-Ofun: xrl.us/hxhk | pugscode.org | pugs.kwiki.org | paste: paste.lisp.org/new/perl6 or sial.org/pbot/perl6
Set by apple-gunkies on 11 November 2005.
clkao seen ibleach 00:02
jabbot clkao: I havn't seen ibleach, clkao
rafl clkao: Do you mean iblech? 00:03
clkao ya
rafl clkao: aka Ingo Blechschmidt.
clkao: He's not that active on IRC but he answers mails in <24h normally.
clkao ok, thansk
trying to find out how jsbackend returns a perl6 object 00:04
rafl iblech is the right person to ask that, indeed. 00:05
q[Skud] what's perl6's equivalent for $#array? 00:12
Alias_ @array.something 00:13
q[Skud] i got that far :)
Alias_ ah
q[Skud] it's not length, that's string-only 00:14
integral eval: my @a = (1, 2, 3); $#a;
?eval my @a = (1, 2, 3); $#a;
q[Skud] and i can't find anything in S29
00:14 evalbot_8225 is now known as evalbot_8233
evalbot_8233 Error: unexpected "#" expecting "::" 00:14
buu ?eval @x=1..4; @x.length
evalbot_8233 Error: Undeclared variable: "@x"
buu ?eval my @x=1..4; @x.length
evalbot_8233 Error: No such method: "&length"
buu ?eval my @x=1..4; @x.keys
evalbot_8233 (0, 1, 2, 3)
buu heh
rafl elems?
buu ?eval my @x=1..4; @x.elems
evalbot_8233 4
rafl yay!
buu eval: 1
hrm 00:15
Yay!
integral ?eval my @a = (1, 2, 3); @a.end
evalbot_8233 2
integral (the trick is to grep ../src/Pugs/Prim.hs for elems, then go up one line...)
stevan_ ?eval my @a = (1, 2, 3); +@a; 00:16
evalbot_8233 3
q[Skud] ok, elems it is.
integral q[Skud]: umm, that's not the same as $#a, @a.end is the same as $#a.
q[Skud] oh, right.
sorry. i am not quite awake here
huh. 00:19
here's a slightly wacky thing.
say "Sorted: ", @animals.sort; 00:20
say "Reversed: ", @animals.reverse;
Sorted: aardvark bat bear cow dog zebra
Reversed: zebraaardvarkcowdogbatbear
integral heh
q[Skud] i presume that's not meant to happen. 00:21
integral it's something todo with arrays vs. lists
the sort seems to be working with an array, but the reverse with a list 00:22
q[Skud] yeah.
integral curiously, ~@a.sort and ~@a.reverse produce a spaced string each 00:24
q[Skud] ok, you'll have to excuse me if i kind of blink and go "duhhhh..." like i'm a bit short on neural function here, because i only started playing with this yesterday. 00:26
integral np :-)
q[Skud] just kinda starting as a "dumb user" and working through perlintro trying to figure out what each thing there is, in perl 6 idiom. 00:27
errrr. how do i declare a hash and access one of the elements? like, in perl5, it would be $days{'February'} 00:28
ah, got it 00:29
i hadn't quoted the key
integral yeah, {} doesn't autoquote now. You need to use <> or <<>> to get that: %hash<<$interpolated>>, %hash<notinterpolated> 00:31
(<> and <<>> are like the old qw)
q[Skud] ok, but i can still say %days{'February'} 00:32
integral yes. 00:34
it's so that {} is always more like a closure
clkao implmemented slurp 00:37
buu Is there an ioctl function in pugs? 00:45
?eval ioctl 00:46
evalbot_8233 Error: No such sub: "&ioctl"
rafl buu: No. Not yet.
integral if it's in System.POSIX, it should be a trivial exercise to add :-) I think there was a tutorial on the wiki at one point 00:47
buu Bummer =[
clkao runjs is printing 1..0 for some reason 00:59
that looks like skipall
svnbot6 r8234 | skud++ | Some new stuff I encountered. 01:07
r8235 | skud++ | Errr, lots of new stuff. Intro embiggened. Differences likewise. Moved p5 stuff into a 01:09
r8235 | skud++ | subdir.
clkao nopaste.snit.ch:8001/5868 01:14
clkao grins
svnbot6 r8236 | clkao++ | Implement slurp() with Perl5/IO::File.
