Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2007-05-19

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:49 thoughtpolice joined #perl6
00:50 buetow joined #perl6
00:52 japhb Finally figured out a syntax that would work for the dispatch tables, so now have current benchmark results:
00:52 japhb 'step075' benchmark (values in FPS):
00:52 japhb REVISION    NOGFX   STUBBED    FULL
00:52 japhb r15908     13.796     9.390   5.170
00:52 japhb r16189     14.003     9.534   5.084
00:52 japhb r16322     11.784     8.228   3.890
00:52 japhb r16421     18.164     8.828   error
00:52 japhb r16447     22.089     9.861   4.595
00:52 dduncan and my smoke is done
00:53 dduncan oddly enough, smokeserv-client.pl seemed to hang after it uploaded the data, and eventually died with error: 500 read timeout ... but prior to that, I could already see the smoke on the server
01:00 mjk joined #perl6
01:14 polettix joined #perl6
01:19 Aankhen`` Is there an easy way to test out changes to smokeserv-server.pl?
01:19 lambdabot Aankhen``: You have 2 new messages. '/msg lambdabot @messages' to read them.
01:20 SubStack joined #perl6
01:21 thoughtpolice what exactly is the difference between 'use v6' and 'use v6-alpha'? from the pod, v6 is used for pure perl6 code that is not valid perl5, and v6-alpha is just a perl5 module to implement perl6; is there a real difference?
01:23 japhb joined #perl6
01:23 thoughtpolice v6-alpha is a p5-2-p6 compiler front end (experimental) that implements a small fraction of p6 semantics and syntax; does pugs implement more than this, thus the distinction?
01:24 Caelum joined #perl6
01:25 japhb pugs does implement more than v6.pm.  pugs is p6-on-haskell.  v6 is p6-on-p5.  mp6/kp6 are p6-on-p6 (ask fglock for all the gory details).  And then there's parrot's native perl6 implementation, which is bottom up instead of top down (as pugs more or less is)
01:26 japhb but the line 'use v6;' tells p6 or p5 that you want to speak p6 now.
01:26 thoughtpolice so basically v6-alpha is used when you want perl6, but you may lack pugs since perl5 will just call the module to execute properly?
01:28 thoughtpolice easy enough; I saw the two use clauses used many places so I didn't exactly know if there was a definitive difference or somesuch
01:28 japhb when you say 'use v6-alpha;' you are actually making use of valid p6 syntax and "just happens to work" p5 syntax to do the right thing no matter which actual executable you use
01:28 thoughtpolice IOW, v6.pm is just 'perl on perl action'
01:29 japhb Because pure 'use v6;' under p5 would tell you that your perl is too old.  But 'use v6-alpha;' loads v6.pm and starts compiling your p6 code to p5.
01:30 Aankhen`` Hmm, I thought v6-alpha indicated that the code was written using unstable P6 syntax.
01:30 japhb Notionally, you're correct.
01:30 Aankhen`` Yay.
01:30 thoughtpolice ah. so basically the difference is if you would rather use pugs as your implementation (v6) as opposed to just compiling to p5 (v6-alpha)
01:30 japhb But it has to be that way to make p5 not just error out.  Happy accident.
01:30 Aankhen`` Heh.
01:31 japhb thoughtpolice: I suppose you could think of it that way, sure.
01:31 thoughtpolice japhb: thanks.
01:31 thoughtpolice japhb++
01:31 japhb But I'm sure TimToady or fglock would give you some deeper concept.  :-)
01:31 thoughtpolice delicious.
01:32 thoughtpolice japhb: thanks for the distinction; the difference is clear.
01:37 thoughtpolice japhb: i'm mainly asking due to the fact I wish to add my little prompt app to examples/, but I simply did not know which would be more appropriate; since I'm more interested in using pugs, I'll just go the 'use v6;' route I suppose.
01:40 svnbot6 r16452 | thoughtpolice++ | added examples/promptr.p6
01:40 thoughtpolice that would be the one.
01:42 Jmax do i need ghc-extralibs?
01:46 svnbot6 r16453 | Aankhen++ | util/smartlinks.pl:
01:46 svnbot6 r16453 | Aankhen++ | * removed spurious `display: block` CSS.
01:46 svnbot6 r16453 | Aankhen++ | * made JS logic a bit more intelligent about replacing the trailing :.
01:46 svnbot6 r16453 | Aankhen++ | * tweaked the display of line numbers and test results in the snippet labels.
01:47 Aankhen`` I think my commit messages are longer than the lines of code I write.
