»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.pugscode.org/ | UTF-8 is our friend! | Rakudo Star Released!
Set by moderator on 29 July 2010.
masak when you want something more advanced than that, you can use (a) macros, (b) a sublanguage, or (c) eval. 00:00
there is very limited support for the former two, so I use eval at present.
jnthn masak: I suspect you should be able to stick stuff into namespaces too.
00:00 Alias_ joined
masak jnthn: could you give me and frooh a proof-of-concept? 00:01
jnthn: does it involve pir:: calls?
frooh well, I can use eval as a shim, but if that's really the recommendation....that's totaly gross
like, long term I mean
masak eval isn't that gross. 00:02
jnthn masak: I meant in ideal Perl 6 rather than what's in Rakudo today.
frooh masak: to me it is :-)
masak what's gross is if you connect user input to eval somehow.
frooh I'm not talking about security
masak but as for the rest, it's just a really powerful tool with a bad rep, just like recursion.
frooh right
which is fine
jnthn masak: Can you go a bit deeper with that point? 00:03
frooh but why would you take access to the symbol table away in p6?
masak jnthn: are you genuinely curious, or just being punny?
jnthn frooh: I suspect it should be along the lines of SomePackage::<&name> := sub { ... }
masak: I thought you'd know me well enough by now. :P
frooh paste.scsys.co.uk/47459
jnthn: the former is what I *thought* would work 00:04
masak jnthn: good, because I don't really have more to say on the issue. :)
frooh jnthn: any idea if that's supported?
masak g'night, #perl6.
jnthn night, masak
safe travles
:-)
*travels
masak likewise.
frooh jnthn: yeah, doesn't seem to be supported 00:06
jnthn No, we don't implement any of the interpolated package stuff yet.
frooh I'm fine leaving it out as this isn't that big of a deal
but I don't think that eval is really a good long term solution
jnthn Something *like* that (but I think not exactly looking that way) should work.
Yeah
frooh it's kindav a giant hammer you know?
jnthn Aye
I agree
We just don't have a good way yet to avoid it. 00:07
frooh sure, and that's fine
and even if you never did
giving us that hammer means we can wrap tools around it
jnthn *nod* 00:08
shirtless Is twigil to make a variable immutable? 00:13
Is there a*
frooh star: eval { die 1 }; say 2; 00:21
p6eval star 2010.07: OUTPUT«Nominal type check failed for parameter '$code'; expected Str but got Block instead␤ in 'eval' at line 1128:CORE.setting␤ in main program body at line 22:/tmp/S8BxE8IkJN␤»
frooh star: try { die 1 }; say 2;
p6eval star 2010.07: OUTPUT«2␤»
tylercurtis shirtless: I think you want "my $foo is readonly;" 00:22
shirtless tylercurtis, how would it look if I had a has %commands =
tylercurtis shirtless: oh. I see. You want an immutable hash. 00:24
shirtless I guess for now it's safe to let it be mutable
tylercurtis Presumably there's a way to do that.
shirtless I'll figure out out eventually when i have some time
tylercurtis I don't know that it is.
shirtless going through the source or what not
tylercurtis shirtless: EnumMap appears to be what you want. perlcabal.org/syn/S32/Containers.html#EnumMap 00:25
rakudo: my EnumMap $foo .= new(a => 5); $foo.perl.say; $foo<a> = 6; $foo<a>.say; 00:26
p6eval rakudo 9b6189: OUTPUT«{"a" => 5}␤Cannot modify readonly value␤ in '&infix:<=>' at line 1␤ in main program body at line 22:/tmp/OigwFODmdv␤»
shirtless cool 00:27
I need more time :-O
tylercurtis Whoa. C99 supports escape sequences in identifiers? 00:36
00:37 [Coke] joined
melte hm. okay, related to the above, how would I call an anonymous sub that's in a hash? e.g. in perl5: &{$table{$item}} 00:47
00:48 quietfanatic joined
quietfanatic That'd just be $table{$item}() 00:49
It's the () at the end that calls subs
not the &
00:50 ruscoekm joined
melte star: my %table = ('a' => sub { say "hello world" }); $table{'a'}(); 00:51
p6eval star 2010.07: OUTPUT«===SORRY!===␤Symbol '$table' not predeclared in <anonymous> (/tmp/dOxWBbXjzo:22)␤»
melte star: my %table = ('a' => sub { say "hello world" }); %table{'a'}();
p6eval star 2010.07: OUTPUT«hello world␤»
melte :)
tylercurtis quietfanatic: are you some sort of ninja or clairvoyant or IRC-log-reader? 00:54
Oh my... apparently, the C99 preprocessor considers 12324.34023_afbjkej43e+342p-234..............342342345 to be a number. 00:58
melte star: my %table = ('a' => sub { 1 }); %table{'a'}() 01:02
p6eval star 2010.07: ( no output )
melte weird. I get an error "Could not find sub &a in main program body at line 1" 01:03
blah, my typo 01:04
pmichaud good evening, #perl6 01:05
...was it something I typed? 01:07
quietfanatic tylercurtis: The latter, I'm afraid.
tylercurtis Good evening, pmichaud.
sorear hello pmichaud 01:08
01:08 ashleydev joined
jnthn o/ pmichaud 01:09
No, just packing ;-)
pmichaud yeah, I did that earlier. I'm at the airport now. 01:10
jnthn :-)
Leaving in 2 and a bit hours
pmichaud scheduled to board in 19 mins 01:11
trying to get last-minute updates via wifi before boarding :)
01:13 phenicsl joined
jnthn :) 01:18
Have a safe flight. 01:19
pmichaud you also :)
jnthn Thanks. :)
Hopefully see you in 16.5 :-)
tylercurtis std: 5++ 01:20
p6eval std 31886: OUTPUT«ok 00:01 114m␤»
melte perl6 -e 'my $a = getc($*IN); $a.perl.say' # gives me "\\x[0]" everytime 01:25
pmichaud getc NYI, I think. 01:26
melte sheeet
pmichaud I'm surprised you don't get an error, tbh 01:27
rakudo: my $a = getc($*IN); $a.perl.say
p6eval rakudo 9b6189: OUTPUT«src/gc/gc_ms.c:1256: failed assertion 'PObj_is_movable_TESTALL(str)'␤Backtrace - Obtained 20 stack frames (max trace depth is 32).␤/home/p6eval//p2/lib/libparrot.so.2.6.0 [0x2ab0b900f203]␤/home/p6eval//p2/lib/libparrot.so.2.6.0(Parrot_confess+0x87)
..[0x2ab0b900f337]␤/home/p6eval//p2…
pmichaud well, that's not helpful.
melte heh
pmichaud star: my $a = getc($*IN); $a.perl.say
p6eval star 2010.07: OUTPUT«"\\x[0]"␤»
pmichaud that's.... surprising.
melte glad I'm not the only one :) 01:28
ingy greetings
melte easier question then: how do I read from $*IN in general? =$*IN, <$*IN>, I haven't hit upon the working syntax yet
pmichaud $*IN.get # read a line
$*IN.lines # read all of the lines
tylercurtis pmichaud: getc appears to be implemented: "my $c = $!PIO.read(1); fail if $c eq ''; $c;". 01:29
pmichaud I see it there.... but obviously it's failing.
tylercurtis Whether it works, may be a different question.
melte thanks. I googled and searched synopses, didn't find the relevant spec. I'm still bad at finding perl6 stuff :(
pmichaud anyway, boarding now -- bbl.
tylercurtis melte: for IO stuff, try S32 IO 01:30
pmichaud melte: yes, I'm not sure getc() is documented for perl6 yet, or that it's even "for real"
anyway, bbl.
ingy \\o
tylercurtis melte: and assume that much of it may be inaccurate or NYI.
cxreg ponders how to instrument callbacks with zavolaj 01:31
tylercurtis cxreg: last time I looked at Gtk (assuming that's still what you're working on), I don't think Parrot had the right callback signature available. Not certain, though. 01:32
01:32 molaf joined
cxreg tylercurtis: it is. the callback is a GClosure* 01:33
tylercurtis: obviously it will take some magic :) 01:34
tylercurtis cxreg: good luck. :) 01:35
ingy is masak nini? 01:37
cxreg tylercurtis: i'm thinking there might be a way of standardizing function pointers and callbacks for C librarys to call HLL code 01:38
although in this case, the fp is buried in another struct :/
01:40 colomon joined
cxreg (maybe such a thing already exists, actually) 01:41
jnthn ingy: nini?
ingy nigh nigh
jnthn oh
yes
It's like, 3:40am here in Sweden. :-)
ingy jnthn: why are you in Sweden? 01:42
teh grrlz?
I just spent the last 24 hours updating the TestML specification: testml.org/specification/language/ 01:43
jnthn ingy: :P
ingy now the grammar is in completely in Pegex
and looks 90% like p6rulez 01:44
jnthn ingy: No, came to join some startup company here that a friend was creating
ingy now I need to write the Pegex syntax grammar in Pegex
jnthn ingy: I kinda work there some of the time, and Rakudo rest of the time :-) 01:45
ingy and once a p5regexp/pcre engine is in rakudo
I can implement a YAML parser once
for p5 and p6 01:46
and every other language I port pegex to
life is about to get really good for ingy and his mini languages
jnthn :-)
cxreg and i wonder if parrot -> nci -> C -> ??? -> parrot -> hll would cause reentrancy issues 01:47
ingy right now the pegex engine is working for TestML parsing, but the grammar is in a funky YAML+Pegex syntax 01:48
bootstrapping is teh awesome
cxreg Pegex?
01:48 orafu joined
tylercurtis cxreg: Parrot has builtin NCI support for some callbacks. So, maybe not, though I don't know how much they've been tested. 01:48
ingy jnthn: that's cool. did you pick up all your stuff and move there? 01:49
jnthn ingy: Yes
cxreg tylercurtis: any pointers where to look for those?
jnthn ingy: I move to furnished appartments
ingy: So I can keep my amount of stuff quite low :-)
ingy jnthn++
jnthn: I need to follow you
tylercurtis cxreg: docs.parrot.org/parrot/latest/html/...l.pod.html is the PDD, though, I don't know how accurate it is.
cxreg tylercurtis++ 01:50
ingy cxreg: Pegex is like 90% of p6rules built over a standard regex engine
cxreg: ie an Acmeist p6rules
cxreg o_O 01:51
ingy cxreg: this is a (the only existing) Pegex grammar for now
testml.org/specification/language/
but it uses most of the features and totally works with a minimum of code 01:52
it's so I can have the same parser technology for projects like TestML in a dozen languages
cxreg is this the bastard love child of your epiphany about parsing plus p6rules?
ingy perl6++ for creating p6rules 01:53
perl6-- for keeping it to itself
cxreg needs a "p6rules" t-shirt
ingy cxreg: yes :)
cxreg rock on
ingy www.pegex.org/ 01:54
needs a little content :)
I have enough to write a spec now I think
01:54 colomon joined 01:56 colomon_ joined 02:00 kid51 joined 02:01 fridim joined 02:03 colomon joined
exodist ingy: I have spent all day pondering writing a prl6 grammars implementation in c or c++, then I check here and see your messages. what are you planning? 02:06
ingy exodist: join #pegex 02:07
02:25 schobes joined
shirtless I feel stupid asking this but I can't figure out how to create a ticket in RT 02:27
02:29 ingy left, ingy joined
shirtless Looks like I have to email it. 02:31
jnthn shirtless: yes, email
(there's no web interface to submit)
02:31 mrsaturns joined 02:32 samlh joined
mrsaturns Good evening everybody 02:32
lue and as far as I can tell, you have to be signed up to submit a ticket (jnthn: am I right here?) 02:33
samlh lue: no
jnthn lue: no
lue that's weird. Before I signed up for the RT, I tried once and it didn't work. 02:34
jnthn lue: YOu can submit
lue: But not manage the tickets.
02:36 LaVolta joined
shirtless rakudo: my @array; @array.^methods.sort(); 02:44
p6eval rakudo 9b6189: ( no output )
shirtless rakudo: @array.^methods.sort();
p6eval rakudo 9b6189: OUTPUT«===SORRY!===␤Symbol '@array' not predeclared in <anonymous> (/tmp/WQd_d0Roge:22)␤»
shirtless try it. On windows it produces duplicate elements after calling sort
Don't know about linux yet.
jnthn rakudo: Array.^methods>>.name>>.sort>>.say 02:46
p6eval rakudo 9b6189:
..OUTPUT«new␤splice␤delete␤perl␤at_pos␤new␤rotate␤pop␤Capture␤elems␤Str␤list␤push␤exists␤unshift␤batch␤ACCEPTS␤eager␤at_pos␤sort␤iterator␤Bool␤Num␤perl␤of␤munch␤shift␤map␤hash␤flat␤fmt␤Str␤Int␤list␤item␤Num␤elems␤fmt␤flip␤lc␤p5chop␤to-radians␤Numeric␤asinh␤acotan␤chr␤cotanh␤acotanh␤comb␤flo…
diakopter heh
jnthn fail
shirtless Is that the exact equivalent of my code?
jnthn rakudo: Array.^methods>>.name.sort>>.say
p6eval rakudo 9b6189:
..OUTPUT«␤ACCEPTS␤ACCEPTS␤BUILD␤BUILDALL␤Bool␤Bool␤CREATE␤Capture␤Capture␤IO␤Int␤Num␤Num␤Numeric␤Numeric␤PARROT␤Seq␤Str␤Str␤Str␤WALK␤WHENCE␤WHERE␤WHICH␤abs␤acos␤acosec␤acosech␤acosh␤acotan␤acotanh␤all␤any␤asec␤asech␤asin␤asinh␤at_key␤at_pos␤at_pos␤at_pos␤atan␤atan2␤atanh␤batch␤bless␤bytes␤c…
jnthn shirtless: Well, it's doing it to the type objects rather than an instance 02:47
shirtless: Anyway, it's not surprising.
shirtless perhaps it could be a windows IO issue
jnthn shirtless: Consider class A { method m { } }; class B is A { method m { } }; B.^methods # two methods called m in the list
And it's just the same here
Some are overrides.
