Pugs 6.2.6 uploaded by cpan-upload.p6! | pugscode.org/ <Overview Journal Logs> | pugs.kwiki.org | paste at sial.org/pbot/perl6
Set by integral on 2 June 2005.
clkao ah, so chained comparison is not working yet. 00:26
arcady ?eval my $x = 3; 0 < $x < 5 00:42
evalbot6 bool::true
arcady ?eval my $x = 3; 0 < $x < 2
evalbot6 bool::true
obra ?eval 0E0 00:43
evalbot6 0
obra ?eval 0E0 ?? 'true' :: 'false';
evalbot6 'false'
obra wonders
obscurite I see TPF put up ideas for the google summer of code grants. (www.perlfoundation.org/gc/grants/proposals.html) 00:45
perl6 mason, core yaml, and perl6-rules-per-parser aren't on it 00:46
make that, perl6-rules-perl-parser
obra obscurite: ideas listed doesn't mean "we don't want other stuff" 00:47
obscurite I forget the actual, real, acronymn.
obscurite nods
Just saying.
obra "perl6 mason" is flamebait, sadly 00:48
obscurite giggles
I didn't come up with that one. 00:49
What do you think of the weblist? 00:50
obra It's a reasonable set of starter projects
I think the whole summer of code thing is a bit crazily organized
obscurite Ya, kind of a free for all. 00:51
obra They should have had it up and running by march.
clkao ?eval 1 < 10 < 3
evalbot6 bool::true
obra ?eval 1 < (10 < 3) 00:52
evalbot6 bool::false
obra ?eval (1 < (10) < 3
evalbot6 Error: unexpected end of input expecting digit, fraction, exponent, term postfix, operator, ">>=:=<<", ">>!=<<", ">>==<<", ">><=<<", ">>>=<<", ">>~~<<", ">>!~<<", ">>eq<<", ">>ne<<", ">>lt<<", ">>le<<", ">>gt<<", ">>ge<<", ">><<<", ">>><<", "\187=:=\171", "\187!=\171", "\187==\171", "\187<=\171", "\187>=\171", "\187~~\171", "\187!~\171", "\187eq\171", "\187ne\171", "\187lt\171", "\187le\171", "\187gt\171", "\187ge\171", "\187<\171", "\187>\171", "=:=",
obra ?eval (1 < 10) < 3
evalbot6 bool::true
kelan woo that ones fun
clkao actually what i want is $a < any(@foo) < $b 00:55
any more expressive way?
autrijus: can you try make chained comparison work before 6.2.7? it's crucial to me 01:12
obra for svk6? 01:13
clkao: did you submit failing tests?
clkao there's already todo test
no, for a tool for my own leisure
obra ah 01:15
SOAP::Real::6
meppl gute nacht - good night 01:22
vel hi. i've looked at the office at autrijus presentation, but forgot the link. anyone can make my evening interested? 02:59
s/interested/interesting/ 03:00
obra www.pugscode.org ?
vel no, i think it is one for YAPC NA. something on autrijus.org... 03:01
mugwump the one in the pugs svn? 03:08
vel yeah, i was wrong... i was thinking of pugscode.org/talks/apw 03:09
sorry, obra
03:21 knotty_ is now known as knotty
QtPlatypus pugs -V should tell you the version of parrot you are embedding. 03:43
05:03 khisanth_ is now known as Khisanth
revdiablo $ pugs -e 'my $foo = { bar => { baz => [ "one", "two", "three" ] } }; say ref $foo<bar><baz>' Str 05:05
shouldn't that be Array? 05:06
pasted kind of weird -- it prints Str
QtPlatypus ?eval my $foo = { bar => { baz => [ "one",two", "three" ] } }; $foo<bar><baz>.perl 05:11
evalbot6 Error: unexpected "[" expecting term
QtPlatypus ?eval [ "one",two","three" ] 05:14
evalbot6 Error: unexpected "[" expecting program
QtPlatypus ?eval ("one",two","three")
evalbot6 Error: unexpected "t" expecting term postfix, operator, ",", ":" or ")"
QtPlatypus ?eval \("one",two","three")
evalbot6 Error: unexpected "t" expecting term postfix, operator, ",", ":" or ")"
locksy ?eval my $foo = [ "one", "two", "three" ]; 05:15
?eval my $foo = [ "one", "two", "three" ];
evalbot6 ['one', 'two', 'three']
locksy ?eval my $foo = { bar => [ "one", "two", "three" ] };
evalbot6 {('bar' => 'one'), ('two' => 'three')}
QtPlatypus Thats wrong
It looks like pugs is unrolling [] into the {} 05:17
gaal hola 05:35
in Prim, op2 "Scalar as", there's an interesting idiom:
str <- fromVal x :: Eval VStr 05:36
in another place, (reverse), i see that string arguments are passed by reference. does this code auto deref somehow? (can't see how) 05:37
theorbtw1 We're off. See you from the US. 05:50
gaal have a good flight!
too late :)
hmmm, to we have Maybe-like utils for VUndef? 05:52
i want to do something with a Val, but short-circuit to Nothing if it's not defined. 05:53
batkins_ will parrot still be developed without dan leading it? 06:08
acme_ chip is leading it!
batkins_ sweet 06:10
it would be sad to see parrot go
obra chip has been leading it for several months
batkins_ is this the same chip who was writing topaz?
obra yes 06:11
QtPlatypus Is there a multyline quoting convention? 07:23
Khisanth heredocs?
QtPlatypus Is it implemented in pugs yet? 07:24
knewt_ QtPlatypus: not been implemented yet, but see the end of examples/cookbook/01strings/01-00introduction.p6 for the new heredoc syntax 07:36
QtPlatypus Thanks knewt_++ 07:37
autark-jp thought I saw someone commit a patch for simple heredocs... 07:51
07:52 Lopo_ is now known as Lopo
QtPlatypus ?eval $a = "Con" ~ "Cat"; 08:01
evalbot6 Error: Undeclared variable: "$a"
QtPlatypus ?eval my $a = "Con" ~ "Cat";
evalbot6 \'ConCat'
chip Buenos Dias 08:02
autrijus_ there's heredoc support
but not _correct_
you have to say
$a = q:to/END/
foo
END;
instead of
$a = q:to/END/;
foo
END
that actually may be construed as a feature ;)
(but it's against spec, and hence a bug)
QtPlatypus Thanks autrijus_ 08:07
Hi scook0 08:14
scook0 howdy QtPlatypus
much happening in pugs today? 08:15
QtPlatypus kisses wamrly "I love you as well. See you later" 09:24
MAV
09:27 Lopo_ is now known as Lopo
nothingmuch morning 10:56
xerox 'morning 10:59
11:18 sbkhh is now known as Odin-
svnbot6 r4501, fireartist++ | * new tests for sub signature matching 11:53
r4501, fireartist++ | * more tests for BUILD args
r4502, scook0++ | Fixed Haddock-confusing comments in Parser.hs
chip Haddock? 11:54
scook0 chip: Haddock generates documentation from annotated Haskell source code 11:56
www.haskell.org/haddock/
chip ok 11:58
Limbic_Region can now use + at work 12:15
Juerd++
svnbot6 r4503, autrijus++ | * hw2005 - review from Chip Salzenberg. 13:01
13:34 nothingmuch_ is now known as nothingmuch
PerlJam "metacomputational monads"? 13:35
xerox Eval (Eval a)
nothingmuch bah! 13:38
it turns out my hands are obstructing the antena 13:39
Aankhen`` perlbot karma Aankhen 13:41
perlbot Karma for Aankhen: 74
Aankhen`` Wow.