01:36 blm_ is now known as blm_work
svnbot6 r8237 | clkao++ | * Add --run to runjs.pl, supporting js, jspm, jssm. 01:57
r8237 | clkao++ | * Add --perl5 to runjs.pl, which initialises perl5 related
r8237 | clkao++ | functions for jspm.
r8237 | clkao++ | * Move stuff from jsperl to PIL2JS/JSPM.
tewk Does pugs support here docs? can someone point me to a synopsis or test example? 02:23
meppl gute nacht 02:43
buu Stupid perl5 02:45
Scary fucking vodoo magic.
sili_ what
meppl its technic 02:46
no voodoo
buu Lies.
sili_ just wait until i finish this parrot lisp.
buu meppl: I'm trying to make C use perl file handles.
meppl hm 02:47
i would like xs for perl6
have
but im very patient 02:48
for c++ 02:50
*sleeping* 02:51
Andy obra: I can't /msg you becaue I'm not registered. See AIM. 03:32
sri_ audreyt: please increment the version of Module::Install::Admin in the next release, Catalyst has a prereq on M::I 05:28
gaal morning all... 05:29
blm morning gaal 06:12
avar pugs -e 'my Num @a = 0..Inf; say @a[100]' takes forever;/ 06:14
blm Does perl 6 use lazy evaluation like Haskell?
Sorry is it supposed to do lazy evaluation? 06:15
avar it is 06:17
which is why it shouldn't take forever
(last I read the specs)
sili__ there was talk about laziness and Inf on the lists recently 06:21
i'm not sure how/what was implemented in pugs
blm Is audrey around? 06:44
Alias_ Goddam it... PPI still leaking despite %PARENT hash being cleaned up properly 06:45
blm Hi Alias_
Alias_ heya
blm Is Audrey en route?
Alias_ hmmm about 20 megabytes per 25,000 documents
Probably
blm Do you know Module::Install? 06:46
Alias_ yep
blm Trying to install it from CPAN and it seems to depend on Module::Scandeps. When I try to install Module::Scandeps it seems to find the old version of M::I and bombs out "Can't locate method "_top" via package "PACK001" 06:47
Alias_ hmm... dunno 06:48
blm at Module::Insall::Base line 52
Alias_ The system one or the bundled one
blm Might move old Module::Install out of way.
Get this error! says /usr/local/lib/perl5/site_perl/5.8.7/Module/Install/Base.pm line 52 and there is no dir /usr/local/lib/perl5. 06:50
I don't like cygwin :-(
Alias_ Personally I'm putting my bets on PXPerl being sane 06:51
blm What is PXPerl? (excuse my ignorance :-) ) 06:52
Alias_ An attempt to write a sane Windows Perl install
blm Excellent!
Alias_ One that doesn't rely on a proprietary PPM package
blm You on win32 much? 06:53
Alias_ Win32 desktop, Linux server
avar how do I access <obj:Time::Local> in perl6? doing pugs -e 'say for localtime'
blm Does par work?
avar perl -le 'print for localtime' would work 06:54
Alias_ Reasonably well for bundling applications
blm Includes dependancies?
Alias_ I think so...
blm Cool
svnbot6 r8238 | tewk++ | Simple perl6 test harness for pil. 07:32
avar how can I see which rfc's got accepted? 07:47
dev.perl.org/perl6/rfc/39.html <- I like this one for example, but I don't think it got accepted;/ 07:48
nothingmuch avar: every apocalypse 07:49
they list the RFCs which got considered
as rejected or accepted
but most RFCs were munged 07:50
IMHO 'say' is much more parsable than >foo<, too
and that mixes up with hyper operators and stuff
avar select STDERR;
nothingmuch you can define it though:
avar say >"goes to -err and -out"<;
I just thought it was neat as an inverse of <> 07:51
nothingmuch &circumfix:{'>', '<'} ::= say;
avar does that work in pugs? 07:52
nothingmuch let's try 07:57
?eval &circumfix:{'>', '<'} ::= &say; >
?eval &circumfix:{'>', '<'} ::= &say; >"foo"<;
?eval "foo"
hmm... i think it's dead
07:57 evalbot_8233 is now known as evalbot_8238
evalbot_8238 Error: unexpected ">" expecting ";", statements or end of input 07:57
nothingmuch is not really in touch with pugs lately
evalbot_8238 "foo"
nothingmuch been having a "useful" phase (as opposed to an "interesting" phase) - lots of Catalyst work
avar nothingmuch: I tried, it was more of a "should it work or did I fuck anything up?" question;) 07:58
nothingmuch avar: i guess the grammar isn't dynamic yet 07:59
but it will work
we have an internals lecture at work
see you
avar cya
Skud evening pjf 08:12
PJF Good evening Skud! How are you this fine day?
svnbot6 r8239 | yiyihu++ | Renew the broken links in series of `01Overview.html`. 08:13
PJF Good day, blm. 08:26
Skud: Wow! Lots more documentation. 09:37
Alias_ PPI 1.07 is uploaded! 09:38
Now it can parse line noise!
wolverian yay :) Alias++
Alias_ I leaks a bit still... 09:39
1k per document I think
But I can certainly confirm it isn't me leaking circulars now
So it's now someone else's problem
PJF Congrats, Alias. 09:41
Alias_ At least it's editor-compatible now
Which takes a surprisingly large load off my mind 10:05
wolverian editor-compatible? 10:07
Alias_ As in you can type any old shit and it won't crash the parser
wolverian oh, okay. I think I see.
right. :)
Alias_ Which is a big step
wolverian so, where's the vim plugin?
