't think so
the JS backend is long broken
since I joined #perl6 I haven't seen a single commit to it
cosimo_ JS? 23:07
moritz_ pugs had a javscript backend, yes
that's what the $*OS eq "browser" tests are for 23:08
cosimo_ wow!
probably it *would* make sense to build a t\harness_js then... :-) 23:09
moritz_ last time I tried not a single test passed 23:10
cosimo_ ok, clear 23:11
pugs_svn r19767 | cosimo++ | [t] Test suite refactoring. 23:17
r19767 | cosimo++ | Moved chmod tests under t/spec/S16-filehandles
r19768 | cosimo++ | [t] Refactored say/print tests undef t/spec/S16-io. 23:18
r19768 | cosimo++ | Converted pod markers to pod6.
r19769 | cosimo++ | [t] Test suite refactoring. 23:36
r19769 | cosimo++ | Moved readline tests under t/spec/S16-unfiled.
r19769 | cosimo++ | Ok, it's unfiled. Maybe we can move to better folders later?
cosimo_ bye, good night! 23:37
23:55 cyberpass joined 23:56 Khisanth 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!