Hacking away at Pugs is good for your karma. :-P
nothingmuch perlbot highest karma 13:45
perlbot The top 5 karma entries: autrijus: 660, C: 475, iblech: 457, stevan: 128, gaal: 87
nothingmuch wow, C is not the highest anumore =) 13:46
Aankhen`` Heh.
?eval sub foo () returns Void { }; foo();
evalbot6 undef
Aankhen`` ?eval sub foo () is unsafe { }; foo(); 13:47
evalbot6 undef
Aankhen`` ?eval sub foo () is unsafe { 42 }; foo();
evalbot6 42
Aankhen`` ?eval $?PUGS_VERSION
evalbot6 \'Perl6 User\'s Golfing System, version 6.2.6, June 2, 2005 (r4503)'
Aankhen`` Hrm...
?eval sub foo () is unsafe returns Void { 42 } foo()
nothingmuch ?eval say "nothingmuch" ~ "++"
evalbot6 Error: unexpected "r" expecting trait or block
Error: No compatible subroutine found: "&say"
Aankhen`` ?eval sub foo () is unsafe returns Void { 42 }; foo();
evalbot6 Error: unexpected "r" expecting trait or block
Aankhen`` Heh.
Nice try.
?eval sub foo () returns Void is unsafe { 42 }; foo(); 13:48
evalbot6 42
Aankhen`` ?eval "nothingmuch" ~ "++"
evalbot6 'nothingmuch++'
Aankhen`` ?eval <1 2 3 4 5>.pick
evalbot6 '2'
Aankhen`` ?eval <1 2 3 4 5>.pick
evalbot6 '4'
Aankhen`` ?eval <nothingmuch evalbot6>.pick ~ "++"
evalbot6 'evalbot6++'
nothingmuch ?eval "nothingmuch" ~ "++"
evalbot6 'nothingmuch++'
nothingmuch ?eval "nothingmuch" ~ "-- " x 10 13:49
evalbot6 'nothingmuch-- -- -- -- -- -- -- -- -- -- '
nothingmuch beh
Aankhen`` You really want to do that? :-) 13:50
?eval 1|2|4
evalbot6 (1 | 2 | 4)
nothingmuch ?eval ("Aankhen``" ~ "++ ") x 10
evalbot6 'Aankhen``++ Aankhen``++ Aankhen``++ Aankhen``++ Aankhen``++ Aankhen``++ Aankhen``++ Aankhen``++ Aankhen``++ Aankhen``++ '
nothingmuch perlbot karma Aankhen``
perlbot Karma for Aankhen``: 19
Aankhen`` o_O
nothingmuch =) 13:51
Aankhen`` ?eval ("Aankhen``" ~ "--") x 100
evalbot6 'Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhen``--Aankhe
Aankhen`` perlbot karma Aankhen``
perlbot Karma for Aankhen``: -21
nothingmuch perlbot karma Aankhen``
perlbot Karma for Aankhen``: -21
nothingmuch tee hee
Aankhen`` Would you believe that was accidental?
nothingmuch you meant to do ++?
Aankhen`` No, I meant to do 10.
nothingmuch ah =)
Aankhen`` ?eval ("Aankhen``" ~ "++") x 21 13:52
evalbot6 'Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++'
Aankhen`` perlbot karma Aankhen``
perlbot Karma for Aankhen``: -1
Aankhen`` ?eval ("Aankhen``" ~ "++") x 10
evalbot6 'Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++Aankhen``++'
Aankhen`` perlbot karma Aankhen``
perlbot Karma for Aankhen``: 8
Aankhen`` I believe that ought to be right.
PerlJam this much bot twiddling is the beginning of the end of #perl6
Aankhen`` Such dramatics! 13:54
nothingmuch one more 13:56
Aankhen``++
that's how it was before, right?
PerlJam: how so? we're not interrupting any conversation
Aankhen`` nothingmuch >> Yeh, but I figured I'd leave one off. :-)
Thanks, though.
Oh, I heard the best quote yesterday. 13:57
Or rather, I was present when it became a quote. :-P
[01:33:03] * Topic is '<topic decent="decent" />'
[01:40:49] * BananaServ (none@r0xX0rs.your.b0xX0rs.net) has joined #TheGathering 13:58
[01:41:04] <@BananaServ> PHP topic?
nothingmuch ?
PerlJam nothingmuch: no, but what happens is that others see what you've done and try to imitate it. And then others see that and think "you can do that?" and then they try it. And so forth. The next thing you know #perl6 is just a channel for playing with the bot. 14:06
nothingmuch In that case perhaps we should start a #perl6bot channel, which is the only place evalbot can write 14:07
and under special circumstances you can ask evalbot to nopaste the output of the last command you eval'ed
Aankhen`` PerlJam >> You do realise it was inevitable? :-) 14:08
nothingmuch but i think we should do that only if/when it really is a problem
integral the easier solution it to tell perlbot and jabbot to ignore *bot (or just evalbot)
PerlJam nothingmuch: agreed. I was just dooming and glooming. I'm not into premature optmization. Address only real problems. etc. 14:09
nothingmuch ?evalbot6 "PerlJam++"
;-)
QtPlatypus ?eval perl++
evalbot6 Error: unexpected end of input expecting term
nothingmuch integral: here's a workaround
14:10 nothingmuch is now known as someguy
QtPlatypus ?eval "perl"++ 14:10
evalbot6 Error: Can't modify constant item: VStr "perl"
someguy nothingmuch++
14:10 someguy is now known as nothingmuch
Aankhen`` ?eval "perl++" 14:10
evalbot6 'perl++'
PerlJam I think QtPlatypus is trying to prove me right
nothingmuch nothingmuch--
QtPlatypus Does perl6 do the magical increment strings thing.
?eval my $a="perl";++$a 14:11
evalbot6 'perm'
PerlJam QtPlatypus: yes.
wolverian neat, eh? :)
Aankhen`` ?eval Array.perl 14:12
evalbot6 '\\{obj:Class}'
Aankhen`` ?eval ::Array.perl
evalbot6 '\\{obj:Class}'
QtPlatypus I see, a string by itself is imutable, I have to mutate a scalar verable
?eval ++13
evalbot6 Error: Can't modify constant item: VInt 13
QtPlatypus Ok that makes sence.
wolverian sense 14:14
Aankhen`` Cents!