Alias_ My queue now is basically, 1. Wait for PXPerl to release 2. Get PCE actually running 3. Write Perl::Editor 4. Then you can write a vim plugin 10:08
The parser in it's standalone is surprisingly useless :)
wolverian PCE?
Alias_ Editors need the next layer of support infrastructure too, caching and metrics and transforms, oh my!
The cross platform pure Perl Wx+Scintilla Perl editor 10:09
Which has some pretty big issues, but I'm helping the author to work through them, then I'm going to use it as my base platform for the refactoring Perl editor grant
Anyways
Alias_ home
(It also means you can now parse Perl 5 into pugs btw) 10:10
audreyt: You want 1.107. It's over to you now.
wolverian right. this is exciting :) Alias++ 10:11
rafl Is there a freenode server that listens to a port != 6667? My ISP seems to block that port.. 10:38
avar does perl6 have some quick way to create accessors like ruby? 10:46
wolverian has $.foo; 10:48
see S12
kane_ rafl: cgi:irc installations would work.. like this one: webchat.xs4all.nl/cgi-bin/ircnet/irc.cgi 10:57
integral feather already runs a CGI:IRC :) 10:58
rafl kane_: I actually don't want to use CGI:IRC but irssi and not on feather but on my host :-)
integral rafl: most freenode boxes listen on a few ports around 6667, something like 6665-9 I think
rafl integral: OK, I'll try that. 10:59
kane_ rafl: yeah, this was more 'last resort'
you could tunnel it via feather too btw
rafl kane_: That's what came to my mind as well. But I'll try using IPv6 first. 11:00
kane_ bbl, off to the office
integral 6to4++
rafl integral: Do you know a good tunnel provider? sixx maybe? 11:02
integral I don't. I have decent 6to4 access so I just use that 11:03
avar how do I call my own method within a class, self.mehod() doesn't work (as seen in S12) 11:12
wolverian self.method or just .method if $_ =:= self (as is the default in methods) 11:13
?eval class Foo { method bar { "hi" } method baz { .bar } } Foo.new.baz 11:14
11:14 evalbot_8238 is now known as evalbot_8239
evalbot_8239 "hi" 11:14
PJF ?eval class Foo { method bar {"hi"} method baz { self.bar } } Foo.new.baz 11:15
evalbot_8239 "hi"
PJF ?eval class Foo { method bar {"hi"} method baz { self.bar() } } Foo.new.baz
evalbot_8239 "hi"
PJF Good. You had me worried for a moment there, avar. ;) 11:16
svnbot6 r8240 | asavige++ | Evil documentation plan (first cut based on Audrey caffeine free coke brain-dump) 11:17
avar ?eval role P {method lang { "poo" }} class AE does P {method hello() { self.lang }} AE.new.hello 11:18
evalbot_8239 Error: No compatible subrountine found: "&lang"
avar ?eval class P {method lang { "poo" }} class AE is P {method hello() { self.lang }} AE.new.hello
evalbot_8239 Error: No compatible subrountine found: "&lang"
avar ?eval class AE is P {method lang() { "poo"} method hello() { self.lang }} AE.new.hello 11:19
evalbot_8239 Can't exec "./pugs": Permission denied at examples/network/evalbot//evalhelper.p5 line 46.
avar ?eval class AE is P {method lang() { "poo"} method hello() { self.lang }} AE.new.helloo
evalbot_8239 Can't exec "./pugs": Permission denied at examples/network/evalbot//evalhelper.p5 line 46.
PJF Oh my god! You killed evalbot!
avar ?eval class AE is P {method lang() { "poo"} method hello() { self.lang }} say AE.new.lang
11:19 evalbot_8239 is now known as evalbot_8240
evalbot_8240 OUTPUT[poo ] bool::true 11:19
pasteling "avar" at 212.30.222.56 pasted "woo" (26 lines, 550B) at sial.org/pbot/14952 11:21
avar *** No compatible subroutine found: "&self""
rafl avar: my $langs = self.langs; looks wrong. 11:31
Huh.. no. It doesn't. 11:32
PJF Good evening Andrew. 11:47
asavige hi PJF Audrey left Sydney at 1:00 pm on China Airlines flight C152 11:48
should have just arrived in taipei about now
PJF I have no doubt we'll see her active on IRC again shortly then. ;)
asavige wow, skud has hit this channel like a ... skud missile ;-) 11:49
integral and written lots of documentation!
