Prelude was borken (worked around now): sial.org/pbot/11497 | pugscode.org/ <Overview Journal Logs> | pugs.kwiki.org | paste: sial.org/pbot/perl6 | Chip needs help!!! www.geeksunite.net
Set by autrijus on 4 July 2005.
mugwump (but with Set::, etc ;)) 00:00
svnbot6 r5222, autarch++ | make Date.now alias work 00:14
mugwump autarch: we will be needing to differentiate our modules via a namespace. our opposing views seem to be; 00:40
1. the shorter namespaces should be used for the most used pieces 00:41
vs
2. the shorter namespaces should be used for the simpler concepts
would you agree with that? 00:42
(btw, by namespace, I mean the third part of the module name, as in dev.perl.org/perl6/doc/design/syn/S...Versioning 00:46
svnbot6 r5223, autarch++ | some docs 05:06
Aankhen`` Darren_Duncan++ # DBI v2 discussion 06:13
mugwump Darren_Duncan++ # "accomodate such abhorrent computer programming practices" 06:18
Aankhen`` Heh. 06:19
Darren_Duncan hello
Aankhen`` Hiya.
Darren_Duncan you said my name, so Colloquy alerted me to come here 06:20
Aankhen`` Colloquy?
Darren_Duncan my IRC client for Mac OS X
Aankhen`` Ah.
Darren_Duncan while being fully GUI, its open source too 06:21
mugwump ducks off to www.chansmartialarts.com &
Darren_Duncan not that I've bothered to take advantage of that, I did give the developer some money
Aankhen`` ?eval gmtime 06:22
evalbot6 Error: No compatible subroutine found: "&gmtime"
Darren_Duncan doeth
gaal Aankhen``: working on it 06:24
good morning btw
Aankhen`` Good morning. :-D
Very good morning. :-D 06:25
?eval unless (0) { "foo" } 06:31
svnbot6 r5224, Aankhen++ | * added HTTP::Date (stubs only!).
r5225, Aankhen++ | * HTTP::Headers needed `use v6;`.
evalbot6 'foo'
masak morning 06:32
Aankhen`` Morning. 06:34
?eval if 1 { "foo" }
evalbot6 'foo'
06:39 shachaf_ is now known as Shachaf
Aankhen`` ?eval my @foo = <bar 1 baz 3 quux 2>; my %foo = @foo.map:{ .uc => 1 }; %foo 06:41
evalbot6 Error: cannot cast from VList [VList [VStr "PERL6",VInt 1]] to Pugs.AST.Internals.VCode (VCode)
Aankhen`` ?eval my @foo = <bar 1 baz 3 quux 2>; my %foo = @foo.map:{ (.uc => 1) }; %foo
evalbot6 Error: cannot cast from VList [VList [VStr "PERL6",VInt 1]] to Pugs.AST.Internals.VCode (VCode)
theorbtw1 Interesting; it thinks that @foo is the code. 06:44
svnbot6 r5226, Aankhen++ | * made HTTP::Headers compile.
r5226, Aankhen++ | * HTTP::Headers -- went back to `our @header_order`, et al.
theorbtw1 ?eval my @foo = <bar 1 baz 3 quux 2>; my %foo = { (.uc => 1) }.map @foo; %foo
evalbot6 Error: unexpected "@" expecting operator, ";" or end of input
theorbtw1 ?eval my @foo = <bar 1 baz 3 quux 2>; my %foo = { (.uc => 1) }.map(@foo); %foo
evalbot6 Error: cannot cast from VList [VStr "bar",VStr "1",VStr "baz",VStr "3",VStr "quux",VStr "2"] to Pugs.AST.Internals.VCode (VCode)
Aankhen`` ?eval class Foo { method bar (::?CLASS $stuff) { $stuff } } Foo.new.bar 07:28
evalbot6 Error: unexpected ":" expecting formal parameter, end of input or ")"
svnbot6 r5227, Aankhen++ | * HTTP::Headers: couple of workarounds to remove errors. 07:29
r5228, Aankhen++ | * HTTP::Status: fixed `lc` -> `uc`.
nothingmuch hola 07:34
Aankhen`` iblechbot >> I thought you were able to get ::?CLASS and ::?ROLE to parse? 07:36
D'oh
.
svnbot6 r5229, Aankhen++ | * HTTP::Request need only `require` HTTP::Message. 07:47
r5230, Aankhen++ | * HTTP::Headers: commented out all references to MIME::Base64.
r5231, qtplatypus++ | More work on self hosting grammar. However this will go on hold
r5231, qtplatypus++ | untill PGE is more mature. The PGE doesn't handle :w and on my
r5231, qtplatypus++ | computer PGE buss errors on more complex grammars.
Aankhen`` Buss errors? 07:52
svnbot6 r5232, Aankhen++ | * HTTP::Message now compiles.
masak Aankhen``: according to dictionary.com, a "buss" is a kiss, which I can understand would be a problem in complex grammars :P 08:39
Aankhen`` Heh.
nothingmuch i beg to differ 08:40
kisses are very universal tokens
they work as expected in both the sexual and parental contexts, under almost all platforms
regardless of language
masak: how is your monadish business coming along? 08:41
masak nothingmuch: i haven't read anything since yesterday
but kudos to you for writing the tutorial, it really helps 08:42
gaal you know what e e cummings said...
