»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org | UTF-8 is our friend!
Set by sorear on 25 June 2013.
00:04 hoverboard left 00:10 salv0 left 00:13 logie_ joined, rurban joined 00:18 rurban left 00:20 dayangkun joined 00:27 salv0 joined 00:28 logie_ left 00:29 gdey_ left 00:40 kbaker joined 00:51 ivan`` joined
TimToady rosettacode.org/wiki/Handle_a_signal#Perl_6 :-) 00:58
01:03 klapperl_ joined 01:06 klapperl left 01:12 logie_ joined 01:14 rurban joined, btyler joined 01:18 gdey joined 01:19 rurban left 01:24 logie_ left 01:28 Sqirrel left 01:29 Sqirrel joined 01:33 gdey left 01:39 rurban joined 01:43 atroxaper joined 01:49 atroxaper left 01:50 hoverboard joined
timotimo apparently not sleeping :| 01:51
that code is pretty. but it could just as well have used supply.zip to get the numbers to "line up" with an interval, right? 01:52
01:54 atroxaper joined
timotimo i don't actually know if there's an operation to turn a lazy list into a supply 02:02
02:05 atroxaper left 02:08 WJB joined 02:17 Alula_ left 02:18 Alula_ joined 02:25 kbaker left 02:26 gdey joined, gdey left
avuserow r: say Buf.new eq "" 02:31
camelia rakudo-{jvm,moar} cc017e: OUTPUT«(timeout)» 02:32
..rakudo-parrot cc017e: OUTPUT«maximum recursion depth exceeded␤current instr.: 'print_exception' pc 145552 (src/gen/p-CORE.setting.pir:61201) (gen/parrot/CORE.setting:11309)␤called from Sub 'Stringy' pc 223521 (src/gen/p-CORE.setting.pir:91883) (gen/parrot/CORE.setting:1044)␤ca…»
avuserow n: say Buf.new eq "" 02:33
camelia niecza v24-109-g48a8de3: OUTPUT«False␤»
timotimo Buf.new.Stringy returns a buf 02:34
so eq trying to coerce Buf into a str and call eq on the result will call the very same eq candidate
yeah, the Stringy method of Buf just returns self 02:35
02:35 xragnar_ joined, xragnar left, xragnar_ is now known as xragnar
timotimo though technically ... a Buf is Stringy already, because it "does Stringy" 02:35
hrm.
avuserow thanks for looking at it timotimo++ :) 02:36
I don't need that to work, I just accidentally hit it with Test's "is" function
timotimo what's the right solution then? :\
it could give the same error as this: 02:37
r: say Buf.new(25, 26, 27, 28).Str
camelia rakudo-moar cc017e: OUTPUT«Cannot use a Buf as a string, but you called the Str method on it␤ in method Str at src/gen/m-CORE.setting:5291␤ in block at /tmp/tmpfile:1␤␤»
..rakudo-jvm cc017e: OUTPUT«Cannot use a Buf as a string, but you called the Str method on it␤ in method Str at gen/jvm/CORE.setting:5291␤ in block at /tmp/tmpfile:1␤␤»
..rakudo-parrot cc017e: OUTPUT«Cannot use a Buf as a string, but you called the Str method on it␤ in method Str at gen/parrot/CORE.setting:5295␤ in method Str at gen/parrot/CORE.setting:1038␤ in block at /tmp/tmpfile:1␤␤»
avuserow m: say Buf.new eq Buf.new # is this also bad?
camelia rakudo-moar cc017e: OUTPUT«True␤»
avuserow huh.
02:38 Sqirrel left
BenGoldberg I think you have to call .decode(something) to turn a Buf into a Str 02:38
timotimo oh
Buf.pm also defines eq for Buf to B\=
Buf*
avuserow yeah, that makes sense, cause that's similar to comparing arrays 02:39
so yeah maybe comparing Bufs and most things with eq is wrong... 02:40
timotimo okay, so a Buf + Any and an Any + Buf eq candidate may be acceptable 02:41
just to throw an exception rather than infinilooping on coercion 02:42
avuserow sounds good to me
02:47 BenGoldberg left 02:48 raiph left
timotimo if you want to, you can patch it and propose it as a pull request :) 02:49
i'd like to get comments from people "higher up" before merging it, though
02:50 thou left
avuserow maybe, we'll see. I'm hacking on a few modules at the moment 02:52
timotimo (if you patch that, also add candidates for ne)
02:58 WJB left 03:12 raiph joined 03:22 SamuraiJack_ joined
lue (The best solution would be to not declare Buf as doing Stringy ...) 03:37
03:40 cxreg joined 03:43 kaare_ joined 03:54 btyler left, thou joined 03:58 slavik joined 03:59 thou left 04:03 todell is now known as segomos, segomos left, segomos joined 04:11 plobsing joined
segomos i dunno i can't remember ever missing or wanting autocomplete 04:16
04:18 anaeem1_ joined 04:23 anaeem___ joined 04:27 anaeem1_ left 04:31 wbill joined
dalek osystem: 0bfe389 | segomos++ | META.list:
Update META.list
 04:32
moritz m: say 42 04:34
camelia rakudo-moar cc017e: OUTPUT«42␤»
04:34 Psyche^ left
segomos m: 42.say 04:34
camelia rakudo-moar cc017e: OUTPUT«42␤»
moritz was just interested in the rakudo revision
04:36 Psyche^ joined
raiph m: say 42.WHY # to entertain segomos 04:38
camelia rakudo-moar cc017e: OUTPUT«Life, the Universe and Everything␤»
04:38 kaare_ left
segomos lol, nice 04:38
i'm enjoying the shit out of perl6, just an fyi for you folks developing it 04:39
04:48 woosley left
avuserow segomos++ # adding more modules 04:48
raiph segomos: you've done more p6 dev than i have :)
segomos god, what a great next version from p5. the module i just submitted is kind of a hack right now :-) i don't the proper way to do sgml 04:52
raiph just started a grammar to parse www.cpan.org/MIRRORED.BY 04:53
04:58 xenoterracide_ left
segomos what should it produce? i've been playing around a lot with grammars 04:58
raiph Just a simple parse tree would do as a start 04:59
m: grammar G { rule TOP { <foo>* }; rule foo { \N* \n }; }; say G.parse("xxx\nyyy\n")
camelia rakudo-moar cc017e: OUTPUT«(Any)␤»
segomos what is it for? 05:00
raiph FROGGS needs to pull something out of it for his CPAN integration work 05:03
I (perhaps foolishly) thought I would try build a P6 grammar for it without worrying about what data/data structure is actually needed for his current use case 05:05
05:06 woosley joined
raiph segomos: why do I get the (Any) response to the evalbot line I just did above? 05:07
05:07 woosley left
segomos m: grammar G { rule TOP { <foo>* }; rule foo { \N* \n }; }; say G.parse("xxx\nyyy\n").say; 05:09
camelia rakudo-moar cc017e: OUTPUT«(Any)␤True␤»
05:11 woosley joined, woosley left
raiph m: say Grammar.WHAT 05:12
camelia rakudo-moar cc017e: OUTPUT«(Grammar)␤»
avuserow "say G.parse("xxx\nyyy\n").say" ?