01:48 [particle1 joined #perl6
02:12 justatheory joined #perl6
02:20 charsba__ joined #perl6
02:31 [particle] joined #perl6
02:44 perlDreamer joined #perl6
02:46 kanru joined #perl6
02:49 mako132_ joined #perl6
02:53 kanru joined #perl6
02:54 perlDreamer Has anyone noticed changes in the scoping of $_ since the last big merge?
02:54 perlDreamer There are some closure_trait tests failing because $_ looks like it's undef all of a sudden
03:02 dmwaters {global notice} Good day all! It's that time again, I've got 4 major rotation servers to upgrade. They've been out of rotation for a while so shouldn't be too noisey, but will cause some major disturbence as this effects the tor servers to. I will give further notice in wallops, so /mode your_nick +w if you are interested. Thank you for your patience, and thank you for using freenode!
03:03 dduncan there were some $_ scope things checked in according to comments
03:03 jvoorhis1 joined #perl6
03:03 dduncan for eval blocks for one thing
03:09 perlDreamer sounds right
03:09 perlDreamer the test wraps an eval
03:10 perlDreamer it's the very first test in t/closure_traits/start.t
03:16 araujo joined #perl6
03:17 jql joined #perl6
03:17 nelhage joined #perl6
03:17 araujo joined #perl6
03:17 araujo joined #perl6
03:18 buubot joined #perl6
03:18 jql joined #perl6
03:18 buu joined #perl6
03:18 QtPlatypus joined #perl6
03:18 kolibrie joined #perl6
03:18 TreyHarris joined #perl6
03:18 takesako___ joined #perl6
03:18 SamB joined #perl6
03:18 arguile joined #perl6
03:18 allbery_b joined #perl6
03:20 pasteling joined #perl6
03:21 shachaf joined #perl6
03:21 LCamel joined #perl6
03:21 Lunchy joined #perl6
03:21 Maddingue joined #perl6
03:22 nelhage joined #perl6
03:22 nothingmuch joined #perl6
03:22 takanori joined #perl6
03:22 gaal joined #perl6
03:22 cernd joined #perl6
03:22 rgs joined #perl6
03:23 obvio171 joined #perl6
03:23 haelix joined #perl6
03:23 cls_bsd joined #perl6
03:23 orafu joined #perl6
03:23 Gothmog_ joined #perl6
03:25 Caelum joined #perl6
03:26 mtve joined #perl6
03:26 Tene joined #perl6
03:28 TreyHarris joined #perl6
03:36 Yaakov joined #perl6
03:39 REPLeffect joined #perl6
03:44 zgh joined #perl6
03:44 svnbot6 r16454 | Aankhen++ | t/syntax/unicode.t:
03:44 svnbot6 r16454 | Aankhen++ | * added a Hindi test
03:44 drupek12 joined #perl6
03:53 svnbot6 r16455 | Aankhen++ | util/smartlinks.pl:
03:53 svnbot6 r16455 | Aankhen++ | * removed `name` attribute on anchors in favour of `id` attribute.
03:58 offby1 joined #perl6
04:00 zperl joined #perl6
04:01 weinigLap joined #perl6
04:17 bonesss joined #perl6
04:22 justatheory_ joined #perl6
04:28 [particle] joined #perl6
04:28 bonesss joined #perl6
05:09 BooK_ joined #perl6
05:59 meppl guten morgen
05:59 Aankhen`` Namaste meppl. :-)
05:59 meppl :D
06:00 meppl good morning aankhen
06:00 Aankhen`` Kaise hai aap? ^_^
06:00 Aankhen`` (Translation: how are you?)
06:01 meppl im fine
06:01 meppl i cant speak indian
06:01 Aankhen`` It's Hindi, actually.
06:01 meppl k
06:02 Aankhen`` India has about 20 official languages, and hundreds more spoken all over, so calling a language "the Indian language" would be a tad inaccurate. ;-)
06:03 meppl i see
06:05 perlbot joined #perl6
06:15 spinclad TimToady: S12:1650 could perhaps explain 'basal' as 'non-final' or such; i tripped over its lack of definition for a while today.
06:26 Averell joined #perl6
06:33 devogon joined #perl6
06:43 tokuhirom_ changed their nick to toku_away
07:00 zperl joined #perl6
07:06 cybercobra joined #perl6
07:06 cybercobra could anyone give an example of how UNDO would be used?