asavige she is putting me to shame (though I just uploaded Audrey's brain-dump evil plan)
PJF Night all. I seem to have landed myself with real work for the next couple of weeks, and have to sleep at awkward times as a result. Happy hacking. 12:00
Odin-LAP avar: Hmm. Ertu að elta mig? :p 12:09
avar maður mætti spurja þig af því sama! 12:10
Odin-LAP avar: Ég kom á undan. ;)
avar viss? Ég hef ekki verið hérna í einhvern tíma 12:11
ég mun hinsvegar aldrei elta þig á #emacs ;)
Odin-LAP Haha.
Viss?
avar haha 12:12
já
haha 12:16
$ pugs -CParrot -e 'say "foo"'|wc -l 12:17
355
3558
meppl guten morgen 12:43
GeJ servus
clkao nothingmuch: did you mention there's a tapmatrix diff util or something like tha? 12:46
meppl good morning gej 13:03
nothingmuch clkao: ? 13:57
oh
yes
i have that
i'll talk to you about it when I get home
nothingmuch.woobling.org/Test-TAP-HTMLMatrix
nothingmuch.woobling.org/Test-TAP-Model
get those darcs repos
they have working code
(example.pl has dummy failures) 13:58
i have to go pick up my mom from the univ 13:59
ciao!
xinming how to get a line from stdin, and process it, and get another line? 14:03
pd clocks pugs at less than 2000x slower than perl5, woo hoo! 14:17
avar 1 for =<>; ? 15:20
gaal process(=$*IN); $next = =$*IN; 15:30
process($^line) for =$*IN;
xinming try it in pugs first please 15:31
gaal ./pugs -e 'print ">>> $_" for =$*IN' 15:32
this works
but you may be Suffering from Buffering <tm>.
xinming for =$*IN -> $line { print "myline: " ~ $line;} 15:33
Khisanth that works
too :P
xinming the problem is, It will hold all input in a "temperary" list, and after you enter EOF, it then process the list 15:34
Khisanth eh?
xinming Khisanth: You can save it in a file, and run it
Khisanth well I guess laziness isn't implemented yet?
gaal yes, IO buffering.
xinming gaal: How to get rid out of this? 15:35
gaal eval '$| = 1', :lang<perl5>
just kidding :-)
good question. I wonder if getContents or whatsits name can be configured to work in line buffered mode; I but it can, but there's a chance that requires POSIX :-( 15:36
Khisanth mumbled $*OUT is autoflushed; ? :P
gaal uh, the immediate answer for you though is, sorry I don't know. 15:37
pd on perl5: "Perl does not support truly unbuffered output (except insofar as you can syswrite(OUT, $char, 1))." Heh.
gaal pd, in this case linebuffered mode would be sufficient. 15:39
xinming So, It's the problem of haskell. :-)
fglock xinming: it's not a problem with buffering - this works: for 1..10 { print "myline: " ~ =$*IN }
gaal fglock: what is the problem then? 15:40
xinming: it's not Haskell's fault, probably, if that what's you mean.
fglock I think it is a problem with laziness not being implemented yet - it is trying to construct the full =IN list before using it
gaal ah, wonderful: haskell.org/ghc/docs/latest/html/li...IO.html#12 15:41
there are buffering operators including linemode in non-posix haskell io.
Oh. :-/
that is unfortunate. :( 15:42
avar ?eval say​"boo"​if​1
evalbot_8240 OUTPUT[boo ] bool::true
avar harhar
gaal if anyone wants to add simple buffering, btw, it's a nice and easy addition. 15:43
avar ?eval say"boo"if1
evalbot_8240 Error: unexpected "1" or "i" expecting term postfix, operator, ":", ",", postfix conditional, postfix loop, postfix iteration, ";" or end of input
avar >;)
xinming fglock: for your example, you have to write exit the for loop yourself
fglock xinming: yes. I'll try with -Bperl5 15:45
avar ?eval say​​1 15:46
evalbot_8240 OUTPUT[1 ] bool::true
avar ?eval say​1
evalbot_8240 OUTPUT[1 ] bool::true
avar ?eval say1
evalbot_8240 Error: No such sub: "&say1"
gaal parsefail if rand > .5; # wtf 15:47
pd heh.
avar doesn't that bot take ?eval in query?
gaal hmm? 15:48
Khisanth don't think so
gaal fwiw I have an old r6995 lying about that doesn't exhibit this interesting behavior.
xinming why the first `?eval say1` doesn't fail?