"kisses are a better fate than wisdom"
masak (except for 06.kwid which is a nightmare)
(in a good way, of course) 08:43
Aankhen`` Oh, I thought it might be something along the lines of "One McChicken burger, no cheese".
gaal i can't quite picture him saying that, A :-)
Aankhen`` :-P
masak gaal: kisses, wisdom... can't i have both?
gaal it made sense in the context when he said it. 08:44
masak usually does 08:45
gaal i find myseif using a pattern, and i wonder if it can be reused by others: 08:46
the Prelude often constructs an object from an array
say you have class Time { has $.year; has $.month; has $.day }
and an internals command called localtime that returns @tm = <2005 7 5> 08:47
so my facade returns
Time::Local.new( year => 2005, month => 7, day => 5 ) 08:48
(this is simplified of courtse)
08:48 whitay_ is now known as whitay
gaal now if there were a standard way to say 08:48
@ids = <year month day>
i could return
Time::Local.new ( map { ; @ids[$_] => @dat[ $_ ] ;} 0 .. @ids.end ) 08:49
except without the explicit map
:-)
(btw the semicolons are just there for paranoia)
and i meant s/dat/tm/ there
so this works, but it's a common idiom that doesn't *look* pretty: therefore, i propose a builtin to do it 08:50
masak gaal: so what you would like is a construct of some kind instead of the explicit map?
why not use zip in some way? 08:51
or reduce [=>] ?
gaal zip! that works.
masak :)
masak likes zip 08:52
gaal i don't know why, but [=>] doesn't work
ah i think i do know 08:53
i don't want year => 2005 => month => 7
masak i don't think that is what [=>] gives you
gaal zip however isn't enough
masak: in pugs it just gives the first elem, dunno if that's correct 08:54
i need to coerce the list into a hash, no?
masak are we talking about zip?
i think it takes two lists and returns a hash 08:55
gaal we're talking about using the result of zip as the input to new
masak yes
gaal new( year => 2005, month => 7)
do i need a pairs construcor? is there one?
ah, there is indeed. 08:56
let's see if pugs has it
masak seems like zip returns a list.... hm
gaal yes!
so i need pairs zip @ids, @tm 08:57
cool :)
masak yes, that should work :)
gaal <<3 p6
masak perl6 is spookily convenient 08:58
gaal pugs is spookily "already implemented" 08:59
masak "accidentally implemented" :)
f0rth_ is reading channel9.msdn.com/wiki/default.aspx...QuickStart 09:00
Aankhen`` goink out. 09:06
BBL. &
gaal ?eval class Foo has $.x; has $.y; sub mkFoo { my @ids = <x y>; my @vals = <1 2>; return Foo.new(pairs zip @ids, @vals) } mkFoo.perl 09:28
evalbot6 Error: unexpected "p" expecting block construct, term, ":", term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
gaal that's not quite the error i've been getting :-) 09:29
?eval class Foo has $.x; has $.y; sub mkFoo { my @ids = <x y>; my @vals = <1 2>; return Foo.new(pairs zip @ids, @vals) } mkFoo().perl
evalbot6 Error: Can't use positionals in default new constructor
gaal there. the problem is that new is getting confused. however, sadly this doesn't work either: 09:30
?eval class Foo has $.x; has $.y; sub mkFoo { my @ids = <x y>; my @vals = <1 2>; my %temp = pairs zip @ids, @vals; Foo.new( %temp ) } mkFoo().perl
evalbot6 Error: Can't use positionals in default new constructor
gaal so new needs some fixin'. 09:31
also: is it a known bug that class variables don't work? 09:32
class Foo; my $bar; sub get_bar { $bar } 09:33
?eval class Foo; my $bar; sub get_bar { $bar }
evalbot6 undef
gaal class Foo; my $bar = 8; sub get_bar { $bar } get_bar() 09:34
grrrr
?eval class Foo; my $bar = 8; sub get_bar { $bar } get_bar()
evalbot6 \8
gaal !?
this didn't workforme, honest :)
?eval class Foo; my @bar = <8 9>; sub get_bar { @bar } get_bar() 09:35
evalbot6 ['8', '9']
gaal hmmmm
castaway_ why are those refs? 09:58
gaal i don't rightly know, but autrijus once assured me that's correct. :-) 09:59
something about how it's a container, but context still confuses me in p6. 10:00
castaway_ umm, ok 10:02
looks odd from here ,)
nothingmuch =D 10:09
perl -e 'no UNIVERSAL::isa';
that works =)
10:10 _meppl_ is now known as meppl
gaal why is clocksPerSecond = 1000000 in Compat? that's way off 10:10
the correct value is *square* that.
...why's it in Compat anyway? 10:11
ugh! time and time uses it incorrectly :-( 10:12
it should always use 10^12 10:13
broquaint What version, nothingmuch? It doesn't work here ...
gaal oh wait no, it squares it
why would it do that?
what's the spec for 'is copy' on optional args? 10:16
that don't have defaults? 10:17
?eval sub demo(Int ?$arg is copy) { $arg = 4 } demo()
evalbot6 \4
gaal wtf? this throws an error on my box 10:18
?eval $?PUGS_VERSION
evalbot6 \'Perl6 User\'s Golfing System, version 6.2.7, June 13, 2005 (r5232)'
gaal hmm, no it doesn't; but equivalent-looking code in prelude does. 10:19
svnbot6 r5233, gaal++ | localtime builtin. Needs speccing. two TODOs: 12:13
r5233, gaal++ | 1. localtime() fails loudly. use localtime(time) for now.
r5233, gaal++ | 2. scalar localtime doesn't work yet; we need to work out context better.
Aankhen`` gaal >> You're not working on precompiling Prelude? 12:24
gaal it already is precompiled, but pugscc isn't complete (need autrijus for that) so that's disabled for now.