07:22 Aankhen`` Hmm.  This isn't the best example, but: try { $successful_transactions++; ...some code that might blow up...; UNDO { $successful_transactions--; } }
07:22 Aankhen`` Er, forget the `try`.
07:25 Aankhen`` Meh, I'm not good at useful examples  when I'm sleepy.
07:25 Aankhen`` s/  / /
07:40 iblechbot joined #perl6
07:55 demq joined #perl6
07:59 elmex joined #perl6
08:05 jisom joined #perl6
08:07 dduncan left #perl6
08:10 Aankhen`` From S04: "Note though that the first expression within parens or brackets is parsed as a statement".  So does that also work for, e.g. my @foo = 'bar', ($_ for <a b c d e>), 'baz'
08:10 Aankhen`` ?
08:10 Aankhen`` (Yes, I realize what a contrived example that is.)
08:33 svnbot6 r16456 | Aankhen++ | util/smartlinks.pl (JS):
08:33 svnbot6 r16456 | Aankhen++ | * combined a couple conditionals.
08:33 svnbot6 r16456 | Aankhen++ | * moved the location.hash bit outside the loop.
08:33 svnbot6 r16456 | Aankhen++ | * fixed the line anchors!
08:36 larsen_ joined #perl6
08:56 Aankhen`` Ah, reading about `do` in rvalues answered my question.
08:56 araujo joined #perl6
08:59 Aankhen`` ?eval my @uniq = gather for <a a b a c d d> { state $previous = take $_; next if $_ === $previous; $previous = take $_; }
08:59 Aankhen`` Aiye.
09:00 Aankhen`` Where's evalbot?
09:00 Tene dead, apparently
09:00 Aankhen`` O noes, it's on another server.
09:00 Aankhen`` Or something.
09:01 Aankhen`` * evalbot_r16443 (i=audreyt@feather.perl6.nl) Quit (anthony.freenode.net irc.freenode.net)
09:01 Aankhen`` BTW, need someone to help me figure this out... in the snippet I was eval'ing above, it basically discards consecutive duplicates, right?
09:02 Aankhen`` So if the input were <a a b a c d d>, I'd get <a b a c d>?
09:02 Aankhen`` Er, actually.
09:02 Aankhen`` Now I'm confused.
09:02 Aankhen`` I just don't understand that code.
09:03 Aankhen`` Too many `take $_`s in there.
09:04 rindolf joined #perl6
09:08 penk joined #perl6
09:17 kanru joined #perl6
09:44 buu Take three $_ and call me in the morning.
09:45 VanilleBert joined #perl6
09:48 prism joined #perl6
09:58 zperl hi
09:59 zperl im trying to build pugs from ubuntu distro... cant i do it using ghc 6.6?
10:00 CindyLinz joined #perl6
10:01 audreyt joined #perl6
10:01 DarkWolf84 joined #perl6
10:09 avar zperl: afaik yes
10:09 avar well, I'm building it with 6p6 on debian so yes
10:13 trunix joined #perl6
10:15 BooK joined #perl6
10:16 elmex joined #perl6
10:19 chris2 joined #perl6
11:15 bernhard joined #perl6
11:20 buetow joined #perl6
11:50 VanilleBert left #perl6
12:02 pbuetow joined #perl6
12:15 lisppaste3 joined #perl6
12:25 Jedai joined #perl6
12:31 Jedai666 joined #perl6
12:32 Jedai666 changed their nick to Jedai
12:42 Limbic_Region joined #perl6
12:48 pmurias joined #perl6
12:48 pmurias hi
12:49 pmurias how difficult would it be to have a progress bar during the pugs build process?
12:50 pmurias (thought about it while watching ghc being built from source)
12:51 * Limbic_Region doesn't see the need
12:52 Limbic_Region the N out of 112 works for me
12:52 pmurias explain the phrase please
12:55 Limbic_Region pmurias, when I build pugs it tells me that it is on 1 of 112, then 2 of 112, then 3 of 112, etc
12:55 Limbic_Region that is sufficient for me to know how much progress it has made
12:58 pmurias sorry for a stupid suggestion (I last built pugs a long time ago)
12:58 buetow joined #perl6
12:58 * pmurias is waiting for ghc6.6 to compile to check if having 10x more ram will speed it up
13:02 Limbic_Region why 6.6 then?
