Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2008-01-04

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:00 Auzon joined #perl6
00:02 amoore joined #perl6
00:14 skids joined #perl6
00:31 araujo joined #perl6
00:43 justatheory joined #perl6
00:51 Southen_ joined #perl6
00:53 Southen_ joined #perl6
00:53 cookys_ joined #perl6
00:53 Southen joined #perl6
00:55 jjore-m joined #perl6
00:58 TimToady ruoso_: see new text at S09:1161 regarding autovivification
00:59 TimToady I don't think \% should be treated different from \(%)
01:04 araujo hello
01:04 Auzon Hi
01:10 Treeform joined #perl6
01:10 go|dfish joined #perl6
01:10 Treeform ah more!
01:10 alester oh whoops
01:10 alester not here, Treeform.  go to irc.perl.org and the #parrot room
01:10 buu Is this a treasure hunt?
01:11 buu Is there a prize at the end?
01:11 buu "I'm sorry Treeform, but the princess is in another irc channel"
01:11 Treeform alester, why there?
01:11 alester Because they're discussing a pastebot using Mechanize in ther.e :-)
01:18 lyokato_ joined #perl6
01:28 cnhackTNT joined #perl6
01:35 Schwern joined #perl6
01:47 meppel-san good night
01:58 Schwern joined #perl6
02:16 jrm_ joined #perl6
02:18 Squern joined #perl6
02:23 jiing joined #perl6
02:28 Squern joined #perl6
02:29 dmq joined #perl6
02:38 Schwern joined #perl6
03:14 BinGOs joined #perl6
03:20 BinGOs joined #perl6
03:53 jjore-m joined #perl6
04:00 kanru joined #perl6
04:08 c9s joined #perl6
05:10 rhr joined #perl6
05:28 Andy_ joined #perl6
05:37 Aankhen`` joined #perl6
05:45 alc joined #perl6
05:58 luqui joined #perl6
06:21 Southen_ joined #perl6
06:21 cookys_ joined #perl6
06:22 cookys joined #perl6
06:24 Alias_ joined #perl6
06:29 dlocaus left #perl6
07:02 iblechbot joined #perl6
07:11 Southen_ joined #perl6
07:28 kupopo joined #perl6
08:18 devogon joined #perl6
08:22 BinGOs joined #perl6
08:23 franck___ joined #perl6
08:28 Southen_ joined #perl6
08:32 xinming_ joined #perl6
08:40 Southen_ joined #perl6
08:46 pochi joined #perl6
08:59 gaal Hi, folks. Has anyone been trying to build Pugs itself recently with a HEAD GHC?
09:01 avar Probably not, pugs is pretty much abandonware now ;/
09:02 avar need more lambdacamels
09:04 gaal I'm trying to treat bitrot. The compilation is failing very early with bad package paths etc., and I don't really remember how our wrapped ghc-pkg was supposed to work
09:04 gaal Does it build OK with a release GHC? What version?
09:15 avar I haven't built it in over a year
09:21 gaal One problem is that we use ghc-pkg-wrapper in configuration, but that uses the third-party/installed/packages.conf, which is initially empty
09:21 gaal So base GHC packages aren't seen, and configuration fails.
09:22 gaal Now, I don't remember if there was some way to say "use this packags.conf to read, and that packages.conf to write/install to"
09:28 gaal Okay, I have a one-character patch to do that
09:28 penk joined #perl6
09:28 gaal But now I get inconsistencies with hidden base packages :/
09:32 masak joined #perl6
09:36 jisom joined #perl6
09:41 ebassi joined #perl6
09:47 ruoso_ TimToady, nice... It's simpler now... it's always on the binding :)
09:55 Southen_ joined #perl6
09:58 luqui gaal, yeah, it's going to suck kindof.  there was a big library package shift for 6.8.  some of the breakage falls back on the package authors...
10:07 gaal luqui: (hi!) I don't understand how what we have now _could_ have worked
10:09 gaal i.e. Setup --with-hc="wrapper that does ghc-pkg -f empty-file"
10:09 luqui I have no idea what we're doing now :-)
10:10 luqui nor how to use ghc-pkg...
10:10 luqui it was just an ill-informed tidbit.
10:11 gaal what we have now == what we've had for a year or so, presumably
10:11 luqui i never knew
10:15 njbartlett_ joined #perl6
10:21 asprillia joined #perl6
10:29 meppuru joined #perl6
11:09 franck____ joined #perl6
11:10 franck____ joined #perl6
11:18 rindolf joined #perl6
11:21 franck___ joined #perl6
11:49 cognominal_ joined #perl6
12:19 jan__ joined #perl6
12:26 drrho joined #perl6
12:26 jan__ joined #perl6
12:27 ruoso_ hmmm...