Aankhen`` Ah, okay. 12:25
gaal nice bughunting task: see why localtime() crashes while similar code outside prelude doesn't. 12:26
Aankhen`` In a minute.
Rebuilding Pugs.
gaal i didn't mean i was assigning this to you :) though if you want to pick it up, go for it! :) 12:27
Aankhen`` I know you didn't, but I've got some time. :-)
gaal Aankhen``++
Aankhen`` Shh!
Only after I do it!
gaal async { sleep $n ; Aankhen``++ } 12:28
Aankhen`` Heh.
Um... where is Prelude? 12:32
gaal src/perl6/Prelude.pm 12:33
Aankhen`` Ah, right.
gaal this whole time issue needs more speckwork.
Aankhen`` *** Can't modify constant item: VUndef 12:36
at <prelude> line 159, column 13-42
gaal ah! /me is happy to find that autarch is the maintainer of p5's Time::Local
Aankhen`` That's the error you're getting? 12:37
gaal yes.
backlog a little bit, and you'll see i don't get the same error for similar code in regular circumstances
kungfuftr castaway_: about? 12:42
castaway_ yup 12:43
Aankhen`` ?eval 10^12 12:52
evalbot6 10
gaal oops
Aankhen`` ?eval 10^^12
evalbot6 bool::false
Aankhen`` ?eval 10**12
evalbot6 1000000000000
gaal nice catch
Aankhen`` :-)
Aankhen`` is checking to see if it works now. 12:53
Hmm.
Now the line has moved.
Oh, no, that's just my debugging.
gaal fixed r5234
svnbot6 r5234, gaal++ | 10^12 => 10**12, Aankhen``++10^12 => 10**12, Aankhen``++10^12 => 10**12, Aankhen``++10^12 => 10**12, Aankhen``++10^12 => 10**12, Aankhen``++10^12 => 10**12, Aankhen``++10^12 => 10**12, Aankhen``++10^12 => 10**12, Aankhen``++10^12 => 10**12, Aankhen``++10^12 => 10**12, Aankhen``++
gaal errr 12:54
Aankhen`` stares at gaal.
gaal i didn't say that
really i didn't
must have hit '10' in vi before editing, and ESC:wq quickly. 12:55
Aankhen`` Aankhen``-- Aankhen``-- Aankhen``-- Aankhen``-- Aankhen``-- Aankhen``-- Aankhen``-- Aankhen``-- Aankhen``--
gaal but that's not the souirce of the problem, anyway
lol
i mean it was certainly a bug
Aankhen`` Yeh, I'm still working on it.
gaal right, i'm p6-ling. 12:56
Aankhen``++ :)
Aankhen`` ?eval sub demo (Int ?$foo = 5 is copy, Int ?$bar = 10 is copy) { $bar += $foo; }; demo(10, 10); 12:57
evalbot6 Error: Can't modify constant item: VInt 10
Aankhen`` `is copy` is being interpreted as `is rw` there? 12:58
?eval sub demo (Int ?$foo = 5 is copy, Int ?$bar = 10 is copy) { $foo += 5; }; demo(10, 10);
evalbot6 Error: Can't modify constant item: VInt 10
Aankhen`` ?eval sub demo (Int ?$foo = 5 is copy, Int ?$bar = 10 is copy) { $foo += 5; }; my ($foo, $bar) = (10, 15); demo($foo, $bar);
evalbot6 Error: Can't modify constant item: VRef <Scalar>
Aankhen`` ?eval sub demo (Int ?$foo is copy, Int ?$bar is copy) { $foo += 5; }; my ($foo, $bar) = (10, 15); demo($foo, $bar);
evalbot6 \15
Aankhen`` Wha? 12:59
?eval sub demo (Int ?$foo is copy, Int ?$bar is copy) { $foo += 5; }; my ($foo, $bar) = (10, 15); demo(foo => $foo, bar => $bar);
evalbot6 \15
Aankhen`` ?eval sub demo (Int ?$foo = 10 is copy, Int ?$bar = 10 is copy) { $foo += 5; }; my ($foo, $bar) = (10, 15); demo(foo => $foo, bar => $bar);
evalbot6 Error: Can't modify constant item: VRef <Scalar>
Aankhen`` Pugsbug?
?eval sub demo (Int ?$foo = 10 is copy, Int ?$bar = 10 is copy) { ($foo, $bar); }; my ($foo, $bar) = (10, 15); demo(foo => $foo, bar => $bar); 13:00
evalbot6 (\10, \15)
gaal I think so! add it?
Aankhen`` OK.
gaal $n has passed.
Aankhen`` ?eval sub demo (?$foo = 10 is copy, ?$bar = 10 is copy) { ($foo, $bar); }; my ($foo, $bar) = (10, 15); demo(foo => $foo, bar => $bar); 13:01
evalbot6 (\10, \15)
Aankhen`` Committed bug test. 13:07
svnbot6 r5235, Aankhen++ | * added bug test: `is copy` ne `is rw`.
Aankhen`` Please take a look at it in case I missed something.
BBL. &
gaal thanks!
do stringification methods work? what's the syntax? method ~ ? 13:18
method ~ is not it. 13:20
mugwump prefix:<~> I've seen 13:22
gaal ?eval class Foo; method prefix:<~> { "mmmm!" } ~Foo.new 13:23
evalbot6 Error: Wrong number of invocant parameters: 0 actual, 1 expected
gaal ?eval class Foo; method prefix:<~> (:) { "mmmm!" } my $x = Foo.new; "$x" 13:24
evalbot6 '<obj:Foo>'
gaal ?eval class Foo; method prefix:<~> () { "mmmm!" } my $x = Foo.new; "$x"
evalbot6 '<obj:Foo>'
gaal oh well.