raiph m: say Any
camelia rakudo-moar cc017e: OUTPUT«(Any)␤»
05:13 kaare_ joined
segomos it returns Any because it doesn't parse 05:13
m:grammar G { rule TOP { <foo>* }; regex foo { <data> \n }; regex data { \N* }; }; say G.parse("xxx\nyyy\n").say 05:14
m: grammar G { rule TOP { <foo>* }; regex foo { <data> \n }; regex data { \N* }; }; say G.parse("xxx\nyyy\n").say
camelia rakudo-moar cc017e: OUTPUT«｢xxx␤yyy␤｣␤ foo => ｢xxx␤｣␤ data => ｢xxx｣␤ foo => ｢yyy␤｣␤ data => ｢yyy｣␤␤True␤»
raiph segomos: thanks. why does regex work and not rule? isn't the difference that rule does backtracking and sigspace? 05:17
05:19 wbill left, denis_boyun_ joined
segomos i believe, but am probably wrong, that doing <data> \n requires look ahead 05:20
05:20 woosley joined
raiph huh. looks like i got it the wrong way round -- its regex that backtracks, which is obvious in retrospect 05:22
ie regex works like p5 regex in that regard whereas rule ratchets 05:23
TimToady it's not the backtracking, it's that rule will match whitespace before your \n
and eat the newline, so \n can' tmatch
segomos TimToady++
05:23 woosley left
TimToady rule foo { <data>\n } would work 05:23
(as long as you don't also make data a rule) 05:24
segomos indeed it does
TimToady in which case teh space after \N* would also eat <ws>
raiph TimToady++, segomos++ # thanks 05:25
segomos raiph: use Grammar::Trace..it has helped me quite a bit 05:26
05:26 xinming__ joined
segomos grammar::Tracer 05:26
raiph I tried both that and the full debugger; tools can help but can't stop me being me :) 05:27
05:29 xinming_ left 05:32 darutoko joined 05:35 xinming_ joined 05:37 xinming__ left 05:43 thou joined 05:47 SHODAN left, thou left 05:49 SHODAN joined 05:52 btyler joined 05:54 rurban left 05:55 btyler left, xinming__ joined 05:58 xinming_ left 06:05 kaleem joined 06:08 denis_boyun_ left, IllvilJa left 06:15 xinming__ left 06:16 xinming_ joined, IllvilJa joined 06:21 hummeleB1 joined
sergot morning o/ 06:39
masak morning, sergot, #perl6 06:46
06:47 denis_boyun_ joined 06:48 cooper joined
sergot hi masak :) 06:48
06:59 FROGGS joined
moritz \o 07:03
07:05 hummeleB1 left
dalek osystem: b4a9da0 | Adrian++ | META.list:
Add Compress::Snappy
 07:05
osystem: 9a272fa | moritz++ | META.list:
Merge pull request #29 from avuserow/patch-2

Add Compress::Snappy
07:05 hummeleB1 joined, zakharyas joined
FROGGS morning 07:08
07:10 hummeleB1 left, hummeleB1 joined 07:11 denis_boyun_ left, hummeleB1 left 07:13 Ven joined
dalek kudo/nom: 280736a | (Elizabeth Mattijsen)++ | src/core/signals.pm:
Supply.merge(@s) reverts to noop on 1 Supply
 07:28
07:28 dmol joined 07:31 thou joined
lizmat and good *, #perl6! 07:31
lizmat still feels signal() should be Supply.signal(), as it returns a Supply 07:33
07:35 sqirrel joined, thou left
Ven o/, lizmat 07:37
dalek kudo/nom: 383ba3b | (Elizabeth Mattijsen)++ | src/core/signals.pm:
Further simplification of signal()
arnsholt o/ 07:38
07:48 fhelmberger joined 07:51 kivutar joined, hoverboard left
jnthn lizmat: And file watchers too? And sockets? And async file operations? 07:52
Just because something returns a Supply doesn't mean it should have a method in Supply...
dir returns a list of files, but we don't write List.dir :) 07:53
07:53 btyler joined
Ven [\op] gets a lazy reduction ? nice 07:56
FROGGS fwiw, I do not really like constructor methods other than .new
because these usually do not mix well with methods that are not constructors 07:57
in SDL (P5 module) we ended up with ->from_surface or ->video_surface which kinda implies being a constructor 07:58
but only kinda
07:58 btyler left
Ven if you want to have a `is cached` sub, can you use it on the proto ? or do you have to is cached on every multi sub? 08:02
m: my @a = (0..5).roll(5) xx 5; say @a[*]>>[0]; 08:04
camelia rakudo-moar cc017e: OUTPUT«4 1 3 2 3 0 4 4 1 3 3 4 3 1 5 4 3 2 0 5 4 0 2 2 3␤»
Ven m: my @a = (0..5).roll(5) xx 5; say @a[*] eqv @a; 08:05
camelia rakudo-moar cc017e: OUTPUT«False␤»
jnthn Ven: For now I'd not try putting it on a proto. It'll interact rather badly with the fact that protos get optimized out in the common case if they're onlystar ones. 08:06
lizmat jnthn: fair enough :-) 08:10
08:10 rindolf joined
lizmat (wrt to signal not being in Supply) 08:11
Ven m: sub a($a, $b) { say $a.perl ~ $b.perl; }; say 5.&a(5); 08:12
camelia rakudo-moar cc017e: OUTPUT«55␤True␤»
Ven m: sub a($a, $b) { say $a.perl ~ $b.perl; }; say 5.&a('fo'); # ;_;
camelia rakudo-moar cc017e: OUTPUT«5"fo"␤True␤»
lizmat jnthn: wrt is_cached on a proto: could the existence of the trait be the reason *not* to optimize it away? 08:14
jnthn lizmat: Yeah, but it doesn't convey that at present.
lizmat: I can't remember how is cached works; if it wraps, then we can maybe make wrap take off the onlystar flag. 08:15
lizmat it's a wrap
how would you take off the onlystar flag?
jnthn It's just an attribute declared in Routine iirc 08:16
So I think you'd just bindattr_i it.
lizmat ok, will check
Ven (not sure what's going on with @a VS @a[] vs @a[*]) 08:18
08:19 atroxaper joined
nwc10 jnthn: t/spec/S17-concurrency/promise.t failed test 64 for me when run in parralel 08:26
ok 64 - got the right order 08:27
twice now when running on its own
nwc10 runs it in a shell loop until it fails 08:28
lizmat I think it has to do with a thread being joined *before* it finishes running
nwc10 more accurately, while it passes, do it again
jnthn lizmat: I've never actually been able to observe that happen.
lizmat nwc10: you're on OSX as well, right? 08:29
nwc10 that's C<my> machine.
most real work is done by C<"my"> machine
and many of those are Linux
jnthn I wonder if it's that under pthreads, join can return spuriously? 08:30
nwc10 and this "my" machine is linux
lizmat afk for a few hours&
nwc10 not ok 64 - got the right order 08:31
# got: '1 0 2 3 4 5 6 7 8 9'
# expected: '0 1 2 3 4 5 6 7 8 9'
# Looks like you failed 1 tests of 64
And, botherit, tests seem to return exit success even when they fail
jnthn The documentation doesn't suggest so
08:41 denis_boyun joined 08:46 thilp joined 08:48 LLamaRider left 08:52 Ven left 09:09 spider-mario joined 09:11 daniel-s joined 09:19 thou joined, kivutar left 09:24 thou left
FROGGS TimToady: do we need to change S02:1064? there is an inconsistency between `my $x; (*, *, $x) = (1, 2, 3);` and `my ($, $, $x) = (1, 2, 3);` 09:25
synopsebot Link: perlcabal.org/syn/S02.html#line_1064 09:26
FROGGS TimToady: if I had the choice I'd make * in signatures behave like $ 09:27
jnthn FROGGS: I don't think we can make * do that in sigs... 09:30
ferry &
FROGGS hmmm 09:31
but it would fit... "Slurp one thing to nowhere" 09:32
09:36 sqirrel left, donaldh joined
tadzik FROGGS: hey, what's the status of creating moarvm fakexecutables? How far did you take the idea/ 09:47
I'm very interested in this, for the gmaes on steroids purposes
FROGGS tadzik: I can create a foo executable that is a maorvm binary that executes perl6.moarvm + your custom code as .moarvm files included 09:51
currently it needs a "-e 1" to not run the repl
and, it still tries to load things like Perl6/BOOTSTRAP.moarvm and other internal modules, even when they are bundled 09:52
tadzik ah, I stumbled upon that BOOTSTRAP too 09:53
and it insisted on looking for it in /home/tadzik/blablaabla
FROGGS so, currently I have made branchen in MoarVM, nqp and rakudo called "execname" 09:54
and two unpushed diffs for MoarVM and rakudo
tadzik: these need to applied to the execname branches gist.github.com/FROGGS/3e5bee1ef8e122c9b6fe 09:56
in the gist you can see how the .moarvm files are embedded in the your-game.c file 09:57
but somehow we need to trick rakudo's module loader
(and disable the repl)
tadzik I see 09:58
FROGGS brb 10:00
10:03 Ven joined
Ven m: say $bool = True but False; if $bool { say 'hello'; } 10:04
camelia rakudo-moar 383ba3: OUTPUT«===SORRY!=== Error while compiling /tmp/ac6p8xKgiY␤Variable '$bool' is not declared␤at /tmp/ac6p8xKgiY:1␤------> say $bool⏏ = True but False; if $bool { say 'hello␤ expecting any of:␤ postfix␤»
Ven m: my $bool = True but False; if $bool { say 'hello'; }
camelia ( no output )
masak FROGGS: the '$' syntax was introduced way back when because '*' couldn't do that.