12:28 ruoso kp6: my %a; my $b := %a{'b'}{'c'}; say exists %a{'b'}{'c'};
12:28 exp_evalbot r19326: OUTPUT[no method 'LOOKUP' in Class 'Bit'␤ at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 345␤ KindaPerl6::Runtime::Perl5::MOP::__ANON__('HASH(0x8e777f4)', 'LOOKUP', 'HASH(0x8e77c38)') called at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 169␤
12:28 exp_evalbot ..main::DISPATCH('HASH(0x8e777f4)', 'LOOKUP', 'HASH(0x8e77c38)') c...
12:29 ruoso kp6: my %a; my $b := %a{'b'}{'c'}; say exists %a{'b'};
12:29 exp_evalbot r19326: OUTPUT[0␤]
12:29 ruoso kp6: my %a; my $b := %a{'b'}{'c'}; say %a{'b'}.keys;
12:29 exp_evalbot r19326: OUTPUT[no method 'keys' in Class 'Bit'␤ at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 345␤   KindaPerl6::Runtime::Perl5::MOP::__ANON__('HASH(0x8e7cd44)', 'keys') called at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 169␤
12:29 exp_evalbot ..main::DISPATCH('HASH(0x8e7cd44)', 'keys') called at - line 53␤␤]
12:29 ruoso kp6: my %a; my $b := %a{'b'}{'c'}; say %a.keys;
12:29 exp_evalbot r19326: OUTPUT[␤]
12:30 ruoso it seems that the autovivification is broken in kp6 now... it used to work...
12:37 ruoso kp6: my %a; my $b := %a{'b'}{'c'}; $b{'d'} = 1;
12:37 exp_evalbot r19326: RESULT[error in Block at compiled/perl5-kp6-mp6/lib/KindaPerl6/Grammar/Sub.pm line 753, <> line 1.␤*** Syntax Error in Block: missing closing curly bracket  ␤]
12:38 ruoso kp6: my %a; my %b := %a{'b'}{'c'}; %b{'d'} = 1;
12:38 exp_evalbot r19326: RESULT[error in Block at compiled/perl5-kp6-mp6/lib/KindaPerl6/Grammar/Sub.pm line 753, <> line 1.␤*** Syntax Error in Block: missing closing curly bracket  ␤]
12:38 ruoso kp6: my %a; my %b := %a{'b'}{'c'};
12:38 exp_evalbot r19326: RESULT[error in Block at compiled/perl5-kp6-mp6/lib/KindaPerl6/Grammar/Sub.pm line 753, <> line 1.␤*** Syntax Error in Block: missing closing curly bracket  ␤]
12:38 ruoso kp6: my %a; my $b := %a{'b'}{'c'}; say %a.keys;
12:38 exp_evalbot r19326: OUTPUT[␤]
12:38 ruoso kp6: my %a; my $b := %a{'b'}{'c'}; $b{'d'} = 1;
12:39 exp_evalbot r19326: RESULT[error in Block at compiled/perl5-kp6-mp6/lib/KindaPerl6/Grammar/Sub.pm line 753, <> line 1.␤*** Syntax Error in Block: missing closing curly bracket  ␤]
12:39 ruoso kp6: my %a; my $b := %a{'b'}{'c'}; $b = 1;
12:39 exp_evalbot r19326: RESULT[error in Block at compiled/perl5-kp6-mp6/lib/KindaPerl6/Grammar/Sub.pm line 753, <> line 1.␤*** Syntax Error in Block: missing closing curly bracket  ␤]
12:57 gaal Okay, good, pugs still builds with 6.6.1.
13:04 jan__ joined #perl6
13:05 Alias_ joined #perl6
13:15 cmarcelo joined #perl6
13:18 masak joined #perl6
13:19 jferrero joined #perl6
13:32 ispy_ joined #perl6
13:35 araujo joined #perl6
13:56 jhorwitz joined #perl6
14:03 TJCRI joined #perl6
14:22 marmic joined #perl6
14:31 rindolf joined #perl6
14:58 acmoore joined #perl6
15:01 polettix joined #perl6
15:15 blindfish_ joined #perl6
15:29 skids joined #perl6
15:34 skids my @w;
15:34 skids map { @w[$^i] = coro { "a".say; yield; "b".say; return; } }, (0,1,2);
15:34 skids map { @w[$^i]($i) }, (0,1,2,0,1,2);
15:35 skids ... that does what I want to happen...
15:35 skids but I can't figure out how to get the coro code into a neater place.
15:36 skids calling a sub that yields --> "cannot yield from routine"
15:36 skids calling a ref to a block that yields --> "cannot yield from prim"
15:36 skids calling a ref to a coro --> calls the same coro six times, not three coros 2 times.
15:38 skids Is there some way to "copy a closure"?