QtPlatypus Was the whole precompiile Prelude thing rejected? 13:29
gaal oh, no, it just needs some fixes in the compiler.
svnbot6 r5236, gaal++ | localtime() works, as does localtime(3.14). 13:39
QtPlatypus reads S29 "How likely is this to be?" 13:52
svnbot6 r5237, Stevan++ | Perl6-MetaModel : removed the add_class_attribute and cleaned up accessor generation; removed the MethodTable object (I didnt need it)
gaal what do you mean, Qt?
QtPlatypus The uglyness of the names. 13:53
gaal there's plenty of work on the draft 239, but it's slowly being done
you mean class names? they're exported by default
QtPlatypus Arh ok.
gaal s/they're/builtins/
QtPlatypus Its just the line "Not all functions are guaranteed to be imported into the global package" 13:56
threw me
gaal i think that refers to things like reset
that nobody wants except for p5 compat
QtPlatypus or dump
gaal heh. 13:57
QtPlatypus Should reduce's definition should be changed so to makes use of identy defaults? 14:02
gaal was there a last word on that thread?
QtPlatypus I think it was one of thouse threads that never end 14:03
gaal me too.
QtPlatypus But I know that larry blessed identy defaults for [...] and was wondering about its use in >>...<< 14:05
Limbic_Region seen autrijus 14:41
jabbot Limbic_Region: autrijus was seen 1 days 10 hours 50 minutes 44 seconds ago
Limbic_Region hmm
I am playing hookie today and have time to debug why Prelude.hs fails to compile on my home Win32 machine
gaal worksforme, L~R 14:42
make clean?
bbiab &
Limbic_Region gaal - try an entire clean partition starting from scratch 14:43
and it only appears to be b0rk on home machine
interesting to note - making unoptimized blows up at a different point 15:01
Compiling Data.Yaml.Syck ( src/Data/Yaml/Syck.hs, src/Data/Yaml/Syck.o )
Aankhen`` gaal >> prefix:<~> defines what happens when the ~ operator is used; I believe what you're looking for is something like this: multi sub *coerce:<as> (Foo $self, Str ::to) { "mmmm!" } 15:39
gaal rehi 15:52
Aankhen``, i don't understand your signature: why the Str::to? 15:53
Aankhen`` Well, AFAICT, the second parameter is the class which is being coerced to.
In this case, you're handling Str, or any derivative of it. 15:54
gaal oh, i wasn't paying attention. hmm, need to reread on coercions.
they don't work yet, do they? in pugs that is
Aankhen`` Not that I know of. 15:55
The sub is parsed, but not used.
gaal ok
L~R, are you getting 'not in scope' errors on PreludePC? 16:02
i've started getting them also, but they were just warnings before i took a break. 16:03
Limbic_Region no gaal
this has been going on for a long time now 16:04
gaal what's the error then?
Limbic_Region blow up compiling Prelude.hs (with optimized build)
gaal is there an error message?
Limbic_Region and blow up compiling Data.Yaml.Syck ( with unoptimized build)
NMAKE : U1077: 16:05
Limbic_Region will do the optimized build again to see if it is different, but I don't think so
gaal syck is no different in unoptimized. 16:07
Limbic_Region gaal - now that's interested
pasteling "iblech" at 80.81.16.68 pasted "Hm... OO question" (32 lines, 1.1K) at sial.org/pbot/11540
Limbic_Region err interesting even
I can't properly make clean
it complains about a line being too long
gaal rm -rf * and svn co again?
Limbic_Region so I will wipe out and start from scratch
yeah
Aankhen`` Limbic_Region >> You need a newer version of NMAKE. 16:08
Are you using 1.5-something?
Limbic_Region Aankhen`` - I am using the latest one provided by MS
but I can verify it isn't a path issue though
Aankhen`` Could you show me the prolog it spits out?
Limbic_Region it doesn't spit out any prolog that I know of 16:10
if that was directed to me
Aankhen`` It ought to say "Microsoft (R) Program Maintenance Utility Version Blah Blah"...
Limbic_Region heh
prolog != programming language in this context 16:11
1.50
Aankhen`` Yep.
Right.
That's the latest one?
Limbic_Region I am checking now
Aankhen`` Microsoft (R) Program Maintenance Utility Version 7.10.3077
Copyright (C) Microsoft Corporation. All rights reserved.
Limbic_Region think that came with a compiler
gaal no, i have 7.
Limbic_Region where do you get it? 16:12
Aankhen`` Older versions of nmake can't handle long command lines.
I believe you can download the VC++ 2003 toolkit or something like that.
I got mine with VS.NET 2003.
Limbic_Region right
gaal yes. see the README.win32 file in a perl5 tarball.
Limbic_Region that's the point
I believe unless you download a larger application 16:13
the latest you can get of just nmake is 1.5 something
Aankhen`` Yes, if you download the standalone version, that's the latest one.
Limbic_Region ok - so that's an interesting side point 16:15
but not relavent to the problem at hand
since I can use the 1.5 version at work and have it compile just fine
just started optimized build
Aankhen`` Dinner. & 16:16
gaal different subversion maybe?
Limbic_Region NMAKE : fatal error U1077: 'C:\WINNT\system32\cmd.exe' : return code '0x1'
gaal - I have verified that the downloaded file is the same both at home and work 16:17
gaal hmm. sorry, no clues then.