shirtless So dupes could be caused by overrides? 02:48
The problem is that if I do @array.^methods; before the sort, and pipe through unix sort, there are no dupes
jnthn They very likely are
shirtless that's how I know sort() is the culprit
02:49 schobes left
tylercurtis rakudo: Array.^methods>>.name>>.say 02:50
p6eval rakudo 9b6189:
..OUTPUT«at_pos␤new␤splice␤delete␤perl␤at_pos␤sort␤iterator␤Bool␤Num␤perl␤of␤munch␤shift␤map␤hash␤flat␤fmt␤new␤pop␤rotate␤Capture␤elems␤Str␤list␤push␤exists␤unshift␤batch␤ACCEPTS␤eager␤elems␤fmt␤Str␤Int␤list␤item␤Num␤acos␤eval␤round␤chop␤tanh␤split␤match␤exp␤acosh␤words␤cotan␤atan2␤lcfirst␤…
jnthn rakudo: Array.^parents>>.say 02:51
p6eval rakudo 9b6189: OUTPUT«List()␤Iterable()␤Cool()␤Any()␤Mu()␤»
tylercurtis Sure about that? Looks as though at least a few duplicates are there.
diakopter rakudo: Array.^methods>>.name>>.sort>>.say
p6eval rakudo 9b6189:
..OUTPUT«splice␤perl␤delete␤at_pos␤new␤Num␤perl␤of␤munch␤shift␤map␤hash␤fmt␤flat␤new␤pop␤rotate␤Capture␤elems␤Str␤list␤push␤exists␤unshift␤batch␤ACCEPTS␤eager␤at_pos␤sort␤iterator␤Bool␤item␤Num␤elems␤fmt␤Str␤Int␤list␤match␤exp␤acosh␤words␤cotan␤atan2␤lcfirst␤uc␤sec␤cis␤log␤substr␤from-radia…
lue could it be listing multis more than once?
jnthn No, it's just inheritance. 02:52
shirtless How so
diakopter perl of munch shift map
jnthn diakopter: Getting to eager with the >> considered harmful. :-)
shirtless when you don't run the .sort() you see no duplicates
does it interpolate something?
jnthn shirtless: Can you nopaste what you're running and the output you're getting? 02:53
shirtless yeah
rcfox Hey guys, I want to read a binary file, but it keep complaining about malformed UTF-8 strings. 02:54
sorear google is paying masak to implement binary file support 02:56
you can help if you want
jnthn rcfox: Needs :bin option passed to open
tylercurtis shirtless: There are duplicates even without sort.
shirtless tylercurtis, I didn't get them 02:57
I'm doing it again anyway
rcfox jnthn: It didn't help...
lue I can't help but wonder what this binary file is. 02:58
shirtless tylercurtis, oh, yeah there are.
tylercurtis, disregard that bug.
I must have lost the dupes some how the first time I did it when I copied it into an ssh session.
jnthn rcfox: masak++ is working on binary IO at the moment - I know there are some tests in svn.pugscode.org/pugs/t/spec/S16-fi...ndles/io.t but also that there's some more work to go. 02:59
rcfox lue: It's a PNG file, but that shouldn't matter.
(I don't actually want to open it as an image.)
02:59 Thomas_ joined
Thomas_ Hello 02:59
rcfox jnthn: I see... Thanks. 03:00
03:00 jaldhar joined
lue rcfox: you're right, it shouldn't. Just curious :) 03:00
Thomas_ I just started programming in perl this week. Is there anything a newbie can do to help the community? 03:01
rcfox I just wanted to see if I could parse a image file with these new-fangled rules.
Gah, my typing sucks tonight. 03:02
jnthn rcfox: They're not really been aimed at binary files, tbh, more strings, though I guess it may be workable. 03:04
rcfox jnthn: I know, but it was just a crazy idea that I wanted to play with. 03:05
jnthn Thomas_: hi! Yes, probably. :-) You interested in Perl 5 or Perl 6 or both? :-)
03:07 TiMBuS joined
rcfox Is there anything like perldoc for Perl 6? 03:08
jnthn rcfox: I think the closest thing is "grok"
OK, I should go get ready to head off to the airport. :-)
03:09 masonkramer_ joined
jnthn afk for some hours (train, flight, train, another train, trying to find hotel... :-)) 03:09
o/
tylercurtis o\\ 03:10
o/
rcfox Hm. 03:24
Can you use map in a for loop?
rakudo: my @test = 1,2,3; for map { $_+1 } @test { say; } 03:25
p6eval rakudo 9b6189: OUTPUT«===SORRY!===␤Missing block at line 22, near "@test { sa"␤»
rcfox Yeah, that's what I get too.
Oh, I'm using map wrong. 03:27
And apparently, say.
tylercurtis rakudo: my @test = 1, 2, 3; for map(@test, { $_ + 1 }) { .say; } 03:29
p6eval rakudo 9b6189: OUTPUT«No candidates found to invoke for method 'map' on object of type 'Array'; available candidates have signatures:␤:(Mu : &block;; *%_)␤␤ in 'map' at line 1715:CORE.setting␤ in main program body at line 22:/tmp/de6D1IMPhY␤»
tylercurtis rakudo: my @test = 1, 2, 3; for map({ $_ + 1 }, @test) { .say; } 03:30
p6eval rakudo 9b6189: OUTPUT«2␤3␤4␤»
rcfox .say?
Ah, neat.
tylercurtis rcfox: Perl 6 doesn't really do implicit use of the topic like Perl 5 does. Instead you do things like ".foo" which is equivalent to "$_.foo".
kthakore rcfox: oh hai! 03:33
rcfox kthakore: Hi.
kthakore rcfox: what up?
rcfox: how is your project?
rcfox Procrastinating from studying. :P 03:34
kthakore rakudo: $a = 'foo'; $a.say();
p6eval rakudo 9b6189: OUTPUT«===SORRY!===␤Symbol '$a' not predeclared in <anonymous> (/tmp/k0gnR11xqq:22)␤»
kthakore rcfox: cool
rakudo: my $a = 'foo'; $a.say();
p6eval rakudo 9b6189: OUTPUT«foo␤»
rcfox My project is good. We demoed it, and did very well.
kthakore rcfox: video?
rcfox Ah, we should have made a video...
No, there's no video.
I'll probably do a write up after my exams are done. 03:35
kthakore cool 03:38
shirtless ooh: 03:42
rakudo: 1..3.say();
p6eval rakudo 9b6189: OUTPUT«3␤»
shirtless rakudo: (1..3).say();
p6eval rakudo 9b6189: OUTPUT«123␤»
shirtless So which is further along in implementing perl6 pugs or rakudo? 03:44
03:45 drbean joined
LaVolta rakudo: @*INC.perl.say 03:46
p6eval rakudo 9b6189: OUTPUT«["lib", "/home/p6eval/.perl6/lib", "/home/p6eval//p2/lib/parrot/2.6.0-devel/languages/perl6/lib", "."]␤»
03:52 Italian_Plumber joined
frooh 5/3 03:56
rcfox So what's up with the postcircumfix operator?
allbery_b pugs is no longer under active development
rcfox I tried to have it take 2 arguments, and then it tells me it's expecting 3. 03:57
rakudo: class foo { method postcircumfix:<[ ]>($a,$b) { say $a*$b; }}; my $bar = foo.new; $bar[1,2]; 03:59
p6eval rakudo 9b6189: OUTPUT«Not enough positional parameters passed; got 2 but expected 3␤ in 'foo::postcircumfix:<[ ]>' at line 22:/tmp/ThBeJnOXv9␤ in main program body at line 1␤»
rcfox It works with one argument... 04:00
avuserow rakudo: class foo { method postcircumfix:<[ ]>($a,$b) { say $a*$b; }}; my $bar = foo.new; $bar[(1, 2)];
p6eval rakudo 9b6189: OUTPUT«Not enough positional parameters passed; got 2 but expected 3␤ in 'foo::postcircumfix:<[ ]>' at line 22:/tmp/e2cqsBDFIN␤ in main program body at line 1␤»
avuserow rakudo: class foo { method postcircumfix:<[ ]>($a) { say $a.perl; }}; my $bar = foo.new; $bar[1, 2];
p6eval rakudo 9b6189: OUTPUT«(1, 2)␤»
avuserow rakudo: class foo { method postcircumfix:<[ ]>(@a) { say @a.perl; }}; my $bar = foo.new; $bar[1, 2]; 04:01
rcfox Oh, it treats everything as an array?
p6eval rakudo 9b6189: OUTPUT«(1, 2)␤»
avuserow not sure if it's supposed to or not, but it's passing them as one value, from what I can tell
rakudo: class foo { method postcircumfix:<[ ]>($a) { say $a.perl; }}; my $bar = foo.new; $bar.[1, 2]; 04:02
p6eval rakudo 9b6189: OUTPUT«(1, 2)␤»
avuserow rakudo: class foo { method postcircumfix:<[ ]>(($a, $b)) { say $a*$b; }}; my $bar = foo.new; $bar.[1, 2]; 04:04
p6eval rakudo 9b6189: OUTPUT«2␤»
avuserow rakudo: class foo { method postcircumfix:<[ ]>(($a, $b)) { say $a*$b; }}; my $bar = foo.new; $bar.[3, 2];
p6eval rakudo 9b6189: OUTPUT«6␤»
avuserow whoa, Perl 6 is DWIMmier than I recalled
rakudo: class foo { method postcircumfix:<[ ]>(($a, $b)) { say $a*$b; }}; my $bar = foo.new; $bar.[3, 2, 5]; 04:05
p6eval rakudo 9b6189: OUTPUT«Too many positional parameters passed; got 3 but expected 2 in sub-signature␤ in 'foo::postcircumfix:<[ ]>' at line 22:/tmp/EXcSi6yB__␤ in main program body at line 1␤»
avuserow rcfox: doe sthat make sense?
rcfox Oh, hm. 04:06
Yeah.
Unintuitive syntax though...
avuserow not really. it's kind of unpacking it 04:07
rcfox I guess so.
avuserow rakudo: my @a = 1, 2; my ($a, $b) = @a; say $a; say $b;
p6eval rakudo 9b6189: OUTPUT«1␤2␤»
avuserow many functional languages let you do this with pattern matching in assignments.
(note that my terminology is not quite perfect) 04:08
rcfox Hm, is there a way to assign values with this operator?
Or just take in another argument... 04:09
$bar[2,3] = 5
avuserow I imagine there is, but I don't know how 04:10
rcfox Oh, I'm just dumb. 04:15
shirtless which is more popular amongst perl6 devs, undescores or camel case? 04:22
tylercurtis shirtless: for class names, CamelCase, for method/function/variable names, hypenated-names
shirtless tylercurtis, cool thanks 04:23
tylercurtis, i don't want my code to be disliked because it doesn't follow popular convention
:D
04:38 LaVolta joined 04:39 s1n joined
rcfox Hm, does Perl 5 have the xx operator? 04:43
I'm thinking no. 04:45
So I tried to use it in Perl 6, but it doesn't quite behave as I expect it to.
tylercurtis rcfox: Perl 5's "x" operator is both Perl 6's "x" and "xx", I think.
rcfox: what did you expect it to do? 04:46
rcfox Well, I did this, to initialize a 2D array: @.data = [0 xx $width] xx $height;
And it seems to have only evaluated [0 xx $width] once, and then reused that for each xx $height. 04:47
melte isn't that what you want
rcfox So now, if I modify @.data[0][0], it also modifies @.data[1][0]
04:48 phenicsl left
rcfox melte: Is there a better way to do it? 04:49
tylercurtis rcfox: I think xx is supposed to only evaluate its arguments once. maybe ([0 xx $width] for ^$height).
rcfox What is ^? 04:50
tylercurtis rakudo: (^5).perl.say
p6eval rakudo 9b6189: OUTPUT«0..^5␤»
rcfox Oh... 04:51
Hm.
tylercurtis rakudo: .say for ^5
p6eval rakudo 9b6189: OUTPUT«0␤1␤2␤3␤4␤»
shirtless How do I call a method defined inside a class from within that same class? I tried .foo() but had no luck 04:52
tylercurtis shirtless: on self?
shirtless I tried self.foo()
didn't work either
tylercurtis, i just want to invoke a member of the class from inside the constructor 04:53
tylercurtis shirtless: inside the constructor? new or BUILD?
shirtless new
tylercurtis self in new is the class usually.
shirtless self doesn't work 04:54
rcfox Should you play with new?
Try using BUILD.
shirtless I have no idea how. Is it covered in the guide?
rcfox perlcabal.org/~azawawi/html/spec/S1...ILD.t.html 04:55
tylercurtis shirtless: why are you using new, by the way? Unless you're doing something really weird(like a singleton or some such), you should probably either just override BUILD or override new to just call bless with some argument-munging.
rcfox perlcabal.org/syn/S12.html#Construc...ialization
shirtless tylercurtis, I just want to call some functions that initialize the state of the calss
ok cool
tylercurtis shirtless: then you want BUILD. 04:56
shirtless I guess the real docs are the standard
rcfox These Synopses are not easy to read though...
I found the code much more useful.
shirtless I want to avoid swimming in the code at least for the next 2 hours
I'll try my luck with the synopses
thanks
rcfox shirtless: I mean the code that's included in the synopses. 04:57
shirtless oh
melte the good thing is that as the years go on there will be actual other sources out there
shirtless this is a bit difficult for me probably because I don't know perl5
I am hoping that I won't have to learn it too.
maybe I'll learn a bit, but I'm not diving into it. 04:58
tylercurtis shirtless: I don't know Perl 5 either. :) Perl 6 is definitely learnable without Perl 5 knowledge.
rcfox Hm, I had to end up doing push @.data, [0 xx $width] for ^$height; 05:00
But it worked.