15:44 pochi_ joined #perl6
15:48 Limbic_Region joined #perl6
16:00 jedai joined #perl6
16:11 rdice joined #perl6
16:17 |Jedai| joined #perl6
16:39 barney joined #perl6
16:51 Psyche^ joined #perl6
16:57 thoughtpolice joined #perl6
17:13 Andrew__ joined #perl6
17:14 Andrew__ left #perl6
17:15 Andrew__ joined #perl6
17:18 jedai joined #perl6
17:19 Aankh|Clone joined #perl6
17:23 Aankh|Clone joined #perl6
17:24 dlocaus joined #perl6
17:25 dlocaus .
17:25 dlocaus left #perl6
17:28 _Jedai_ joined #perl6
17:40 pugs_svn r19327 | ruoso++ | [yap6] The autovivification schema requires a full bootstrapped object system after all. So I decided to work a little more in the YAP6 bootstrap concept, take a look at NOTES_BOOTSTRAP.txt to understand how YAP6 will bootstrap itself and support any object system.
17:42 ruoso @tell fglock please take a look at http://svn.pugscode.org/pugs/v6/yap6/NOTES_BOOTSTRAP.txt
17:42 lambdabot Consider it noted.
17:43 ruoso I really would like considerations on the NOTES_BOOTSTRAP.txt file I've just written. This may be the last architectural sketch of YAP6 before having it completely planned.
17:46 ruoso TimToady, if you have some time, could you please take a look at this last notes I wrote?
17:54 hexmode` joined #perl6
17:55 hexmode` left #perl6
17:56 hexmode joined #perl6
18:02 thoughtpolice joined #perl6
18:02 rindolf joined #perl6
18:04 pbuetow joined #perl6
18:05 kupopo joined #perl6
18:06 nothingmuch joined #perl6
18:12 mncharity Has anyone created a full table of contents for the sNN?  Ie, something which scrapes the sNN for title and section headers, and combines them into a single document, linking to the assorted docs?
18:12 [particle] i've only seen that per synopsis
18:13 mncharity There's http://feather.perl6.nl/syn/ , but that's not it.
18:13 lambdabot Title: Official Perl 6 Documentation
18:13 mncharity hmm
18:13 PerlJam mncharity: I don't think so.  (at least I haven't seen anything close other than the link you just mentioned)
18:14 [particle] pod::perldoc::totoc could help
18:15 meppuru joined #perl6
18:19 mncharity re totoc, ah, so it doesn't need to be a kludge. ;)
18:20 [particle] bdfoy++
18:23 mncharity Looks like the german Perl6Tutorial is gone  http://wiki.perl-community.de/bin/view/Wissensbasis/Perl6Tutorial .
18:23 lambdabot Title: PerlCommunityWiki � Wissensbasis � Perl6Tutorial
18:23 mncharity no, just mislinked http://wiki.perl-community.de/bin/view/Wissensbasis/PerlTafel
18:23 lambdabot Title: PerlCommunityWiki � Wissensbasis � PerlTafel
18:24 mncharity http://www.perlfoundation.org/perl6/index.cgi?perl_table "my german, nicely linked Perl 6 Overview" is the mislink
18:25 lambdabot Title: Perl Table / Perl 6
18:25 mncharity wasn't there a copy of the sNN with line numbers?  I don't see it listed in http://www.perlfoundation.org/perl6/index.cgi?documentation
18:25 lambdabot Title: Documentation / Perl 6
18:27 [particle] i don't know of one, i always used svn+gvim to get line numbers
18:34 mncharity http://perlcabal.org/syn/ is it.
18:34 lambdabot Title: Official Perl 6 Documentation
18:34 mncharity eg http://perlcabal.org/syn/S09.html#line_1161
18:34 lambdabot Title: S09
18:35 [particle] ah, i see. they don't display, though
18:36 mncharity good point.  having some sort of mouseover "this is line N" might be a nice enhancement.
18:43 mncharity [particle]: PerlJam: thanks
18:43 Andrew__ left #perl6
18:45 pmurias joined #perl6
18:46 pmurias mncharity: i use grep '=head' for a table of content
18:51 mncharity right.  in this case, I'd like to create a combined table of contents, html/pod, linking into the sNN.  use case is
18:52 mncharity I seem to frequently end up wandering around the sNN, trying to re-find information I know I've seen before.
18:52 mncharity maybe it's just me? :)
19:00 Schnueff joined #perl6
19:02 pmurias mncharity: i just ack around
19:03 [particle] creating an index would be super helpful
19:03 [particle] X<>++
19:04 pmurias [particle]: yes
19:04 pmurias it would be very nice
19:08 mncharity an index?  rather than just the google sitesearch at the top of http://feather.perl6.nl/syn/ ?
19:08 lambdabot Title: Official Perl 6 Documentation
19:09 [particle] we do need to build a book out of it
19:09 iblechbot joined #perl6
19:14 mncharity Hmm.  I don't really know what the state of automated book index generation is.  But quality index creation by an professional index writer has been a very time intensive and ai-complete process.