Limbic_Region in fact - what is the md5sum of your src/Pugs/CodeGen/PIR/Prelude.hs on Win32 ? 16:18
gaal 4ccb5e280a7b67529ebcfa3c297eabd0 16:19
Limbic_Region 4ccb5e280a7b67529ebcfa3c297eabd0 *Prelude.hs 16:20
yep - same file
cognominal_ ?eval (1|2)<<.say 16:21
evalbot6 Error: unexpected "<" expecting term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
Limbic_Region the only difference that I know of between work and home is WinXP Pro vs Win2K
cognominal_ I thougt << was a fallback for the opening french quote
iblech ?eval (1|2)>>.say 16:22
evalbot6 Error: unexpected ">" expecting term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
gaal cognominal, it's also the hyperop. 16:23
cognominal_ ho, I got them backward, but that does not change anything
gaal ?eval (1,2,3) >>+ 1
evalbot6 Error: unexpected ">" expecting operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
gaal hmmm. that used to work.
iblech ?eval (1,2,3) »+ 1
evalbot6 Error: unexpected "\\" expecting operator, postfix conditional, postfix loop, postfix iteration, ";" or end of input
gaal ?eval (1,2) >>+<< (5,6)
evalbot6 (6, 8)
gaal at least that still does. 16:24
svnbot6 r5238, gaal++ | update localtime sig according to lwall's decree.
cognominal_ oukaze? :)
that would be ukase in English 16:25
gaal exactly. :) 16:26
cognominal_ you tried to Rat on him... :) 16:27
gaal what he disliked wasn't the Rat, it was the double Int.
coral ?eval &localtime 16:33
evalbot6 \sub {...}
coral neat
16:39 Limbic_Region_ is now known as Limbic_Region
Limbic_Region Aankhen`` fwiw - Microsoft Visual C++ Toolkit 2003 does NOT come with a copy of nmake.exe 16:39
gaal L~R, one of the supporting kits does then. 16:40
.NET platform SDK or whatever they call it
Aankhen`` Wha?
I coulda sworn it does. :-\
Limbic_Region I think it was an oversite
Aankhen`` I'm really sorry if I wasted your time...
Limbic_Region I read an msdn blog article after I installed it
and the .NET framework
and told it no - it wasn't allowed to contact the internet 16:41
no worries
so now I am back to square 1
cognominal_ ?eval my @a=777; my $a=666; say << 1 @a $a >>
evalbot6 Error: No compatible subroutine found: "&say"
Aankhen`` Did you get the .NET SDK?
Limbic_Region a Prelude.hs that refuses to compile for no apparent reason
gaal L~R, want mine?
Limbic_Region Aankhen`` - no
gaal - that would be great
gaal L~R, please also read perl5's README.win32
cognominal_ how come, say does not work with the bot? 16:42
Aankhen`` I think the .NET SDK has it.
Limbic_Region Aankhen`` - I really don't want to have to download the whole thing for 1 file (since I have no intention of using it for anything else)
Aankhen`` Hmm.
What's your e-mail address?
Limbic_Region gaal - I am sure I have in the past - but when I build Perl on Win32 it is usually mingw or Cygwin
cognominal_ eval my @a=777; my $a=666; print << 1 @a $a >>
Limbic_Region [email@hidden.address] 16:43
cognominal_ ?eval my @a=777; my $a=666; print << 1 @a $a >>
evalbot6 Error: No compatible subroutine found: "&print"
Limbic_Region goes to look at the readme though since gaal has mentioned it twice now
iblech ?eval say 42
evalbot6 Error: No compatible subroutine found: "&say"
iblech That used to work
(print won't, however, but say should)
Aankhen`` ?eval my @a=777; my $a=666; << 1 @a $a >> 16:44
gaal forum2.org/gaal/pugs/nmake.exe
evalbot6 3
Limbic_Region ?eval say "42"
iblech ?eval *say 42
evalbot6 Error: No compatible subroutine found: "&say"
Aankhen`` Bleh.
evalbot6 Error: No compatible subroutine found: "&say"
Limbic_Region hmm
Aankhen`` "nmake.exe is an executable file. For security reasons, GMail does not allow you to send this type of file."
gaal see url above.
iblech I think I know why it broke -- it's because the declaration of the safe &say isn't exported correctly or so
Aankhen`` I saw.
BTW, L~R, you want a GMail invite?
cognominal_ where << >> is documented? the implemantation interpolates scalars but not arrays
Limbic_Region Aankhen`` - no, I have 50 or so at any given time 16:45
Aankhen`` OK.
Limbic_Region I use that address exclusively for list subscriptions though
gaal s03 probably, cognominal?
Limbic_Region Aankhen`` - just change the extension - I will change it back
cognominal_ ho, I did not know that thru gmail, i belonged to a private club...
Limbic_Region real good security you know
Aankhen`` gaal put it up at forum2.org/gaal/pugs/nmake.exe 16:46
Limbic_Region oh
even better
let me lower the firewall
Aankhen`` Does Kwid have any mechanism for marking up edits? 16:47
Limbic_Region good news is I can now nmake clean 16:48
Limbic_Region attempts a new build
Aankhen`` Can roles have BUILD and the like? 16:49
Limbic_Region same error - thanks gaal and Aankhen`` though 16:50
Aankhen`` Limbic_Region >> Check that it's using the new version.
Oh.
I thought you were talking about clean.
Could you nopaste the full output? 16:51
pasteling "Limbic_Region" at 24.35.57.240 pasted "src/Pugs/CodeGen/PIR/Prelude.hs is still blowing up on Win32 (home machine only)" (45 lines, 3K) at sial.org/pbot/11541 16:52
Limbic_Region out of curiosity - did the ghc requirements change and no one told me? 16:53
Aankhen`` I am using 6.4 without problems.