QtPlatypus dformosa.zeta.org.au/~dformosa/Spelling.html 14:15
PerlJam btw, check out feather.perl6.nl/~duff/qotw/007e/frost.p6
pugs is *slow*
QtPlatypus hasn't expremented with pugscc "Does it give any speed up?" 14:17
QtPlatypus tries it "Not much" 14:18
integral hmm, did anyone ever get the profiling stuff working with pugs?
PerlJam I experienced something quite odd last night. I was showing someone how to do something in perl5 and I wrote perl6 code without thinking about it. perl6 is starting to get into my brain.
nothingmuch integral - IIRC pjcj tried to use hat for code coverage metrics, but hat didn't support all the stuff pugs uses 14:19
integral pointy blocks are *so* brilliant! :-)
nothingmuch: I was meaning the stuff built into GHC
to just figure out which bits of Eval.hs are expensive
nothingmuch ticky ticky metering?
integral yeah, something like the ticky ticky thingie
(of which I know little)
hmm, sloccount needs perl6 support now 8) 14:21
hmm, pugs is about 13 person-years and 1.8e6 USD
nothingmuch seen autrijus_ 14:27
jabbot nothingmuch: autrijus_ was seen 45 seconds ago
nothingmuch autrijus_: i have some harrorth to toss your way 14:28
aufrank hey all 14:45
nothingmuch hello aufrank 14:46
aufrank can anyone tell me if any of the apocalypses address how/whether perl6 will support multi-dimensional arrays?
Khisanth shape! 14:47
aufrank is that a mailing list thing or an apocalypse thing? I wanna read up :) 14:48
Aankhen`` dev.perl.org/perl6/doc/design/syn/S09.html
aufrank heh 14:49
Aankhen`` Read under "Compact arrays".
aufrank I just found a pointer to S09 in A2 ;)
thanks :)
Aankhen`` Ah well. :-)
aufrank ok, off to read docs 14:50
thanks all
nothingmuch seen Corion 15:02
jabbot nothingmuch: Corion was seen 4 days 17 hours 29 minutes 23 seconds ago
nothingmuch oh my
Shillo Hullo, all 15:09
QtPlatypus Hi 15:10
15:37 sbkhh is now known as Odin-
svnbot6 r4504, ninereasons++ | cookbook: literal whitespace in quoted strings 16:05
QtPlatypus What causes the release number to be incremented? 16:08
ninereasons QtPlatypus, do you mean "revision number" as in r4504 ? or release version number as in "Pugs 6.2.6" 16:10
QtPlatypus The r\d\d\d\d number 16:11
ninereasons that's a subversion number. the version control software takes care of that, with each submission. -- or do you mean, more technically, "by what mechanism?" (in which case, I don't know) :-) 16:12
QtPlatypus So every check in increments it? 16:15
corneli yes 16:19
every check-in, every of single files, increments the subversion revision number
s/ever of single files/even of a single file/ 16:20
nothingmuch seen lightstep 16:21
jabbot nothingmuch: lightstep was seen 2 days 7 hours 23 minutes 45 seconds ago
nothingmuch "Tedium being the stepmother of invention" 16:32
steveq Hi there 16:42
I seem to be having a problem with want
I keep getting ' No compatible subroutine found: "&List"'
is this a known problem?
PerlJam show the code. 16:43
Perhaps you want ::List
or maybe you're using List rather than "List"
or something.
steveq return want.List ?? @ret :: @ret[0];
I basically copied the test from want.t 16:44
sub obj_ok_in_list { want.List ?? 42 :: 0 }
PerlJam hrm. That's an odd syntax.
or I just haven't put on my hat of sufficient imagination yet. 16:45
steveq It seems right from SO6 but then i'm very new to this perl6 stuff 16:47
PerlJam yeah, I guess it does
But note that those tests in want.t fail :)
(at least on my pugs)
want ~~ "List" works though 16:48
as does want ~~ ::List
(or should anyway, I didn't test those exact things) 16:49
steveq right, i'll do it that way then
thanks
rjbs I was re-reading some synopsis (or other doc) recently, and saw the statement that $a .. $z were automatically available as args to a closure. It doesn't seem true, and I can't find it again.
Was I hallucinating? Was that changed? Am I doing something wrong?
(I know about placeholder args.)
PerlJam rjbs: Well, that's what $^a, $^b, etc are.
Limbic_Region wish me luck 16:50
Limbic_Region is about to send a message to the list that is likely going to be highly contraversial
rjbs Yeah, I'm sure I read something about without-^. Maybe it was a rejected RFC or something. :)
Aankhen`` rjbs >> You can use placeholders anywhere in a block and they'll be bound to the parameters passed to the block, in Unicode order.
rjbs Aankhen``: yeah, those I know. oh well! I'm happy with those, just wondering what I'd read
Aankhen`` (i.e. it's not just restricted to a..z)
PerlJam Limbic_Region: awesome!
rjbs has been finally putzing around writing perl6, and is having fun. 16:51
Aankhen`` rjbs >> Isn't it? ^_^
Limbic_Region of fuck
accidently sent out my work number
PerlJam rjbs: Me too. Earlier I commented that I was showing someone something in perl5 and inadvertently wrote perl6 code. It's starting to infect my brain :)
Limbic_Region kicks himself in the ass
rjbs anybody have a perl6.vim? :) 16:52
Limbic_Region believes there is one in Pugs rjbs
PerlJam Limbic_Region: you have an assumption that probably isn't true: "I am sure that a great deal of that unpublished work has already been decided" 16:53
rjbs Limbic_Region: Oh, I'll look!
Khisanth rjbs: there should be one in the repo
Limbic_Region perlmonks.org/index.pl?node_id=332117
grrr
rjbs there it is!
Limbic_Region svn.openfoundry.org/pugs/util/perl6.vim
Khisanth especially since so many commit messages mention it :)
Limbic_Region my copy/paste is broke
first I accidently paste my work sig
rjbs adds it to his ftdetect.
Limbic_Region now post the wrong link 16:54
Khisanth first your keyboard now your mouse
PerlJam Limbic_Region: and anyway, your "solution" is already the case. 16:55
Limbic_Region: svn.perl.org/perl6
Limbic_Region PerlJam - so where do I checkout the docs?
where do I submit patches? 16:56
PerlJam that's the svn repository of the actual documents that @Larry work from.
Limbic_Region: you can submit patches to p6l probably.
Limbic_Region that's the point PerlJam
I was playing dumb for a reason
it isn't general public knowledge
PerlJam oh, you were trying to incite discussion and dissemination? 16:57
Limbic_Region and no one, that I know of, has specifically asked for patches to help unburden the load of keeping them up to date
Limbic_Region expects the first reply to be a pointer 16:58
my reply is going to be - where do I send patches
PerlJam If you like, I could start the ball rolling :-)
er, give it the second push
Limbic_Region I am not sure 16:59
basically what I would like from @larry is buy in that they think it is a good idea
no sense in telling everyone there is a publicly available resource if they can't do anything about it 17:00
of course, if you worded your email in such a way as to say
the hard part is already taken care of <repository link>. We just need @larry to decide if they want to accept patches and decide where to submit them" 17:01
that would be ok
Limbic_Region really wants an "unsend" feature to his email 17:02
PerlJam well, what we really needs is a more pugsian effort.