19:35 jisom joined #perl6
19:37 wolverian it seems to be relatively hard to add showing line numbers on the page on hover into smartlinks.pl ... easiest way to me looks parsing the html as an extra pass to gather the line numbers from the html that way
19:38 * wolverian investigates more
19:42 pugs_svn r19328 | ruoso++ | [yap6] Entering a very extensive deep refactoring. After studying more
19:42 pugs_svn r19328 | ruoso++ | the specs and ellaborating NOTES_BOOTSTRAP.txt, the mechanism was very
19:42 pugs_svn r19328 | ruoso++ | simplified and we have now a implementation independent API that the
19:42 pugs_svn r19328 | ruoso++ | implementations will bind to. The first implementation will be the
19:43 pugs_svn r19328 | ruoso++ | YAP6__LOWL__... objects which will comply to the new API. The
19:43 pugs_svn r19328 | ruoso++ | structure didn't change much, but now I understood that I should start
19:43 pugs_svn r19328 | ruoso++ | by the metaclass object even for the lowlevel objects that will be
19:43 pugs_svn r19328 | ruoso++ | implemented in C. After that, the higher level objects will be
19:43 pugs_svn r19328 | ruoso++ | implemented in a stage1 language that knows how to use this lowlevel
19:43 pugs_svn r19328 | ruoso++ | objects.
19:45 pbuetow wow!
19:46 ruoso pmurias, please take a look at the new NOTES_BOOTSTRAP.txt and the new yap6.h
19:46 mncharity my first experience using socialtext.  first impression: it's very much weaker than mediawiki.  eg,
19:46 ruoso I'm in despair for reviews
19:48 mncharity eg, I changed a line or two of someone else's text.  there was no way to comment on intent (no change log message, no associated talk/discussion page).  the mediawiki social context and conventions were missing.  it felt like using a usemod.
19:50 mncharity do we *really* want to use this instead of http://perl.net.au/ ?
19:50 TimToady ruoso: the thing I get from it is that the protoobject is basically caching the responder interface that is computed by the metaclass
19:51 ruoso TimToady, not only caching, but holding the prototype information in itself
19:51 TimToady using "responder interface" as defined by nothingmuch
19:51 ruoso the  metaclass doesnt hold the class composition
19:51 ruoso the prototype *is* the responder interface
19:51 TimToady right, metaclass is generic
19:51 ruoso the metaclass is the object layout
19:51 ruoso in nothingmuch's terms
19:52 TimToady well, I think it's more than just layout
19:52 TimToady but we can split hairs later
19:52 ruoso yeah... the metaclass and the prototype are quite coupled
19:52 ruoso but that's hidden for each metaclass
19:53 ruoso so we can decouple later
19:53 TimToady nevertheless, your analysis comforts me that I did something right to invent protoobjects
19:53 ruoso :)
19:53 ruoso I need reviews, but it seems that this model can implement all of the specs
19:53 ruoso because it doesn't implement anything
19:53 TimToady because we want Dog to think it's a dog, even when it isn't.  :)
19:53 ruoso :)
19:54 chruck joined #perl6
19:54 ruoso in fact, I could even host PMCs and SV*s
19:54 ruoso inside the objects
19:55 TimToady indeed, we trying to keep layout indirection in mind for that reason
19:55 TimToady s:2nd/<ws>/ are /
19:56 JarJarBinks joined #perl6
19:56 TimToady of course, if you're gonna mix p5 and p6 runloops, you have other problems...
19:57 TimToady I mean p5 and parrot runloops
19:57 ruoso hmm
19:57 ruoso I see p5 integration as using perlembed
19:57 TimToady hard to maintain stackless when multiple runloops
19:58 TimToady s/when/with/
19:58 mncharity wolverian: just for myself, I would find a combined table of contents much more useful than the line numbers.  though line numbers might be nice too.
19:58 TimToady and as soon as you blow stackless, you make continuations very much harder.
19:58 ruoso hmm... sure...  but I wouldn't expect p5 interaction to be stackless
19:58 ruoso I mean
19:58 ruoso I wouldn't expect to make p5 continuations from p6
19:59 ruoso I can only do what p5 supports (by p5 I really mean libperl)
20:00 mncharity wolverian: the http://search.cpan.org/~bdfoy/Pod-Perldoc-ToToc-1.07/ code is quite small, and could be a good foundation.  perhaps a multi-pod combined toc module?
20:00 TimToady sure, but if we can fiddle both p5 and parrot to use an alternate runloop, we might find a way to shim both into one level. just something to bear in mind
20:01 ruoso I see.. maybe using different interpreter instances instead of stacking them
20:01 ruoso for p5
20:01 cognominal_ joined #perl6
20:02 ruoso s/them/in it/
20:02 TimToady hmm, that would make a rather heavy continuation, I suppose
20:02 ruoso hmmm
20:02 ruoso or using the trick kp6 does
20:02 ruoso which is holding closures
20:02 ruoso and running everything inside that closure
20:03 TimToady trick is to get p5 regex engine to cough up a continuation closure for you...