13:02 Limbic_Region I would think 6.7 would be the way to go
13:06 pbuetow joined #perl6
13:09 pmurias i used the newest one in portage
13:10 pmurias the newest on haskell.org/ghc is 6.6.1
13:11 pmurias bike trip&
13:14 xern joined #perl6
13:19 prism joined #perl6
13:31 pasteling "Limbic_Region" at 68.48.99.41 pasted "Build failure and subsequent smoke test issues on Win32/MinGW as of r16456" (59 lines, 4.5K) at http://sial.org/pbot/24983
13:31 Limbic_Region @tell audreyt Build failure and subsequent smoke test issues on Win32/MinGW as of r16456 - http://sial.org/pbot/24983
13:31 lambdabot Consider it noted.
13:43 Jmax Makefile.PL in pugs svn has a pod error
13:59 ggoebel joined #perl6
14:24 moritz how do you validate pod?
14:25 Jmax Test::Pod iirc
14:25 Jmax but that's not how I know :)
14:25 moritz cpan++
14:25 moritz so how do you know?
14:26 wolverian podchecker if you don't need to automate it
14:26 Jmax perldoc warns one
14:26 Jmax *** ERROR: Spurious text after =pod at line 3 in file Makefile.PL
14:26 moritz debian++ # for providing libtest-pod-perl ;-)
14:27 Jmax neat, podchecker++
15:03 bonesss joined #perl6
15:05 maskd joined #perl6
15:52 araujo joined #perl6
15:54 franck___ joined #perl6
15:59 pmurias i get
15:59 pmurias Can't call method "load_all_extensions" on an undefined value at inc/Module/Install.pm line 128.
15:59 pmurias BEGIN failed--compilation aborted at Makefile.PL line 49.
16:00 pmurias when i run Makefile.PL
16:07 TimToady whoops, forgot to spinclad++ on s:g/basal/super/
16:11 rindolf Hi all.
16:11 rindolf Hi moritz, TimToady
16:11 moritz hi rindolf ;)
16:11 rindolf moritz: what's up?
16:11 rindolf I submitted three grant applications for TPF.
16:11 moritz rindolf: cool ;)
16:11 moritz rindolf: for what projects?
16:12 justatheory joined #perl6
16:12 rindolf moritz: here they are: http://www.shlomifish.org/F​iles/files/text/tpf-grants/
16:12 moritz rindolf: I'm just coding glue scripts atm, nothing special ;)
16:12 lambdabot Title: Index of /Files/files/text/tpf-grants
16:12 rindolf moritz: OK.
16:12 rindolf NEXT:: is causing me trouble.
16:13 justatheory joined #perl6
16:14 moritz rindolf: I noticed that File::Find::Rule is sometimes horribly slow - can you confirm that?
16:15 Ixodes joined #perl6
16:17 rindolf moritz: no, I never benchmarked it.
16:18 moritz well, I didn't profile it, so I don't really know if it really is slow, or if the calling application made some mistakes
16:21 Jmax uploading openbsd smoke :)
16:21 moritz Jmax++
16:21 Jmax any other requests? does haskell run on SCO?
16:22 xinming joined #perl6
16:22 Yappo joined #perl6
16:24 moritz SCO is evil ;)
16:26 Yappo joined #perl6
16:27 Jmax evil, yes
16:27 Jmax but that's what makes it fun
16:27 Jmax much like VMS
16:27 moritz ;)
16:27 moritz did you do some parrot smokeing?
16:27 Jmax i'm going to do that now
16:28 Jmax i 'make fulltest', right?
16:28 Jmax or just make smoke?
16:29 [1]jaffa joined #perl6
16:29 moritz make smoke does the trick
16:29 [1]jaffa hi
16:29 [1]jaffa how is perl6 going?
16:29 moritz hi [1]jaffa
16:29 moritz well, it's developing ;)
16:29 [1]jaffa who is it made for?
16:30 moritz for every programmer how is lucky enough to be allowed to use it ;)
16:31 [1]jaffa why do you say that?
16:32 moritz well, there will always be companies that find perl "scary"
16:32 moritz because it gives you enough rope to hang yourself if you want to
16:32 moritz and a few kilometers more ;)
16:33 [1]jaffa well, that is not good, is not it?
16:33 VanilleBert joined #perl6
16:33 moritz that was always perl's philosphy - to allow everything you could possibly want
16:34 [1]jaffa it does not sound bad...
16:34 [1]jaffa but it is looking like an abandoneware
16:35 [1]jaffa why is it going to be finished?
16:35 moritz did you mean "when"?
16:35 Tene moritz: how many commits have we had in the past couple of days? ;)
16:35 [1]jaffa when. when.
16:35 [1]jaffa that is right
16:35 moritz Tene: perhaps 10 I'd guess...