Limbic_Region PerlJam - I specifically avoided Pugs in that email 17:03
but I agree 17:04
PerlJam I see that. But the same model is what would work for the design docs. Let the "general public" of the perl6 community update the docs willy nilly then have @Larry or even $Larry as release manager to periodically review the docs and post releases.
Limbic_Region trust people that show effort and a degree of sense and intervene when necessary
PerlJam - I envision the process to be more of a negotiation 17:05
PerlJam If I were more dedicated to the task, everytime I saw a "let's go with that" from $Larry on p6l, I'd post a patch that updates the docs.
Limbic_Region right - exactly what I proposed
only it wouldn't get committed until blessed 17:06
which leaves room to negotiate
ok - good idea, but let's do it this way
Jedi mind trick - let the people making decisions think it was their idea
I just see that on paper, we are really only about 1/3 of the way through the design process and I know that isn't the case 17:08
it is unfair to ask the handful of people that are already putting blood sweat and tears into it to work harder
so - offer to help while continuing to give them final authority 17:09
this by the way is a result of starting to go through the Synopses for Luke with the timeline of what's supposed to be in each one to find missing things 17:11
it was too much stuff for one person to do it alone
Aankhen`` ?eval module Foo { } 17:15
evalbot6 Error: unexpected "{" expecting trait, ";" or end of input
svnbot6 r4505, iblech++ | ext/libwww-perl -- Updated svn:ignore.
Aankhen`` iblech >> Think you could get the block-style module declarations to pass? :-) 17:16
s/pass/parse/
Limbic_Region ?eval module Foo {};
evalbot6 Error: unexpected "{" expecting trait, ";" or end of input
iblech You mean module Foo {...}?
Ah
Limbic_Region ?eval module Foo {...};
evalbot6 Error: unexpected "{" expecting trait, ";" or end of input
Aankhen`` Indeed.
?eval module Foo { sub bar () { ... } } 17:17
evalbot6 Error: unexpected "{" expecting trait, ";" or end of input
Aankhen`` ?eval sprintf("%s", "foo") 17:23
evalbot6 pugs: Printf.printf: bad argument
Aankhen`` ?eval sprintf("%3d", 1)
evalbot6 ' 1'
Aankhen`` ?eval sprintf("%c", "abcd")
evalbot6 '
Aankhen`` ?eval sprintf("%s %s", "foo", "bar")
evalbot6 pugs: Printf.printf: bad argument
Aankhen`` What am I doing wrong?
iblech Re. I'll look into module Foo {...} :)
PerlJam Aankhen``: nothing probably. It just might be that the %s specifier is sub-implemented :)
iblech Yep, it's because of Haskell
As Haskell doesn't know arbitrary number of args (out of the box), it's currently hardcoded to work with one arg or so only
Aankhen`` Ah.
iblech It's trivial to make it work with more args, I think, but it's non-trivial to generalize it
Aankhen`` ?eval sprintf("%d %d", 1, 2)
evalbot6 '1 2'
iblech According to Pugs.Prim:887, sprintf() should work with 0-3 args
Ok, then the real reason why your earlier example doesn't work is because %s isn't yet implemented, at PerlJam said
Aankhen`` Ah. 17:24
sprintf() is being reimplemented in Haskell?
s/is being/has been/ as appropriate.
iblech Pugs uses the Haskell Text.Printf module, dunno if this is a pure-Haskell implementation or if it uses the libc one. 17:26
Aankhen`` OK.
xerox Aankhen``, I know of a printf implementation...
Let me search
www.cse.unsw.edu.au/~dons/hs-plugin...de_sec_7.2 17:27
Aankhen`` Thanks. :-) 17:32
?eval sub foo () returns Str is primitive { } 17:36
evalbot6 undef
Aankhen`` ?eval sub foo () is primitive returns Str { }
evalbot6 Error: unexpected "r" expecting trait or block
clkao mmm i don't seem to be able to build PUGS_EMBED=perl5 on osx 17:37
svnbot6 r4506, Aankhen++ | * added skeleton for URI::Escape.
r4507, Aankhen++ | * added C<is export> and C<returns Foo> as appropriate to URI::Escape.
Aankhen`` ?eval class Foo { multi method prefix:<~> () { $?CLASS } }; my $foo = Foo.new(); ~$foo; 17:41
evalbot6 Error: Wrong number of invocant parameters: 0 actual, 1 expected
Aankhen`` ?eval class Foo { multi sub prefix:<~> () { $?CLASS } }; my $foo = Foo.new(); ~$foo;
evalbot6 '<obj:Foo>'
Aankhen`` ?eval class Foo { multi method prefix:<~> () { $?CLASS } }; my $foo = Foo.new(); "$foo";
evalbot6 '<obj:Foo>'
Aankhen`` ?eval class Foo { multi method prefix:<~> () { "An object of class " ~ $?CLASS } }; my $foo = Foo.new(); "$foo"; 17:42
evalbot6 '<obj:Foo>'
iblech Aankhen``: module Foo {...} is now parsed, checking in in a moment :)
Aankhen`` ?eval class Foo { multi method *prefix:<~> () { "An object of class " ~ $?CLASS } }; my $foo = Foo.new(); "$foo";
evalbot6 '<obj:Foo>'
Aankhen`` iblech++
clkao src/syck/syck.o: unknown symbol `_printf$LDBLStub'
Aankhen`` ?eval class Foo { multi sub *prefix:<~> () { "An object of class " ~ $?CLASS } }; my $foo = Foo.new(); "$foo"; 17:43
evalbot6 '<obj:Foo>'
Aankhen`` ?eval class Foo { multi sub *prefix:<~> () { "An object of class " ~ $?CLASS } }; my $foo = Foo.new(); ~$foo;
evalbot6 '<obj:Foo>'
Aankhen`` Hrm.
clkao ?eval 1 < 10 < 5
evalbot6 bool::true
iblech Pugs seems to (incorrectly) expand this to (1 < 10) && (1 < 5) 17:44
Ah, no, it parses this as (1 < 10) < 5 17:45
(1 < 10) resulting in true, which is then numified to 1, which is less than 5 17:52
Aankhen`` iblech >> Since you're on a roll, could you also make it parse C<is export(:DEFAULT :foo :bar)>? :-)
*puppy eyes*
iblech :) I'll look into it, but this one seems to be a little more tricky
Aankhen`` :-D
svnbot6 r4508, iblech++ | Pugs.Parser -- Made module Foo {...} parse.