10:07 salv0 left 10:08 dalek left, dalek joined, ChanServ sets mode: +v dalek
Ven m: say first Int, ("a", 1, 2); 10:14
camelia rakudo-moar 383ba3: OUTPUT«1␤»
10:15 telex left 10:16 rindolf left 10:20 salv0 joined 10:26 telex joined, telex left 10:28 Alina-malina left, telex joined 10:32 rindolf joined 10:44 prevost joined 10:53 eternaleye joined 10:59 rindolf left
dalek kudo/nom: d2dfb68 | (Elizabeth Mattijsen)++ | src/core/Routine.pm:
is cached now resets the "onlystar" flag if needed

So that proto's with "is cached" will not be optimized away.		 11:04
lizmat jnthn: ^^^ this doesn't regress anything, but am not sure this is what you meant 11:05
afk for another hour or so
11:08 thou joined
Ven lizmat++ # fixing stuff I just asked 11:08
11:12 thou left 11:14 rindolf joined 11:25 atroxaper left 11:36 kurahaupo joined
Ven p6: say (^Inf)[*-10..*-5].perl 11:38
camelia rakudo-{parrot,moar} d2dfb6: OUTPUT«()␤»
..niecza v24-109-g48a8de3: OUTPUT«Unhandled exception: Cannot use value like WhateverCode as a number␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /home/p6eval/niecza/lib/CORE.setting line 300 (Any.Numeric @ 6) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /home/p6eval/niecz…»
..rakudo-jvm d2dfb6: OUTPUT«Unhandled exception: java.lang.RuntimeException: Missing or wrong version of dependency 'src/Perl6/Grammar.nqp'␤ in (gen/jvm/main.nqp)␤␤»
Ven m: say Inf - 5 11:39
camelia rakudo-moar d2dfb6: OUTPUT«Inf␤»
Ven m: say (^5)[Inf..Inf]
camelia rakudo-moar d2dfb6: OUTPUT«␤»
11:40 Bucciarati joined 11:45 ivanshmakov left, ivanshmakov joined
FROGGS m: say Inf cmp Inf 11:49
camelia rakudo-moar d2dfb6: OUTPUT«Same␤»
11:54 a3gis joined
Ven there's no shortcut for ^$x X ^$x (or with Z), right ? 12:03
m: say (gather for ^5 { take 'line' => $_ })[3].perl 12:04
camelia rakudo-moar d2dfb6: OUTPUT«"line" => 3␤»
Ven m: say (gather for ^5 { take 'line' => $_; take 'foo' => 3 })[3].perl 12:05
camelia rakudo-moar d2dfb6: OUTPUT«"foo" => 3␤»
12:06 Alina-malina joined
Ven m: say do given False { when True { 5 } } 12:07
camelia rakudo-moar d2dfb6: OUTPUT«5␤»
Ven m: sub foo(True) { 5 }; foo(False)
camelia ( no output )
Ven m: sub foo(True) { 5 }; say foo(False) # just checking, but that should DWIM
camelia rakudo-moar d2dfb6: OUTPUT«5␤»
Ven Is there a HTML generation lib in p6 yet ? 12:10
12:12 SamuraiJack_ left
Ven Can I generate names to export ? ie `for <a b c> -> $as { sub foo is export(:$as) {} }` 12:17
lizmat lotrrotk 12:19
oops, ww :-)
Ven wut
12:19 kbaker joined
Ven m: anon sub($) { 5 } 12:20
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!===␤Type 'sub' is not declared. Did you mean 'Sub'?␤at /tmp/KlhtG8oJBL:1␤------> anon sub⏏($) { 5 }␤Malformed anon␤at /tmp/KlhtG8oJBL:1␤------> anon ⏏sub($) { 5 }␤ expect…»
Ven m: say (anon($) { 5 }).() 12:21
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/x2hg380uPi␤Unsupported use of $) variable; in Perl 6 please use $*EGID␤at /tmp/x2hg380uPi:1␤------> say (anon($)⏏ { 5 }).()␤»
Ven m: say (anon($a) { 5 }).()
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/2FcONxHaGT␤Variable '$a' is not declared␤at /tmp/2FcONxHaGT:1␤------> say (anon($a⏏) { 5 }).()␤ expecting any of:␤ postfix␤»
Ven m: say (sub ($a) { 5 }).()
camelia rakudo-moar d2dfb6: OUTPUT«Not enough positional parameters passed; got 0 but expected 1␤ in sub at /tmp/HOGrcbO1ne:1␤ in block at /tmp/HOGrcbO1ne:1␤␤»
Ven m: my @closed = <a img>; say any(@closed) eq 'a'; 12:22
camelia rakudo-moar d2dfb6: OUTPUT«any(True, False)␤»
Ven r-p: my @closed = <a img>; say so any(@closed) eq 'a';
camelia rakudo-parrot d2dfb6: OUTPUT«True␤»
12:23 SamuraiJack joined
Ven r-p: my @closed = <a img>; say so any(@closed) eq 'h1'; 12:23
camelia rakudo-parrot d2dfb6: OUTPUT«False␤»
Ven I'm getting some WATs from parrot. 12:26
p: my Str $foo; say "Hey $foo";
camelia rakudo-parrot d2dfb6: OUTPUT«use of uninitialized value of type Str in string context in block at /tmp/5QFjS4wghV:1␤␤Hey ␤»
Ven p: my Str $foo = 'abc'; say "Hey $foo";
camelia rakudo-parrot d2dfb6: OUTPUT«Hey abc␤»
Ven p: my Str $foo = 'abc'; say "Hey <a>$foo</a>";
camelia rakudo-parrot d2dfb6: OUTPUT«postcircumfix:<{ }> not defined for type Str␤ in method Str at gen/parrot/CORE.setting:12622␤ in method Str at gen/parrot/CORE.setting:1038␤ in method Stringy at gen/parrot/CORE.setting:1048␤ in method Stringy at gen/parrot/CORE.setting:1047…»
Ven p: my Str $foo = 'abc'; say "Hey <a>{$foo}</a>"; 12:27
camelia rakudo-parrot d2dfb6: OUTPUT«Hey <a>abc</a>␤»
Ven p6-- # too agressive interpolation :(
m: sub gensub($n) { my multi sub dispatch($ where * eq $n) { True }; my multi sub dispatch($) { False } }; my &l = gensub('foo'); say l('a'), l('foo'); 12:30
camelia rakudo-moar d2dfb6: OUTPUT«FalseFalse␤»
12:30 sftp left
Ven m: sub gensub($str) { my multi sub dispatch($ where { $_ eq $str}) { True }; my multi sub dispatch($) { False } }; my &l = gensub('foo'); say l('a'), l('foo'); 12:30
camelia rakudo-moar d2dfb6: OUTPUT«FalseFalse␤»
12:30 raiph left 12:31 sftp joined
Ven m: gist.github.com/Nami-Doc/15ac60bb18fb1914ba04 12:32
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/xLIvQEvJBT␤Slurpy positionals with type constraints are not supported.␤at /tmp/xLIvQEvJBT:18␤------> my multi sub htmlize(Str *@content⏏, *%attrs) {␤ expecting any o…»
12:33 treehug88 joined
FROGGS masak: perl6-m -e 'my (*, *, $x) = (1, 2, 3); say $x' 12:33
3
12:34 salv0 left
Ven Wtf perl 6 ._. 12:35
FROGGS Ven: problems?
Ven m: sub gen($name) { my multi sub d(Str $str) { "Hey $name - $str" }; my multi sub(Int $n) { callwith($n.Str) } }; my &g = gen('foo'); say g(5), g("navi"); 12:36
camelia rakudo-moar d2dfb6: OUTPUT«Type check failed in binding $n; expected 'Int' but got 'Str'␤ in sub sub at /tmp/yoXMoq6lFV:1␤ in block at /tmp/yoXMoq6lFV:1␤␤»
Ven FROGGS: gist.github.com/Nami-Doc/15ac60bb18fb1914ba04
FROGGS typo in the second multi name
Ven m: sub gen($name) { my multi sub d(Str $str) { "Hey $name - $str" }; my multi sub d(Int $n) { callwith($n.Str) } }; my &g = gen('foo'); say g(5), g("navi");
FROGGS my multi sub(Int $n) <-- subname is missing
camelia rakudo-moar d2dfb6: OUTPUT«Type check failed in binding $n; expected 'Int' but got 'Str'␤ in sub d at /tmp/1Vj8pFklFm:1␤ in block at /tmp/1Vj8pFklFm:1␤␤»
Ven m: sub gen($name) { my multi sub d(Str $str) { "Hey $name - $str" }; my multi sub d(Int $n) { callwith($n.Str) }; return &d; }; my &g = gen('foo'); say g(5), g("navi"); 12:37
camelia rakudo-moar d2dfb6: OUTPUT«NilHey foo - navi␤»
FROGGS ahh
Ven better.