Khisanth probably to avoid having people abuse it for various purposes
avar xinming: because it's magical;) 15:49
gaal avar: are you asking if eval works? yes, for p6 eval. not for other :langs.
?eval say eval "hola"
evalbot_8240 OUTPUT[ ] bool::true
gaal ahem.
?eval eval "hola" 15:50
evalbot_8240 undef
gaal er, yes i should learn perl.
Khisanth heh
gaal ?eval say eval "'hola'"
avar is there anything in place to secure that bot?
evalbot_8240 OUTPUT[hola ] bool::true
avar ?eval "foo"
evalbot_8240 "foo"
gaal avar: yes, look into safe mode.
there are both pugs-specific and OS measures. 15:51
some prims are marked as "unsafe" so you can't do them from the bot:
?eval open "/etc/passwd"
evalbot_8240 Error: No such sub: "&Pugs::Internals::openFile"
xinming ?eval $*IN
avar ?eval fork while fork
evalbot_8240 \undef
Error: No such sub: "&fork"
gaal and the thing that dispatches pugs in the bot uses a limit on CPU and time: 15:52
xinming ?eval my $a = =$*IN
evalbot_8240 Error: Can't use readline() or getc() in safemode.
gaal ?eval 1 while 1
evalbot_8240 (no output)
gaal also, there are limits on output size and RAM.
xinming ?eval sub x { y }; sub y { x }; y
evalbot_8240 pugs: out of memory (requested 1048576 bytes)
xinming gaal: I think so
gaal to which please join me in saying, iblech++ :-) 15:53
xinming ?eval system "free"
evalbot_8240 Error: No such method: "&system"
gaal please see examples/network/evalbot/README.pod 15:54
xinming ?eval say1
evalbot_8240 Error: No such sub: "&say1"
avar ?eval $x = 1; $x *= 2 while 0..Inf
evalbot_8240 Error: Undeclared variable: "$x"
gaal say "my".
avar ?eval my $x = 1; $x *= 2 while 0..Inf
evalbot_8240 pugs: out of memory (requested 1048576 bytes)
xinming wonders why avar's example works
gaal xinming: he said "my".
avar ?eval my $x = 1; $x *= 2 while 1 15:55
evalbot_8240 (no output)
xinming avar ?eval say​​1
evalbot_8240 OUTPUT[1 ] bool::true
is there a "my" ? :-/
pd ?eval say "what";
evalbot_8240 OUTPUT[what ] bool::true
xinming ?eval say"what" 15:56
evalbot_8240 OUTPUT[what ] bool::true
xinming ?eval say1
evalbot_8240 Error: No such sub: "&say1"
avar xinming: see, magic;)
pd ?eval say 1;
avar actually, it's not magic
evalbot_8240 OUTPUT[1 ] bool::true
avar you can use any unicode whitespace in perl6 to delimit tokens, including zero-width unicode whitespace
gaal pfff
avar (this should result in lots of clever code) 15:57
"clever"
gaal remind me to hack that out of the parser :-)
pd ?eval sub say1 { say "what"; }; say1;
evalbot_8240 OUTPUT[what ] bool::true
Khisanth but it won't lower your golf score!
avar Khisanth: maybe,
Khisanth: not if you count it by bytes
but if you count by characters...
it'll be the same
but less understandable 15:58
gaal the possibilities for ACME::Bleach are interesting.
pd Remind me to store data in UNICODE characters then.
gaal *Anything* becomes a one-liner.
Khisanth Acme::Onliner?