20:03 wolverian mncharity, yeah. unfortunately, the code is rather obtuse. :(
20:03 ruoso TimToady, fglock was dealing with it in kp6 p5rx
20:03 ruoso I'm not sure how it ended
20:03 wolverian hmm
20:03 wolverian actually, it could be separate from smartlinks.pl
20:04 TimToady well, I think he basically did his own matching with closures
20:04 TimToady at least, until the ratchet came
20:04 ruoso that's the procedural version
20:04 ruoso kp6 has a p5rx based version
20:04 TimToady but ratchet merely reduces the need for continuation
20:05 TimToady I don't profess to know how kp6 does it
20:05 ruoso I'm not sure... but I think kp6-p5rx was supporting gather-take inside tokens using continuations
20:05 ruoso or corountines
20:05 ruoso hmm
20:05 ruoso not sure
20:05 TimToady iirc it was coro
20:06 ruoso yeah... probably
20:06 ruoso anyway
20:06 TimToady but coro is in some ways not as powerful as gather/take, since you have to tie it to a sub
20:06 TimToady I suspect gather/take in kp6 is limited for that reason
20:07 ruoso this is to when we'll be implementing yap6-p5 integration...
20:07 TimToady can you gather in one dynamic scope and take in a lower dynamic scope?
20:07 ruoso yap6 needs to exist first ;)
20:07 ruoso TimToady, I'm not sure, but I think you can
20:07 mncharity wolverian: ok
20:08 mncharity anyone: perhaps remove -noindex from whatever generates the sNN html?  Then at least one would have tables of contents in the individual sNN.
20:08 TimToady well, just bear in mind that storing anything in the C stack is just going to bite you later
20:08 DaGo joined #perl6
20:08 ruoso kp6: sub a { take 10 }; my $a = gather for ( 1 .. 10 ) { a() }; say $a[1];
20:08 exp_evalbot r19328: OUTPUT[no method 'APPLY' in Class 'Undef'␤ at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 345␤        KindaPerl6::Runtime::Perl5::MOP::__ANON__('HASH(0x824c874)', 'APPLY', 'HASH(0x8e7cb88)') called at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 169␤
20:08 exp_evalbot ..main::DISPATCH('HASH(0x824c874)', 'APPLY', 'HASH(0x8e7cb88)') ca...
20:08 mncharity perhaps a bit less pretty.  but more usable.
20:08 ebassi joined #perl6
20:09 ruoso TimToady, that's why yap6 will need a full interpreter
20:09 ruoso and I'm still hoping KP6 evolves to one
20:10 TimToady well, one way or another, we'll get there
20:10 TimToady might have to discover 999 ways not to invent a lightbulb first, of course
20:11 lorn_ joined #perl6
20:11 ruoso heh.. but that's the problem of saying that your lightbulb must support an interface to any lamp
20:11 PerlJam TimToady: I just started paying attention to this window and already you're depressing me  :)
20:11 TimToady or maybe by the time we're done they'll have added continuations to C  :)
20:12 TimToady PerlJam: take it from me, it doesn't matter if you're depressed as long as you're also persistent.  :)
20:13 PerlJam indeed.
20:13 PerlJam That's sometimes a hard lesson to learn though
20:14 wolverian mncharity, hm? the individual SNNs do have ToCs.
20:14 PerlJam wolverian: He wanted an all-inclusive ToC
20:15 wolverian oh. what's that, compared to what we have now?
20:15 wolverian re: "22:08  mncharity : anyone: perhaps remove -noindex from whatever generates the sNN html?  Then at least one would have tables of contents in the individual sNN."
20:16 wolverian is he talking about the dev.perl files?
20:16 pmurias ruoso: the newest kp6 is bits of regex inside subs
20:16 PerlJam oh, I'm not sure what the end goal was beyond that.
20:16 pmurias s/kp5/kp6p5rx/
20:16 wolverian mncharity, do you mean http://spec.pugscode.org or http://dev.perl.org/perl6/doc/synopsis.html ?
20:16 lambdabot Title: Official Perl 6 Documentation
20:19 TJCRI joined #perl6
20:20 mncharity the http://feather.perl6.nl/syn/ google search box form isn't quite right, so the search always fails (eg, search for "perl").
20:20 lambdabot Title: Official Perl 6 Documentation
20:21 * ruoso going home
20:21 pmurias ruoso: isn't a prototype just a undefined object?
20:22 ruoso pmurias, no, a prototype is a wanabe object,
20:22 ruoso that happens to answer false to defined()
20:22 pmurias ruoso: yes
20:23 pmurias why is there a special YAP6__Prototype type?