12:38 cogno joined
Ven still incorrect, though, mmh 12:38
12:38 xenoterracide_ joined, cogno is now known as Guest898
Ven m: sub h(%h) { %h.perl } sub xd(*%a) { say h(%a) }; xd(); 12:39
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/vwL1T8wWWc␤Two terms in a row␤at /tmp/vwL1T8wWWc:1␤------> sub h(%h) { %h.perl } ⏏sub xd(*%a) { say h(%a) }; xd();␤ expecting any of:␤ method arguments␤ …»
Ven m: sub h(%h) { %h.perl }; sub xd(*%a) { say h(%a) }; xd();
camelia rakudo-moar d2dfb6: OUTPUT«().hash␤»
Ven r-p: sub h(%h) { %h.perl }; sub xd(*%a) { say h(%a) }; xd();
camelia rakudo-parrot d2dfb6: OUTPUT«().hash␤»
Ven r-p: sub h(%h) { %h.perl }; multi sub xd(Int, *%a) { callsame(|%a); }; sub xd(*%a) { say h(%a) }; xd(5); 12:40
camelia rakudo-parrot d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/hMLDJ_NWKQ␤Redeclaration of routine xd␤at /tmp/hMLDJ_NWKQ:1␤------> same(|%a); }; sub xd(*%a) { say h(%a) }⏏; xd(5);␤ expecting any of:␤ postfix␤ …»
Ven r-p: sub h(%h) { %h.perl }; multi sub xd(Int, *%a) { callsame(|%a); }; multi sub xd(*%a) { say h(%a) }; xd(5);
camelia ( no output )
Ven r-p: sub h(%h) { %h.perl }; multi sub xd(Int, *%a) { callwith(|%a); }; multi sub xd(*%a) { say h(%a) }; xd(5); 12:41
camelia ( no output )
FROGGS that is too much spam for me to see what is going on :o)
you could talk with camelia in privmsg, and then only paste stuff here for discussion
Ven FROGGS: ha, sorry. gist.github.com/Nami-Doc/15ac60bb18fb1914ba04 -- fails with "not enough positional parameters, got 1 but expected 2 in sub attributes at Generate.pm:4" 12:42
12:43 treehug88 left
FROGGS m: for { a => 1, b => 2 } -> $k, $v { say "$k $v" } 12:43
camelia rakudo-moar d2dfb6: OUTPUT«Not enough positional parameters passed; got 1 but expected 2␤ in block at /tmp/lW1NBkRnxJ:1␤␤»
FROGGS m: for { a => 1, b => 2 }.kv -> $k, $v { say "$k $v" }
camelia rakudo-moar d2dfb6: OUTPUT«a 1␤b 2␤»
FROGGS Ven: ^^ 12:44
Ven m: gist.github.com/Nami-Doc/15ac60bb18fb1914ba04 12:45
camelia rakudo-moar d2dfb6: OUTPUT«Too many positional parameters passed; got 1 but expected 0␤ in block at src/gen/m-CORE.setting:645␤ in sub htmlize at /tmp/7TD_BafJ8J:19␤ in sub htmlize at /tmp/7TD_BafJ8J:10␤ in block at /tmp/7TD_BafJ8J:31␤␤» 12:46
Ven s/same/with/, I guess. 12:47
m: gist.github.com/Nami-Doc/15ac60bb18fb1914ba04
camelia rakudo-moar d2dfb6: OUTPUT«Nil␤»
Ven m: sub gen-for($str) { sub { say $str; } }; my &hello = gen-for('hello'); my &bar = gen-for('foobar'); hello(); bar(); hello(); # I don't even ... 12:49
camelia rakudo-moar d2dfb6: OUTPUT«hello␤foobar␤hello␤»
12:50 salv0 joined
Ven m: sub gen-for($str) { my multi sub named { say $str; }; &named }; my &hello = gen-for('hello'); my &bar = gen-for('foobar'); hello(); bar(); hello(); # I don't even ... 12:50
camelia rakudo-moar d2dfb6: OUTPUT«foobar␤foobar␤foobar␤»
Ven ^ is that a bug ?
or is it "declare vs compile-time" stuff again ?
(and if so -- fair enough, but I'm not sure how I can get a multi anonymous sub to be returned) 12:51
lizmat sub a { multi sub foo { }; &foo } ? 12:52
Ven (I'm really starting to think I'm just not smart enough for p6 considering how much time I have to go back and how unhelpful web resources are ...)
lizmat: check my previous paste. 12:53
it doesn't correctly close over variables, it seems
12:54 Guest898 left
Ven m: sub gen-for($str) { multi sub named { say $str; }; &named }; my &hello = gen-for('hello'); my &bar = gen-for('foobar'); hello(); bar(); hello(); # try without `my` 12:55
camelia rakudo-moar d2dfb6: OUTPUT«foobar␤foobar␤foobar␤»
Ven Nope, same thing.
FROGGS Ven: that seems to about the lexpads, yes
that seems to be exactly what masak++ blogged about
Ven yes.
12:55 anaeem___ left, pochi joined
Ven A workaround would be a class overriding postcircumfix:<( )> I guess, but ew. 12:56
colomon Ven: why do you want an anonymous multi? (or to put it another way, I'm not sure why not being able to figure out how to get a weird edge case to work would mean you're not smart enough for p6.) 12:58
Ven colomon: gist.github.com/Nami-Doc/15ac60bb18fb1914ba04 that's my use-case
colomon: and because I keep hitting "weird edge cases" it seems, considering how much I spam `r:` and friends 12:59
lizmat just had a déjà vu experience 13:00
Ven lizmat: what do you mean ? 13:01
lizmat your gist reminded me a lot of CGI.pm 13:03
at least, some of it 13:04
Ven doesn't know CGI.pm
Ven github searched but only got on p5 stuff
lizmat metacpan.org/pod/CGI
indeed, perl 5
it's one of the oldest Perl 5 modules 13:05
13:05 kaare_ left
lizmat cpanratings.perl.org/dist/CGI 13:05
you seem to be doing 2. generating HTML using function calls.
Ven I'm not fond of this kind of html-generation, but it gets the job done to start with 13:06
lizmat "It seems to be fairly widely agreed that the HTML generation part is best forgotten about"
Ven will probably port HAML later ...
lizmat but if it gets the job done for you, more power to you :-) 13:07
Ven with masak++'s indent parser
lizmat need to run some errands and pick someone up &
Ven Actually, even with that p6 bug which seems unfixable, I still don't get the correct result 13:08
13:09 guru joined 13:10 guru is now known as Guest59030, Guest59030 is now known as ajr_
Ven Actually, even with that p6 bug which seems unfixable, I still don't get the correct result. I'll just give up that one 13:12
colomon: that's an actual example of what I mean. that strange lexpad "bug" aside, I don't understand at all why I don't get my other multi 13:14
's return value when I use {call,next}with
13:21 kshannon joined
Ven It looks like I can't find {next,call}{same,with} anymore, to add it. I'd expect it to be in S06 13:23
[Coke] what advantage is there to having the 2 multi subs over a single sub that takes a slurpy arg? 13:24
(I'm not saying it shouldn't work the way you're doing it, of course) 13:25
just wondering why it's preferable to go that route. 13:26
13:26 SamuraiJack left
FROGGS TimToady / jnthn / masak: this makes * in signatures work: gist.github.com/FROGGS/a78ff92ec5e7ce2f8fda 13:37
TimToady / jnthn / masak: no breaking spectests btw
13:38 cogno joined 13:39 cogno is now known as Guest15734, xenoterracide_ left
Ven [Coke]: no idea. You'd ~~ List and .join if so ? 13:40
m: my s(*$slurp){ say $slurp.perl }; s(1, 2); 13:42
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!===␤Type 's' is not declared␤at /tmp/4ef0fx8NgR:1␤------> my s⏏(*$slurp){ say $slurp.perl }; s(1, 2);␤Malformed my␤at /tmp/4ef0fx8NgR:1␤------> my ⏏s(*$slurp){ say $slurp.perl }; …»
Ven m: sub s(*$slurp){ say $slurp.perl }; s(1, 2);
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/zCCKp7S9Sq␤Calling 's' will never work with argument types (Int, Int)␤ Expected: :(Any $slurp)␤at /tmp/zCCKp7S9Sq:1␤------> sub s(*$slurp){ say $slurp.perl }; ⏏s(1, 2);[…»
Ven m: sub s(*@slurp){ say @slurp.perl }; s(1, 2);
camelia rakudo-moar d2dfb6: OUTPUT«Array.new(1, 2)␤»
Ven (what's *$ supposed to do ?)
13:42 bluescreen10 joined 13:43 Guest15734 left
[Coke] I dunno, *@ slurps positions, *% slurps named. 13:43
*positionals
so I'd use *@, get a list of args, and then .join it and use the logic from the first multi sub. 13:44
I'm just wondering if you're expecting a performance benefit from doing it the other way.
Ven [Coke]: yeah, but I just changed it to do that
[Coke] (given that slurps all the time can be slow if the main path is single arg)
Ven I didn't want to use slurpy params for the main usage which is more often than not only one arg
[Coke] fair enough. Just curious.
Ven but I'll leave it the "slow" way since the "fast" one doesn't work (yet ?) 13:45
[Coke] also fair. :) 13:47
13:48 btyler joined
Ven [Coke]: the other thing I wanted to do is generate a function for each tag and "export" it automatically. Can I do that ? 13:49
PerlJam Ven: with EVAL ;) 13:50
Ven I'd really, really, really, really like not to use EVAL or anything like that.