05:00 Mowah joined
rcfox cperl-mode really does not like unparenthesized if statements... 05:06
shirtless anyone mind clearing this up for me? nopaste.info/830c3c16e2.html 05:11
05:11 kaare joined
shirtless I tried to call BUILD like I saw in the doc, and got that. 05:11
tylercurtis shirtless: 'foo.new(arg => "bar")' would be to correct way to do that. 05:18
shirtless facepalms
thanks
I must have read that and it went in one eye and out my nose
tylercurtis shirtless: If you want to take positional parameters, you need to override new and do some munging in what you pass to bless. 05:23
shirtless yeah
TiMBuS i get the feeling it's going to be very common for people to trip up on perl6's constructors
05:25 drbean joined 05:27 Su-Shee joined
rcfox rakudo: class foo { method infix:<+>($left,$right) { say "$left, $right"; } }; my $bar = foo.new; $bar+$bar; 05:28
p6eval rakudo 9b6189: OUTPUT«Can't take numeric value for object of type foo␤ in 'Any::Numeric' at line 1348:CORE.setting␤ in 'infix:<+>' at line 6544:CORE.setting␤ in main program body at line 22:/tmp/eGk6eo4xRq␤» 05:29
05:29 phenicsl joined
rcfox ^ What's going on there? 05:29
05:31 PZt joined
rcfox I looked at the synopses, and it looks like I'm doing the same thing... 05:31
TiMBuS can methods be infix? 05:33
rcfox Changing it to sub has the same effect.
Or do you mean that it has to be outside of the class? 05:34
TiMBuS take the sub out of the class and it works
tylercurtis rcfox: also, it should be a multi to avoid clobbering normal infix:<+> 05:35
rcfox: and assuming it's in a module, you'll probably want to export it.
shirtless I am getting a type check error expected Int but got Any instead, and I declared my member to be Int 05:36
what could be wrong?
TiMBuS it's Any if it hasn't been initialized 05:37
rcfox Thanks guys.
shirtless TiMBuS, I initialized it at the top of my class like this: has Int $!response-limit = 1024;
did it not actually get initialized because I did it wrong?
TiMBuS are you getting the error in BUILD
or new
shirtless from a function that is invoked from BUILD
yeah
TiMBuS or any place before that default would be called 05:38
yeah thats why
shirtless oh ok
so I will be to initialize it in BUILD instead
rcfox tylercurtis: Hrm, how do I export it?
TiMBuS it's initialized last, so that you can do things like.. has $.init; has $.thing = $.init*5;
then you can set $.init in BUILD 05:39
tylercurtis shirtless: if you don't touch $!response-limit in BUILD, it should be initializzed to teh default, but I don't think it is in Rakudo at the moment.
rcfox: multi infix:<+> (Foo $a, Foo $b) is export {...}
shirtless ok, so as a workaround I will have to set it in BUILD
05:47 drbean joined 05:54 justatheory joined
lmistura exit 05:58
05:59 uniejo joined 06:00 tadzik joined
tylercurtis github.com/ekiru/nativecallgen is the very early beginnings of my Zavolaj-based bindings generator. And now to sleep! 06:32
Good night, #perl6.
tadzik phenny: tell masak the yapsi README says about running 'make', but there is no Makefile. Also big kudos for it, reading the code is an awesome experience 06:41
phenny tadzik: I'll pass that on when masak is around.
rcfox Is it possible to use a monkey-typed method only in the context of a certain class? 07:01
Or maybe just add a method to a specific object? 07:04
07:05 wayland76 joined 07:10 eternaleye joined
rcfox Ah-ha, I was able to do it with a role. 07:18
ingy greetings 07:20
rcfox Hi. 07:21
ingy hi rcfox 07:27
rcfox How's it going?
ingy rcfox: well! I just figured out how to do some insanity in p5 07:33
it took me a week, but it's so simple
\\o/
rcfox Heh. 07:38
07:40 kiffin joined 07:48 meppl joined 07:50 araujo joined 08:00 Alias_ joined 08:04 asakra joined 08:07 rhebus joined
asakra hi, can you please point me to a blog/article/text that describes the point behind "star"? are this going to be a Rakudo budles? will it we *, **, *3, etc in the future? 08:07
snarkyboojum asakra: rakudo.org/announce/rakudo-star/2010.07 use.perl.org/~pmichaud/journal/39411 08:09
rcfox Woo, I just implemented (naive) matrix multiplication with coderefs (er, I guess they're called Blocks now?) as the matrix elements. 08:16
And the blocks aren't evaluated until you actually print the matrix. 08:17
I had to give the blocks a Num() function so that math wouldn't break. ;) 08:18
asakra snarkyboojum: I've read the first one alredy and now the second one. still the future or "stars" is unclear to me. at some point in the future there will have to be some numbers, or? 08:19
08:19 timbunce joined 08:20 thebird joined
snarkyboojum asakra: as I understand it, there will definitely be more than one, probably monthly to start, and then perhaps every three months once thing settle down. Currently the releases seem to be of the form yyyy.mm - not sure how they'll be named. 08:21
asakra: that's just my understanding from what I've gleaned reading the IRC logs etc 08:22
asakra: is that as clear as mud now? :D 08:25
asakra snarkyboojum: exactly :) this will be a mess once you start asking users what star they use... 08:27
snarkyboojum asakra: well it seems that Rakudo Star is billed as a "distribution", i.e. compiler bundled with docs/book, useful modules etc, so that's how I look at it, "Rakudo Star" is the name of the distro, and the version numbers are whatever they're decided to be.. doesn't really worry me a great deal, as long as it's obvious which is earlier than another :) 08:28
asakra: I think there are plans to improve --version to show that it is a "Rakudo Star" or not, so it'll be fairly obvious from that imo 08:29
08:31 drbean joined
asakra snarkyboojum: i wanted to write a message to our pm group about rakudo *, then i stoped so that i make it clear what the star actually means. so it is a bundle with an installer and some useful stuff, ready to use, right? 08:33
moritz_ right 08:34
it's compiler + modules + docs
08:45 daxim joined 08:50 Trashlord joined 08:51 darkop joined, jfried joined 09:09 clintongormley joined
snarkyboojum asakra: yep - remembering that it's for early adopters who want to use an early but "usable" distribution which implements large chunks of the Perl 6 spec :) 09:12
or something :)
09:12 tadzik joined 09:18 err_or joined 09:21 err_or left 09:23 muru joined, walfisch joined 09:33 d4rt joined 09:36 [1]Casan joined
bbkr good morning #perl6 09:45
09:45 plol joined
moritz_ good localtime to you too 09:45
szbalint splendid, indeed :) 09:46
wayland76 Does anyone know if there are plans for a gtk-rakudo? 09:53
moritz_ hasn't heard of any
frettled What would a gtk-rakudo do, come with annoying pop-ups for each error message? ;) 09:55
moritz_ lol 09:56
wayland76 I was thinking more along the lines of allow you to write Rakudo programs that display things in windows
szbalint can attest that moritz_ did laugh out loud :)
09:56 vpm joined
frettled :) 09:57
moritz_ wayland76: should be possible with Zavolaj
ruoso wayland76, gtk would be a very interesting bind to make
because it implements its own object orientation
frettled Hmm, "Zavolaj", that has to be one of jnthn's projects. ;)
wayland76 googles zavolaj
moritz_ otoh it doesn't support callbacks yet, afaict
which is kinda its biggest limitation 09:58
ruoso so the "only" thing you need is to properly implement the metaclass (a variant of ClassHOW)
frettled Maybe GTK requires support in Parrot?
ruoso frettled, it certainly do...
bbkr std: my $foo = "foo"; say $foo R~= "foo"
p6eval std 31886: OUTPUT«ok 00:01 118m␤»
frettled It's definitively both interesting and useful, though.
09:59 timbunce joined, jferrero joined
wayland76 jnthn++ for Zavolaj :) 09:59
tadzik frettled: why should it require Parrot suppoer? 10:00
*support
10:00 Italian_Plumber joined
bbkr rakudo: my $foo = "foo"; say $foo R~= "foo" 10:00
p6eval rakudo 9b6189: OUTPUT«Cannot modify readonly value␤ in '&infix:<=>' at line 1␤ in 'reverseargs' at line 130:CORE.setting␤ in main program body at line 1␤»
ruoso frettled, moritz_, wayland76, but gtk support probably needs something more specific...
bbkr above behavior is correct and expected?
tadzik why should it?
moritz_ ruoso: you mean some kind of OO wraping/morphing 10:01
tadzik It's just a C library, zavolaj may even handle it right now
to some extend
ruoso moritz_, yes... you just need to implement the metaclass translating glib oo to perl 6 oo
moritz_ rakudo: my $foo = "foo"; say $foo [R~]= "foo"
dalek kudo: f8b73df | moritz++ | src/core/Match.pm:
Match.new() (no subcaptures yet)
ruoso in gtk2-perl (perl5) you can extend a gtk class in perl and use its objects as an argument to a c call
10:01 gbacon joined
p6eval rakudo 9b6189: OUTPUT«===SORRY!===␤Confused at line 22, near "say $foo ["␤» 10:01
moritz_ ruoso: why? you can just call methods on it
ruoso moritz_, it won't give you "subclassing a gtk class and sending an instance of your subclass to a gtk method" 10:02
moritz_ bbkr: I think it's correct.
bbkr: 'abc' ~= $x also fails
ruoso: yes, but I don't think a basic wrapper needs that
a more complete one, yes
ruoso but doing a basic wrapper will probably be harder then doing the OO one 10:03
gtk is OO
it's just glib OO
bbkr moritz_: so what should i do with rt.perl.org/rt3/Ticket/Display.html?id=64818 ? reject, because second case atempt is incorrect (it modifies read-only value)
moritz_ bbkr: test if it works with a variable on the $HS 10:04
bbkr $HS ?
moritz_ sorry, RHS
>./perl6 -e 'my $x = "a"; my $y = "b"; $x R~= $y; say $y' 10:05
a
./perl6 -e 'my $x = "a"; my $y = "b"; $x R~= $y; say $x'
a
not good
the ticket is valid, even if the test case provided isn't
bbkr moritz_: thanks, leaving unchanged
moritz_ bbkr: I'll comment on it 10:06
bbkr moritz_++
moritz_ rakudo: my ($x, $y) = <a b>; $x R~= $y; say "$x $y"
p6eval rakudo 9b6189: OUTPUT«a a␤»
10:07 briang joined 10:08 vpm joined 10:09 amkrankruleuen joined
moritz_ what's the difference between 'require' and 'need'? 10:11
is 'require' run-time? 10:12
ah yes, it is 10:15
10:20 Alias joined 10:21 azert0x joined
bbkr std: constant ($a, $b) = 1,2; 10:23
p6eval std 31886: OUTPUT«[31m===[0mSORRY![31m===[0m␤Malformed constant at /tmp/xCgHKZ1I4w line 1:␤------> [32mconstant [33m⏏[31m($a, $b) = 1,2;[0m␤ expecting any of:␤ trait␤ type_declarator␤Parse failed␤FAILED 00:01 113m␤»
10:24 wamba joined
bbkr that constant definition has changed since rt.perl.org/rt3/Ticket/Display.html?id=69740 , is there any new syntax now for compact constant declaration? 10:24
moritz_ constant a = 3;
my constant $x = 5; 10:25
iirc
std: constant a = 5; a
p6eval std 31886: OUTPUT«ok 00:01 116m␤»
moritz_ std: my constant $x = 5; $x
p6eval std 31886: OUTPUT«ok 00:01 117m␤»
bbkr moritz_: ticket refers to declaring multiple constants in one line. which was valid then, but it is not now. 10:26
10:26 barika joined
moritz_ std: my contant ($x, $y) = (1, 2); say $x 10:26
p6eval std 31886: OUTPUT«[31m===[0mSORRY![31m===[0m␤Malformed my at /tmp/hcnyL11DxB line 1:␤------> [32mmy [33m⏏[31mcontant ($x, $y) = (1, 2); say $x[0m␤ expecting any of:␤ scoped declarator␤ typename␤Parse failed␤FAILED 00:01 114m␤»
moritz_ std: my constant ($x, $y) = (1, 2); say $x
p6eval std 31886: OUTPUT«[31m===[0mSORRY![31m===[0m␤Malformed constant at /tmp/6XjhLVJODB line 1:␤------> [32mmy constant [33m⏏[31m($x, $y) = (1, 2); say $x[0m␤ expecting any of:␤ trait␤ type_declarator␤Parse failed␤FAILED 00:01 114m␤»
Su-Shee *hihi* "
Reading #perl6 documentation is like reading a fiction novel. Reading #python3 documentation is like reading the newspaper."
bbkr so what status should ticket have? rejected, because STD has changed? 10:27
moritz_ yes
wayland76 Hey all. I asked the gtk-rakudo question on the gtk-perl channel on the Gnome IRC server, and I'll paste the answer below.
moritz_ bbkr: and please make sure the tests are gone
bbkr moritz_: thanks. i'll also remove fudged test
reading my mind ;)
wayland76 I asked: Does anyone know if there are plans to do a gtk-rakudo, possibly using Zavolaj? 10:28
They answered: I doubt it. We need a C API to create bindings for gobject-introspection. We might have to go down to parrot to implement callables :-/
That might be what ruoso was saying 10:29
bbkr star: constant a = 5; # this breaks S04-declarations/constant.t, i'll create another ticket for it and fudge test in test suite 10:32
p6eval star 2010.07: OUTPUT«===SORRY!===␤Constant type declarator not yet implemented at line 22, near "= 5; # thi"␤»
moritz_ yes, constant mostly NYI 10:33
gfldex i happen to think about a perl6shell last night 10:42
it even made me start a blog! gfldex.wordpress.com/2010/08/02/thi...erl6shell/
tadzik oh, you reminded me of size parameter for File::Find 10:45
moritz_ what should 10:48
require foo;
do?
sorear according to S11, it's an error 10:50
require needs an explicit import list, so that the symbols to be stubbed are known at compile time before the pads are frozen
if you want to get really fancy, :OUR<:DEFAULT> should be legal (but not the default) 10:51
sorear out
moritz_ sorear++ # thanks
10:53 PacoLinux joined
moritz_ rakudo: use Foo <a b c>; 10:55
p6eval rakudo f8b73d: OUTPUT«===SORRY!===␤Unable to find module 'Foo' in the @*INC directories.␤(@*INC contains:␤ lib␤ /home/p6eval/.perl6/lib␤ /home/p6eval//p1/lib/parrot/2.6.0-devel/languages/perl6/lib␤ .)␤»
10:55 PhatEddy joined
PhatEddy rakudo: class T {has $!a; method a {return $!a}}; my $x = T.new; $x.a = 'abc'; say $x.a 10:57
p6eval rakudo f8b73d: OUTPUT«abc␤»
PhatEddy If I wanted to prevent the assignment to the private attribute - how would I do it?