Aankhen`` I'm just concerned about letting it atleast parse so that the declarations can be put in place, even if no action is taken.
iblech Is "is export[...]" allowed too (like it's the case with normal :foos?)
nothingmuch Limbic_Region: ping
Aankhen`` Otherwise, it'll be a pain to go back to every module and change it all once the proper behaviour is implemented.
iblech Yep, right...
Aankhen`` S11 just mentions is export(...).
Hmm.
I think once I've ported URI::Escape, I'll take a look at the diffs for your "foo is now parsed" commits.
I should learn to make them parse myself rather than keep bugging you.
iblech No worries
nothingmuch autrijus_: www.ultratechnology.com/meta.html
iblech What's allowed inside the (...)-expression? I.e., is "is export (42 + 13)" allowed (from a purely syntax point of view)?
s/export (/export(/
Aankhen`` No.
iblech Ok, that simplifies things :)
Aankhen`` One sec.
pasteling "Aankhen``" at 203.101.1.1 pasted "C<is export> (from S11)" (5 lines, 325B) at sial.org/pbot/10919 17:53
iblech Ok, I'll make it parse those 17:54
Limbic_Region nothingmuch: pong 17:55
nothingmuch how's your haskell progressing? 17:56
Aankhen`` Be back in 10 minutes.
Limbic_Region nothingmuch - I am going to spend all day Saturday playing
I think it is time to stop reading and start doing
nothingmuch that took me a long while to realize 17:57
i played with the YAHT exercises, but, ehh...
i tried doing the meet-the-monads sheep tutorial
and even reimplemented msum along the way 17:58
but I still didn't feel like i actually did something
Limbic_Region right - it is like I can follow along and I can do the things that aren't much different then what I just read 17:59
but it doesn't feel like it is a part of me
I still have to go refer to the docs for anything but the most trivial
so - just time to sink or swim
iblech Aankhen``: "is export(...)" is parsed now. 18:03
Aankhen``: Making things parse is usually just finding the appropriate rule and then adding appropriate function calls to slurp abit of the source, i.e. "optional $ verbatimParens $ many $ satisfy (/= ')')" will parse "(...)" 18:05
svnbot6 r4509, iblech++ | Pugs.Lexer -- Make "is export(...)" parse (quick hack for now).
Aankhen`` What does $ mean? 18:06
iblech ($) is just a nice way to avoid parens: You can always write "a (long expression)" as "a $ long expression" 18:07
nothingmuch damnit! i hate concrete
iblech (Its definition is "a $ b = a b" or so)
Aankhen`` Ah, so basically, what you wrote is a bunch of function calls?
iblech Yep.
nothingmuch the way it gets clean is a (b c (d e f (g h))) looks like lisp 18:08
iblech That line is equivalent to optional (verbatimParens (many (satisfy (/= ')'))))
Aankhen`` OK.
nothingmuch a $ b c $ d e f $ g h -- doesn't
not that that's clear, due to the fact that the function names are single letters
despite my bad example it does help readability 18:09
Aankhen`` BRB. 18:10
nothingmuch oof, stupid chat program 18:14
nothingmuch decides to switch to irssi 18:15
Limbic_Region PerlJam ping 18:18
steveq Is there a way to say "only test this single test file"?
iblech steveq: There is: prove t/.../.t 18:20
You may need to set an environment var before running prove, though -- PERL_HARNESS IIRC
PERL_HARNESS=/path/to/pugs prove t/.../...t
steveq aha
Not seen that before
iblech And ./pugs t/.../.t will work, too, given a correct %*ENV<PERL6LIB> 18:21
steveq yeah, I couldn't work out te correct PERL6LIB 18:22
iblech make install or PERL6LIB=/pugs_sourcedir/blib6/lib 18:23
steveq cheers
Reading the tests is an excellent way to learn perl6 I find 18:25
iblech Yep :) 18:27
t/++
18:27 Aankh|Clone is now known as Aankhen``
ninereasons In perl 5, if I say my @ar = ('x') x 10; # then @ar is defined as containing 10 elems of 'x' 18:32
what about in perl 6?
?eval my @ar = ('x') x 10; @ar;
evalbot6 ['xxxxxxxxxx']
ninereasons is that a bug?
or is the old syntax replaced with the xx operator? 18:33
vcv- yep
ninereasons yep replaced?
vcv- x = scalar xx = list
?eval my @ar = 'x' xx 10; @ar;
evalbot6 ['x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x', 'x']
ninereasons ok, so x is now more strict ? 18:34
PerlJam "more strict"?
ninereasons should have used a different character than 'x' for the example
PerlJam The special case for LIST x EXPR was removed is all. 18:35
ninereasons PerlJam, that's what I mean. no special case for LIST
vel hi. anyone knows what's going on with www.jwcs.net/~jonathan/perl6/ (windows pugs repository for binaries) 18:40
svnbot6 r4510, Aankhen++ | * HTTP::Message -- added C<is copy> to string parameters, added multi subs handling C<Str $unsafe>.
Limbic_Region vel - apparently it is an apache misconfiguration problem 18:41
Aankhen`` ?eval true 18:42
evalbot6 Error: unexpected end of input expecting term
iblech ?eval bool::true
evalbot6 bool::true
Aankhen`` ?eval bool::true
evalbot6 bool::true
vel 10Limbic_Region: thanks. i am sure i downloaded snapshot from there few days ago...01 18:43
Limbic_Region as did I 18:44
PerlJam vel: and the apache config couldn't have changed in the mean time?
Limbic_Region seen Jonathan
jabbot Limbic_Region: Jonathan was seen 2 days 7 hours 59 minutes 12 seconds ago
Limbic_Region seen Jonathan_
jabbot Limbic_Region: Jonathan_ was seen 10 days 18 hours 28 minutes 42 seconds ago
Aankhen`` ?eval my %foo; for (my $i = 0; $i < 5; $i++) { state %foo = (foo => "bar"); } %foo.perl 18:45
evalbot6 Error: unexpected "f" expecting ";", statements or end of input reserved word
Aankhen`` ?eval my %foo; loop (my $i = 0; $i < 5; $i++) { state %foo = (foo => "bar"); } %foo.perl
evalbot6 Error: No compatible subroutine found: "&my"
Limbic_Region PerlJam - I posted an empty framework for S17 and asked if it was an acceptable patch
Aankhen`` ?eval my (%foo, $i); loop ($i = 0; $i < 5; $i++) { state %foo = (foo => "bar"); } %foo.perl
evalbot6 '{}'
PerlJam Limbic_Region: yeah, and I saw your PM post too. Limbic_Region++
iblech Aankhen``: The inner state %foo hides the outer my %foo
Aankhen`` iblech >> Yeh, I thought it would... I just figured it wouldn't hurt to try. ;-) 18:46
Does C<FIRST { ... }> work?
iblech ?eval sub abc { my $x; FIRST { $x++ }; $x } abc(); abc(); abc()
evalbot6 \undef
iblech Hm.. bad test 18:47
Aankhen`` ?eval sub abc { my $x; FIRST { $x++ }; $x } (abc(), abc(), abc())
evalbot6 (\1, \undef, \undef)
Limbic_Region best way to learn is to RTFM right? What do you do if TFM is missing information - find out what it is an fix it
iblech So FIRST {...} works, as should INIT and CHECK 18:48
Aankhen`` OK.