[Coke] I don't know how to programmatically do "is export" 13:51
Ven That's basically saying "I'm defeated, the language can't do that for me"
Maybe that'd be possible with a macro, but I can't picture how 13:52
Ulti are there any modules using LibraryMake I can take a look at? 13:57
retupmoca Ulti: the simplest one I'm aware of is Auth::PAM::Simple 13:58
Ulti okedoke I will take a look 14:00
thinking of trying for a libccv binding
14:05 logie_ joined
timotimo you can programmatically apply is export to a function by calling trait_mod:<is>($routine, :export) i believe 14:11
14:11 kaare_ joined
Ven timotimo: yeah. but how can I name `our sub` programmatically ? 14:12
timotimo Routine.new? :) 14:13
hm, though the package is runtime read-only 14:14
but you can have an export sub that generates stuff on the fly
and not have to rely on the package
PerlJam Why does it need to be an "our" sub?
14:17 cognominal left
Ven It doesn't need to. I need to give it a name 14:18
m: $MY::a = 5; say $a; 14:19
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/cNdjfe2uB1␤Variable '$a' is not declared␤at /tmp/cNdjfe2uB1:1␤------> $MY::a = 5; say $a⏏;␤ expecting any of:␤ postfix␤»
14:19 treehug88 joined 14:20 donaldh left
timotimo if you look at ADT, you'll see a way to dynamically create classes (and their methods) of arbitrary names; the metamodel probably offers something similar for creating routines with names 14:20
and there's apparently set_name 14:21
Ven m: my &s = sub { 5 }; s.set_name: 'foo'; say foo();
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!===␤Unrecognized regex metacharacter ; (must be quoted to match literally)␤at /tmp/qeHo1Gd73N:1␤------> my &s = sub { 5 }; s.set_name: 'foo'⏏; say foo();␤Couldn't find terminator .␤at /tmp/qeHo1Gd73…»
Ven m: my &s = sub { 5 }; s.set_name('foo'); say foo();
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!===␤Unrecognized regex metacharacter ; (must be quoted to match literally)␤at /tmp/pTmHBDsbK_:1␤------> my &s = sub { 5 }; s.set_name('foo')⏏; say foo();␤Couldn't find terminator .␤at /tmp/pTmHBDsbK…»
timotimo hehe 14:22
calling things "s" is problematic :P
the problem is, that set_name will not cause the package that used to contain it to change
you'll still need to have a custom sub EXPORT to assign the things 14:23
that would be much more convenient for you in any case
Ven m: my &bazbar = sub { 5 }; bazbar.set_name('foo'); say foo();
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/OyhJcAAwEg␤Undeclared routine:␤ foo used at line 1␤␤»
Ven didn't read, my bad
14:24 donaldh joined
timotimo m: my &bazbar = sub { 5 }; &bazbar.set_name('foo'); say &bazbar.name 14:25
camelia rakudo-moar d2dfb6: OUTPUT«foo␤»
timotimo you also forgot an & there
bazbar.set_name('foo') would have tried to call set_name on 5
14:25 benabik joined 14:26 rurban joined
Ven m: my &bazbar = sub { 5 }; &bazbar.set_name('foo'); say foo(); 14:27
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/Ds6EnkO5lk␤Undeclared routine:␤ foo used at line 1␤␤»
Ven timotimo: seems to be the same
timotimo well, yeah
these things operate on different levels
you can't expect set_name (run time) to change the lookup of foo (compile time)
14:27 thou joined
retupmoca m: our &foo is export = sub { 42 }; say foo(); # just set the name to begin with? 14:27
camelia rakudo-moar d2dfb6: OUTPUT«42␤»
retupmoca oh wait, I see what you're doing - nevermind 14:30
timotimo retupmoca: and now do that with a "foo" that is not known at compile time :)
yeah
Ven: as i said, sub EXPORT will give you exactly the flexibility you're going to need
also, i don't think EVAL would let you modify the package either
14:40 cognominal joined
Ven timotimo: I'd rather not do a project than using EVAL - that's to say how much I hate it ! 14:41
(A side-project of course ...) 14:44
timotimo yeah
hence the tip to use sub EXPORT.
it seems to me you've been ignoring that hint :P
so, please acknowledge it? or ask me how exactly it works? 14:49
14:50 FROGGS left 14:55 Exodist joined 14:58 thou left
Ven timotimo: I'm doing some IRL stuff too, and since I know if I plan to find something in the "docs" it'll take me 3 hours, I'm not going through that yet :) 14:58
timotimo oh 14:59
Ven (unless you have a doc to point me to, since synsearch doesn't yield anything) 15:00
timotimo just put an "our sub EXPORT" in your module (actually ... at some point it actually had to be outside the package, but inside the compunit) and take a slurpy argument list and return a hash (or pseudostash) 15:01
when your compunit gets pulled in with "use", it'll pass over the arguments from the use line to the export sub and incorporate everything the export sub returns into its namespace 15:02
15:02 Alina-malina left, Alina-malina joined
Ven timotimo: our sub EXPORT { my @tags = <html head body title h1>; (do for @tags { $_ => create-tag($_) }).hash }; ? 15:03
timotimo something like that should work
Ven m: sub foo($x) { $x x 5; }; my @tags = 'a'..'f'; say (@tags Z=> foo).perl; 15:04
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/WwRAK9Wvc7␤Calling 'foo' requires arguments (if you meant to operate on $_, please use .foo or use an explicit invocant or argument)␤ Expected: :(Any $x)␤at /tmp/WwRAK9Wvc7:1␤------> [32…»
Ven m: sub foo($x) { $x x 5; }; my @tags = 'a'..'f'; say (@tags Z=> @tags>>&foo).perl;
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/9QVSVq9Ard␤Missing << or >>␤at /tmp/9QVSVq9Ard:1␤------> tags = 'a'..'f'; say (@tags Z=> @tags>>&⏏foo).perl;␤ expecting any of:␤ postfix␤ infix stop…»
Ven m: sub foo($x) { $x x 5; }; my @tags = 'a'..'f'; say (@tags Z=> @tags>>.&foo).perl;
camelia rakudo-moar d2dfb6: OUTPUT«("a" => "aaaaa", "b" => "bbbbb", "c" => "ccccc", "d" => "ddddd", "e" => "eeeee", "f" => "fffff").list␤»
Ven p6++ 15:05
is there a module which contains the list of HTML tags?
15:05 denis_boyun left
Ven timotimo: gist.github.com/Nami-Doc/15ac60bb18fb1914ba04 whew. I got it. 15:06
15:06 thou joined 15:07 FROGGS joined
Ven m: sub are-if-statements-or-expressions { return if True { 1 } else { 0 } } 15:07
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/rvV1WM9wrt␤Unexpected block in infix position (two terms in a row, or previous statement missing semicolon?)␤at /tmp/rvV1WM9wrt:1␤------> tements-or-expressions { return if True ⏏…»
Ven m: sub are-if-statements-or-expressions { return (if True { 1 } else { 0 }) }
camelia ( no output )
15:07 PerlJam joined
Ven alrighty. that's cool. 15:07
timotimo you can put a "do" in front instead of parenthesis around 15:08
btw: 15:09
r: say <a href="google.com/">
camelia rakudo-{parrot,jvm,moar} d2dfb6: OUTPUT«a href="google.com/"␤»
timotimo %)
15:09 salv0 left
Ven timotimo: ha ! 15:09
timotimo: and yeah, I should know better, I use `do for` in several places 15:10
timotimo r: say <a href="google.com/"\> what is this sorcery </a>.perl
camelia rakudo-{parrot,jvm,moar} d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/tmpfile␤Unable to parse expression in quote words; couldn't find final '>'␤ ␤at /tmp/tmpfile:1␤------> e.com/"\> what is this sorcery </a>.perl⏏<EOL>…»
timotimo aaw
r: say <a href="google.com/"\> what is this sorcery \</a>.perl
camelia rakudo-{parrot,jvm,moar} d2dfb6: OUTPUT«("a", "href=\"google.com/\">", "what", "is", "this", "sorcery", "</a")␤» 15:11
FROGGS r: say <body><blink>WAT</blink></body> # synopsebot is funny :o)
camelia rakudo-{parrot,jvm,moar} d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/tmpfile␤Two terms in a row␤at /tmp/tmpfile:1␤------> say <body><blink>⏏WAT</blink></body> # synopsebot is funny␤ expecting any of:␤ postfix␤ …»
FROGGS m: say <body><blink>WAT</blink></body> # synopsebot is funny :o)
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/RV5UlKDWMP␤Two terms in a row␤at /tmp/RV5UlKDWMP:1␤------> say <body><blink>⏏WAT</blink></body> # synopsebot is funny␤ expecting any of:␤ postfix␤ i…»
timotimo ?
FROGGS m: say <body><blink>WAT</blink></body>
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/9HZOGTgjoq␤Two terms in a row␤at /tmp/9HZOGTgjoq:1␤------> say <body><blink>⏏WAT</blink></body>␤ expecting any of:␤ postfix␤ infix stopper␤ …»
FROGGS damn
see #moarvm
I pasted there instead of privmsging camelia
timotimo yeah, depends on the filename
FROGGS seems I am confused too -.- 15:12
Ven FROGGS: it just works in timotimo++'s case because of <words syntax>, y'know :P ?