20:23 ruoso because prototype is the one that have a reference to the metaclass
20:24 devogon joined #perl6
20:24 mncharity re ToC, http://dev.perl.org/perl6/doc/design/syn/S02.html , for example, does -not- have ToC at the top.  I suggest it should.  Perhaps less pretty, but more usable.  Whatever is generating it is likely being told not to create one, as at least for pod2html, it defaults on.
20:24 lambdabot Title: Synopsis 2: Bits and Pieces - perl6
20:24 pmurias ruoso: i'll think about it
20:24 jjore-m joined #perl6
20:25 ruoso TimToady, re c stack: I've been thinking about it for some time already... and in fact there are ways to implement continuations in C
20:25 wolverian mncharity, I think everyone uses spec.pugscode.org nowadays
20:25 mncharity further re ToC, it would be nice to have an html page with a combined table of contents for all the synopses.
20:25 ruoso but I really need to go home
20:25 jjore-m skids, do you still care about copying closures?
20:25 wolverian mncharity, good job spotting the google search. I don't know how to fix it, unfortunately. site:feather.perl6.nl doesn't help.
20:25 * mncharity goes to look at spec.pugscode.org ...
20:25 wolverian it's the same as feather.perl6.nl/syn
20:26 mncharity ah, a ToC.  happiness. :)
20:27 wolverian yeah. the index page could use some love though
20:27 wolverian e.g. instead of "Packages <a...>Synopsis</a> ...rest..." it could just be "<a href="syn...">Packages</a> ...rest..."
20:27 wolverian or do we want to keep the term synopsis on the index page?
20:28 mncharity and http://dev.perl.org/perl6/doc/synopsis.html a link to http://perlcabal.org/syn/ (/me wonders how many aliases feather has...)
20:28 lambdabot Title: Synopses - perl6
20:28 wolverian a few :)
20:29 pugs_svn r19329 | szabgab++ | remove unused code
20:30 pugs_svn r19330 | szabgab++ | fix smartlinks
20:31 mncharity re link, I always suggest going for simplicity and clarity.  having a column of boldface repetitions of the word "Synopsis" is just wasted ink and distraction.
20:31 wolverian I agree.
20:32 mncharity re google search, let's see...
20:32 Juerd_ mncharity: Too many
20:32 Juerd_ There are too few canonical URLs for all of Perl.
20:32 Juerd_ 5 and 6
20:32 wolverian I agree. :)
20:33 Juerd_ There's a gazillion websites about Perl, gazillion minus a dozen of them are outdated
20:33 Juerd_ perl.org is said to be canonical, but is also a lot of outdated crap
20:33 Juerd_ Then there's the commercially controlled perl.com, which is even worse
20:34 Juerd_ The owner of perl6.org has "bigger plans" for it: a migration site
20:35 Juerd_ But it does have perl6 documents right now, probably to gain some google points for it.
20:35 Juerd_ Those are synced, but that has been broken before
20:35 skids jjore-m: figured out sub { return coro {}}.  But always like to hear about funky voodoo.
20:38 jjore-m Oh, I don't know how coro interacts with magic. I was just going to say that Data::Dump::Streamer serializes closures to perl that can be evaled to duplicate them.
20:39 mncharity googling for -"FAIL Perl6-Doc" "However, these Synopses also contain updates to reflect the" suggests of the sNN sites, only dev.perl.org/perl6/doc/design/syn/ is being indexed?
20:40 gaal Hi there!
20:40 mncharity hi gaal :)
20:40 wolverian yes, it seems to be google doesn't index feather.perl6.nl/syn
20:40 wolverian I don't know why. Juerd?
20:40 gaal long
20:40 gaal time
20:40 gaal no
20:40 gaal ?eval [~] <m oo se>
20:40 mncharity lol
20:41 wolverian heh
20:41 gaal pugs: [~] <m oo se>
20:41 exp_evalbot RESULT["moose"]
20:42 [particle] gaal++
20:42 gaal http://pugscode.org/ | nopaste: http://sial.org/pbot/perl6 | pugs: [~] <m oo se> | We do Haskell, too | > reverse . show $ foldl1 (*) [1..4] | irclog: http://irc.pugscode.org/
20:42 gaal what? I didn't do anything :)
20:42 gaal oh er
20:43 gaal how does one sudo /topic on irc?
20:43 [particle] and if you continue to do nothing, we'll always love you
20:43 gaal that sounds pretty tempting
20:45 rindolf gaal!
20:45 rindolf gaal: hi.
20:46 rindolf gaal: you didn't attend the Israeli Perl Workshop, right?
20:46 gaal I did not.
20:46 rindolf gaal: ah. Why?
20:46 gaal I was busy being loved!
20:47 rindolf gaal: long time.
20:47 rindolf gaal: what's up?