Sorry - not doing much p6 lately 10:58
moritz_ rakudo: class T {has $.a; method a {return $!a}}; my $x = T.new; $x.a = 'abc'; say $x.a
p6eval rakudo f8b73d: OUTPUT«===SORRY!===␤A method named 'a' already exists in class 'T'. It may have been supplied by a role.␤»
10:59 synth joined
moritz_ it's a bug that rakudo allows assignment to the return value by default 10:59
rakudo: class T {has $.a}; my $x = T.new; $x.a = 'abc'; say $x.a
p6eval rakudo f8b73d: OUTPUT«Cannot modify readonly value␤ in '&infix:<=>' at line 1␤ in main program body at line 22:/tmp/sPR6zXR4Yi␤»
PhatEddy moritz: we looked at the 'method already exists' issue with overriding the accessor yesterday ... 11:00
moritz_ do you want/need a custom accessor that also does other stuff?
PhatEddy If possible ... yes - trying to do some work on URI module. Thx. 11:01
szabgab I wonder if HTML::Template works? 11:14
I tried o load it into memory and it blows up
and masak is not around I think 11:15
anyone knows if there is a working templating system for perl 6? 11:16
tadzik tried this one from Web.pm, szabgab?
szabgab is there a templating system within the Web project? 11:17
tadzik I think so
szabgab looking
11:18 am0c joined
szabgab hmm, I can't find it 11:19
tadzik let me see 11:20
zby I there - I am trying out the HTTP daemon example - but I get "Method 'method' not found for invocant of class 'HTTP;Request'"
baest szabgab: there is this one, was this the one you've tried: github.com/masak/html-template/tree/master/t/ 11:21
-/t/
tadzik zby: oh hi :)
zby tadzik: czesc
tadzik zby: if you're working on WebNano you might want to try HTTP::Server:Simple::PSGI 11:22
not in proto, but it's on Github
szabgab baest: I was trying to use that one but it is broken now
zby thanks - but not yet, just trying out
11:22 jhuni joined
zby this is for both the examples in HTTP::Daemon 11:24
tadzik HTTP::Daemon is in Star, no? 11:25
bbkr std: say "foo" ~~ /<[d..b]>? foo/ 11:26
p6eval std 31887: OUTPUT«ok 00:01 119m␤»
bbkr rakudo: say "foo" ~~ /<[d..b]>? foo/
p6eval rakudo f8b73d: OUTPUT«foo␤»
bbkr are reversed ranges officially allowed in spec? 11:27
tadzik The last commits are from 2009, not even Configure.pl runs for me, maybe it's just outdated and incompatibile now.
zby tadzik - yes
tadzik I wonder how it got into Star 11:28
zby I just cut and pasted the examples from modules/mainline/lib/HTTP/Daemon.pm
tadzik zby: try HTTP::Server::Simple maybe, looks like HTTP::Daemon is a bit outdated now 11:29
Rakudo is changing, some things written in the past tend to not work now
zby OK - I will - but was that enough of a bug report? 11:31
can you recreated the problem?\\
tadzik looks so, HTTP::Daemon does not work for me at all 11:32
sounds reasonable for a Star bug, imho it should be removed before it's working again
hmm. use File::Find; find(dir => '.') gives me "Not enough positional parameters passed; got 0 but expected at least 1" -- what can cause this? 11:33
oh ok, it's just REPL not printing backtraces 11:35
snarkyboojum what is modules/mainline/* ? 11:36
11:36 colomon joined
bbkr std: say 1.++( 11:36
p6eval std 31888: OUTPUT«[31m===[0mSORRY![31m===[0m␤Unable to parse argument list at /tmp/4KPoL3z6Bx line 1:␤------> [32msay 1.++[33m⏏[31m([0m␤Couldn't find final ')'; gave up at /tmp/4KPoL3z6Bx line 1 (EOF):␤------> [32msay 1.++([33m⏏[31m<EOL>[0m␤ expecting argument list␤Parse
..failed␤FAILED…
bbkr std: say 1.++()
p6eval std 31888: OUTPUT«ok 00:01 115m␤»
tadzik snarkyboojum: HTTP::Daemon
snarkyboojum tadzik: right, and Test::Harness and Test::Differences 11:37
but why under mainline.. i.e. what does it mean? :)
tadzik No idea :)
snarkyboojum I'm betting mberends knows :) 11:38
11:38 asakra left 11:40 spacebat joined, barika joined, pnu joined, rcfox joined 11:41 au|zzz joined, molaf joined
tadzik nopaste.snit.ch/22518 -- why a different result? 11:41
cognominal rakudo: sub a( %h?, *%h1) { say "{%h.defined}" }; a( :b ) 11:42
p6eval rakudo f8b73d: OUTPUT«Method 'defined' not found for invocant of class 'Hash'␤ in 'a' at line 22:/tmp/fxjlHLAAea␤ in main program body at line 22:/tmp/fxjlHLAAea␤»
tadzik star: say [1, 2, 3] == [2, 1, 3] 11:43
p6eval star 2010.07: OUTPUT«1␤»
tadzik is it ok?
cognominal rakudo: sub a( %h?, *%h1) { say {%h, %h1}.keys }; a( :b ) # I would have expected this code to behave 11:44
p6eval rakudo f8b73d: OUTPUT«Method 'HOW' not found for invocant of class 'Hash'␤ in '!STORE' at line 1␤ in 'circumfix:<{ }>' at line 598:CORE.setting␤ in 'a' at line 22:/tmp/mc0TDvBevx␤ in main program body at line 22:/tmp/mc0TDvBevx␤»
daxim I want to package rakudo-star-2010.07 and noticed that building against an external parrot fails because parrot-config item revision == 0 11:51
is this known already?
here's a delicious pasta: paste.scsys.co.uk/47478 11:53
tadzik same without --parrot-config? 11:54
szabgab jnthn: I wonder if on windows when using PERL6LIB, if I should separate the entries using : as on linux or with ; as I think is done on windoows
right now it seems rakudo is splitting on : while at leastthe PATH in my computer uses ; for separators 11:55
daxim yes, same without --parrot-config
bbkr tadzik: looks OK. it forces num context on arrays, same as +[1,2,3]. so [1, 2, 3] == [2, 1, 3] checks if both arrays have the same number of elements
tadzik oh, right
bbkr tadzik: are you from Poland ?
tadzik so how do I check elements equality?
bbkr: yep
oh, eq works fin 11:57
12:00 beejeebus joined
bbkr tadzik: if you want to check equality with order use smartmatch: [1,2,3] ~~ [1,3,2] 12:00
tadzik bbkr: looks like eq works too. But smart matching looks more elegant indeed
bbkr tadzik: smartmatch does not stringify. so it's faster 12:01
tadzik star: use Test; is [1, 2, 3], [1, 2, 3], 'foobar';
p6eval star 2010.07: OUTPUT«ok 1 - foobar␤»
tadzik star: use Test; is [1, 2, 3], [2,1, 3], 'foobar';
p6eval star 2010.07: OUTPUT«not ok 1 - foobar␤# got: '1 2 3'␤# expected: '2 1 3'␤»
tadzik bbkr: even better, thanks
12:02 ruoso joined 12:03 envi^home joined
bbkr and if you want to check that array contains only specific values then junctions are useful: [1,3,2,3,2] ~~ any(1,2,3) 12:03
tadzik yeah, I know about Junctions 12:04
hmm, assigning lazy list to an array stringifies the whole list into just one array element. Weird 12:05
bbkr tadzik: example please 12:06
snarkyboojum star: say [1,2,3] == [1,'2',3]
p6eval star 2010.07: OUTPUT«1␤»
snarkyboojum star: say [1,2,3] eqv [1,'2',3] 12:07
p6eval star 2010.07: OUTPUT«0␤»
snarkyboojum star: say [1,2,3] ~~ [1,'2',3]
p6eval star 2010.07: OUTPUT«0␤»
tadzik hmm, or maybe not
colomon star: say [1,2,3] == [1,'blueberry',3]
p6eval star 2010.07: OUTPUT«1␤»
snarkyboojum best not to use ==
12:07 bluescreen joined
bbkr == is OK. as far as you remember it forces numeric context on array and thus means 'compare number of elements' 12:08
star: say +[1,2,3]
p6eval star 2010.07: OUTPUT«3␤»
tadzik hmm, this is strange anyway 12:09
snarkyboojum oh didn't read the history properly
tadzik oh, it just created an array with the only element being the lazy list
bbkr tadzik: please paste your example
snarkyboojum goes back into his hole :)
tadzik bbkr: wklej.org/id/371733/ 12:10
12:10 amkrankruleuen joined 12:15 colomon joined, timbunce joined
bbkr tadzik: use my @t = @(find :dir<tools>); so List is not flattened into one element. 12:16
tadzik oh right, thanks bbkr 12:17
snarkyboojum I just realised the point I was trying to make was about eq not == ;) 12:18
star: [1,2,3] eq [1,'2',3]
p6eval star 2010.07: ( no output )
snarkyboojum star: say [1,2,3] eq [1,'2',3] 12:19
p6eval star 2010.07: OUTPUT«1␤»
snarkyboojum goes back into his hole again
12:20 drbean joined
szabgab phenny: tell masak I am trying to fix HTML::Tempolate that seems to be totally broken, see you in Pisa 12:20
phenny szabgab: I'll pass that on when masak is around.
moritz_ has proper internetz!
szabgab: when are you coming? 12:21
bbkr star: say ~[1, "2", 3] # snarkyboojum - eq compares stringified values. and stringified [1, "2", 3] is the same as [1, 2, 3]
p6eval star 2010.07: OUTPUT«1 2 3␤»
moritz_ sorry, I always forget the times
szabgab I am at LV Airport
shoudl arrive at midnight
TLV airport
moritz_ will you stay at the venue hotel?
szabgab nope
moritz_ pity
12:22 colomon joined
szabgab yes 12:22
moritz_ (not beccause it's so great, but because I won't see you for breakfeast)
12:22 dolmen joined
szabgab will you have a hackathon tomorrow? 12:22
tadzik moritz_: any idea who included HTTP::Daemon in Star? 12:23
moritz_ don't know.. maybe it was me :-) 12:26
szabgab: yes
12:27 PhatEddy left
tadzik moritz_: does it work for you? 12:28
moritz_ tadzik: haven't tried it
tadzik moritz_: looks like it does not work at all :?
:/
moritz_ tadzik: maybe mberends knows, he's been working on HTTP server stuff 12:29
12:29 beejeebus left 12:34 envi^home joined
szabgab moritz_: is the hackathon in your hotel or at the venue? 12:35
moritz_ szabgab: they are the same
szabgab wow, 5 test are already passing in HTML::Template! 12:36
so I'll try to come by and join you guys
moritz_ \\o/
szabgab lovely, I get an error " 12:37
moritz_ std: use Test <plan ok>;
p6eval std 31888: OUTPUT«Compiled ./lib/Test.pm6␤ok 00:01 115m␤»
szabgab Could not find sub &escape
and the comment in the code is "Rakudo segfaults here"
moritz_ used to be "could not find non-existant sub &escape" :-)
szabgab I think it is just natural that it will segfault if the sub is not implemented yet :) 12:38
moritz_ szabgab: some of the web stuff is pretty old, when rakudo was *very* immature
szabgab anyway, I am hungry now
moritz_: is there any working templating system you know about?
to generate some static html files 12:39
moritz_ szabgab: say()
aka "no"
szabgab say "no" 12:40
now boarding
moritz_ have a safe and pleasant flight
szabgab I hope I get food and then I can hack some more on this :)
moritz_ where this = ?
szabgab HTML::Template
and then my slide generator
moritz_ ah 12:41
snarkyboojum bbkr: yep - that's what I was pointing out
snarkyboojum sleeps
cosimo String::CRC32 ported to Perl 6. \\o/ 12:45
github.com/cosimo/perl6-string-crc32 12:46
Guest93426 Rakudo mentioned in www.h-online.com/security/news/item...48464.html
Whatever The H Week is...
moritz_ it's the English offspring of heise.de
cosimo++
cosimo would be cool to move it to PIR though 12:47
cache::memcached uses string::crc32, so I needed it :)
i learned quite a few operators +>, +&, +^, and friends 12:48
tadzik hmm. Separating .pm and .pod seems like a good idea to me
not in all cases probably
cosimo i have tons of questions on perl6 pod 12:49
moritz_ cosimo: discuss them with masak++ when he's online
frettled cosimo: I'm sure someone will weigh in with their answers ;)
moritz_ cosimo: are you coming to YAPC::EU?
cosimo moritz_: unfortunately, no 12:50
moritz_ :(
cosimo would have been nice to meet a few perl6 hackers...
moritz_ indeed
pmichaud, jnthn, masak, szabgab and a few others are coming 12:51
and I'm already there :-)
cosimo nice and warm, I guess 12:52
moritz_ we have air condition :-) 12:55
Su-Shee don't forget to get icecream while you're in italy ;) 12:57
szbalint Su-Shee: are you coming to YAPC::EU? 12:58
13:00 drbean joined 13:01 drbean_ joined
sjn cosimo, frettled: Oslo.pm meeting tomorrow is about Rakudo. Wanna come and do a Show & Tell? :) 13:02
moritz_ you can have live IRC stream from the Rakudo hackathon in Pisa :-)
cosimo sjn: i guess we need a specific topic, otherwise it would be a bit "dispersive" (if that's english) 13:03
[Coke] twitter.com/ssutch/status/20098567855
(from #perl)
cosimo: "scattered" is probably more idiomatic. 13:04
frettled disentangled
;)
tadzik well, at least it's a science fiction. And not as boring as a newspaper
frettled (actually, I'd use "disjointed" or something like that, I think) 13:05
sjn cosimo: "Show and Tell" is a good topic, isn't it? :)
bbkr what does using token name with dot before mean? like <.tag_start> in github.com/masak/html-template/blob...Grammar.pm ?