FROGGS Ven: yes, sure :o) 15:13
m: say <blink>..</blink> 15:14
camelia rakudo-moar d2dfb6: OUTPUT«"blink".."/blink"␤»
FROGGS m: say <blink>...</blink>
camelia rakudo-moar d2dfb6: OUTPUT«(timeout)»
FROGGS ahh, now I know why it times out
15:16 plobsing left
timotimo %) 15:16
this is pretty terrible.
Ven doesn't know why 15:18
15:19 kaleem left, treehug88 left
timotimo r: class List { method new(*@foo) { say "created a list with @foo[]" } }; <a href="test">; 15:19
camelia ( no output )
timotimo r: say <a href="test">.WHAT 15:20
camelia rakudo-{parrot,jvm,moar} d2dfb6: OUTPUT«(Parcel)␤»
timotimo r: class Parcel { method new(*@foo) { say "created a list with @foo[]" } }; <a href="test">;
camelia ( no output )
timotimo not quite doable apparently
15:20 Alina-malina left, Alina-malina joined 15:21 atroxaper joined 15:22 kurahaupo left 15:23 salv0 joined 15:27 kurahaupo joined
hoelzro morning #perl6 15:29
did I hear that all of the star modules test ok on mokudo now? =) 15:30
15:32 treehug88 joined
tadzik yup :) 15:32
15:33 donaldh left 15:37 hoverboard joined 15:38 Vlavv left, zakharyas left
TimToady Ven: True always indicates a successful smartmatch, so you'd have to write that signature (Bool where .so) or some such 15:38
Ven TimToady: I know. I just find it very weird 15:39
`given $bool { when True {}; when False {} }` definitely doesn't DWIM
TimToady yes, but at least that usually warns 15:40
std: given my $bool { when True {}; when False {} }
camelia std ec3d853: OUTPUT«Potential difficulties:␤ Smartmatch against True always matches; perhaps you should use :so or *.so or ?* instead at /tmp/O5JUDDkAsK line 1:␤------> given my $bool { when ⏏True {}; when False {} }␤ Smartmatch against False always fail…»
TimToady or is supposed to
std: sub foo (True) {}
camelia std ec3d853: OUTPUT«ok 00:01 125m␤»
TimToady probably should there too
but smartmatches wouldn't be very smart if they couldn't use booleans to return success 15:42
15:43 atroxaper left
masak Ven: we basically need it for file tests et al. 15:45
er, etc.*
15:45 atroxaper joined 15:46 SamuraiJack joined
TimToady as for interpolating things that look like $foo</a>, it's not clear how much we should pander to other languages' syntactic foibles 15:46
how long will tag syntax last? probably a good while yet, so might be worth dwimming that one
and the wat is unlikely, and easy to work around 15:47
masak (or issuing a "probably not what you meant" warning)
15:48 SamuraiJack left, SamuraiJack_ joined
TimToady that's also a valid approach 15:48
std: sub ($) {} 15:49
camelia std ec3d853: OUTPUT«ok 00:01 125m␤»
TimToady m: sub ($) {}
camelia ( no output )
15:52 Vlavv joined
Ven TimToady: I just think {} isn't so bad you want to try and recognize everything inside quotes 15:53
Ven would even be okay for #{{} 15:54
15:55 pecastro joined
TimToady qc// will recognize *only* {} 15:55
timotimo TimToady: you're suggesting $foo</bar> would not be interpreted as accessing the key "/bar" in $foo, because that looks like a tag?
Ven qc//++ then.
TimToady timotimo: that, or what masak++ suggests 15:56
Ven m: qc/<a href="/">hey</a>/;
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/iRR40isthi␤Two terms in a row␤at /tmp/iRR40isthi:1␤------> qc/<a href="/⏏">hey</a>/;␤ expecting any of:␤ postfix␤ infix stopper␤ infix or m…»
Ven m: qc//<a href="/">hey</a>//;
camelia rakudo-moar d2dfb6: OUTPUT«===SORRY!=== Error while compiling /tmp/OHGYxz5hWM␤Two terms in a row␤at /tmp/OHGYxz5hWM:1␤------> qc//<a href="/">⏏hey</a>//;␤ expecting any of:␤ postfix␤ infix stopper␤ infix or…»
Ven ^ that's what I'm wary of
TimToady you can't doubled // like that
Ven Too bad :p 15:57
TimToady only brackets can double/triple...
15:57 atroxaper left
flussence m: qc\<a href="/">hey</a>\; 15:57
camelia ( no output )
TimToady but that's a LTA error message
timotimo which was what exactly?
TimToady std: qc//<a href="/">hey</a>// 15:58
camelia std ec3d853: OUTPUT«===SORRY!===␤Two terms in a row at /tmp/rIFUOqtR2p line 1:␤------> qc//<a href="/">⏏hey</a>//␤ expecting any of:␤ POST␤ feed_separator␤ infix or meta-infix␤ infixed function␤ postcircumfix␤ postfix␤ po…»
15:58 atroxaper joined
TimToady oh, it's not a regex, so // is legal, nevermind 15:58
timotimo: masak++ was suggesting a warning rather than a dwim 15:59
Ven ah nice, flussence
15:59 treehug8_ joined
TimToady thinks backslash is kind of bad to use for delimiters 15:59
flussence m: say 'foobar' ~~ /b/
camelia rakudo-moar d2dfb6: OUTPUT«｢b｣␤␤»
flussence m: qc｢<a href="/">hey</a>｣; 16:00
camelia ( no output )
Ven thinks backslash is better than having to escape slash everytime in html
TimToady m: say qc[<a href="/">hey</a>]
camelia rakudo-moar d2dfb6: OUTPUT«<a href="/">hey</a>␤»
flussence thinks not having to write raw html in the first place is better-er...
TimToady less worser, anyway 16:01
Ven plans to port Jade or HAML
16:01 treehug88 left 16:03 thou left
timotimo okay, so how do you get away from that warning? use {' '} instead of < >? 16:04
16:04 btyler left
flussence
.oO( I read a fascinating article a while back that made me uncomfortably aware that HTML/CSS are just as much a big ball o' mud as PHP... )
 16:05
huf no duuhu
the web is a ghetto 16:06
masak FROGGS++ # 4a8dac6
Ven At least we ought to do our job better with HTML stuff
masak flussence: oh, and JavaScript is all pure and perfect? :P
flussence JS has font/color formatting methods on strings too, but the article didn't mention it :) 16:07
huf yeah. in comparison to browserjs, yeah.
Ven hahahahahha
"hey".blink(), anyone ? 16:08
FROGGS masak: No such commit 4a8dac6 without a repository or gist reference at -e line 1 # :P
Ven (js' cool, dom is pretty bad)
FROGGS (I found the moarvm commit btw)
16:09 thou joined 16:10 Rotwang joined
timotimo r: say "foo" ~~ <baz bar foo> 16:13
camelia rakudo-{parrot,jvm,moar} d2dfb6: OUTPUT«False␤»
16:13 btyler joined
timotimo r: say "quux" ~~ <baz bar foo> 16:13
camelia rakudo-{parrot,jvm,moar} d2dfb6: OUTPUT«False␤»
timotimo sadly, "foo" eq any(@list) is rather slow in rakudo :\
seems like it's somewhat of a pattern, though
Ven timotimo: should be stopping after first occurence though 16:15
what'd be the "other" way to do it ?
timotimo i don't know
FROGGS Ven: it should not stop
timotimo i thought smartmatch against a list would do it. otherwise, a set probably would do it
Ven FROGGS: in boolean context
timotimo ah, i was refering only to boolean context at this moment 16:16
FROGGS Ven: ahh, okay, that is the job of the optimizer then
16:16 treehug8_ left
timotimo r: say "foobar" ~~ set <foobar bazquux plopp> 16:16
camelia rakudo-{parrot,jvm,moar} d2dfb6: OUTPUT«False␤»
Ven FROGGS: and the optimizer should be doing it :p
timotimo hmm.
i thought about that very optimization in the past
sadly, it's not quite as easy as you might think; especially since there's some flattening involved 16:17
maybe if the @self-closing-tags were a constant ...