20:48 gaal everything is just peachy
20:50 mncharity Juerd_: sigh.
21:09 Juerd_ mncharity: ?
21:10 wolverian Juerd_, do you know why google doesn't index feather.perl6.nl/syn?
21:15 ruoso joined #perl6
21:25 mncharity Juerd_: sigh re scattered broken state of perl web sites.  especially having two wikis.
21:31 mncharity I suppose it isn't the two-ness of the wikis which is the problem.  Python had that for a while.  It's that the one which is "official", is also rather crippled.  So instead of two healthy competing wikis, both lose.
21:33 szabgab joined #perl6
21:34 szabgab evening
21:34 szabgab it seems to me that util/prove6 loads the perl5 version of Test.pm ..
21:35 szabgab I tried ./util/prove6  t/02-test-pm/3-script.pl
21:35 szabgab and got t/02-test-pm/3-script.......pugs: user error (*** user error (***
21:35 szabgab Unexpected "(@"
21:35 szabgab ...
21:36 szabgab can anyone point me what am I doing wrong?
21:40 pasteling "szabgab" at 192.117.127.193 pasted "prove6 failure" (19 lines, 808B) at http://sial.org/pbot/29590
21:40 alester joined #perl6
21:41 hexmode` joined #perl6
21:41 pugs_svn r19331 | szabgab++ | set executable bit on pastebot
21:44 mncharity szabgab: no idea - sorry.
21:44 mncharity "Currently all of the builtins for perl6 are written in PIR. However, it would make more sense if some builtins were written in Perl 6." - Patrick R . Michaud  -  Dec 28 2007 on p6c
21:47 [particle] mncharity: what about it?
21:49 mncharity [particle]: no comment.
21:50 mncharity http://perl.net.au/wiki/Perl_6#External_Links now has links to synopses, irc, and p6l archives, and links to google them.  these seem a minimum requirement to get spec work done.  additional suggestions would be most welcome.
21:50 lambdabot Title: Perl 6 - PerlNet
21:52 Auzon joined #perl6
21:52 REPLeffect joined #perl6
22:00 meppuru good night
22:02 szabgab ?eval 1 === "1"
22:02 Auzon pugs: eval 1 === "1"
22:02 exp_evalbot RESULT[undef]
22:02 Auzon kp6: 1 === "1"
22:02 exp_evalbot r19331: OUTPUT[syntax error at position 2, line 1 column 2:␤1 === "1␤  ^ HERE␤]
22:02 Auzon pugs: 1 === "1"
22:02 exp_evalbot RESULT[Bool::False]
22:02 Auzon The topic is kind of misleading ;)
22:02 szabgab pugs: 1 !== "1"
22:03 exp_evalbot RESULT[Bool::False]
22:03 szabgab Auzon: thanks
22:03 szabgab so 1 and "1" are both not equal and non not equal
22:03 Auzon No problem ;)
22:04 Auzon Hm. You're right, that doesn't make sense
22:04 diakopter wolverian: the search engines were bringing (the original) feather to its knees.... so I told robots.txt to block all... until the performance issues could be resolved.  if it could be set to allow indexing of everything but the trac pages, then I think indexing should be reallowed
22:04 Auzon pugs: not(1 === "1")
22:04 exp_evalbot RESULT[Bool::True]
22:04 Auzon pugs: 1 !=== "1"
22:04 exp_evalbot RESULT[Bool::True]
22:04 Auzon pugs: 1 !== "1"
22:04 exp_evalbot RESULT[Bool::False]
22:04 wolverian diakopter, who can do that?
22:04 Auzon pugs: 1 != "1"
22:04 exp_evalbot RESULT[Bool::False]
22:05 diakopter Juerd_: see above
22:05 mncharity It doesn't look like svn.perl.org/perl6/pugs/trunk/  is currently in google.  Though curiously, svn.perl.org/perl6/pugs/trunk/docs/01Overview.html is.
22:05 Auzon It appears as though "!==" means "!=" :-/
22:05 szabgab ah, the negation of === is !=== or !== ?
22:05 mncharity Has anyone asked google code search to include it?  http://www.google.com/codesearch/addcode?hl=en
22:05 lambdabot Title: Google Code Search
22:05 Auzon szabgab: I'll have to look at the synopses
22:06 szabgab I see !=== mentioned in S03
22:06 szabgab under Negated relational operators
22:07 szabgab "Any infix relational operator may be transformed into its negative by prefixing with !"
22:07 szabgab so it should be !=== I guess
22:07 gaal szabgab: do you maybe have some cruft in your environment? prove6 worksforme
22:07 Auzon So the negation of "==" is both "!=" and "!=="
22:07 szabgab gaal: cruft, always but I don't know what
22:08 mncharity google code does seem to have old 2006 cvs snapshots though.  sigh.
22:08 szabgab I used to have PERL6LIB set but removed it now , either way it did not work.