[Coke] frettled: +1
moritz_ bbkr: it surpresses capturing 13:06
bbkr moritz++
kthakore morning moritz_ 13:07
moritz_ oh hai
kthakore moritz_: I had a few questions for ya
moritz_: Are there any bug reports for learn.perl6.org thing?
moritz_ kthakore: I don't know of any learn.perl6.org thing 13:08
kthakore erm ... hold on looking for github.com/moritz/try.rakudo.org
moritz_: sorry just woke up
frettled sjn: I think I would be a poor presenter right now, I'm way too scatter-brained these days.
moritz_ kthakore: you mean besides "it's not running yet"? 13:09
kthakore moritz_: :P
moritz_: ok
but I don't even know how to ... install this stuff 13:10
kthakore sucks at web stuff
sjn frettled: too scattered to point at something and tell "this is cool"? :)
tadzik kthakore: it's not to install
kthakore: this is about to be working one day, on the web
moritz_ right; the idea is that there will be a central installation
kthakore I know
moritz_ it's too hard to automate the install
kthakore hmm?
moritz_ because it requires actions under different user names
kthakore so how can I help? 13:11
moritz_ for security reason
kthakore I see
moritz_ kthakore: talk to ash_ when he's online, he was the one last working on it
kthakore will do! o\\
frettled sjn: I suspect it includes a bit more than drawing the Rakudo logo on a blackboard/whiteboard, saying «this is cool», and stepping down :)
tadzik frettled: this might actually be tempting. Stepping up, saying "sorry, I'm too busy hacking Perl 6 to talk about it, but you're welcome to join me" and step down :) 13:12
frettled hee-hee
That's a cool one.
kthakore moritz_: ok. Another thing
13:13 Axius joined
kthakore moritz_: I wanted to get your permission to use perl6 book to make SDL::Manual 13:13
moritz_: sdlperl.ath.cx/SDL_Manual.letter.pdf
moritz_: chromatic already said ok
moritz_ kthakore: it were pronik and lucs on #perl6book who did most of the layout stuff... if you need another license than the one from the p6 book, talk to those two folks 13:14
kthakore ok 13:15
moritz_ (it's currently a CC share-alike non-commerical, but we want to drop the non-commercial at some point)
kthakore ok
frettled sjn: I'll think a bit about it. I hope cosimo will think a bit about it, too, maybe we can find a nice way to segway out of it ;) 13:16
sjn hehe 13:17
13:17 macroron joined 13:18 timbunce_ joined
[Coke] std: barf() ?? 1 :: 3 13:24
p6eval std 31888: OUTPUT«[31m===[0mSORRY![31m===[0m␤Please use !! rather than :: at /tmp/O3sVxpM7qm line 1:␤------> [32mbarf() ?? 1 :[33m⏏[31m: 3[0m␤ expecting any of:␤ coloncircumfix␤ signature␤Undeclared routine:␤ 'barf' used at line 1␤Parse failed␤FAILED 00:01 116m␤»
13:25 ashleydev joined
cosimo sjn: where's the meeting? 13:27
sjn cosimo: at Nordaaker 13:28
cosimo: check your Oslo.pm inbox :)
[Coke] rt.perl.org/rt3/Ticket/Display.html?id=76856 - can someone claim/apply that patch?
moritz_ is that the cygwin Makefile.in thing?
[Coke] cjm++
moritz_: no, it fixes up unicode during file gen.
(we're assuming utf8 instead of being explicit in at least one case.) 13:29
moritz_ ok, will test that after the Makefile.in patch
13:32 wtw joined, drbean joined 13:33 Holy_Cow joined
wtw hi :) i just played around a bit with rakudo star. am i right in guessing that "custom meta operators" don't work (yet?)? 13:33
i tried: 13:34
multi sub infix:<!>($x, $y) { return $x + $y;
}
say [!] 1,2,3,4,5;
but get
===SORRY!===
Could not find sub &infix:<!>
moritz_ rakudo: our multi sub infix:<!>($x, $y) { return $x + $y; }; say [!] 1, 2, 3, 4, 5
p6eval rakudo f8b73d: OUTPUT«15␤» 13:35
moritz_ wtw: yes, it's a well-known issue with visibility - by default subs are only stored in the lexical scopes
wtw: so the routine that handles [!] can't look it up yet
wtw ah ok :)
moritz_ wtw: with 'our multi sub ...' you can work around it
wtw cool, i love the kind of flexiblity this provides :) 13:36
moritz_ too
wtw yep works, thx a lot :)
moritz_ there's a build-in which should do the lexical lookup in the caller, but it's not yet implemented
it's called 'lift'
13:36 REPLeffect joined
wtw i already love p6 even if it's not fully there yet :D 13:38
moritz_ too
dolmen cosimo?
[Coke] do we want a Perl6 Bug Summary email like p5p gets? 13:39
parrot was getting one for a while when we were on RT.
dolmen rakudo: my $x ::= 3; $x = 5; 13:40
p6eval rakudo f8b73d: OUTPUT«Cannot modify readonly value␤ in '&infix:<=>' at line 1␤ in main program body at line 22:/tmp/FnLJUy8Syz␤»
dolmen cosimo: in String::CRC32, you @CRC_TABLE should be defined as readonly with ::= 13:41
snarkyboojum moritz_: lastest change to the star README says "If you just want to use Rakudo Star, please download a pre-built tar.gz from github.com/perl6/book/downloads ***" should that perhaps refer to github.com/rakudo/star/downloads?
dalek kudo: 65eb876 | moritz++ | src/core/Any-list.pm:
attempt to produce more awesme error message when you do map { hash => 1}
kudo: 8f8c519 | rurban++ | build/Makefile.in:
Fix installation on cygwin
kudo: 1a5d4a3 | (Christopher J. Madsen)++ | build/gen_core_pm.pl:
Explicitly request :utf8 layer in gen_core_pm.pl to fix RT #76856
13:42 d4rt joined 13:46 LaVolta joined 13:47 proller joined 13:54 mjk joined 13:55 _jaldhar joined
mjk I execute a demo in <UsingPel6-draft.pdf> 13:55
dolmen cosimo: I've submitted a review on GitHub for String::CRC32
cosimo: github.com/cosimo/perl6-string-crc3...d65d18665d
mjk In pages 33, the sub is "order-steak", but is say ERROR! 13:56
sub order-steak($how = 'msg'){
say "$how";
}
and then execute:order-steak() 13:57
will raise a ERROR
if execute &order-steak(),it is OK
13:58 plobsing joined
mjk or rename the name of sub , ie aorder-steak(), it's ok 13:59
or delete 'o' (rder-steak()), it's ok
dolmen rakudo: my $x-3 = 2; 14:00
p6eval rakudo f8b73d: OUTPUT«Use of uninitialized value in numeric context␤Cannot modify readonly value␤ in '&infix:<=>' at line 1␤ in main program body at line 22:/tmp/r2znVFk992␤»
14:00 redicaps joined
dolmen rakudo: sub x-3 { }; 14:00
p6eval rakudo f8b73d: OUTPUT«===SORRY!===␤Confused at line 22, near "sub x-3 { "␤»
dolmen rakudo: sub x-3 { 5 };
p6eval rakudo f8b73d: OUTPUT«===SORRY!===␤Confused at line 22, near "sub x-3 { "␤»
dolmen std: sub x-3 { 5 }; 14:01
p6eval std 31888: OUTPUT«[31m===[0mSORRY![31m===[0m␤Malformed block at /tmp/MDwN4F7SZx line 1:␤------> [32msub x[33m⏏[31m-3 { 5 };[0m␤ expecting any of:␤ new name to be defined␤ routine_def␤ trait␤Parse failed␤FAILED 00:01 114m␤»
dolmen std: sub x-a3 { 5 }; 14:02
p6eval std 31888: OUTPUT«ok 00:01 114m␤»
dolmen rakudo: sub x-a3 { 5 };
p6eval rakudo f8b73d: ( no output )
[particle] less than awesome error messages there
x-3 is not a valid identifier
14:02 _jaldhar joined
[particle] std: sub x-3 {}; 14:03
p6eval std 31888: OUTPUT«[31m===[0mSORRY![31m===[0m␤Malformed block at /tmp/d3_rU0Avjh line 1:␤------> [32msub x[33m⏏[31m-3 {};[0m␤ expecting any of:␤ new name to be defined␤ routine_def␤ trait␤Parse failed␤FAILED 00:01 114m␤»
mjk Can anyone point out why?
[particle] oh, missed that one....
dolmen std: sub order-anything { }; my $x = order-anything();
[particle] yes, hang on...
p6eval std 31888: OUTPUT«ok 00:01 117m␤»
dolmen rakudo: sub order-anything { }; my $x = order-anything();
p6eval rakudo f8b73d: ( no output )
cosimo dolmen: saw your comments, thanks!
[particle] from STD.pm6: token ident { <.alpha> \\w* } token apostrophe { <[ ' \\- ]> } token identifier { <.ident> [ <.apostrophe> <.ident> ]* } 14:04
apostrophe must be followed by alpha
mjk dolmen:is ok! but why after renamed, the sub is OK? 14:05
14:05 irclogbot_backup joined
mjk dolmen: and this sub isn't return value 14:06
dolmen mjk: this looks like a bug. Maybe 'order' is special token in the grammar.
mjk: can you reproduce it here?
[particle] it's likely a rakudo parser bug
snarkyboojum star: sub ord {}; ord();
p6eval star 2010.07: ( no output )
kthakore moritz_: the bot works now 14:07
moritz_: :P
[particle] rakudo: sub order-steak ($how = 'msg') { say $how; }; order-steak()
mjk dolmen: ?
p6eval rakudo f8b73d: OUTPUT«msg␤»
[particle] star: sub order-steak ($how = 'msg') { say $how; }; order-steak()
p6eval star 2010.07: OUTPUT«msg␤»
[particle] i don't see an error 14:08
mjk dolmen: do you need me to do some thing?
snarkyboojum for some reason it's reproducable in a file with sub ord {} on the first line and ord(); on the second
[particle] mjk what is your rakudo version?
cosimo dolmen: both your suggestions are fine, but the test suite fails if I apply them... 14:09
mjk dolmen: rakudo-star.2010.07.msi
dolmen cosimo: the problem is probably that Buf doesn't have a split() method... 14:11
14:12 Mowah joined
mjk [particle]: use your code is OK, but if use:sub order-steak($how='medium'){say $how;};order-steak() ,not OK 14:14
snarkyboojum [particle]: the issue doesn't manifest if the sub and the calls to the sub are on the same line
dolmen star: my $buf1 = "This is the test string".encode("UTF-8");
p6eval star 2010.07: ( no output )
mjk [particle]: you can test it
[particle] aha 14:15
snarkyboojum [particle]: seems to have something to with new lines after the defintion of a sub starting with ord
[particle] if you put a semi-colon after the sub definition, it works, too
snarkyboojum gist.github.com/504699 # golfed
mjk [particle]: yes 14:16
dolmen cosimo: added comments
snarkyboojum [particle]: can confirm that adding a ; to the end of the sub definition in that gist fixes the problem too :)
mjk is strange!
14:17 isBEKaml joined
isBEKaml hi, Star6! :) 14:17
snarkyboojum isBEKaml: o/
[particle] works in nqp 14:18
isBEKaml From what pmichaud++ said in the announce, R* release will be monthly. Does that mean there won't be dev releases of rakudo like we did pre-star? 14:19
[particle] no 14:21
the compiler releases will continue monthly
R* distros will likely be based on the compiler releases
isBEKaml right. I was wondering since I already submitted a slackbuild script for rakudo compiler to SBo. :)
[particle] mjk: definitely a parser error: 14:22
cosimo dolmen: Buf.split missing yes 14:23
14:23 slavik joined
mjk [particle]: aha 14:23
[particle] gist.github.com/504714 14:24
x3nU is there any perl6 module for md5 or sha1? 14:31
tadzik md5, yes
digest-md5
x3nU Digest::MD5 doesn't work for me: "load_bytecode" couldn't find file 'Digest/MD5.pir'
tadzik x3nU: poke cosimo then, iirc he's the author 14:32
snarkyboojum [particle] fwiw, it's not ord that is special, it's any sub name starting with a token infix symbol and having one more chars appended to it.. i.e. or. and. xor. err. leg. but. etc
one or* more 14:33
[particle] snarkyboojum: aye, and i believe this is a known bug, but i haven't searched the queue
snarkyboojum [particle]: ah ok 14:34
[particle] it looks quite familiar...
x3nU tadzik: i've leaved message for him on priv 14:35
dolmen star: Buf.^methods().perl.say 14:38
p6eval star 2010.07: OUTPUT«Method 'perl' not found for invocant of class 'ResizablePMCArray'␤ in main program body at line 22:/tmp/GeFw3FNSML␤»
dolmen cosimo: Buf implementation is incomplete (and not even specified): see above 14:39
tadzik dolmen: nah, it's just a Role
like Hash
14:39 GeneralMaximus joined
tadzik star: Hash.^methods 14:39
p6eval star 2010.07: ( no output )
tadzik star: Hash.^methods.perl 14:40
p6eval star 2010.07: OUTPUT«Method 'perl' not found for invocant of class 'ResizablePMCArray'␤ in main program body at line 22:/tmp/m59tt5n9g1␤»
tadzik at least that's what moritz_ told me :)
14:40 Guest23195 joined
dolmen cosimo: perlcabal.org/syn/S32/Containers.html#Buf 14:42
snarkyboojum yeah, can't introspect those types :|
dolmen star: buf.new([65, 66, 67]).HOW.methods().say 14:44
p6eval star 2010.07: OUTPUT«Could not find sub &buf␤ in main program body at line 22:/tmp/WsyJoyUPkI␤»
dolmen star: Buf.new([65, 66, 67]).HOW.methods().perl.say
p6eval star 2010.07: OUTPUT«too few positional arguments: 1 passed, 2 (or more) expected␤ in main program body at line 22:/tmp/bDMhJkr_ZD␤»
dolmen star: Buf.new([65, 66, 67]).^methods().perl.say
p6eval star 2010.07: OUTPUT«[{ ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ... }, { ...