gaal At least, I propose Acme::Bleach::Unary 15:59
which encodes your program in zero-width chars
the number of chars is a unary representation of the data
fglock xinming: '=$*IN' is not implemented in -Bperl5 yet, so I can't test it
gaal so the module needs not use pesky tabs, spaces, and newlines
and create programs that truly fit on one line! 16:00
I'm racing against myself here, see, because I fully intend to disallow this. I just don't know what I'll get to first. 16:01
xinming the console need to be utf-8 compatible. :-)
gaal brb 16:02
avar gaal: acc. to the synopsis it's there on purpose
?eval 1<2 16:17
evalbot_8240 bool::true
avar ?eval 1>2
evalbot_8240 bool::false
avar ?eval my $x = "foo"; my $y := $x; $y = "bar; $x 16:18
evalbot_8240 Error: unexpected "b" expecting block, "\\", "$/", "$", "$!", "\"", word character, "::", ".", ":", "=", subroutine name, array subscript, hash subscript or code subscript 16:19
avar ?eval my $x = "foo"; my $y := $x; $y = "bar"; $x
evalbot_8240 \"bar"
avar ?eval my $x = "foo"; my $y ::= $x; $y = "bar"; $x
evalbot_8240 \"bar"
avar ?eval my $x = "foo"; my $y ::= $x; $y = "bar"; \$x
evalbot_8240 \"bar"
avar ?eval my $x = "foo"; my $y ::= $x; $y = "bar"; $x =:= $y
evalbot_8240 bool::true
avar ?eval my $x = "foo"; my $y ::= $x; $y = "bar"; my $z = "bar"; $x =:= $z 16:20
evalbot_8240 bool::true
avar I have ¥ on my ibook keyboard somewhere, I just can't find where.. 16:23
gaal avar: Y works wherever ¥ does 16:30
svnbot6 r8241 | kane++ | r484@coke: kane | 2005-12-14 13:28:31 +0100
r8241 | kane++ | * eliminate the use of one global $Obj for the config
r8241 | kane++ | * it's still a singleton though
r8242 | kane++ | r485@coke: kane | 2005-12-14 17:29:25 +0100
r8242 | kane++ | * make the installation variables in config.pm private
r8242 | kane++ | * they are wrapped from Installation.pm
r8242 | kane++ | * add ls_all_accessors that returns private accessors as well in config.pm
r8242 | kane++ | * try and use path::class everywhere
r8242 | kane++ | * every dir/file returned by config.pm is a p::c object 16:31
r8242 | kane++ | * rewrite all code to use the new installation wrappers rather than
r8242 | kane++ | the old config.pm accessors
rafl kane++; #using Path::Class everywhere. 16:39
fglock re minilang.subset(p6) - codespeak.net/pypy/extradoc/talk/os...paper.html - about the pypy project 17:20
avar ?eval "x" x 3 18:02
18:02 evalbot_8240 is now known as evalbot_8242
evalbot_8242 "xxx" 18:02
avar ?eval "x" xx 3
evalbot_8242 ("x", "x", "x")
avar ?eval "x" xx 3 ~ "x"
evalbot_8242 "x x xx"
avar ?eval ("x" xx 3) ~ "x" 18:03
evalbot_8242 "x x xx"
avar ?eval scalar "x" xx 3
evalbot_8242 Error: No such method: "&scalar"
avar ?eval scalar ("x" xx 3)
evalbot_8242 Error: No such method: "&scalar"
avar oh my
?eval scalar str("x" xx 3) 18:10
evalbot_8242 Error: No such method: "&scalar"
avar ?eval str("x" xx 3)
evalbot_8242 Error: No such method: "&str"
avar ?eval (str)("x" xx 3)
evalbot_8242 Error: No such sub: "&str"
xinming avar: In perl 6, I think scalar context is splited into Number context and String context 18:27
gaal ?eval ~("x" xx 3) 18:28
evalbot_8242 "x x x"
gaal ?eval +("x" xx 3)
evalbot_8242 3
gaal rehi btw
xinming so, you can use ~ to force it into string context, and use + to force it into numberic context
gaal ?eval ?("x" xx 3)
evalbot_8242 bool::true
gaal and boolean context :)
xinming by the way, anyone here knows about m4?
how to supress the #!/usr/bin/m4 line if chmod a .m4 file into executable 18:29
integral under linux you might be able to fiddle with binfmt_misc I suppose 18:31
but a shebang line is standard to invoke a script
integral misses the point, oops
avar xinming; gaal: thx;) 18:32
I mostly know perl5, just typing stuff into pugs & evalbot while reading dev.perl.org
xinming integral: yes, But the problem is, for m4, it will print the shebang line
integral to be honest I didn't even know you could access @ARGV in m4 :) 18:33
xinming #!/usr/bin/m4 18:34
foo here
xinming@Laptop:~/tmp$ cat asdf.m4
#!/usr/bin/m4
define(`foo', ``foo' here')
foo
I'd think using m4 as a cgi script would be fun. :-P
avar ?eval reverse <a b c> 18:55
evalbot_8242 "c b a"
avar ?eval reverse ~(<a b c>)
evalbot_8242 "c b a"
avar ?eval reverse "abc" 18:56
evalbot_8242 "cba"
gaal avar: at this point please run your own pugs locally :-) 19:06
j0sephi picasa.hu/1134587387-al.kaida.xmas.jpg ahaha idiots :):):) :) 19:21
gaal oh, someone kick j0sephi already 19:42
svnbot6 r8243 | gaal++ | * docs/p6doc/ - a few small fixes to skud++'s notes 19:45
stevan__ fglock: the Object Space is very much inspired by PyPy (hell we even took their name :) 19:50
tewk stevan__: I've got a very simple test harness, I'll paste a diff and get your comments 19:52
stevan__ tewk: did you commit it? 19:56
tewk I'm about too. An old version is in t/pil/simple.t ./pugs t/pil/simple.t
I'
I've factored out the code to ext/Test/lib/TestPil.pm, but I don't know what to do with the build process. 19:57
I've been copying ext/Test/lib/TestPil.pm to blib6/lib manually.