16:36 moritz Tene: a few irclog patches, a few testcases on operator precedence, and a few fixes in ext/
16:36 moritz [1]jaffa: christmas, but we don't know which christmas ;)
16:36 Tene [1]jaffa: http://cia.vc/stats/project/perl6
16:36 lambdabot Title: perl6 - CIA.vc
16:36 moritz Tene: whey are you asking?
16:36 Tene moritz: he was suggesting it was abandoned.
16:38 [1]jaffa ok, I see something is happening.
16:38 moritz [1]jaffa: pugs and parrot are both moving
16:39 moritz [1]jaffa: pugs in a top down fashion, parrot bottom up
16:39 moritz and hopefully they'll meet in the middle ;)
16:40 [1]jaffa I though they already met
16:40 [1]jaffa just more connections are needed
16:40 moritz pugs has a parrot backend, but atm that's bit rotten
16:41 [1]jaffa I see
16:41 pmurias audreyt has done work on her private branches and was begining to merged them recently
16:42 [1]jaffa ok
16:42 [1]jaffa I am going.
16:42 [1]jaffa bye all
16:43 moritz bye
16:43 Tene Odd.
16:43 moritz maybe we should write and FAQ on perl6 status?
16:43 Tene Perhaps.
16:46 prism joined #perl6
16:53 moritz why do have the feeling that I've seen such a FAQ somewhere?
16:58 Jmax parrot r18582 has been smoked on openbsd
16:59 TimToady yay
17:00 Jmax http://smoke.parrotcode.org/smoke/parrot-s​moke-0.4.12-devel-r18582-unknown--i386-ope​nbsd-cc-default--1179593746-955--7175-7155​-20-253-609-0--2d443430202d2d67632d6465627​567--727e9a5660ab8ec0311177b365a1dd9e.html not a bad success rate either
17:00 lambdabot Title: TAP Matrix - Sat May 19 16:55:04 2007 GMT, http://tinyurl.com/yoh8dt
17:00 jisom joined #perl6
17:02 moritz Jmax: yes, the parrot++ folk are very serious about quality ;)
17:03 Jmax i'm surprised that the smoke util doesn't acquire more hardware information
17:03 Jmax i imagine things like CPU arch/speed/build and ram/swap can put the times in proper perspective
17:03 xinming_ joined #perl6
17:04 moritz right
17:04 moritz you could propose that on #parrot on irc.perl.org or parrot-porters@perl.org
17:04 semifor joined #perl6
17:04 moritz here your parrot ideas are likely to be forgotton :(
17:05 TimToady unless [particle]++ sees them...
17:05 prism joined #perl6
17:05 moritz of course ;)
17:08 TimToady but we mustn't observe [particle] or we'll break the quantum entanglement between pugs and parrot
17:20 Psyche^ joined #perl6
17:24 moritz how feature complete is STD.pm?
17:29 amnesiac joined #perl6
17:32 weinigLap joined #perl6
17:33 tzoa joined #perl6
17:34 TimToady it's feature complete except for whatever I've forgotten.  :)
17:34 TimToady but it's only a parser
17:34 moritz ok
17:35 buetow__ joined #perl6
17:35 TimToady I also need to go back and retrofit some of the decisions there back into the synopses, which are a little out-of-date on the grammatical categories
17:37 TimToady the todo at the beginning still lists four things
17:37 moritz "only a parser" - I just remeber a quote by damian about parsing perl... ;)
17:38 Psyche^ changed their nick to Patterner
17:38 moritz any ideas for a status faq? I currently have "is perl 6 dead?", "what about perl 6 implementations?", "when will it be released?" and "why pugs? why haskell?"
17:39 TimToady status faqs tend to go out of date rather rapidly, which says something about the status :)
17:39 moritz TimToady: you're right about that, but I keep answering the same questions here on #perl6 over and over again
17:41 TimToady but how people respond to the answers often tells us whether they would be good contributors, so it might be worth it
17:41 TimToady especially if we can share the load
17:42 obra Has anyone played with http://www.cs.rit.edu/~bja8464/lambdavm/ ?
17:42 lambdabot Title: LambdaVM - The Haskell to Java Translator
17:42 TimToady sending people to a faq immediately sort of says "go away and don't bother me"
17:43 justatheory joined #perl6
17:43 moritz TimToady: ok, you convinced me
17:43 moritz *not commiting status-faq.html*
17:44 TimToady maybe we can tell people they can get a status update for free if they sign up for a commit bit :)
17:45 moritz ;-)
17:45 ingy good morning
17:47 moritz good evening ingy ;)
17:48 buetow joined #perl6
17:48 ingy :)
17:49 TimToady obra: looks interesting as an intermediate solution, though we can perhaps do better with direct P6-to-Java.  But it at least shows it's possible, I think.