Do you have Parrot installed?
Limbic_Region no
Aankhen`` You could try building that and putting parrot.exe in your path.
Limbic_Region has always worked before without it - as well as working at work without it 16:54
but sure
attempting build now with parrot in path 16:58
same error 17:00
;-)
Aankhen`` :-\ 17:01
Limbic_Region you will never guess in a million billion years what the problem was 17:04
svnbot6 r5239, Aankhen++ | * modules/Grammars/Perl6.pm: numerous small changes and a few fixes.
Limbic_Region does a little dance for figuring it out
gaal disk space? :)
Limbic_Region GHC-- for 1 .. Inf;
no
gaal well? 17:05
Limbic_Region ghc ships with a perl.exe and perl56.dll in the root directory
gaal in *its* root dir, yes
Limbic_Region it was constantly interfering with my installed perl since . takes precedence when searching %PATH
so I deleted it
no harm no foul right 17:06
gaal but you shouldn't have that dir in your path
Limbic_Region gaal - it isn't
gaal you should have bin/ in the path
Limbic_Region I mean when I am in GHC's root directory
as in
Aankhen`` o_O
gaal why are you ever? :)
Aankhen`` You were trying to build Pugs from GHC's root directory?
Limbic_Region that's where I install Pugs
Aankhen`` Woo.
Limbic_Region I have everything Pugs related under a single directory 17:07
to include parrot now btw
so anyway
gaal L~R, please add this to the infrequently asked questions list :-)
Limbic_Region you can still do perl in ghc's root directory since it checks %path
gaal but you should not have deleted ghc's perl. i'm sure it needs it.
Limbic_Region nope
any ole perl will work
it just needs to be in a hardcoded path 17:08
which is why I said GHC--
since I didn't have 5.6 lying around
gaal also i don't see what perl had to do with your syck breakage.
Limbic_Region neither do I
if you feel like bug hunting - just rename your perl.exe in GHC's root directory ;-) 17:09
note the rename and not delete
Aankhen`` - fwiw, pugs is a sub-directory in the ghc root directory (not IN the root directory itself) 17:10
Aankhen`` I figured. :-)
gaal oh, c'mon you wimps! install everything in C:\ :-P 17:11
Aankhen`` Hehe.
Aankhen`` considers porting MP3::Info.
Limbic_Region ok - so now I forgot what exactly it was I wanted to work on 17:12
oh - I wanted to see if any of the pugsbugs I had submitted tests for were working now
Limbic_Region really doesn't feel like running nmake test
are any of the Win32 smokes up to date?
nope - both are at least a month old 17:13
Aankhen`` Looks like `seek` hasn't been implemented.
gaal mine is a day or two old if that's any help
Aankhen`` No MP3::Info for now, I guess. 17:14
gaal perlcabal.org/~gaal
Aankhen`` wanders off to find a decent game to play.
gaal Aankhen``, want me to look into it?
ah :)
Aankhen`` gaal >> Actually, what I would really like is a working Time::Local. :-)
And MIME::Base64.
Limbic_Region gaal - if you update regularly you should get autrijus to add it to the list on the pugscode homepage
Aankhen`` So if you need help with your priorities, there you go. ;-) 17:15
gaal you have a working localtime now, if that's what you mean
Aankhen`` BBL, jumping around and shooting people. &
gaal ducks
Limbic_Region gaal - thanks - it looks like a few have been fixed, but not all of them 17:16
gaal need a newer smoke? i'll start one.
Limbic_Region no 17:17
thanks though
gaal well i started one anyway, so there :-)
Limbic_Region heh
I am just looking at removing some bug workarounds from an example before committing it
gaal as for the smoke index, i think it's mentioned in the wiki, but i don't automate the build so it's not crucial that it be advertised widely. 17:18
Limbic_Region actually I think a new smoke is in order 17:19
the past smoke doesn't list one of the tests I am checking for
gaal one will be up in 30 minutes.
Limbic_Region hmmm - but I thought you said a few days 17:20
June 29th is when I uploaded the test
*shrug* - I am going to take Jean shopping and will bbiab 17:21
afk &
cognominal_ clkao: this is not Perl6... I think I understand most or Data::hierarchy, but I don't get the sticky stuff... 17:28
gaal hey autarch :)
cognominal_ clkao, can you give me a hint?
autarch gaal: hey, thanks for localtime 17:29
gaal sure! the string form doesn't work yet, as you may have noticed
autarch not an issue for me
gaal thanks for the nice mail on p6-l btw 17:30
want to work on timegm in the prelude?
autarch should it be in the prelude or in a Time::Local module?
it's a fairly involved chunk of code (take a look at the P5 Time::Local)
gaal i need to add another field, to represent the opaque haskell structure
clkao cognominal_: it's sticky. not inheritable for children
cognominal_: why arey ou trying to use d::h?
autarch OTOH, I can probably more or less copy the P5 code 17:31
gaal looks
hah, i forgot macos used a weird epoch 17:32
autarch yeah, it's insane
I'm guessing P6 will not support MacOS <X 17:33
gaal p6 hides the epoch anyway.
autarch thank god
epochs are evil
gaal hmmm, we don't have pack yet. is that critical? 17:34
autarch I don't think so, let me look
no, that's just for caching
gaal line 54
autarch I'd probably implement a non-caching version first
however, can I force integer math with Pugs? That's important, or else I need to use do int( $year / 100 ) everywhere 17:35
gaal you can say Int $year...