I've also got a t/pil/bit.t ready to go. 19:58
stevan__ is it in p6? or p5?
tewk: cool!
19:58 stevan__ is now known as stevan
tewk p6, but I can make it p5. The p6 stuff took me a while to come up to speed on, so I kept it simple to start. 19:59
stevan tewk: simple is good, and p6 is good too
self-hosting++
tewk so is t/pil a good place for tests, and do you have any guidance about the build process and what I need to do there?
If not I'll check in and ask for build process help. 20:00
stevan t/pil is good yes
I think maybe it should be Test::Pil though, not TestPil
but thats a detail
we can easily change if we want too
tewk Test::Pil I agree, I'm just getting comfortable with the pugs tree there are so many exts, blibs, libs, I'm a little lost. 20:01
Here it comes
stevan tewk: ext is where this should go,.. and when people run the Makefile.PL it should get copied to the blib dir 20:02
tewk That what I though, I just haven't got it to work how I expect yet.
stevan you can also do ./pugs -I ext/Test/lib t/pil/bit.t 20:03
svnbot6 r8244 | tewk++ | pil test code factored out to it's own module. 20:04
r8244 | tewk++ | tests for NativeBit not, and, and or operations
gaal or export PERL6LIB=blib6/lib
hey stevan 20:05
stevan hey gaal
tewk I found this on hack.pod at pugscode.org and have been using it, export HARNESS_PERL=./pugs export PERL6LIB=blib6/lib 20:06
Where is the HARNESS_PERL used? 20:07
stevan tewk++ # very nice stuff 20:08
gaal by prove
tewk ahh
gaal tewk: prove is a p5 tool, but it can use pugs to generate TAP data
did anyone write a TAP module for other languages alredy? Java, C# etc? 20:09
*already
should be just a few days work, and then they can benefit from the matrix etc.
tewk ?paste 20:10
gaal perlbot nopaste
perlbot Paste your code here and #<channel> will be able to view it: sial.org/pbot/<channel>
gaal why isn't perlbot clever enough to s/<channel>/$channel/? :-) 20:11
stevan gaal: I think there are some other TAP modules out there, there is a JS one I know, and IIRC someone on perl-qa mentioned a C/C++ one
gaal whoa, a c one could have helped me a while ago
tewk thanks, I have an idea for test descriptions, but I'm have p6 woes :) I'll paste in a minute
gaal sure 20:12
c is hardly a language where TDD is fun though :)
stevan wonders what about C is fun at all :P 20:13
gaal finds it hard to express
certainly not strings :)
but some of the time I find a great elegance in it. 20:14
some of the time it's plain ridiculous though :)
GLib has some beautiful c. 20:15
stevan I haven't had the benefit of a CS education, so the low level stuff you need to fiddle with in C is just frustrating to me
I much more prefer higher level abstractions :)
I do see some beauty in C though 20:16
I used to read a lot of books on hard-real-time systems, and the C code written for that stuff is (usually) really nice compact and elegant code
I guess I can appreciate it, but can't ever see writing it :) 20:17
gaal funny, I don't have a CS education either, and abstractions are often difficult for me. at least the first couple of years :)
stevan was a painting major, abstract painting mostly :)
I "see" code in my head visually, and I never could "see" C very well 20:18
gaal has a rothke reproduction he likes in the living room
stevan does too :)
gaal: which one? 20:19
gaal it's not titled. Red, from 1956.
stevan wonders if we have the same one :)
gaal :) 20:20
heh, i mixed him up with the poet
www.prints4all.com/cgi-bin/commerce...ey=IMA3918 20:21
stevan nope.... different ones 20:23
mine has more browns in it
have you ever seen his work in person
amazing stuff
I have gotten yelled at by many a MoMA guards for standing too close :)
gaal Distinctly I remember a very magical half-hour in the Tate :-) 20:24
Might have been less. Anyway... we're spamming the channel :-) 20:25
stevan prefers to think of it as educating the masses,.. but okay ;) 20:26
pasteling "tewk" at 71.32.232.64 pasted "Ugly Test::Pil, syntatic errors, but it has my ideas for test descriptions" (44 lines, 1.4K) at sial.org/pbot/14958 20:30
tewk Now to try to get it to work :)
stevan tewk: what is the need for generating the description? 20:32
as opposed to writing it with the test itself?
tewk I have the benefit of a CS education, which has taught me to write dozens of lines to code to avoid typing a description for each test :) 20:34
stevan LOL 20:35
tewk I'm just trying to get up to speed on Perl6 and I'm laxy.