Could anyone explain to me just when CHECK { ... } is called, and what it is used for? 18:49
Hmm, wait.
I just realised I was looking at the S instead of the A.
Aankhen`` goes to read the A.
Limbic_Region Aankhen - if the A disagrees with the S then trust the S 18:50
Aankhen`` Yeh, but I figure the A would explain it, the S would just offer a brief summary. :-)
iblech The order is: BEGIN (at compile time) - CHECK (compile time, too, but ALAP) - INIT (run time, ASAP) - ... - END (run time, ALAP)
Aankhen`` I saw the order later, I couldn't understand what it's for. 18:51
Limbic_Region "The synopses are now in the process of becoming our design documents, so you'll find a lot of details here that aren't mentioned elsewhere. If there's a conflict between any other documents and these documents, believe these documents (unless the other document is email from one of us, of course, in which case you should yell at us for not keeping these up to date.)"
So when we say "S5", we mean "spec" as well as "synopsis".
So really, the S no longer means synopsis - it means Spec
s/longer means/longer just means/
Aankhen`` OK.
So, you wanna tell me what CHECK is all about?
Is it only a question of being called at a particular time? 18:52
s/time/point/
Limbic_Region Aankhen - in p5 it had specific meaning - I haven't familiarized myself with p6
Aankhen`` OK. 18:53
Limbic_Region the rule of thumb though is - if it isn't spelled out differently assume p5
Aankhen`` goes to CHECK the Perl 5 meaning.
iblech :d
Limbic_Region AES18 isn't written yet is it?
nor is 21 18:54
iblech Right, both not yet written 18:55
Limbic_Region I am hoping that if submitting patches for outlines is acceptable
filling in the blanks will come faster because we will be able to visually see what is missing
we don't have to hop around randomly 18:56
Aankhen`` I feel sorry for anyone who bought P6E...
Limbic_Region oh, that was addressed way back when and $larry had this to say
Aankhen`` - why - it was just as much about parrot as it was p6
Aankhen`` But both evolve so fast... 18:57
Limbic_Region probably more so about Parrot
I bought it (and read some of it) - I don't feel bad about it
Aankhen`` You poor thing. :-( 18:58
Aankhen`` ducks.
Limbic_Region If only I were as smart as the people around me seem to think I am 18:59
I would be able to do more to help the project then just advocacy and a few tests 19:00
Aankhen`` - reading the book, even if 30% of it is no longer applicable is still a heck of a lot better than trying to catch up on 5 years of posts on the list
Aankhen`` Ahh, lighten up. :-)
Limbic_Region was mixing threads of thought 19:01
only 1 of those lines was intended for you
Aankhen`` I know.
Aankhen`` is just fooling around anyway. 19:02
Limbic_Region yeah, Perl is sexy that way
Aankhen`` installs the Microsoft Platform SDK.
BRB. 19:03
svnbot6 r4511, Aankhen++ | * implemented &URI::Escape::uri_escape. 19:19
r4511, Aankhen++ | * don't need %subst in URI::Escape.
r4512, iblech++ | docs/talks/README -- Added link to the HTML version of autrijus++' APW slides.
iblech Aankhen``: Just FYI, as I see you don't stop porting/creating URI/CGI-related modules :) -- I've already ported URI (with most of its subclasses) at tpe.freepan.org/repos/iblech/URI/lib/URI 19:23
Aankhen``++
Aankhen`` dies. 19:26
nothingmuch dances on Aankhen``'s grave 19:27
iblech (The subclasses are mostly only ten lines or so long, mostly protocol handlers, e.g. URI::http, URI::ftp, etc.) 19:29
iblech waits for Aankhen`` to resurrect :)
19:36 khisanth_ is now known as Khisanth
sri_ is glob implemented now? 19:37
Aankhen`` iblech >> Thank you for reminding me. 19:40
Aankhen`` wonders how many times he must be told that URI has been ported before it finally sinks in. 19:41
Hmmm... how can we integrate that with Pugs?
PerlJam sri_: I don't think so.
sri_ then i'll commit something based on File::Find and Text::Glob 19:42
Aankhen`` pokes iblech. 19:43
iblech Aankhen``: Yep?
Aankhen`` How can your URI distro be integrated with Pugs?
iblech Aankhen``: As I think some idioms don't work yet in Pugs -- simply continue to port URI in ext/, and have an occasional look at my port 19:46
Aankhen``: Or you might want to copy a module.pm and modify it so it works in Pugs
Aankhen`` Alright. 19:48
iblech Aankhen``: I.e. you can orientate yourself at my port
nothingmuch sial.org/pbot/10925
Aankhen`` I was just looking at it from the point of view of keeping both of them synchronised.
Aankhen`` wonders how you get VS.NET to alter the PATH when building a project... 19:49
Oh. 19:50
I just figured it out.
I love me.
iblech No need to keep my FreePAN port synchronised -- that's only a burden for you, I think. We can simply remove my port when your one in ext/ is finished
(I think)
Aankhen`` Oh $#@! 19:52
.
iblech nothingmuch: Nice! :) 19:55
nothingmuch appearantly "here cell allot" is spelled ",", but that doesn't sound as dramatic =/ 19:57
svnbot6 r4513, iblech++ | lib/Perl6/Pugs.pm -- More documentation
Aankhen`` ?eval 128.as("%%%02X") 20:01
evalbot6 pugs: Printf.printf: bad formatting char X
Aankhen`` ?eval 128.as("%%%02x")
evalbot6 pugs: Printf.printf: bad argument
Aankhen`` ?eval 128.as("%%%02d")
evalbot6 pugs: Printf.printf: bad argument
Aankhen`` Bah.
iblech ?eval sprintf "%%%02X", 128 20:02
evalbot6 pugs: Printf.printf: bad formatting char X
iblech ?eval sprintf "%%%02x", 128
evalbot6 '%80'
Aankhen`` iblech >> You seem to have a P5 style s/// in &uri_unescape. 20:10
autrijus_ my laptop's motherboard just died on me (!) 20:18
svnbot6 r4514, Aankhen++ | * EOL at EOF.
r4514, Aankhen++ | * return 1!
r4515, Aankhen++ | - changed C<sprintf> to C<.as>.
Limbic_Region Aankhen`` - does that mean you can say say $str.as('%.2f'); # ??? 20:20
Aankhen`` Apparently. 20:21
?eval 123.as('%04d')
evalbot6 pugs: Printf.printf: bad argument
Aankhen`` But it doesn't seem to work in Pugs.
I got it from iblech's URI::Escape.
Come to think of it, I remember reading about .as. 20:22
autrijus_ .as works only for %s yet
Aankhen`` OK.
