»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by sorear on 4 February 2011.
TimToady_ timotimo: I've now changed it to ow(in) x .print :) 00:01
timotimo not exactly sure why that works the way it should :| 00:04
TimToady_ but if we had a C-style comma operator we could use R, to get the reversed version
.print always returns 1
well, True, but that's 1 numerically 00:05
ordinarily I'd use the ()[0] trick, but for this particular problem one wants to avoid the appearance of using arrays
one version I had: first *, ow(in), .print; 00:06
[//] ow(in), .print; works in rakudo but not in niecza
00:07 fgomez left
timotimo why is [//] ow(in), .print; different from ow(in) // .print? 00:08
TimToady_ reduction form doesn't short-circuit the right side
timotimo or i should say "favourable"
oh, good to know
is that specced to always be the case?
TimToady_ I think so
timotimo that's good 00:09
TimToady_ S03:4585
but the x is cool because even with something like a list operator it gives the appearance of using a composite structure to reorder values 00:10
(it doesn't, it's just changing the order of execution, but still...)
felher "This is still quite useful for performance, especially if the list could be infinite." :D 00:11
TimToady_ well, technically, it's picking what you return
00:12 tokuhiro_ joined 00:22 benabik_ joined, benabik left, benabik_ is now known as benabik, cognominal__ joined 00:25 cognominal_ left
diakopter pmichaud: you around? 00:29
nm; figured it out 00:30
00:31 cognominal__ left, cognominal__ joined 00:36 spider-mario_ joined 00:38 cognominal__ left, cognominal__ joined, spider-mario left 00:40 cognominal_ joined 00:41 colomon joined 00:44 cognominal__ left 00:49 skids joined 00:53 spider-mario_ left 00:55 whiteknight joined 00:56 whiteknight is now known as Guest67415, cognominal__ joined 01:00 cognominal_ left 01:05 curiosity joined
curiosity hi~ perl6! nqp compile error!need a toturial for that. 01:06
diakopter hi. first, what's your build environment? OS, compiler? 01:07
curiosity linux 32 ,gcc 01:08
diakopter could you paste the error somewhere (not here)?
curiosity like ...invoke not implement... 01:09
many people has this problem ..... 01:11
sorear There's very little we can do without knowing the full error and everything you did to get it
TimToady_ it's also going to be difficult to diagonise this, what with the light-speed lag to curiosity... 01:12
curiosity invoke() not implemented in class '2025589745'...that's my error
TimToady_ *diagnose, yow 01:13
sorear curiosity: the full errror should be a dozen lines or more. Do you know how to work a pastebin? 01:14
curiosity hello TimToady! help me!
sorear Help me help you.
TimToady_ knows nothing about nqp
curiosity that's bad 01:15
01:15 TimToady joined
diakopter curiosity: go here and paste your error messages: pastebin.com/ 01:15
benabik "invoke() not implemented" is a Parrot error about a missing viable. Means it tried to run something not runnable.
sorear TimToady cares more about the language design, not NQP
but until curiousity gets back to us with...
01:15 mucker joined
sorear 1. what version of NQP were you trying to build? 01:16
2. what version of Parrot do you have?
curiosity newest!
sorear 3. what kind of computer are you on?
curiosity all newest!
TimToady do you have an old one installed out there somewhere?
benabik Newest release or newest from git?
sorear Or newest from the RHEL 5 repository?
diakopter 4. are you trying to compile NQP as part of compiling rakudo? or by itself? 01:17
curiosity uninstall taht all!
my system is arch...........
diakopter -. -. 01:18
curiosity *.*
sorear which version of arch?
diakopter curiosity: please answer the above question #4 :)
curiosity newest
sorear for the sake of people who readthe logs in 5 years, it would be nice to have numbers for all this
(also, it allows us to check and make sure there isn't a newer version) 01:19
diakopter Obscurity by Posterity
curiosity what's question #4 ? read english not so fast..
diakopter are you compiling Rakudo? 01:20
curiosity no.i must compile nqp before compile rakudo.
diakopter ok. did you tell nqp to build its own parrot? 01:21
or are you using an installed parrot?
sorear Are you compiling manually, or ./configure --gen-nqp ?
curiosity I compile parrot first.it'sok.no error.next...compile nqp...than got errot. 01:22
compiling manually 01:24
sorear Where did you get NQP from?
curiosity git
sorear What commands did you run to build it?
WHich git?
curiosity perl Configure.pl.
01:25 ivan`` joined
curiosity git://github.com/perl6/nqp.git" 01:25
[Coke] you might have better luck starting with latest rakudo and then using that to build nqp & parrot (it will grab known working versions)
(assuming the goal is to eventually build rakudo)
curiosity coke: give me an instrution please. 01:26
[Coke] when you build parrot, did you install it?
curiosity: sure. Try this: 01:27
curiosity yes!
[Coke] git clone [email@hidden.address] ; cd rakudo ; perl Configure.pl --gen-parrot
that will ignore any installed versions, grab nqp and parrot, install them inside the rakudo directory, and then build rakudo using those versions. 01:28
curiosity coke:thanks!I'll try that!
[Coke] good luck.
TimToady well, and then a "make install"
[Coke] we have a lot of linux/gcc users here, so it should work ok.
TimToady: that's not necessary if you're happy running ./perl6 01:29
(if you do want to install rakudo, you might want to add a --prefix=/path/to/install to the Configure.pl invocation.
curiosity thanks! everybody!
TimToady I install in a subdir 01:30
[Coke] otherwise, the install directory is in the same local install directory (which will work fine, but might be an unusual place if you're doing anything long term.)