stevan tewk: I am not sure we have lexical subs yet
tewk Ah 20:36
stevan although it should work anyway
just not define it lexically
tewk Well that was what I was trying to do, I'll keep playing with it. Maybe I should get back to writing test for PIL. :) 20:37
gaal giggles <for at by> tewk's benefit
tewk I think I will just add an optional description arg and call it good. On to tests. By the way how do you make an arg optional in Perl6?
gaal sub X ($arg?) 20:38
sub X ($arg? = "default val")
tewk thanx
gaal: your much faster then greping through docs and synopsis :) 20:39
gaal #perl6 is a kind of an optimized grep
stevan gaal is a storehouse of p6 knowledge... aka.. its an in-memory search :)
gaal you give me waaay too much credit folks :) 20:40
Juerd & luqui are our resident perldocs
and I will NOT get into a drawing contest with audrey :)
tewk stevan: Being CS educated, I'm very interested in the native mini-language and bootstrapping the ObjectModel. I've got to run, but what are he next small tasks that need done. 20:42
stevan tewk: mostly it is implementing the object meta-model using the mini-language,.. 20:43
this is all based on the work in perl5/Perl6-ObjectSpace
tewk I'll continue improveing TestPil. Can I just move it to ext/Test/lib/Test/Pil.pm and change TestPil to Test::Pil ? 20:44
stevan sure
I will probably add some tests later tonight myself
gaal tewk: svn mv, in case you don't know it
tewk thanks gaal: My perl5/6/haskell is pretty pretty pathetic :) but I know svn, C, etc 20:46
gaal cool :) no offence I hope... impossible to tell what people know and don't know these days :) 20:49
tewk No I appreciate the help, if fact I learn fast and I like suggestions, especially in reguards to my perl. 20:50
gaal learning haskell is hard (for me) but extremely enjoyable when things click 20:51
tewk I've played with python, ruby, scheme, haskell, etc. So I get a lot of the concepts. I just don't know perlism syntax very well, so I revert to evil C style syntax in perl, which usually gets the job done. I want to learn though. 20:53
gaal oh, in that case you're in luck cause learning perl is loads of fun :-) 20:54
sili__ tewk: just avoid c-style for loops and you'll be ok 20:56
cause those are the worstestest.
xinming thinks learning haskell is the biggest pain he ever encountered. 20:58
especially the moand, takes me 3 days to understand what the monads is.
svnbot6 r8245 | tewk++ | TestPil becomes Test::Pil
gaal xinming: you are in great shape :-) It's taking me nine *months* and counting 20:59
xinming frustrating me
gaal: I just understand what it is, But the problem is, I can't write without the the example. :-/ 21:00
21:11 stevan_ is now known as stevan
svnbot6 r8246 | stevan++ | ::Class.get_method_list should have no parameters ;) 21:54
gaal where are the shiftTs that relate to the resetT in opEval (Pugs.Prim.Eval)? 22:33
why is position information omitted in failed evals? because $! isn't fully specced? 22:49
?eval eval "asdfadf"; $!
22:49 evalbot_8242 is now known as evalbot_8246
evalbot_8246 \undef 22:49
gaal ?eval eval "asdfadf"; say $!
evalbot_8246 OUTPUT[No such sub: "&asdfadf" ] bool::true
gaal contrast:
?eval asdf 22:50
evalbot_8246 Error: No such sub: "&asdf"
gaal heh.
well on non-evalbot, that also prints coordinates.
but retEvalResult does only
err@(VError e _) -> do writeRef errSV e 22:51
... on a different note: 22:52
opRequire, when a module was found in %*INC already:
opEval style pathName ""
why? why the call to opEval, just to possibly reset $! ? 22:53
stevan wonders who gaal is talking to :)
svnbot6 r8247 | stevan++ | PIL.Native.*
r8247 | stevan++ | - changed attribute names in PIL/Native/Eval.hs to include the
r8247 | stevan++ | twigils (%methods becomes %:methods, etc)
r8247 | stevan++ | - added src/PIL/Native/Bootstrap.pil, this is the skeleton of the
r8247 | stevan++ | meta-model bootstrap code, it needs a /LOT/ of work :)
r8247 | stevan++ | Perl6::ObjectSpace
r8247 | stevan++ | - misc scribbles in the API_sketch.pod, nothing interesting really
stevan audreyt: we need to chat when you get back online 22:55
stevan dinner &
23:05 avarab is now known as avar
gaal what does "VControl $ ControlEnv env" mean? 23:17
what does 'Syn "env" []' do?
eg. EvalResultEnv -> (`mergeStmts` Syn "env" []) 23:18
zzZZ & 23:19
meppl gute nacht 23:30

Powered by Cro and the Raku® Programming Language.

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

Thank you!