r: my constant @sct = <img a>; push @sct, "test"; say @sct;
camelia rakudo-moar d2dfb6: OUTPUT«Cannot call 'push'; none of these signatures match:␤:(Any:U \SELF: *@values, *%_)␤ in sub push at src/gen/m-CORE.setting:8476␤ in block at /tmp/tmpfile:1␤␤»
..rakudo-parrot d2dfb6: OUTPUT«Cannot call 'push'; none of these signatures match:␤:(Any:U \SELF: *@values, *%_)␤ in any at gen/parrot/BOOTSTRAP.nqp:1219␤ in method push at gen/parrot/CORE.setting:1616␤ in sub push at gen/parrot/CORE.setting:8439␤ in sub push at gen/par…»
..rakudo-jvm d2dfb6: OUTPUT«Cannot call 'push'; none of these signatures match:␤:(Any:U \SELF: *@values, *%_)␤ in any at gen/jvm/BOOTSTRAP.nqp:1212␤ in sub push at gen/jvm/CORE.setting:8417␤ in block at /tmp/tmpfile:1␤␤»
Ven timotimo: github.com/rakudo/rakudo/blob/45e2...ion.pm#L11
the optimization is there 16:18
timotimo that's not the part that is slow :)
r: my constant @sct = <img a>; @sct.push("test")
camelia rakudo-moar d2dfb6: OUTPUT«Cannot call 'push'; none of these signatures match:␤:(Any:U \SELF: *@values, *%_)␤ in block at /tmp/tmpfile:1␤␤»
..rakudo-jvm d2dfb6: OUTPUT«Cannot call 'push'; none of these signatures match:␤:(Any:U \SELF: *@values, *%_)␤ in any at gen/jvm/BOOTSTRAP.nqp:1212␤ in block at /tmp/tmpfile:1␤␤»
..rakudo-parrot d2dfb6: OUTPUT«Cannot call 'push'; none of these signatures match:␤:(Any:U \SELF: *@values, *%_)␤ in any at gen/parrot/BOOTSTRAP.nqp:1219␤ in method push at gen/parrot/CORE.setting:1616␤ in block at /tmp/tmpfile:1␤␤»
timotimo well, that's good i suppose.
the problem is that generating junction objects just to call .bool on them over and over again is very costly
and if there's an array/list involved, you have to be very careful not to break the flattening semantics it's supposed to have 16:19
Ven well, if at one point rakudo/... gets that much inlining, it'll be free
timotimo if it's a constant @foo, the compiler could constant-fold it, though, and do the compile-time unfolding of junctions it already does.
FROGGS timotimo: and it is that hard to turn an any junction into infix:<||> when .Bool is called on it?
timotimo we already do it in some cases
FROGGS timotimo: I mean, do we know that .Bool is called on it and that it is a Junction? 16:20
ahh, cool
timotimo currently, we're only looking for | and & infixes, though
the method form has the problem that it may be doing some kind of flattening
Ven but the same worry applies to all of a codebase self-hosted, I think ..
16:21 Alina-malina left 16:26 hoverboard left, anaeem1 joined 16:30 logie_ left 16:31 Alina-malina joined 16:32 logie_ joined
tadzik oh, gpw talks online 16:32
colomon what does gpw say?
16:33 atroxaper left
tadzik :D 16:35
16:43 treehug88 joined
dalek kudo-star-daily: f1935e1 | coke++ | log/ (5 files):
today (automated commit)
 16:44
kudo-star-daily: 77bd29e | coke++ | log/ (5 files):
today (automated commit)
rl6-roast-data: bc216c0 | coke++ | / (6 files):
today (automated commit)
 16:45
rl6-roast-data: f86eabc | coke++ | / (6 files):
today (automated commit)
16:55 ajr_ left
[Coke] $ ./perl6 16:58
Error occurred during initialization of VM
Could not reserve enough space for the card marking array
whoops.
timotimo one of the gpw14 lightning talks ended in "use python, ruby or java if you want to have something that supports ipv6" 16:59
let's be massively better than perl5 in this regard
16:59 Ven left 17:04 atroxaper joined 17:05 anaeem1 left, anaeem1 joined 17:08 treehug8_ joined, treehug8_ left, atroxaper left, treehug88 left, treehug88 joined 17:09 benabik left, anaeem1 left 17:15 anaeem1_ joined 17:16 dmol left 17:17 dmol joined 17:20 hoverboard joined 17:26 fhelmberger left, kaleem joined, guru joined 17:27 guru is now known as Guest70170, Guest70170 is now known as ajr_ 17:28 rurban left
[Coke] fixed java build issue. (runner didn't have enough memory to even launch ./perl6) 17:29
may try to re-push today's commit once this java run finishes. 17:30
s/build/daily roast/ 17:31
17:39 denis_boyun_ joined 17:45 kaleem left 17:49 bjz left
ponbiki ol 17:55
17:55 a3gis left 17:56 a3gis joined 18:00 a3gis left 18:03 benabik joined
segomos how do i get the line numbers causing warnings in perl6? 18:06
moritz segomos: you mean, programmatically? 18:08
r: say 42 + Any
18:08 a3gis joined
camelia rakudo-{jvm,moar} d2dfb6: OUTPUT«use of uninitialized value of type Any in numeric context␤42␤» 18:08
..rakudo-parrot d2dfb6: OUTPUT«use of uninitialized value of type Any in numeric context in block at /tmp/tmpfile:1␤␤42␤»
moritz you use rakudo-p :-) 18:09
segomos moritz: nah, is there a way to get more verbose just through 'perl6' flags? 18:10
moritz segomos: no
FROGGS at least not for warnings 18:11
moritz segomos: but you can catch the exception, and inspect $!.backtrace
segomos try {} will catch warns?
i'll read about it :)
moritz [Coke]: could you please add brrt-to-the-future.blogspot.nl/ to the planeteria.org/perl6/ feed? kthxbye 18:12
segomos once i'm done making this thing turn poorly formed html into xml
18:12 kurahaupo left, kurahaupo joined
moritz segomos: nah, you need CONROL blocks to catch warnings 18:12
18:19 SamuraiJack_ left
nwc10 is jnthn still on a Ferry? :-/ 18:28
18:29 rurban joined 18:34 rurban left 18:37 raiph joined
tadzik aargh 18:38
I want MooseX::Strictconstructor for P6
REWARD
colomon seems like it ought to be easy to write… 18:39
nwc10 "REWARD" is how many hugs? 18:40
18:41 Sqirrel joined
tadzik at least 5 18:41
"what's wrong? My C library? My spritesheets? My makefiles? My animation engine? My renderer? Oh, no, I just changed variable names in one less place that I should have" 18:42
masak :/ 18:43
18:47 darutoko left
colomon tadzik: you still game programming in p6? :) 18:47
tadzik colomon: you bet :)
I have spritesheets and animations going
colomon is now peering at Mu.BUILDALL, trying to see how it works.... 18:52
segomos tadzik: damn..thought i was going hard in p6 with stuff like html to xml parsers and web scrapin 18:53
18:53 raiph left 18:54 rindolf left, raiph joined
tadzik segomos: oh, it's easier than it sounds 18:54
and sometimes harder than it sounds :)
moritz XML? HTML? that's just text :-) 18:55
FROGGS tadzik: especially when it comes to sounds :o)
tadzik today I learned that for proper collision detection I'm gonna need self-balancing binary, or bi-binary trees
FROGGS tadzik: do yourself a flavor and do bindings to Box2d
tadzik FROGGS: bah, and miss all the fun?
but yeah, I may do that and some point : 18:56
:)
FROGGS tadzik: Box2D does it properly, that is the point :o)
segomos moritz: lol.. i received markup from a client that had several iterations of: <td/>some crap</td>
tadzik but those collision trees sounded exciting when I read about them today, now I want to implement them for the fun of it
FROGGS tadzik: then you perhaps should have a look at P5's Alien::Box2D / Box2D
moritz tadzik: so, start by writing a balanced tree data structure module? 18:57
tadzik: masak++ will love you for it :-)
tadzik :) sounds like a good start
FROGGS because we made our own Makefile instead of using Box2D's cmake crap
tadzik oooh, regular makefiles :)
colomon tadzik: don't you really want quadtrees? Or am I misunderstanding the problem? 18:58
tadzik colomon: I suppose I could maybe get away with binaries, but quadtrees would probably be best 19:00
colomon has dreamed of implementing quadtrees for yes. …. and will not be doing it any time soon due to insane schedule. 19:01
*years
19:02 fhelmberger joined 19:04 kurahaupo left, kurahaupo joined 19:05 atroxaper joined
vendethiel m: say 1 if 1 == 1 == 1 19:06
camelia rakudo-moar d2dfb6: OUTPUT«1␤»
segomos does 1 equate to True ? 19:07
m: say 1 if 1 == True
camelia rakudo-moar d2dfb6: OUTPUT«1␤»
FROGGS m: say 2 if 2 == 2 == 2
camelia rakudo-moar d2dfb6: OUTPUT«2␤»
segomos m: say 1 if 0 == 0 == 0
camelia rakudo-moar d2dfb6: OUTPUT«1␤»
segomos say 0 if 0 == True
m: say 0 if 0 == True
camelia ( no output )
FROGGS m: say +True # segemos, that answers your question 19:08
camelia rakudo-moar d2dfb6: OUTPUT«1␤»
19:08 anaeem1_ left 19:10 atroxaper left 19:11 Rotwang left
dalek ast: c6d9377 | (David Warring [email@hidden.address] | integration/advent2012-day13.t:
test ∉ ∌ ⊈ ⊄ ⊉ ⊅
 19:16
19:17 prevost left
dalek rl6-most-wanted: b101595 | segomos++ | most-wanted/modules.md:
added options for a TSV && CSV
 19:18
19:18 fhelmberger left 19:21 dwarring joined
dalek rl6-most-wanted: b4fa47e | segomos++ | most-wanted/modules.md:
Added another option for parsing html -> xml
 19:28
19:30 rurban joined 19:35 rurban left
[Coke] moritz: done 19:35
moritz [Coke]++ # thanks 19:39
masak segomos: True numifies to 1. whether it *equals* 1 depends on your notion of equality :) 19:43
segomos how is 0 == 0 == 0 processed?