..}, { …
14:45 skangas joined
snarkyboojum oh cool 14:47
star: Buf.new().^methods(:local).say 14:48
p6eval star 2010.07: OUTPUT«postcircumfix:<[ ]>contentsdecodeofat_poselemsnew␤»
snarkyboojum star: Buf.^methods(:local).say
p6eval star 2010.07: OUTPUT«Method 'say' not found for invocant of class 'ResizablePMCArray'␤ in main program body at line 22:/tmp/TIFgxOeeHe␤»
snarkyboojum star: Int.^methods(:local).say
p6eval star 2010.07:
..OUTPUT«asinhNumericcosechrealsStrasecacotanhfloorsechACCEPTSasechrandtruncateroundsqrtasinBoolcoshexpacoshsuccsigncotanatan2ComplexseccislogsinIntBridgeto-radiansacosechabsacotanlnceilingcotanhunpolaratanRealacosRattanhatanhcosecNumperlWHICHrootsisNaNtanfrom-radianslog10cospredacosecsinh␤»
snarkyboojum so some quirks
dolmen star: (map { $_ } Buf.new([65, 66, 67])).perl.say 14:49
p6eval star 2010.07: OUTPUT«===SORRY!===␤Confused at line 22, near "(map { $_ "␤»
dolmen star: (map { $_ }, Buf.new([65, 66, 67])).perl.say
p6eval star 2010.07: OUTPUT«(.new(contents => [65, 66, 67]))␤»
dolmen star: for Buf.new([65, 66, 67]) { $_.perl.say } 14:50
p6eval star 2010.07: OUTPUT«.new(contents => [65, 66, 67])␤»
dolmen star: for @{Buf.new([65, 66, 67])} { $_.perl.say }
p6eval star 2010.07: OUTPUT«===SORRY!===␤Non-declarative sigil is missing its name at line 22, near "@{Buf.new("␤»
dolmen star: for @(Buf.new([65, 66, 67])) { $_.perl.say } 14:51
p6eval star 2010.07: OUTPUT«.new(contents => [65, 66, 67])␤»
dolmen cosimo: don't know how to iterate the elements of a Buf
14:51 alester joined, sekon joined 14:52 sekon left 14:53 ispeak_ joined
cosimo dolmen: i'll leave the Str version for now 14:53
dolmen cosimo: it only works on ASCII: code points above 128 will corrupt the CRC 14:55
snarkyboojum star: my $buf = Buf.new([65, 66, 67]); for ^$buf.elems -> $i { say $buf[$i] } 14:57
p6eval star 2010.07: OUTPUT«65␤66␤67␤»
14:59 redicaps left, serged joined, timbunce joined, jdv79 joined
snarkyboojum star: for ^Buf.new([65, 66, 67]).elems -> $i { say $buf[$i] } 15:01
p6eval star 2010.07: OUTPUT«===SORRY!===␤Symbol '$buf' not predeclared in <anonymous> (/tmp/_vfd5rhclV:22)␤»
snarkyboojum heh
15:07 pyrimidine joined 15:10 avuserow joined 15:13 perlygatekeeper joined 15:16 ashleydev joined 15:20 bluescreen joined
cosimo dolmen: so, how do I go on adding a crc32 to buf8 ? 15:21
dolmen cosimo: github.com/cosimo/perl6-string-crc3...ent-121233 15:25
15:25 gbacon joined
dolmen star: my $x = Buf.new([65, 66, 67]); $x[0].perl.say; 15:26
p6eval star 2010.07: OUTPUT«65␤»
dolmen star: my $x = Buf.new([65, 66, 67]); for map { $x[$_] }, 0..($x.elems-1) { $_.say } 15:30
p6eval star 2010.07: OUTPUT«65␤66␤67␤»
PerlJam dolmen: you know you can say 0..$x.end instead of $x.elems-1 15:33
dolmen PerlJam: thanks 15:34
PerlJam or even ^$x.elems instead of 0..$x.end
dolmen star: my $x = Buf.new([65, 66, 67]); for map { $x[$_] }, 0..($x.end) { .say }
snarkyboojum you can also do ^$x.elems to give you the same range :)
p6eval star 2010.07: OUTPUT«65␤66␤67␤»
PerlJam you don't need the extra parens
dolmen star: my $x = Buf.new([65, 66, 67]); for map { $x[$_] }, ^$x.elems { .say }
p6eval star 2010.07: OUTPUT«65␤66␤67␤»
snarkyboojum as per, my $buf = Buf.new([65, 66, 67]); for ^$buf.elems -> $i { say $buf[$i] }
:)
15:35 darkop joined 15:36 tylercurtis joined
PerlJam if someone added support for .values to Buf, you wouldn't need the map or array indexing either :) 15:36
dolmen I agree. In fact, Buf should be Iterable. 15:37
PerlJam oh, Buf already has values, but it's not quite useful enough. 15:38
star: my $x = Buf.new([65, 66, 67]); for $x.values.values { .say }
p6eval star 2010.07: OUTPUT«Buf()<0x702d8a0>␤»
PerlJam oops
star: my $x = Buf.new([65, 66, 67]); for $x.values { .say }
p6eval star 2010.07: OUTPUT«Buf()<0x6fdb640>␤»
d4rt star: my $x = Buf.new([65,66,67]) for $x.values {.values.say } 15:49
p6eval star 2010.07: OUTPUT«===SORRY!===␤Confused at line 22, near "my $x = Bu"␤»
snarkyboojum star: say Buf.new([65, 66, 67]).contents.join(" ")
p6eval star 2010.07: OUTPUT«65 66 67␤»
snarkyboojum star: say Buf.new([65, 66, 67]).contents.WHAT
p6eval star 2010.07: OUTPUT«Array()␤»
d4rt star: my $x = Buf.new([65,66,67]); for $x.contents {.say }
p6eval star 2010.07: OUTPUT«65␤66␤67␤»
snarkyboojum naughty - no idea why contents is available 15:50
d4rt :(
15:50 Guest40089 joined
snarkyboojum oh sorry... no it's fine :) 15:50
15:50 risou joined
d4rt star: my $x = Buf.new([65,66,67]); for $x.values { .WHAT } 15:51
p6eval star 2010.07: ( no output )
d4rt ? 15:52
snarkyboojum there is no .values implemented for Buf
d4rt i thought PerlJam said it did?
snarkyboojum oh ok 15:53
d4rt also
star: my $x = Buf.new([65,66,67]); for $x.values { .say } 15:54
p6eval star 2010.07: OUTPUT«Buf()<0x7007220>␤»
d4rt what up with that then?
snarkyboojum maybe from Any-list
PerlJam d4rt: Buf may have .values accidentally through its inheritance heiarchy or it may have a stub implementation that does what you see
d4rt i see 15:55
pyrimidine re: lives_ok() in Test, shouldn't "lives_ok {warn('foo')}" work?
d4rt thanks
snarkyboojum looks like it comes from Any-list
pyrimidine nopaste? 15:56
15:57 eric1599 joined
snarkyboojum which just does .list on the Buf, which doesn't do what is intended :) 15:58
pyrimidine lives_ok() problem: gist.github.com/504846 16:00
16:06 tadzik joined, timbunce joined, uniejo joined, Pyramide joined 16:15 jaldhar joined
daxim zomg, rakudo's build system sucks. 16:28
16:28 colomon joined
PerlJam daxim: patches welcome :) 16:29
16:30 cdarroch joined
[Coke] daxim: rakudo's in particular, or just the underlying parrot one? 16:31
16:31 Trashlord joined
daxim just rakudo, I am in great need of --build-root, that would make packaging much easier. 16:32
[Coke] ah, yes. parrot could use that particular option too.
someone's working on it on the parrot side, not that that helps you
tadzik packaging Rakudo is ok, Rakudo Star isn't
daxim okay, I'm actually on rakudo-star 16:34
tadzik well, that's broken, it's a fact
daxim what an insight
tadzik but the other fact is, it's hardly possible to fix
daxim I have already resigned to work around it
tadzik The only sensemaking way is to package Rakudo+Parrot+Blizkost and Modules separately 16:35
[Coke] I am still confused by people packaging star.
PerlJam I think pmichaud committed some changes to R*'s repo to help in that area. Maybe wait for the next R* release or build a new one from the current contents of the repo?
(I could be wrong though)
tadzik [Coke]: what OS are you running?
[Coke] tadzik: xp, os x, and ubuntu. 16:36
tadzik [Coke]: so the point is to have R* available via package management, like packages in Ubuntu
16:36 justatheory joined
[Coke] the windows installer justworked for xp, and doing the build from the R* tarball worked fine on ubuntu. I've been doing parrot & rakudo separately on osx for some time, not sure I did * there. 16:36
tadzik: but don't do *, do "parrot", and "rakudo compiler", and "the modules" . no? 16:37
tadzik [Coke]: what do you mean?
16:38 aCiD2 joined
aCiD2 evening perl6 16:38
tadzik [Coke]: the point is to have a package, e.g. in Ubuntu, which users can install w/o compilation. Like a .deb package, or even available in the repos
[Coke] I mean create a a parrot package. and a rakudo compiler package. and a "rakudo modules" package (perhaps not going all the way on this last one as the p5 module packages do.)
tadzik g'afternoon aCiD2 16:39
[Coke] tadzik: yes. why do this from R* and not from the component parts that R* is built from?
tadzik [Coke]: well, I packaged the components (not all of them) on CRUX. But I think it'd be nice to have not-broken build system in Star, so one could package it without mangling numerous things together 16:40
[Coke] assumes that his point makes no sense and gives up. ;) 16:41
PerlJam tadzik: that would have made an excellent grant proposal to TPF I think.
tadzik by the way, I think Fedora wanted to have Star in their repos
PerlJam: what, fixing R* build system?
PerlJam yes
tadzik I can try to do this w/o a grant :) But as pmichaud said yesterday, or was it 2 days ago? It's difficult, for compiling modules needs installed Rakudo, and installed Rakudo needs installed Parrot 16:42
[Coke]: I think the point is that the Star is the collection of Perl6 stuff itself. Making a package which is not using Star, but rather combining the separate components is reinventing Star imho 16:43
PerlJam well, by the time the next grant period rolls around we may have had enough experience with R* releases that it's not a problem anymore :)
[particle] R* didn't use Rakudo 2010.07, it was too buggy 16:44
tadzik I fixed Blizkost Makefile, so I can stand broken R* ;)
16:45 jarek joined
[Coke] tadzik: yes, but if you're packaging things, you're now bundling things which are, in fact, separate items, into a single monster package. R* is, IMO, more a bundle for people who are wish to try things out. Let's say I'm using another language on top of parrot. if I'm getting my packaged parrot as a subset of rakudo, how do I upgrade parrot now? 16:49
shouldn't the R* package just have the appropriate parrot package as a dependency?
16:49 tommyd joined
[particle] [Coke]: when possible, yes, however R* won't always be based on a parrot compiler release or a rakudo compiler release 16:51
PerlJam [particle]: I think Coke is suggesting that the particular parrot release be packaged separately too 16:52
[particle] does ubuntu have the linux core packages as a dependency? i don't know the answer.
eew.
tadzik [Coke]: in fact you rarely package many things into monster package, they're more often small fix bonded by zilions of dependencies, as in the Parrot example. But R* is imho more of a condensed thing than a bunch of small parts
[particle]: define 'core linux packages'
[particle] linux core 2.6.35 16:53
tadzik yes, kernel is a separate package
[particle] ok.
and if ubuntu patches that kernel, does it bundle the patches separately and include them as dependencies? 16:54
tadzik nope
this is their kernel, with their patches and their config
[particle] ok, that's as i thought, too.
looking at it that way, R* could include a parrot-core source pkg as a dependency, and have required patches to that inside the R* pkg, applied after the parrot-core pkg is downloaded and untarred 16:55
PerlJam true 16:56
tadzik nope
PerlJam nope?
tadzik in fact, in this approach, with parrot as a dependency, it would be alredy installed to the fs
16:56 risou_ joined
tadzik what is more, it would more proabaly be binary, compiled alredy 16:56
so no patching and recompiling
[particle] i said source pkg specifically, it'd have to be that way. 16:57
PerlJam tadzik: you can't have a source dep that could be patched/recompiled?
16:57 jarek-s joined
[particle] i don't know if parrot provides source packages, though 16:57
tadzik PerlJam: not always, e.g. in Ubuntu everything is binary, compiling is not happening when you install a package 16:58
gfldex tadzik: until you install emacs, that starts to compile lots of stuff 17:02
or latex that builds lots of font info 17:03
tadzik hmm
huf or any lisp vm basically ;)
tadzik so maybe that's the way. Installing Rakudo, Parrot, Blizkost, and compile modules after that, in this post-install stage?
gfldex but we shouldn't generalise from lisp :D
huf true.
tadzik why? We can write functional stuff to :) 17:04
too
huf yes, but much better.