22:08 gaal env | grep PERL == ?
22:08 gaal HARNESS_PERL might be wrong
22:09 gaal and I think pugs respects PERLLIB also :)
22:09 szabgab PERL5LIB is set to lots of private dirs, nothing else
22:09 gaal (yet another env var nobody uses....)
22:10 mncharity I'm about to ask google code to index  http://svn.perl.org/perl6/pugs/trunk/  .   Any cries of "NOOOOOOO!!" before I push the button?
22:10 lambdabot Title: Revision 14480: /pugs/trunk
22:10 Auzon Hm... I don't think it matters if Google can see an open source repo.
22:11 gaal szabgab: maybe try rm blib6/lib/*yml* and "make" again?
22:11 mncharity Auzon: huh?
22:11 Auzon I wasn't objecting.
22:12 mncharity ok - i just didn't understand your comment...
22:12 szabgab removing PERL5LIB solved the problem...
22:12 gaal szabgab: that's... interesting
22:14 * gaal drifts off to sleep
22:14 gaal z&
22:14 * wolverian tries to wrap his head around parallel polymorphic dispatch
22:15 mncharity g'night gaal
22:16 mncharity button pushed.  http://svn.perl.org/perl6/pugs/trunk/ heads for google code search.
22:16 lambdabot Title: Revision 14480: /pugs/trunk
22:17 LiberX joined #perl6
22:17 pugs_svn r19332 | szabgab++ | some more value identity tests
22:17 LiberX Hello, I am having trouble parsing server commands with Net::IRC. Can anyone help?
22:17 szabgab that's it, night&
22:18 Auzon LiberX: Are you running Perl 6?
22:18 LiberX Yes. I am.
22:18 Auzon Woah, I didn't know you could do that
22:18 wolverian TimToady, will perl 6 define different parallelism modes, that is, distributed vs local, or an api to plug in such things and change the parallelism implementation? not sure how that works with stm though... just thinking about parallel feeds in your post
22:18 LiberX I'm running the latest version of PERL. Whatever it is.
22:19 wolverian LiberX, try freenode #perl
22:19 Auzon Yeah. Perl 6 is in development. You're probably running 5.10
22:19 Auzon or 5.8.8
22:19 LiberX Okay.
22:19 LiberX left #perl6
22:28 ispy_ joined #perl6
22:30 TimToady wolverian: at the moment we'll just let the various implementations duke it out and see if they agree on a low-level approach to parallelism.  I'm trying to keep the language agnostic for now.
22:31 wolverian sounds like a good plan
22:32 mncharity http://feather.perl6.nl/syn/S03.html#Item_assignment_precedence says "$signature OP $capture" for OP of := and ::= , but not for = .  Is that irregularity meaningful, or just noise?
22:32 lambdabot Title: S03
22:36 TimToady = can have either item assignment precedence or list prefix precedence depending on whether the left side looks singular or plural.
22:37 TimToady $a = 1, $b = 2
22:37 TimToady but
22:37 TimToady @a = 1,2
22:37 lambdabot Maybe you meant: activity activity-full admin all-dicts arr ask . ? @ v
22:37 TimToady and maybe I didn't
22:37 mncharity lol
22:37 wolverian @arr
22:37 lambdabot Ahoy mateys
22:37 TimToady @arr = 1,2
22:37 lambdabot Swab the deck!
22:38 wolverian good thing sailors don't say "hash" instead
22:38 TimToady if you say "hash" they just think you sneezed
22:39 * mncharity exploring the connection between Signatures, Captures, and code.
22:40 * TimToady exploring the connection between Signatures, Captures, and naps.
22:40 wolverian heh
22:40 wolverian nap();
22:43 mncharity /me wonders if "$a = $b" the same as "foo(:($a),\($b))" for "sub foo($sig,$cap) { notsurewhatgoeshere($sig,$cap) }"?
22:45 Auzon arr!
22:45 Auzon @arr
22:45 lambdabot Aye Aye Cap'n
22:46 mncharity no, it can't be.  perhaps a foo(<something which captures the lexical and package environment>,:($a),\($b))
22:47 mncharity or maybe, :($a) does that?   /me goes back to groveling over spec...
22:54 thorat joined #perl6
23:01 mncharity "If a signature is assigned to (whether declared or colon form), the signature is converted to a list of lvalue variables and the ordinary rules of assignment apply, except" S03
23:03 mncharity It seems to be referring to compile-time.  Perhaps Signature has some  Env -> "list of lvalue variables" typed method.
23:19 mncharity so foo(:($a).in_current_env(),\($b).in_current_env())  where  sub foo($vars,$vals){ $vars = $vals (not quite right) }  .
23:20 mncharity .in(current_env())
23:46 Caelum joined #perl6
23:48 Andrew__ joined #perl6

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

Perl 6 | Reference Documentation | Rakudo