Limbic_Region ?eval my $str = '123'; $str.as('%.2d')
evalbot6 pugs: Printf.printf: bad argument
autrijus_ hopes the laptop doesn't burn up
Aankhen`` ?eval 123.as('%s')
evalbot6 '123'
Limbic_Region hey autrijus - how are things?
Limbic_Region has been throwing rocks at hornet nests 20:23
Aankhen`` hopes it doesn't burn up either.
autrijus_ <- in the same room as jens, Boogie__, chip, leo
just got the network up
took some work
svnbot6 r4516, Aankhen++ | * &URI::Escape::uri_escape (::Str, ::Str, ::Bool) -- changed $pattern to a regular string.
r4516, Aankhen++ | * &URI::Escape::uri_escape (::Str) -- added C<return>.
r4516, Aankhen++ | * implemented &URI::Escape::fail_hi.
autrijus_ as it's late here we're just going to hold a planning meeting
for the things we can do tomorrow
Limbic_Region autrijus_ - with the help of tall_man, I have found a bug in pugs that is going to have significant impact 20:24
autrijus_ Limbic_Region: good!
Limbic_Region: pray tell
Limbic_Region the bug itself is probably trivial, but the fact there are a good portion of tests that rely on the bad/incorrect behavior - they all need to change 20:25
autrijus_ so I'm pushing for perl5 embed support (as opposed to perl 5 hosting support as ponie)
Limbic_Region: sure. what was the bug?
perl 5 embed support in ponie that is
Limbic_Region ?eval sub bar ( +$foo ) { $foo } bar( foo => 3);
autrijus_ er
evalbot6 \3
autrijus_ perl 5 embed support in parrot that is
Limbic_Region works as expected right
autrijus_ right
Limbic_Region ?eval sub bar ( +$foo ) { $foo } bar( 3);
evalbot6 \3
autrijus_ was about to ask what's the problem with that
Limbic_Region also works as expected???? wrong
autrijus_ it's wrong.
named bindings shouldn't take positional arguments. 20:26
it's a regression
Limbic_Region named parameters must be named
autrijus_ clkao reported that chained comparisons cease to work either
6 > 5 > 4
is no longer true
both worked a few releases back :-/
Limbic_Region but there are apparently a signficant number of tests depending on that incorrect behavior
autrijus_ hates regressions
Limbic_Region: depending how?
Limbic_Region meaning the signature is named but they are passed as positional 20:27
autrijus_ nod... so we need to 1)fix the bug
and 2)unbreak the tests
Limbic_Region and see which previously working tests start breaking
autrijus_ yup. 20:28
obra autrijus_: should 0E0 be 0 but true?
Limbic_Region tall_man has identified two bugs now - really would like to convince him to join the madness
autrijus_ obra: no.
obra :(
svnbot6 r4517, Aankhen++ | - changed C<is export> to C<is export(:DEFAULT)>. iblech++
autrijus_ Limbic_Region: I'll make you a metacommitter and you can send invitations 20:29
Aankhen`` ?eval sub foo (Array of Str *@foo) { gather { for @foo -> $bar { take $bar.uc() } } } foo("foo", "bar", "baz")
evalbot6 Error: unexpected "o" expecting "?", "*", "+", "++" or parameter name
Aankhen`` ?eval sub foo (*@foo) { gather { for @foo -> $bar { take $bar.uc() } } } foo("foo", "bar", "baz")
evalbot6 ('FOO', 'BAR', 'BAZ')
revdiablo ah, that reminds me, pugs -e 'my $foo = { bar => { baz => (<one two three>) } }; say ref $foo<bar><baz>' # print Str, shouldn't it print Array?
maybe my pugs is out of date
Limbic_Region autrijus_ - I don't think that would work in this case - he seems unconvinced about having enough free time
Aankhen`` ?eval sub foo (*@foo of Str) { gather { for @foo -> $bar { take $bar.uc() } } } foo("foo", "bar", "baz")
evalbot6 Error: unexpected "o" expecting trait, ",", ":" or ")"
Aankhen`` sighs.
?eval my $foo = { bar => { baz => (<one two three>) } }; ref $foo<bar><baz> 20:30
evalbot6 ::Str
autrijus_ Limbic_Region: the trick is to send the invitation
and it's up to the other party to take it
revdiablo Aankhen``: ah, you beat me to it
Aankhen`` ^_^
Limbic_Region well - if I find his email address I will ask
revdiablo I can add a test somewhere if there isn't already one
?eval my $foo = { bar => { baz => [<one two three>] } }; ref $foo<bar><baz> 20:31
evalbot6 ::Str
Limbic_Region ok - time for weekend
revdiablo that's actually what I meant
Limbic_Region TTFN all
revdiablo cya Limbic_Region
Aankhen`` ?eval my $foo = { bar => [<one two three>] }; $foo.ref;
evalbot6 ::Hash
Aankhen`` ?eval my $foo = { bar => [<one two three>] }; $foo<bar>.ref;
evalbot6 ::Str
Aankhen`` ?eval my $foo = { bar => [<one two three>] }; $foo<bar>[0].ref; 20:32
evalbot6 ::Str
Aankhen`` ?eval my $foo = { bar => [1, <one two three>] }; $foo<bar>.ref;
evalbot6 ::Int
Aankhen`` Seems to be taking on the ref of the first element.
?eval my $foo = { bar => [1] }; $foo<bar>.ref;
evalbot6 ::Int
Aankhen`` ?eval my $foo = { bar => (1) }; $foo<bar>.ref; 20:33
evalbot6 ::Int
gaal autrijus, hi, i need some help with Reader
revdiablo ?eval my $foo = [ 'bar', [1, 2, 3] ]; $foo[1].ref
evalbot6 ::Array
revdiablo so it works in an arrayref, but not in a hashref?
gaal have been reworking caller per larry's notes 20:34
svnbot6 r4518, Aankhen++ | * fixed signature for &URI::Escape::uri_unescape.
gaal but now i'm stuck with the monad part (which used to work)
Aankhen`` ?eval my $foo = [ <one two three> ]; $foo.ref
evalbot6 ::Array
Aankhen`` ?eval my $foo = [ <one two three> ]; $foo ~~ Ref 20:35
evalbot6 bool::false
pasteling "gaal" at 192.115.25.249 pasted "caller WIP" (32 lines, 1.5K) at sial.org/pbot/10926
Aankhen`` ?eval my $foo = <one two three>; $foo ~~ Ref
evalbot6 bool::false
autrijus_ gaal: what was the error?
gaal it's a type error on return $ Just caller ++ (local (const caller) (fetchCaller))
i know why it's an error
revdiablo t/data_types/mixed_multi_dimensional.t looks to be the place
gaal because local returns m a 20:36
but i don't know how (if) my approach can be fixed
what i'm trying to do is make fetchCaller a generator of deeper and deeper stack frames
autrijus_ foo <- local (const caller) fetchCaller 20:37
return $ Just (caller ++ foo)
gaal what's foo's type? 20:38
nothingmuch autrijus_: did you see the link I sent you?
autrijus_ nothingmuch: no
revdiablo my $h = { a => [ 1,2,3 ] };
is($h<a>.ref, 'Array', "array nested in hashref in one declaration");
does that seem sane?
autrijus_ gaal: "env"?