;)
tadzik :)
gfldex you should compile modules when they are first used or when the user is requesting it like with prelinking
if you insist to precompile stuff when you install the package, we will make you work with a netbook :) 17:05
17:06 timbunce_ joined
tadzik gfldex: look at R* build process then 17:07
it actually _is_ compiling modules :)
daxim in the install target, no less. 17:09
gfldex R* is a source tarball
17:09 itz joined
daxim holy shit, it finished compiling 17:10
itz proto is looking for projects.state and not finding it .. any ideas?
tylercurtis gfldex: how would you implement only compiling modules when first used?
tadzik daxim: but doing it in 'make install' step is wrong 17:11
daxim I agree
tadzik ok
so after make install, the modules are to be compiled
so when we come to the conclusion it's time to nag pmichaud :) 17:12
bbs
daxim no, that should happen in "make"
17:13 Pyramide left 17:14 radu_ joined
gfldex tylercurtis: i would not do that at all. you have to provide precompiled modules in the .deb because the user might not have write access to the place you have store those modules or might not have room in $HOME nor /tmp to store them there 17:14
tylercurtis gfldex: "you should compile modules when they are first used or when the user is requesting it like with prelinking".... so, only when explicitly requesting it? That doesn't seem wise with Rakudo's current performance. Unless I misunderstand you. 17:15
gfldex it was mentioned to compile them when the package is installed 17:16
and that's a bad idea
some ppl run linux on their linksys router
it's a tricky question
but parrot bytecode is portable, so shipping it with the .deb is possible 17:17
i'm quite sure debian will provide 2 variants in the end
one with precompiled and one without precompiled modules
17:18 tommyd joined
gfldex or stick the precompiled modules into another package 17:18
tylercurtis gfldex: but Parrot bytecode isn't portable between versions of Parrot, which means that every precompiled PBC package will have to be updated every time the Parrot package is, even if otherwise unnecessary. Not necessarily too problematic, but still suboptimal. 17:19
17:19 pmurias joined 17:20 d4rt joined
slavik tylercurtis: could it be because parrot is not 'final' ? 17:21
dolmen gfldex: usually you install binary package, and do not build from source
gfldex: the distributor is the one that build the binary package from the source 17:22
daxim build/module-install.pl is an evil hack and must die: it recompiles the pir files every single time
[particle] parrot bytecode is not yet guaranteed portable.
17:22 ggoebel joined 17:23 KenGuru joined
dolmen AFAIK R* is a whole package and should be a single package (.deb/rpm) that includes parrot+rakudo+modules. Installing the package should not alter an existing parrot: everything (Rakudo, Parrot, modules) should be installed into /usr/lib/rakudo-star-2010.07 17:25
17:27 mirrage joined 17:29 smash joined
smash hello everyone 17:29
daxim dolmen, if compiled with an external parrot, it mostly goes to %{_libdir}/parrot/*/languages/perl6 + subdirectories 17:30
x3nU is it possible to create array of bytes in perl6? 17:32
dolmen daxim: the question is: is an external parrot reliable?
x3nU: this called a Buf 17:33
s/this/this is/
daxim that's not the packager's concern. if I used the internal one, the end users would bite off my head for creating a shoddy package that conflicts for no good reason
x3nU dolmen: thanks 17:34
dolmen daxim: if R* embeds an internal parrot, it must not be installed as /usr/bin/parrot but instead as /usr/lib/rakudo-star-2010.07/bin/parrot
x3nU: perlcabal.org/syn/S32/Containers.html#Buf 17:35
daxim that's against FHS, won't do.
x3nU also is it possible that function require for example array of Ints (and only ints!) as argument? 17:36
dolmen daxim: that is not against FHS: in this case the parrot binary is private for rakudo, so it doen't have to be exposed to the whole world in /usr/bin. 17:37
x3nU: yes
x3nU how? 17:38
gfldex x3nU: not implemented yet
dolmen star: sub only-ints(Int @ints) { };
p6eval star 2010.07: ( no output )
17:39 ambs joined
dolmen ambs: already at YAPC::eu? 17:40
ambs nods
dolmen ambs: so see you on wednesday as flesh :)
ambs hehehe :D 17:41
dolmen ambs: you were one of the 2009 organizers, isn't it? 17:43
ambs indeed
dolmen that was my first YAPC. thanks for the great experience 17:44
ambs hehehe
thanks for coming and making it a great conference :)
17:46 desertm4x joined
dolmen daxim: www.pathname.com/fhs/2.2/fhs-4.7.html 17:48
daxim: "/usr/lib includes object files, libraries, and internal binaries that are not intended to be executed directly by users or shell scripts."
daxim: R*'s parrot would be an "internal binary"
x3nU can i specify required size of buf for function? 17:54
tadzik sub foo ($buf where { $buf.size > 50 }) or something? 17:56
x3nU well there's no size method 18:00
(and when i say 'size' i'm thinking about size of byte)
tadzik oh 18:01
no idea then
x3nU well
i will check if all elements of buf are smaller than 256 18:02
it's ugly but what else can i do ;d
tadzik and what do You want to achieve? 18:03
tadzik has never worked with Buffers
x3nU i'm writing script that works on bytes
tadzik binary, huh?
x3nU and i want be sure that buf contains bytes in classical meaning of 'byte' (8 bits)
in c it would be a lot simpler ;d 18:04
tadzik probably
you can write a lib in C and use it via Zavolaj
x3nU that would be overkill imho 18:05
tadzik yep
but no binary files support in Rakudo, afaik
18:06 rokoteko joined
tylercurtis x3nU: Eventually, (i.e., when Rakudo supports native types), buf will be Buf[uint8], and you could use it. 18:06
18:08 timbunce_ joined 18:15 ambs_ joined 18:16 jaldhar joined
[Coke] dolmen - if the binary is a private binary, I withdraw any kibbitzing I already put out there. =-) 18:20
dolmen have to leave 18:21
[Coke] ~~
18:21 tommyd joined 18:40 kst joined 18:46 xinming joined
tadzik remind me, what was this syntax with passing a variable to a block? 18:52
star: { foo => "bar" } -> $var { say $var<foo> }
p6eval star 2010.07: OUTPUT«===SORRY!===␤Confused at line 22, near "{ foo => \\""␤»
tadzik like in for, I don't remember how do I use it w/o loop
tylercurtis star: (-> $var { say $var<foo> })({foo => "bar"}) 18:53
p6eval star 2010.07: OUTPUT«bar␤»
tylercurtis star: (-> $var { say $var<foo> }).({foo => "bar"}) #or with .()
p6eval star 2010.07: OUTPUT«bar␤»
tylercurtis star: -> $var { say $var<foo> }.({foo => "bar"}) # probably doesn't work
p6eval star 2010.07: OUTPUT«bar␤»
tylercurtis Well, I guess it does. 18:54
star: -> $var { say $var<foo> }({foo => "bar"}) # probably doesn't work
p6eval star 2010.07: OUTPUT«bar␤»
tylercurtis tadzik: any of those...
std: -> $var { say $var<foo> }({foo => "bar"})
p6eval std 31888: OUTPUT«ok 00:01 118m␤»
tylercurtis didn't expect the last two to work.
Maddingue star: my @primes = gather { grep { take is_prime($_) } 0 .. Inf } # is lazyness implemented in Rakudo* ? 18:55
p6eval star 2010.07: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 22␤»
tylercurtis Maddingue: I think Rakudo's list assignment is lazy. Use binding. 18:56
std: my @primes = gather { grep { take is_prime($_) } 0 .. Inf }
p6eval std 31888: OUTPUT«[31m===[0mSORRY![31m===[0m␤Strange text after block (missing comma, semicolon, comment marker?) at /tmp/653T1KB1yx line 1:␤------> [32mes = gather { grep { take is_prime($_) }[33m⏏[31m 0 .. Inf }[0m␤ expecting any of:␤ bracketed infix␤ infix or meta-infix␤
..statement modi…
tylercurtis Your grep syntax is wrong, though, Maddingue.
Maddingue maybe. not used yet to Perl 6 :) 18:57
well, I wanted a short yet funny Perl 6 code to paste to www.hetemeel.com/einsteinform.php
tommyd evening all
tadzik oh, this way
tylercurtis Maddingue: also, grep is lazy, so you don't need gather/take there anyway. 18:58
tommyd I've noticed the terms "token", "regex" and "rule" within a Perl 6 grammar - are these terms interchangable (i.e. do all basically mean the same) or are there differences between them?
Maddingue could make a good update on µblogs with a text like: "ok, Perl 6 was a bit long, but we're coming to an end"
tadzik tommyd: there are differences
e.g. rule is whitespace-senvisite
tylercurtis Maddingue: my @primes := grep &is_prime, 1..Inf; 18:59
tadzik but don't believe me too much, I only use token because I never know the differences
:)
tylercurtis Or my @primes := 1..Inf ==> grep &is_prime
Maddingue: 1..* is also the more idiomatic way of doing 1..Inf.
19:00 jfried joined
Maddingue tylercurtis: thanks! 19:00
dear, I really need to start learning Perl 6
tadzik I'm glad to hear that :) 19:02
tylercurtis Maddingue: welcome. List assignment is supposed to be mostly lazy(it eagerly evaluates the finite beginning of the list( e.g. the 1,2,3,4 in 1, 2, 3, 4, 6..*), but doesn't touch the rest), but I think Rakudo's is still eager.
Su-Shee it
it's nice that everybody starts at the same level :)
19:03 gbacon joined 19:04 Trashlord joined, jaldhar joined 19:07 cono joined 19:08 jferrero joined
dalek kudo: 9667975 | moritz++ | src/core/Any-list.pm:
fix typo, noticed by madsen++
 19:13
tylercurtis tommyd, tadzik: token is regex :ratchet, rule is regex :ratchet :sigspace. In other words, rules and tokens turn off backtracking and rules turn on significant whitespace. 19:17
Maddingue tylercurtis++ thanks for the help: twitter.com/maddingue/status/20161383662
tommyd tylercurtis: ok, thanks 19:18
19:19 hercynium joined
tylercurtis Maddingue++ cool! I hope he wasn't expecting it too eagerly, though. :) 19:19
Maddingue ;) 19:21
tommyd is there a way to print the AST of a matching grammar?
tylercurtis tommyd: Foo::Grammar.parse($str).perl.say;
tommyd: If I understand you correctly. 19:22
tommyd tylercurtis: you did - great - this is it what I'm looking for
moritz_ parse tree != AST
in perl 6 speak, an AST is what you attach to a parse tree by calling make($thing) in the action method 19:23
tylercurtis Oh, right, AST.
moritz_ you can access that with .ast on the match object
(just for the record, not what tommyd++ wants)
tylercurtis moritz_: technically, .ast != AST. .ast is an "abstract object" that is "normally used to build up an abstract syntax tree." 19:24
PerlJam I've now got a new-to-me MacBook. Any words of wisdom for building rakudo/parrot on it? 19:26
moritz_ PerlJam: install a Perl 5 from source, into a sseparate directory
PerlJam: add it as the first option to PATH
slavik PerlJam: put Linux on it
moritz_ then follow the "how to build rakudo" instructions 19:27
PerlJam apparently I've no make or C compiler.
19:27 Holy_Cow joined
tylercurtis moritz_: not really necessary unless it has a pretty old OS X. 10.5, at least, ships with 5.8. 19:27
PerlJam (note: I'm not normally a mac person)
tommyd PerlJam: use MacPorts to get the current parrot
PerlJam OSX 10.5.2
19:27 nimiezko joined
PerlJam :-) 19:27
tylercurtis PerlJam: oh, right, need to install Xcode to get that.
tommyd (I packaged that recently)
PerlJam tommyd: I've no idea what that means. 19:28
tylercurtis You can get it from somewhere on Apple's website.
tommyd www.macports.org
PerlJam thanks
tommyd parrot is the VM for rakudo, ie. perl6
because of the (unfortunately unresolved) #344 there is no working rakudo port right now
[Coke] PerlJam: you want to install xcode from the install dvd. 19:29
[Coke] catches up. 19:30
PerlJam heh, apparently this macbook ran windows at one time ... there's a XP disk in the with the Mac disks 19:31
19:34 ggoebel joined 19:35 javs joined 19:36 estrabd joined 19:37 wamba joined, wamba left, wamba joined
ggoebel Can anyone recommend a module with tests which I could use as an example to shamelessly steal\\b\\b\\b\\b\\b\\b learn how people are currently testing their perl6 modules? 19:39
moritz_ ggoebel: my json module has pretty decent test coverage 19:40
github.com/moritz/json/
ggoebel moritz_: thanks!
19:41 justatheory joined
itz moritz_: I'm trying to use proto/plz with star .. is there a recommended version for that? 19:43
s/plz/pls
moritz_ itz: don't think so, just the 'pls' branch of proto 19:44
19:44 KenGuru joined
moritz_ I'm kinda surprised that star doesn't come with it bundled 19:44
tylercurtis moritz_: it has proto, though. 19:45
PerlJam There's always the next star release
tylercurtis moritz_: or at least, it has proto in modules/, but it doesn't seem to be installed. 19:46
itz tylercurtis: I don't see it 19:48
moritz_ in the build directory, not in the install location :( 19:49
rakudo: say ?try { warn "foo"; 1 } 19:53
p6eval rakudo 1a5d4a: OUTPUT«Null PMC access in type()␤ in main program body at line 22:/tmp/dEZc1dHWNA␤»
moritz_ rakudo: try { warn "foo"; say "alive" }
p6eval rakudo 1a5d4a: ( no output )
19:56 m`` left 19:58 tri1 joined 19:59 TSa joined
tommyd I can see a lot of syntax niceties like $<value>>> or $/.hash.values.[] 19:59
here: github.com/moritz/json/blob/master/...Actions.pm
where can I read up on these? Only in the 5th Synopses or is there a shorter overview somewhere? 20:00
moritz_ actually $<values>».thing is nicer 20:01
tylercurtis tommyd: S05 isn't the place to look for most of those.
perlcabal.org/syn/S03.html#Hyper_operators for >>.
tommyd ah ok, thanks 20:02
tylercurtis tommyd: what part of $/.hash.values.[0] do you want to know more about?
moritz_ tommyd: $<value> is accessign the 'value' capture from the current match object, $/
20:02 offerkaye joined
moritz_ and @list>>.method calls the .method on all items in list 20:02
tommyd: the book also talks about the JSON parser, in the "grammars" chapter
tommyd tylercurtis: basically I'm starting with writing a small grammar and I wonder how I can access and iterate over the parsed contents
offerkaye Hi
moritz_ tommyd: then do read the grammar chapter, and probably the regex chapter too 20:03
offerkaye: hi
tylercurtis perl6advent.wordpress.com/2009/12/2...d-actions/ is a pretty good article about grammars and actions. Tene++ 20:04
tadzik star: my @arr = 1, 2, 3; @arr>>.say: 'A $_' # how should it work? 20:06
p6eval star 2010.07: OUTPUT«Too many positional parameters passed; got 2 but expected 1␤ in 'Mu::say' at line 1220:CORE.setting␤ in main program body at line 22:/tmp/daev3P1UTB␤» 20:07
20:07 ashleydev joined, rbaumer joined
Tene tadzik: use the fmt method 20:08
20:08 ashleydev joined
Tene something like... 20:08
star: my @a = 1..3; @a>>.fmt("A %s")>>.say
p6eval star 2010.07: OUTPUT«A 1␤A 2␤A 3␤»
tadzik hmm 20:10
star: my @a = 1..3; @a>>.fmt("A %s\\n") # why the spaces on lines 2-3?
p6eval star 2010.07: ( no output )
tadzik oh you
star: my @a = 1..3; say @a>>.fmt("A %s\\n") # why the spaces on lines 2-3?
p6eval star 2010.07: OUTPUT«A 1␤A 2␤A 3␤␤»
20:10 desertm4x joined
tadzik hrm, nevermind 20:10
moritz_ can anybody still commit to the pugs repo? I get an error here 20:14
Tene tadzik: I see no extra spaces anywhere. Are you referring to the newlines?
tadzik Tene: nah, was just REPL-effect 20:15
Tene the only spaces are between A and the item, as asked for in the fmt.
ah, okay.
moritz_ I guess there was a reason for the "nevermind" :-)
[Coke] moritz_: I just had to rekey my auth and got a big ole warning about storing it plaintext, but yes, I was able to commit. 20:17
moritz_ [Coke]: rekey your auth? what did you do, exactly? 20:18
[Coke] "svn commit TODO". got prompted for my password. entered it. svn complained that I'm an insecure idiot, really do that? (yes). committed.