17:49 TimToady I wonder how much emulation it has to do?
17:49 amnesiac to Java bytecode? or to Java assembly?
17:50 TimToady it says bytecode
17:50 moritz aren't they isomorphic?
17:51 amnesiac not really, you can look at some Java assembly using javap
17:51 amnesiac somewhat more readable
17:53 rashakil joined #perl6
17:54 obra *nod* It was more "eenteresting" than "oh my god. we need to use this Now"
17:54 TimToady "how to make a sow's ear out of a silk purse"
17:55 obra *laugh*
17:55 buetow joined #perl6
17:59 pbuetow joined #perl6
18:09 zperl joined #perl6
18:14 weinigLap joined #perl6
18:15 renormalist joined #perl6
18:24 rindolf Can anyone tell me when overload q{""} => \&_stringify() allows it to return undef?
18:25 rindolf Because from a naive script I'm trying it keeps translating undefs to "".
18:27 Jmax is this p5?
18:29 rindolf Jmax: yes, it is.
18:29 Jmax what package did you "use overload" in?
18:30 rindolf Jmax: let me nopaste my code.
18:31 pasteling "rindolf" at 88.152.120.36 pasted "Trying that stringification will return undef." (64 lines, 741B) at http://sial.org/pbot/24990
18:32 Jmax you're not acutally using package B; are you?
18:32 Jmax if you are, try changing it to C :)
18:35 rindolf Jmax: I am - look below.
18:35 rindolf my $b_instance = B->new("Hello");
18:36 Jmax I don't recall exactly, but there was an issue on perlmonks that I had read a while back regarding a basic example of objects where the user used the B package as an example.  
18:37 Jmax the issue turned out to be that there already exists a package B, and that something he was doing was interfering with it.  just a thought
18:37 rindolf Jmax: oh OK.
18:38 Jmax see perldoc B
18:38 Jmax (it doesn't mention anything about the topic, but it exists)
18:38 rindolf Jmax: OK.
18:39 rindolf Jmax: http://sial.org/pbot/24991
18:39 lambdabot Title: Paste #24991 from "rindolf" at 88.152.120.36
18:43 araujo joined #perl6
18:47 Jmax seems to work fine here, print Child->new->text, "\n"; print Child->new("Hello")->text, "\n"; # undef, "Hello"
18:49 rindolf Jmax: yes, but I'm interested in the stringification that happens at NEXT::AUTOLOAD.
18:50 rindolf Jmax: my problem is that it doesn't fail.
18:52 Jmax set a fallback?
18:52 Jmax maybe I misunderstand, heh
18:53 rindolf Jmax: in line 53 I'm calling NEXT::_initialize before the {text} is inited.
18:54 rindolf Jmax: at this point NEXT:: stringifies $self, and I'd like the "$self" to return undef.
18:54 rindolf Jmax: however, it returns the empty string (q{""})
18:57 Jmax how do you know that it stringifies $self ?
18:58 rindolf Jmax: I studied the code.
18:58 Jmax NEXT.pm ?
18:58 rindolf Jmax: $NEXT::NEXT{$self,$wanted_method}
18:59 rindolf Jmax: yes, of NEXT.pm
18:59 Jmax oh, I see
18:59 Aankhen`` joined #perl6
18:59 Jmax but it doesn't it behave as desired?
19:04 bonesss joined #perl6
19:19 nothingmuch joined #perl6
19:31 dduncan joined #perl6
19:39 isaacd__ joined #perl6
19:55 devogon joined #perl6
20:04 justatheory joined #perl6
20:14 dduncan @tell TimToady thanks for the S02 update to clarify type/role specific sigil binding
20:14 lambdabot Consider it noted.
20:15 dduncan that is such a handy utility
20:18 TimToady though of course I always backlog...
20:18 lambdabot TimToady: You have 1 new message. '/msg lambdabot @messages' to read it.
20:18 TimToady @messages
20:18 lambdabot dduncan said 3m 26s ago: thanks for the S02 update to clarify type/role specific sigil binding
20:19 dduncan sure, and sometimes I have counted on backlogging ... but I also think it unreasonable to assume people will as there can be a lot of stuff to wade through
20:19 TimToady maybe we could have a micro tell that just tells you to backlog with a link to the irc logger
20:19 dduncan I only occasionally backlog myself
20:19 TimToady I figure if someone stays logged in they intend to backlog, probably
20:20 dduncan so that, if your name is mentioned, you are informed?