(i already do has Int $.year for the field) 17:36
autarch and that'll force integer math?
that'll work
gaal ?eval class Foo; has Int $.x; method halve() { $.x / 2 } Foo.new(x=>5).halve 17:37
evalbot6 2.5
gaal oops :)
but that's a bug. 17:38
autarch heh
gaal members aren't in a locked hash at all yet.
you can still use $:undeclared_private_var i think - they're internally just hash elements.
?eval my Int $x = 5; $x / 2 17:39
evalbot6 2.5
gaal that's a worse bug. :-P
actually no:
?eval my Int $x = 5; $x /= 2; $x 17:40
evalbot6 \2.5
gaal oh well.
autarch heh, int(...) it is
gaal with an XXX to clean up when it's done. 17:41
autarch yeah
gaal okay; do you think we need the positional versions of these functions now?
or can we for now always go through the oop interface, i mean. 17:42
autarch for which, timegm & timelocal? 17:43
I like the idea of OO only
the positional version is _so_ ugly and hard to remember 17:44
gaal yes. okay, so these two just do some validation and throw everything at new().
autarch plus if it's a method on some built-in object, caching is simple
PerlJam gaal: For positional, do you mean the same as in P5 or the reverse as in your email to p6l?
autarch either way it sucks to use ;)
PerlJam I think a p5 positional interface would be useful to speed converting p5 to p6 17:45
gaal right. i have to peek at haskell a little, to see if i can sotre their opaque structure in p6.
autarch PerlJam: that's true
gaal it seems that we can't. so instead i'll add an epoch field. 17:53
i'll make it Num :-)
the good news is that if you construct a Time, you can pass either an epoch or everything else, and what you don't supply gets computed. 17:54
at least i think that's good news? please tell me if you think this is slow and bloated. 17:55
meh, you can't get an epoch in haskell without doing calculations. 18:03
autarch that's reasonable, it can always be changed to compute on demand and cache later 18:06
where is the Time interface defined?
gaal nowhere :) we're making it up as we go 18:07
autarch ok
but where are internal objects coded? In haskell or the prelude? 18:08
gaal see src/perl6/Prelude.pm ; it relies on Pugs::Internals::localtime in src/Pugs/Prim.hs 18:09
autarch ok
gaal bbiab 18:14
kolibrie mugwump was working on a Date framework, too: ext/Date/README.pod 18:15
autarch I know 18:16
kolibrie ok, so long as everyone is talking to everyone
brentdax A couple days ago (I'm building a new Pugs right now, can't see if it's still there) lookups like $object.foo.bar were broken. Does anyone know if they still are? 18:41
wolverian Foo.new.bar seems to work 18:42
(when bar is defined.)
brentdax I'll check it again when this finishes compiling... 18:47
btw, does anyone else have trouble with optimized Perl 5 embedding builds?
autarch brentdax: are you on debian, perchance? does it complain about _crypt_struct 18:49
brentdax I'm on Ubuntu, which is Debian-derived. I think that sounds right.
autarch brentdax: yeah, I saw thought on Ubuntu too 18:50
obra hey autarch
autarch my fix was to run make unoptimized, let it compile that one file, stop the compile, then run plain make again
obra: hey there
brentdax What I've been doing is running make, letting it get to that file and crash, then do the rest of the build with make unoptimized. 18:51
Your way is probaby better, though. 18:52
autarch brentdax: it's just that one file, so you can stop the build right after it does that file, delete any .o files it made after that, and then go ahead optimized
gaal there's a compiler directive to turn off optimizations for one file. which one is it, i'll add the flag.
autarch src/Pugs/Embed/Perl5
geoffb *yawn*
hello, all
geoffb has finally sent his article to the editor, and can now relax and spend a couple hours reading IRC backlog. :-) 18:53
BTW, sharing a repository with your editor is the only way to fly 18:54
brentdax Compiling Pugs.Embed.Perl5 ( src/Pugs/Embed/Perl5.hs, src/Pugs/Embed/Perl5.o ) 18:55
/usr/lib/perl/5.8/CORE/reentr.h:611: error: field `_crypt_struct' has incomplete type
gaal brentdax / autarch: please try editing that file and changing the first line to 18:56
{-# OPTIONS_GHC -fglasgow-exts -cpp -O0 #-}
and tell me if it helps / just ci it yourself
(my debian box is too puny to run anything but make unoptimized on anyway :) 18:57
brentdax Change made, compiling now... 18:58
No joy. 18:59
gaal too bad :/
gaal is off again & 19:00
Limbic_Region assumes the integer math is needed to determine epoch and not for returned values since $larry wants fpm 19:15
meppl "$larry"? 19:28
*g 19:29
Limbic_Region doesn't understand the question meppl
if there is one
meppl i thought it was inadvertent 19:31
$larry as variable has no sense ;)
Limbic_Region meppl - to distinguish larry from @larry we use $larry 19:33
meppl achso 19:34
ah okay
in any case its logical senseless to write a "$" for a substantiv of a human-language. or we could say "i go $home and eat a $soup from my $grandma" :D" 19:37
Juerd people with utf8 shells, login on feather please :) 19:39
Limbic_Region how would I know if I have a utf8 shell? 19:40
Juerd Can it display unicode characters? :)
Then very probably it is.
integral it's upside down?
Limbic_Region dunno - never tried
Juerd integral: :)
Limbic_Region has only logged on to feather once
that was to change my password
integral Juerd: the line length is weird though, they look longer than 80 19:41
Juerd Blame thy tty. 19:42
They're 80
integral hmm, must getting some widths wrong then
ah, it is 19:43
Amnesiac yo juerd 19:46
'lo Limbic_Region, integral
Aankhen`` goes to sleep. 19:54
G'night.