(since I'm just using user/pass and it has to store the password in plaintext in ~/.svn/<mumble> 20:19
moritz_ ah 20:20
my git-svn doesn't ask me, just complains
*mumble* 20:21
[Coke] ah, probably because I have 3 different (valid?) pugs urls. 2 on port 80, one on 443, each with a slightly different hostname.
I just never committed from this particular repo before. =-)
s/repo/checkout/
ah, git-svn is even more evil than svn. good luck!
x3nU my first perl 6 module, totally useless hash algorithm implementation, 9 lines of code ;D 20:24
github.com/xenu/Digest--djb2
cosimo x3nU: cool 20:26
tylercurtis x3nU: can you not do for $input? 20:27
tadzik x3nU: why a class? 20:28
cosimo added String::CRC32 to modules.perl.org just now
\\o/
prereq for Cache::Memcached :)
tylercurtis rakudo: say ^5 eqv (0..5-1)
p6eval rakudo 1a5d4a: OUTPUT«1␤»
x3nU tylercurtis: why?
tadzik: dunno, i saw this in many simple modules done like this
(yeah, i know i'm lame programmer, my second day with perl 6) 20:29
tylercurtis x3nU: well, you're not using the index, and manual indexing is kinda error-prone.
tadzik x3nU: come on, stop insulting yourself :)
x3nU: gist.github.com/505260 20:30
tylercurtis tadzik: s/method/sub/
tadzik oh, sure
fixed-faxed 20:31
avar Who's responsible for perlcabal.org
[particle] x3nU: where do you use 33?
moritz_ I can fix stuff there
[particle] i just see it in a comment
moritz_ (to avar )
avar hrm, it's just feather now
moritz_ avar: anything wrong with it? 20:32
avar moritz_: old links like "perlcabal.org/syn/S26.html" are broken now, should probably redirect to the canonical synopsis
20:32 _macdaddy joined, perlygatekeeper joined
avar just noticed it in my own sites (I test outgoing links now), but it might be worthwhile to go through the error log on the server to see if more stuff is broken 20:32
moritz_ avar: yes, I know. I have a local commit that fixes it, currently fighting with pugs svn
avar whee, cool
x3nU [particle]: bitwise left shift gives 33 20:33
tadzik: thanks
[particle] ah, right-o
tadzik x3nU: yw
itz I have to touch "projects.state" but otherwise the proto in star works fine 20:38
20:39 smash joined
x3nU but what can i do to move subroutine to other namespace 20:40
Digest::djb2
becase i think hash() is common name and can cause problems 20:41
tadzik x3nU: remove is export
x3nU k
tadzik oh, but…
x3nU what?
tadzik I think Digest::djb2::hash won't work yet
szbalint just heard moritz_ swear for the first time!
tadzik and the workaround is… using class, as You did :)
x3nU ;D 20:42
tadzik but check, maybe it works :)
moritz_ svn: Commit failed (details follow):
svn: Server sent unexpected return value (403 Forbidden) in response to MKACTIVITY request for '/pugs/!svn/act/c5292047-bcf3-4dce-8001-0878cc47dae1'
has svn over http been disabled? 20:44
20:44 DeathAnchor joined
smash any news on the YAPC::EU Perl 6 Hackathon ? 20:44
x3nU Can not find sub Digest::djb2::hash in main program body at line 1
;)
smash (time and place)
tadzik :( 20:45
moritz_ smash: like, tomorrow after breakfeast
smash: at the venue, either lobby of first floor
tadzik x3nU: commit revert time then :)
but but, haven't you ran your tests before pushing? :P 20:46
smash moritz_: sure, sounds sane
DeathAnchor hey all, I just dl'ed rakudo, I often use perldoc -f <func> or perldoc <keyword> to read up on things. is there an equivilent resource for perl6/rakudo? even an online wiki/doc location would work for me.
tadzik DeathAnchor: grok
from CPAN, for Perl 6
x3nU tadzik: of course not ;d
tadzik * Perl 5
x3nU also how do i revert (i'm new at git)
tadzik x3nU: then you deserved this! ahahahaha 20:47
20:47 Guest23195 left
tadzik x3nU: www.kernel.org/pub/software/scm/git...evert.html 20:47
smash moritz_: maybe update the wiki or something with the info ? 20:48
moritz_ smash: yes, when i'm done swearing at svn
DeathAnchor tadzik: I meant if I want to read about open, in perl5 I would do perldoc -f open, how would I read about perl6 open?
smash moritz_: hehe, fair enough 20:49
moritz_ DeathAnchor: you grep the specification :(
smash: NOT FAIR, FREAKIN' NOT WORKING; SIR!
DeathAnchor are the specs online/
?
moritz_ perlcabal.org/syn/
DeathAnchor thanks moritz_ 20:50
exactly what I was looking for
smash moritz_: which svn ?
moritz_ smash: pugs svn. Can't commit a single thing anymore 20:51
20:51 DeathAnchor left
smash moritz_: want me to try ? 20:51
[Coke] moritz_: what error are you getting?
moritz_ smash: works for [Coke]
403 Forbidden 20:52
[Coke] ok. so, bad creds.
moritz_ [Coke]: it should ask me for creds if they are not accepted
at least svn always did, before
should it work with http? (as opposed to https)
[Coke] that was my commit a moment ago. 20:53
(on http)
moritz_ [Coke]: thanks
[Coke] kerneltrap.org/mailarchive/git/2007/1/15/235895
(which looks like merlyn having a similar issue where he didn't /think/ he was being prompted.) 20:54
moritz_: does the auth in your ~/.subversion look right? 20:55
moritz_ [Coke]: yes. I even renamed the folder, so it wouldn't find it, and ask me for another password
[particle] is the pw rot13?
moritz_ [particle]: no, plain text
[particle] ah, i thought they did rot13... checking red-bean 20:56
moritz_ they do double rot-13
[particle] twice as secure!
x3nU is there repo for perl6 modules or something like that? 20:57
moritz_ x3nU: we just put them on github, and collect them at modules.perl6.org
x3nU what should i do to get my module published there? 20:58
[Coke] thinks it would be nice if we agreed on a standard for "perl6-" vs. not. 21:00
(on modules.perl6.org)
moritz_ x3nU: tell us that it exists 21:02
smash [Coke]: agree
x3nU define 'us' ;)
moritz_: ^
gfldex us ~~ #perl6 21:03
moritz_ x3nU: somebody with commit access to 'proto'
gfldex just spam the link to github until somebody gives in :)
x3nU posted one time already ;d
21:03 ambs joined
x3nU github.com/xenu/Digest--djb2 21:03
if knew who has acces it would be nice ;d 21:04
moritz_ hugme: add xenu to proto
hugme moritz_: You need to register with freenode first
moritz_ hugme: add xenu to proto
hugme hugs xenu. Welcome to proto!
moritz_ x3nU: you now have commit access
x3nU: please add it to poc-projects.list in the 'pls' branch 21:05
x3nU ok
gfldex even the sun is responding to R* ! science.slashdot.org/story/10/08/02...n-At-Earth
it's waving :)
moritz_ -> out 21:13
21:13 pugssvn joined
pugssvn r31887 | bbkr++ | [t/spec] tests for RT #69548 .push allows freaky action-at-a-distance in Rakudo 21:13
r31888 | bbkr++ | [t/spec] tests for RT #69740 Defining several constants with parentheses
r31889 | moritz++ | [t/spec] test for RT #76792, ** and \\w interact badly in regexes
r31890 | coke++ | meaningless commit for moritz_++
r31891 | moritz++ | add "Real World Haskell" to reading list
x3nU so after commit how much time it will take to publication on site? 21:16
tadzik x3nU: it's updated once an hour, iirc
21:18 Casan joined
moritz_ right 21:30
szbalint moritz_++ # hard work for that commit 21:35
21:47 masak joined
masak lolpisa! 21:47
phenny masak: 06:41Z <tadzik> tell masak the yapsi README says about running 'make', but there is no Makefile. Also big kudos for it, reading the code is an awesome experience
masak: 12:20Z <szabgab> tell masak I am trying to fix HTML::Tempolate that seems to be totally broken, see you in Pisa
masak tadzik: oops. thanks. will fix in the blog post; mail is too late. :) 21:48
21:48 tommyd joined 21:56 astrojp joined
masak oh, not the announcement; the README. excellent. 21:56
hm, wonder if it'd be worth implementing 'goto' early in yapsi? :) 22:01
s/worth/worth it/
22:02 pyrimidine left
cognominal In grammar actions, I would like an access to an non existing reduction to blow on me instead of genrating a proxy, later giving rise to cryptic errors. What do you think? 22:03
masak cognominal: $<subrule> // die "blow!" 22:04
dalek psi: b55a18e | masak++ | README:
[README] belatedly updated

on the use-before-declaration thing.
22:06 tri1 left
masak szabgab: yes, HTML::Template is from the alpha days. 22:07
cognominal masak, yes, but a typo happens so easily.
masak szabgab: we should probably not use it much, but instead focus on projects such as Hitomi.
cognominal: if you typo the 'subrule' key, it'll die too, so that's OK.
cognominal no, my point is that it does not die immediately but generates a Proxy. 22:09
22:09 brennen joined, jaldhar joined 22:13 gbacon joined
cognominal good night. 22:14
22:16 oha joined
masak cognominal: not dying is by spec. that goes for all hash-like things, just like in p5. 22:21
cognominal: generating a Proxy object is not spec. 22:22
22:23 jhuni joined 22:25 vpm joined 22:29 d4rt joined 22:43 risou joined 22:49 ruoso joined
Casan what is the p6 equivalent to system in p5 ? 23:13
masak Casan: run
'night 23:14
23:23 whiteknight joined
avuserow rakudo: print join " 1 ", grep {$_ % 2 == 1}, 1 .. 10; 23:25
p6eval rakudo 966797: OUTPUT«1 1 3 1 5 1 7 1 9»
avuserow rakudo: (1 .. 10).grep(*%2).join(" 1 ").print;
p6eval rakudo 966797: OUTPUT«1 1 3 1 5 1 7 1 9» 23:26
avuserow rakudo: ((1 .. 10) Z (1 XX *)).grep(*%2).print;
p6eval rakudo 966797: OUTPUT«11»
avuserow rakudo: ((1 .. 10).grep(* % 2) Z (1 XX *)).print;
p6eval rakudo 966797: OUTPUT«113Whatever()<0x736fcb0>» 23:27
avuserow rakudo: (((1 .. 10).grep(* % 2)) Z (1 XX *))).print;
p6eval rakudo 966797: OUTPUT«===SORRY!===␤Confused at line 22, near ").print;"␤»
avuserow I'm confused too.
rcfox Too many close parens.
avuserow rakudo: (1 XX *).elems.say 23:28
rcfox rakudo: (((1 .. 10).grep(* % 2)) Z (1 XX *)).print;
p6eval rakudo 966797: OUTPUT«1␤»
rakudo 966797: OUTPUT«113Whatever()<0x73ed000>»
avuserow rakudo: (1 X *).elems.say
p6eval rakudo 966797: OUTPUT«1␤»
avuserow okay, how do I get an infinite list again?
sorear 1 xx * 23:29
avuserow rakudo: (1 .. 10.grep(*%2) Z 1 xx *).perl.say 23:31
p6eval rakudo 966797: OUTPUT«()␤»
avuserow rakudo: ((1 .. 10).grep(*%2) Z (1 xx *)).perl.say 23:32
p6eval rakudo 966797: OUTPUT«(1, 1, 3, 1, 5, 1, 7, 1, 9, 1)␤»
avuserow rakudo: (^10.grep(*%2) Z (1 xx *)).perl.say
p6eval rakudo 966797: OUTPUT«()␤»
avuserow rakudo: ((^10).grep(*%2) Z (1 xx *)).perl.say
p6eval rakudo 966797: OUTPUT«(1, 1, 3, 1, 5, 1, 7, 1, 9, 1)␤»
avuserow rakudo: (^10).grep(*%2).join(' 1 ').say 23:33
p6eval rakudo 966797: OUTPUT«1 1 3 1 5 1 7 1 9␤»
avuserow I love the whatever star. :) 23:35
23:37 astrojp left 23:41 dju joined 23:44 dju joined 23:58 Psyche^ 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!