20:20 pjscott joined #perl6
20:20 dduncan well I do to an extent ...
20:21 dduncan I sleep my computer every night, and don't typically backlog stuff during that time ... however, I open IRC when I wake the computer and keep it up while I'm up, and tend to backlog what happened, but then it doesn't get too old
20:21 dduncan sleep=disconnect
20:21 dduncan when I want to check the other period, there is the web logs
20:21 TimToady when I type dd<tab> and it doesn't autocomplete, I figure you're not on and @tell you
20:23 dduncan now I have a question concerning the S02 addition ...
20:23 dduncan you said that, eg, @ is for anything that does the Array role, then including some immutable types in the lsit
20:24 TimToady I'm thinking the Array role doesn't provide rw, just the Array class
20:24 dduncan I see ... that could be confusing
20:24 TimToady but it might be some other name
20:25 dduncan or could "but" or some such be used for an exclusion, to say you just partially compose a role
20:25 dduncan eg, does Array but ro
20:26 TimToady seems not very monotonic to allow removals
20:26 dduncan what is better, when if someone tries to mutate a Seq, that this be caught at compile time as no-such-method or a runtime exception
20:26 dduncan such as when someone mutates a parameter which is a mutable type but marked readonly
20:27 dduncan is a runtime exception, at least now
20:27 moritz if it's possible error detection at compile time is always preferable
20:27 TimToady well, it's potentially a compile-time error if deducable
20:28 TimToady "error: this will certainly fail at run time"  :)
20:28 dduncan yes ... so in the general case it is runtime, but also compile time in limited cases where we can determine
20:29 moritz dduncan: I think that applies to all but syntax errors (which are always at compile time)
20:29 dduncan yes
20:29 TimToady but yes, this comes back to the same question of whether mutability is really a role
20:29 TimToady in the abstract, a mutable object is always just a container for the current immutable value
20:29 dduncan as roles are currently defined, to consist of a set of methods, I would say mutability is not a role
20:30 dduncan I agree
20:30 TimToady however, if you implement it that way, it can be very slow
20:32 TimToady well, I need to wander off, but I'll think about it some
20:32 moritz why does this smell like haskell?
20:32 moritz ;-)
20:32 TimToady perhaps because I was thinking of haskell when I said it?
20:32 dduncan conceptually values and containers are 2 completely different things
20:33 dduncan fyi, my database language tends to emphasize that
20:33 TimToady I noticed... :)
20:34 justatheory joined #perl6
20:34 dduncan while a language can provide the appearance of mutation, it is just a shorthand for the concept of deriving a new value from another one with a small difference ... and could be implemented that way behind the scenes ... but officially it is a new value
20:34 TimToady well, that is the FP view.  the OO view is that a container is a stateful object, not just a reference
20:35 TimToady but the two views are duals of each other
20:36 TimToady anyway, vamoosification &
20:36 dduncan in the terminology of my language, what perl 6 calls a view, I call a "possible representation"
20:36 dduncan s/view/role/
20:37 dduncan mind you, possreps only apply to "scalar" types ... which are analagous to objects with just private attributes
20:37 dduncan collection types that are transparent do not have possreps, just a single representation
20:38 dduncan this is all user-pov of course ... my language doesn't dictate implementation, only what the users see
20:38 dduncan as afaik most/all languages do
20:38 * dduncan back to work
20:46 marmic joined #perl6
20:46 freezone joined #perl6
20:47 svnbot6 r16457 | unobe++ | fixed Makefile.PL pod (JMax++)
20:47 Jmax :)
20:52 iblechbot joined #perl6
20:55 pjscott left #perl6
21:07 moritz diakopter: could you please nopaste that?
21:07 diakopter NOOO
21:07 diakopter sorry :(
21:07 * diakopter begs forgiveness from all
21:07 diakopter moritz: it was an accident
21:07 diakopter didn't mean to paste at all.
21:07 moritz diakopter: ok
21:07 * diakopter requests that it be purged from your log, also :)
21:08 moritz diakopter: ok, I'll do it
21:08 diakopter moritz: thanks.  how embarassing for me. thank you again.