Limbic_Region salutations Amnesiac 19:56
wolverian for the love of god, this game has a copy protection scheme that slows down your computer with an insane process if you bought the game 20:08
kolibrie Juerd: hmm, playing with our minds, are you
Juerd :))
Limbic_Region um - is t/builtins/caller.t b0rk? 20:13
looks like you planned 19 but only ran 3
no such sub ....
eric256_ which game wolverian? 20:25
wolverian eric256_: vampire the masquerade: bloodlines 20:27
eric256_ ahhh. thought maybe you where talking about half life 2. it encrypts the entire contents of like 5 CD's...pain in the but 20:28
wolverian heh.
Limbic_Region pc based games? 20:29
Limbic_Region has considered doing some gaming to bring some balance and wasn't sure what he should get
wolverian yes.
Limbic_Region: what kind of games do you like?
Limbic_Region um - mostly ones that involve either no thinking - (first player shooters) 20:31
or ones that involve almost entirely thinking (Myst - RPGs - etc)
but I am not much of a gamer
autarch then get a PS2 20:32
lots of choices, minimal work ;)
running Windows just for games is too much of a PITA
eric256_ seconds Limbic_Region's games chioces. ;)
Limbic_Region I already own a pc that's dual boot autarch
autarch ah
I used to do that, but I hated leaving linux just to play games 20:33
Limbic_Region autarch - actually Windows is the primary partition
eric256_ ps2++ psp+++ ;)
wolverian Limbic_Region: I can recommend the latter category games mainly
Limbic_Region well - shoot wolverian 20:34
autarch Limbic_Region: eek
wolverian Limbic_Region: Fallout 1 & 2, Arcanum: of Steamworks and Magick Obscura, The Longest Journey, and Planescape: Torment
Limbic_Region autarch - it works out. While I am heavily *nix biased
I use them both equally
wolverian Limbic_Region: recently those genres have been practically dead on the PC, which sucks
(and none of those games are recent)
Limbic_Region I figure it will keep my skills up when/if the need arises - and it is only fair to b1tch about something when you have walked a mile in its shoes
wolverian - you think I could pick them up in a game shop or should I try to order online? 20:35
autarch I've used Windows plenty. Heck, I used to do tech support ;)
wolverian Limbic_Region: They'll be rereleases if they're in the shops.
Limbic_Region: I'd try ordering first.
Limbic_Region gotcha
eric256_ Prince of Persia is a good balance of both shooter and puzzle 20:36
wolverian Limbic_Region: some people also like Morrowind, but it always failed to attract me, in all its sequels 20:37
(it is the prettiest of the bunch, if that matters, though)
(it's a first-person game, too.)
eric256_ cheers when he finds that most if not all his 11 v-mails are junk. 20:38
masak eric256_: prince of persia can hardly be called a "shooter" 20:39
eric256_ i meant as a first person combat, if you count throwing stuff it is a shooter ;)
masak ;)
gaal ooh, autrijus posted. 20:53
masak poor autrijus... "underclocked" is not like him 20:56
Corion datenzoo.de/pugs/win2k.html
(a smoke, long time not smoked :) )
cu all and good night
Limbic_Region thanks wolverian 20:57
and even though Corion quit - thanks to him too - I bugged him about the smoke in the chatterbox
gaal lr, i have that smoke up too. 20:58
Limbic_Region is confused by the results
gaal that != from right now 20:59
Limbic_Region stuff that was b0rk is fixed
and stuff that was working is b0rk
gaal that == the one i started a few hours ago.
Limbic_Region right
gaal i don't have parrot / p5.
Limbic_Region neither do I
or rather I do
but they don't have anything to do with what I was looking at
gaal k 21:00
Limbic_Region gaal - it is
errr
your results don't agree with Corion 21:01
all 4 of parsing_spaceship.t failed for Corion but 2 passed for you
*shrug* - doesn't really matter since at least 2 are still b0rk 21:03
rafl Who is responsible for the debian/ dir in svn? 23:23
meppl what is the sense of this folder? 23:24
kungfuftr rafl: look at the logs?
rafl meppl: It's used to build pugs packages for the Debian GNU/Linux Distribution.
kungfuftr: Good point.
meppl oh 23:25
revdiablo rafl: scw mostly
meppl im debian-user
rafl "idle: 6 days" Hm... 23:27
So, I'd like to fix some things and make some improvements on the package. Shall I ask scw or ask someone else for svn access? 23:28
geoffb rafl, well, I'm no metacommiter, but the general rule around here is to ask for the commiter bit, and then go wild. 23:30
mugwump rafl: you can ask anyone listed as an admin on the openfoundry project page
revdiablo rafl: everything is in source control. if you screw something up, it can just be reverted. 23:35
svnbot6 r5240, mugwump++ | How to shoot yourself in the foot with Perl 6
Odin- O_o 23:36
Interesting commit comment.
mugwump in this case it's literal
revdiablo A /docs/SHOT_IN_THE_FOOT
hehe
oh lord, my working copy is 600 revisions behind 23:37
mugwump++ 23:40
Odin- Good one. :) 23:45
kungfuftr mugwump-- # bloody kiwi 23:47
svnbot6 r5241, mugwump++ | Test for 'is accessor' trait
mugwump runs around kungfuftr flightlessly and pecks at his feet 23:50

Powered by Cro and the Raku® Programming Language.

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

Thank you!