FROGGS yes 19:44
segomos is that not broken into True == 0 ?
FROGGS you an chain it
can*
m: say 1 < 3 < 5
camelia rakudo-moar d2dfb6: OUTPUT«True␤»
FROGGS m: say 1 < 6 < 5
camelia rakudo-moar d2dfb6: OUTPUT«False␤»
FROGGS m: say 8 > 4 < 5
camelia rakudo-moar d2dfb6: OUTPUT«True␤»
FROGGS m: say 8 > 6 > 5
camelia rakudo-moar d2dfb6: OUTPUT«True␤»
FROGGS m: say 8 > 6 > 7 19:45
camelia rakudo-moar d2dfb6: OUTPUT«False␤»
FROGGS see
segomos so..no more 8 > 4 && 8 < 12
FROGGS it is awesome :o)
correct
segomos i've wanted that since the great 90s
retupmoca moritz: Could I get you to look at LWP::Simple PR #29? 19:46
colomon m: say 0 == 0 == 0 19:48
camelia rakudo-moar d2dfb6: OUTPUT«True␤»
FROGGS ohh, you asked about how... 19:50
masak m: say 0 == 0 == 0 == 0 == 0 # you would think eventually Rakudo would get tired of comparing the zeroes, going "I *told* you they're equal already!" 19:51
camelia rakudo-moar d2dfb6: OUTPUT«True␤»
FROGGS the ast is like:
- QAST::Op(call &say) say 0 == 0 == 0
- QAST::Op(chain &infix:<==>) ==
- QAST::Op(chain &infix:<==>) ==
moritz retupmoca: I have looked, but I don't really understand what's going on
FROGGS the inner == gets two elems, and the outer the gets the chained thing and another elem
19:52 rurban joined
FROGGS in nqp is a very nice bit of code that is responsible for the chaining 19:52
vendethiel nqp++ # being nice to code 19:53
colomon hadn't realized that == chained, but is glad that it does
FROGGS I think you can see which infix can do that by looking in the Perl6::Grammar
retupmoca moritz: if the end of the socket read buffer happens to end within the first byte or two of the next chunk header, the chunked-response parsing fails
FROGGS there is a chainy attribute me thinks 19:54
retupmoca moritz: (was causing random/rare test failures for me)
20:12 kaare_ left 20:13 labster joined 20:27 rindolf joined 20:30 Sqirrel_ joined 20:33 Sqirrel left 20:34 Sqirrel joined 20:35 Sqirrel_ left 20:36 benabik left
retupmoca moritz++ # PR merge 20:38
20:38 BenGoldberg joined
lue r: say <A B C> (-) <A C> 20:45
camelia rakudo-{parrot,jvm,moar} d2dfb6: OUTPUT«set(B)␤»
lue ^ I know it's not .perl, but should strings really be unquoted like that in the output? 20:46
FROGGS lue: you got a Set bag, so .gist shows you that 20:48
r: print <A B C> (-) <A C>
camelia rakudo-{parrot,jvm,moar} d2dfb6: OUTPUT«B»
FROGGS s/bag/back/
lue yeah, I'm just not sure about output that looks an awful lot like code not quoting strings 20:49
FROGGS .gist should show you (as a human) what is going on
and I think it does it in this case
lue FROGGS: I realize that, but I still don't like how strings are unquoted. It looks like code, so it seems tempting to use it as such. 20:50
FROGGS hmmmm 20:51
lue r: say set(1,2,3); # it even *is* code!
camelia rakudo-jvm d2dfb6: OUTPUT«set(2, 1, 3)␤»
..rakudo-{parrot,moar} d2dfb6: OUTPUT«set(1, 2, 3)␤»
FROGGS m: say lines.words
camelia rakudo-moar d2dfb6: OUTPUT«No such method 'words' for invocant of type 'List'␤ in block at /tmp/XwAN0jMazR:1␤␤»
20:51 bluescreen10 left
FROGGS m: say lines>>.words 20:51
camelia rakudo-moar d2dfb6: OUTPUT«(signal )» 20:52
FROGGS gah+
m: say lines[0].words
camelia rakudo-moar d2dfb6: OUTPUT«There were three men came out of the West␤»
FROGGS there are no quotes either
lue yes, but nothing about that seems like code :)
FROGGS bah
:P
20:53 rindolf left
lue (I'd prefer something kinda like Match objects, in that they need to convey extra information about themselves, but use corner quotes and generally not-code to do it. I suppose that's not easy to devise for Set though) 20:53
BenGoldberg r: say lines.WHAT
camelia rakudo-{parrot,jvm,moar} d2dfb6: OUTPUT«(List)␤»
colomon m: say lines.map(*.words) 20:56
camelia rakudo-moar d2dfb6: OUTPUT«(timeout)»
colomon m: say lines
BenGoldberg Lines is infinite-ish.
camelia rakudo-moar d2dfb6: OUTPUT«(timeout)»
BenGoldberg m: say .chars for lines 20:57
camelia rakudo-moar d2dfb6: OUTPUT«(signal )41␤25␤37␤24␤0␤55␤26␤37␤24␤0␤40␤35␤38␤22␤0␤42␤32␤46␤19␤0␤43␤26␤42␤28␤0␤43␤28␤48␤30␤0␤47␤26␤33␤23␤0␤43␤25␤48␤38␤0␤42␤28␤42␤32␤0␤35␤30␤43␤27␤0␤0␤0…»
BenGoldberg jp: say .chars for lines
colomon BenGoldberg: how is that possible? 20:58
colomon doesn't think there are that many lines to "John Barleycorn"
raydiak it loops, I'd assume
BenGoldberg For some reason, when it reaches the end-of-file, it's just producing empty strings, instead of ending the list.
colomon p: say lines.map(*.words) 20:59
camelia rakudo-parrot d2dfb6: OUTPUT«There were three men came out of the West Their fortunes for to try And these three men made a solemn vow John Barleycorn must die They've ploughed, they've sewn, they've harrowed him in Threw clouds upon his head And these three men made a solemn vow Jo…»
colomon oh, so moarbug
BenGoldberg p: say .chars for lines
camelia rakudo-parrot d2dfb6: OUTPUT«41␤25␤37␤24␤0␤55␤26␤37␤24␤0␤40␤35␤38␤22␤0␤42␤32␤46␤19␤0␤43␤26␤42␤28␤0␤43␤28␤48␤30␤0␤47␤26␤33␤23␤0␤43␤25␤48␤38␤0␤42␤28␤42␤32␤0␤35␤30␤43␤27␤»
BenGoldberg j: say .chars for lines
camelia ( no output )
BenGoldberg headscratches 21:00
vendethiel r-j is a bit... funny.
BenGoldberg Funny haha, or funny uh oh? 21:01
vendethiel funny WAT
BenGoldberg :)
BenGoldberg ponders a programming language with -Ohumor
21:02 dayangkun left
avuserow thanks for adding me to the perl6 org, moritz++ 21:03
(sadly I'm at work when most others are hacking here)
21:06 kbaker left 21:14 nebuchadnezzar joined 21:18 dayangkun joined 21:20 myfree joined 21:21 myfree left 21:25 kurahaupo_mobile joined 21:26 kurahaupo left 21:36 ajr_ left 21:39 Sqirrel left 21:54 denis_boyun_ left 21:56 go|dfish joined 21:58 kurahaupo_mobile left 22:01 treehug88 left 22:20 d4l3k_ joined, dalek left 22:21 d4l3k_ is now known as dalek 22:26 spider-mario left 22:41 pochi left, pochi joined 22:47 atroxaper joined 22:53 rurban left 23:02 thou left 23:05 xenoterracide_ joined
lizmat .tell nwc10 jnthn and I found the problem that caused t/spec/S17-concurrency/promise.t to fail 23:08
yoleaux lizmat: I'll pass your message to nwc10.
lizmat .tell nwc10 the fix is going to need some thought in ThreadPoolScheduler 23:09
yoleaux lizmat: I'll pass your message to nwc10.
23:14 btyler left 23:22 anaeem1 joined 23:24 rurban joined 23:26 anaeem1 left 23:28 Juerd joined 23:30 rurban left 23:40 bjz joined 23:47 dmol left 23:51 benabik joined 23:52 synopsebot joined, Util joined, [Coke] joined 23:54 Juerd left, Juerd joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!