21:10 moritz mysql> delete from irclog where id >= 20702 and id <= 20732;
21:10 moritz Query OK, 31 rows affected (0.00 sec)
21:27 Psyche^ joined #perl6
21:36 Psyche^ changed their nick to Patterner
21:44 * trippeh got paste-protection in his irc-client
21:44 trippeh :)
21:44 moritz irssi++ ;)
21:53 buetow joined #perl6
22:05 renormalist joined #perl6
22:17 renormalist where do I fix typos in the synopses? In pugs/docs/Perl6/Spec/*.pod? Are these the files that are used for http://perlcabal.org/syn/?
22:17 lambdabot Title: Official Perl 6 Documentation
22:18 moritz renormalist: I'm not quite sure... they live in a different repository, but are somehow synced with the ones in pugs/docs/Perl6/Spec...
22:18 moritz renormalist: easiest way is to /msg, @tell or mail TimToady
22:20 renormalist moritz: hm, it's just a missing newline that breaks an =item list. I don't want to spam someone with minor stuff. I fix it in pugs and wait what happens. :-)
22:20 moritz ok ;)
22:21 renormalist ah, they aren't in pugs, just sync'd from somewhere :-) now I know ...
22:22 renormalist @tell TimToady S11/Modules misses a newline at 249 which breaks the following =over/=item list
22:22 lambdabot Consider it noted.
22:23 renormalist :-)
22:39 svnbot6 r16458 | Darren_Duncan++ | ext/QDRDBMS/ : updated Operators.pm to normalize sum|product as nary functions ; updated AST.pm to add Quasi(Set|Seq|Bag) node types
22:42 dduncan renormalist, it is a limited group of people that can update the synopsis, so you have to ask them to do it, in general ... usually Larry
22:42 dduncan also Audrey
22:45 renormalist dduncan: k, thx
22:47 jisom joined #perl6
22:51 Limbic_Region joined #perl6
22:56 diakopter moritz: I *am* using irssi; putty/right-click is to blame.
22:56 moritz putty-- ;-)
22:57 offby1 *gasp*
22:57 offby1 I _love_ PuTTY
22:57 offby1 what did it do wrong?
22:57 Jmax i've had that happen so many times
22:58 Jmax i just set putty to a different mode (right click == menu, not paste)
22:58 offby1 accidental paste?  _All_ terminal programs that I know of allow that
22:58 meppl good night
22:58 Jmax Settings -> Window -> Selection
22:58 Jmax you can still accidentally mispaste, it's just one more step to do so :)
22:59 moritz offby1: usually when you paste more than $n lines, irssi asks you to confirm it...
23:00 moritz offby1: so if diakopter says that putty is to blame, I assume that it sends the past line by line, so that the warning didn't work
23:01 diakopter Jmax: thanks; good point
23:03 offby1 moritz: hmm.
23:05 moritz offby1: perhaps that's a premature conclusion ;)
23:05 Jmax you're welcome
23:05 offby1 you think "hmm" is a premature conclusion?  I assure you, I pondered long and hard before coming up with that.
23:05 theorb joined #perl6
23:06 moritz offby1: *g*, no my conclusion ;)
23:06 offby1 *swat*    yeah, I knew what ya meant.
23:07 moritz and I started to think that perl hackers where witless ;) *SCNR*
23:08 Limbic_Region salutations all
23:08 offby1 .oO("SCNR"?)
23:08 offby1 Societé Chemins National de Railway?
23:08 moritz hi Limbic_Region ;)
23:08 moritz @google SCNR
23:08 lambdabot http://de.wikipedia.org/wiki/Liste​_der_Abk%C3%BCrzungen_(Netzjargon)
23:08 lambdabot Title: Liste der Abkürzungen (Netzjargon) - Wikipedia
23:09 offby1 aaah
23:09 moritz does lambdabot ask google germany?
23:12 renormalist Google knows who asks because they read this irc log real time and know what you want before you do :-)
23:13 moritz renormalist: that would be funny if it wasn't so close to the truth :(
23:13 renormalist only way to escape is to tell lambdabot commands in Perl6 because Google is still Perl5
23:19 justatheory joined #perl6
23:20 nicodaemos joined #perl6
23:24 renormalist moritz: collecting every public information isn't bad by itself as long as it's really public. It's just unimaginable for us poor humans that some thing really sees all the information as a whole. We are just confused because we thought "public" would mean just the daily reallive "a bit public" for the neighborhood not "that really public" as google makes it. Some day google will just explode once they forget to exclude searching in itself
23:30 ggoebel joined #perl6
23:30 [particle1 joined #perl6
23:55 REPLeffect joined #perl6

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

Perl 6 | Reference Documentation | Rakudo