Env that is
you can't ++ it though
nothingmuch www.ultratechnology.com/meta.html
I'm writing a bare metal forth chapter 20:39
svnbot6 r4519, iblech++ | Added a test for sprintf() and as() (t/builtins/sprintf_and_as.t).
Aankhen`` Is there any clear disadvantage to using MMD?
gaal another thing i tried was to use repeat, outside of fetCaller
Aankhen`` Does it result in a large slowdown?
gaal return . repeat fetchCaller
revdiablo maybe I should :todo<bug>
gaal that didn't work either but i forget why 20:40
Aankhen`` I've been splitting subs into multi subs like crazy... dunno whether that's a bad idea.
arcady probably not a bad idea 20:41
makes it more functional-like and haskelly
nothingmuch arcady: you are probably also interested in the link
arcady huh? 20:42
nothingmuch www.ultratechnology.com/meta.html
Aankhen``: i think multisubs should be the default
read docs/mmd.kwid
and please finalize it if you can, I don't know what exact semantics I would really like 20:43
if/when you finish, post the proposal to p6l =)
autrijus_ jens++ 20:48
svnbot6 r4520, Aankhen++ | * whoops... apparently &URI::Escape::uri_unescape did need that signature; modified it to use MMD.
r4521, iblech++ | examples/ -- Moved some examples to new examples/algorithms/.
r4522, revdiablo++ | added test for HoA declared in one statement
autrijus_ # promise to help me out to get hw2005 in TeX
Aankhen`` ?eval :foo<bar> 20:51
evalbot6 ('foo' => 'bar')
Aankhen`` neatly sidesteps nothingmuch's blame transfer.
nothingmuch Aankhen``: it's not a blame transfer, i just thought you might be interested
Aankhen`` nothingmuch >> Seriously, though, I don't know nearly enough about SMD and MMD, and the relative merits and demerits of each.
nothingmuch nor do i =) 20:52
Aankhen`` Heh.
nothingmuch Class::Events is just an experiment that made it seem like MMD is the way to go
and i tried to think, and it looked like SMD does not conflict with MMD everywhere
but we do get basic pattern matching everywhere
it is very post object oriented
gaal ok, i want to simplify this. env has an envCaller, which i want to locally "be in". can i turn that into a list at all? 20:53
autrijus_ a list?
gaal i wanted to do this:
formatFrame $ last $ take skip $ filter wantFrame $ repeat $ fetchCaller 20:54
more or less
this uses fetchCaller as a generator of
er
monadic inspiration
and filters the first frame based on frame type (wantFrame)
Boogie___ feather.perl6.nl/~boogie/flipchart.html
gaal and takes the correct one (last $ take skip) 20:55
however, these envs aren't really values to be passed around, are they?
autrijus_ they are
you can just take them out
and cast them to Val even
if you want to do that
VControl (ControlEnv env) :: Val
gaal i think it makes it simpler in this case? 20:56
autrijus_ gaal: but sorry, I'm trying to follow chip/leo on the STM-on-parrot discussion here
gaal: that's possible
gaal okay, thanks muchly for so the help so far.
Boogie___ feather.perl6.nl/~boogie/flipchart.html <- so this is what behind Chip currently on the flipchart. ;) 20:57
Aankhen`` ?eval sub foo ({ $bar, $baz, *%quux }) { ... } 20:58
evalbot6 Error: unexpected "{" expecting formal parameter, ":" or ")"
svnbot6 r4523, Aankhen++ | * moved ext/URI-Escape/ to ext/URI/.
Aankhen`` G'night. 21:00
nothingmuch blah
autrijus_ heya nothingmuch. 21:04
nothingmuch: we're figuring out how to add STM to parrot
(with chip and leo)
nothingmuch hi ho
autrijus_: #parrot or IRL?
how does STM get along with JIT? 21:05
Boogie___ IRL, they are together at Leo's place. :)
nothingmuch ah
btw, will any(chip|leo) be at the hackathon? 21:06
autrijus_ nothingmuch: IRL 21:07
nothingmuch: STM is just a way for a SharedFoo PMC todo bookkeeping
so has nothing really to do with JIT
nothingmuch doesn't STM require boxed types to stay sane?
autrijus_ nothingmuch: PMCs are all boxed. 21:08
nothingmuch: even in JIT mode.
nothingmuch and isn't JIT's purpose to do away with boxed types were possible, to gain (hardware dependant) performance?
my $i = 1;
gaal k, i'm passing maybe envs now; how do i enter Reader (?) so that when i do have the Env i want, when i call readVar the lookup happens in the correct context?
nothingmuch what is that compiled down to? a PMC?
I thought Integer, or whatever the builtin type is
autrijus_ nothingmuch: the JIT won't have effects on PMCs 21:09
let alone shared PMCs
nothingmuch ah
arcady I think you have to say "my int $i = 1" to actualy get a native integer 21:10
gaal ahhh, with local?
nothingmuch oh, btw, ask chip if he's seen the bottom half of mmd.kwid
arcady: why not wait with that till we find out it can't be an 'int'?
like when it's shared, or has a method called on it, or has a role attached 21:12
domm autrijus_: et.al.: so you arrived well at Leos place?
Boogie___ domm: well
domm Boogie___: thanks for the APW report
I have slight problems understanding it, though :-)
autrijus_ APW report? 21:13
domm weblabor.hu/hirek/20050609/austrianperlworkshop
Boogie___ domm: wirelesses are on, so we're happy
domm: currently they are planning what they will talk about
domm: and about other topics - cool topics ;) 21:14
doom: well, Hungarian is not easy to understand. Actually, nothing special, it's just a blog.
domm I'm very glad that I finally managed to contribute something to Perl6/Parrot/Pugs
Even if it's not code, but (only) bringing three people together to (hopefully) produce a lot of code (or specs) next week 21:15
(in fact, mor than tree people, how many are staying at Leos place? 4? 5? 21:16
Boogie___ As I see, it will be very-very-very useful. I've heard some "Ahh, now I got the idea" even during travelling here. :)
domm cool!
Boogie____ And as I heard, somebody will join tomorrow 21:19
Unfortunatly, I will leave tomorrow night. :(
nothingmuch showers 21:21
domm nothingmuch: while beeing on irc?
nothingmuch domm: no, only autrijus_ can do that
well, i guess i can do it
domm :-)
nothingmuch but i'm not sure I want to =)
domm goes to bed # workshop organising /is/ exhausting 21:25
Boogie____ feather.perl6.nl/~boogie/flipchart.html update
clkao ?eval 3 < 10 < 6 22:46
evalbot6 bool::true
clkao slackers :p
22:52 khisanth_ is now known as Khisanth

Powered by Cro and the Raku® Programming Language.

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

Thank you!