TimToady a symlink to that works fine
01:38 tokuhiro_ left 01:44 gongyiliao left 01:49 curiosity left 01:50 gongyiliao joined 01:52 gongyiliao left, gongyiliao joined 01:53 Guest67415 left 02:01 mucker left 02:14 jeffreykegler joined 02:18 jeffreykegler left 02:38 Psyche^ joined 02:41 Psyche^ is now known as Patterner
[Coke] search.cpan.org/dist/Regexp-Debugge...ebugger.pm ... wow. 03:00
damian++ 03:01
03:15 sftp left 03:16 localhost joined 03:47 sivoais joined 04:05 seldon left 04:08 benabik left 04:12 benabik joined 04:16 Khisanth joined 04:17 skids left 04:22 revdiablo left, revdiablo joined 04:37 jeffreykegler joined 04:45 jeffreykegler left
TimToady a :$ parameter does not have a name of '' 05:13
it should simply be anonymous, just as a positional $ is
05:14 birdwindupbird joined 05:16 Woodi joined
Woodi hallo today :) 05:25
do we have getop-module for v6 ?
sorear do you need something compatible with getopt_long(3)? 05:28
there is GetOpt::Long in the Niecza source repo, which could probably be turned into a standalone module without much work 05:29
Woodi k, thanx :) 05:30
sorear there is also the standard 'MAIN' routine, which is easy to set up but a bit idiosyncratic
05:53 wtw joined 06:15 thou left 06:18 jaldhar joined, SamuraiJack joined 06:25 jdrab joined 06:37 quietfanatic_ left 06:45 kaleem joined
jnthn morning 06:50
mathw morning 06:52
06:52 crab2313 joined 07:03 fhelmberger joined 07:09 brrt joined 07:13 GlitchMr joined
Woodi hallo mathw :) 07:37
07:45 fgomez joined 07:48 jaldhar left, jaldhar joined 08:00 awwaiid joined 08:01 jaldhar left 08:05 thelazydeveloper joined, jaldhar joined 08:11 bruges joined 08:16 mucker joined 08:17 jdrab left 08:18 jdrab joined, mucker left, fgomez left
moritz \o 08:20
jnthn hi moritz
Woodi o/ 08:23
sorear o/ moritz 08:24
arnsholt o/
diakopter o wait 08:29
I already did altseq 08:30
er, ww
08:33 crab2313 left 08:35 GlitchMr left 08:48 SamuraiJack left 08:49 SamuraiJack joined, GlitchMr joined
moritz phenny: tell TimToady www.wall.org/~larry/perl.html is a 404 08:53
phenny moritz: I'll pass that on when TimToady is around.
diakopter moritz: but where is it linked from
moritz diakopter: many places. Including www.wall.org/~larry/ 08:54
Woodi wow, installing Rakudo installs CPAN6, nice :) 08:57
moritz Woodi: erm, what?
Woodi moritz: modules are bundled now :)
masak top of the antenoon, #perl6 08:59
Woodi: modules are bundled with Rakudo *Star*, yes.
moritz Woodi: they have always been bundled with Rakudo Star
GlitchMr Last time I've lots of links on ~larry page are down
moritz that's the point of having a distribution, versus just having a compiler
GlitchMr For example "1st State of the Onion, 1997 Perl Conference."
But perhaps lots wasn't that much I thought when I've viewed this site 09:00
Woodi moritz: I thinked panda was responsible for module management... 09:02
moritz Woodi: it is. I just didn't know you meant "panda" when you said "CPAN6"
Woodi but, let say, I installed parrot to /opt, then I have /opt/lib/language dir... it is strange. 09:03
* /opt/lib/languages dir 09:04
/opt/lib/parrot/languages it should be ? 09:05
09:07 daniel-s joined 09:08 daxim joined 09:13 Celelibi joined
moritz doesn't see the logic in either of those 09:13
GlitchMr perl: /[^a-z]/ 09:25
perl6: /[^a-z]/
p6eval niecza v19-15-g051783d: OUTPUT«===SORRY!===␤␤Invalid regex metacharacter (must be quoted to match literally) at /tmp/vIrcpaNJxi line 1:␤------> /[^a-⏏z]/␤␤Action method metachar:sym<-> not yet implemented at /tmp/vIrcpaNJxi line 1:␤------> /[^a-…
..rakudo 34e8d4: OUTPUT«===SORRY!===␤Unrecognized regex metacharacter [ (must be quoted to match literally) at line 2, near "^a-z]/"␤»
GlitchMr std: /[^a-z]/
p6eval std f43a358: OUTPUT«Can't use string ("1") as a HASH ref while "strict refs" in use at CursorBase.pm line 881.␤FAILED 00:00 41m␤»
GlitchMr This error is rather helpful... 09:28
What about saying to use <-[a..z]> instead? 09:29
sorear stdbug 09:30
moritz it's an internal error in STD 09:31
09:35 SamuraiJack left 09:41 MayDaniel joined 09:44 HUUB0118408 joined 09:48 crab2313 joined 10:02 gfldex joined
moritz act.yapc.eu/ye2012/schedule?day=2012-08-22 masak++'s talk on macros is parallel to "A MOP for Perl 5" by stevan++ 10:10
masak aww, can someone else hold my talk so I can go listen to stevan++? :/ 10:11
tadzik aww 10:12
10:12 buubot_backup joined
tadzik what do :S 10:12
10:13 cognominal joined
moritz couldn't they have made stevan's talk a keynote? 10:14
tadzik the collisions wouldn't be _that_ bad if one was sure that there'll be a video afterwards 10:15
great, all 3 of those talks wed morning I have starred :/
10:16 cognominal__ left 10:19 mikemol joined, felher joined
masak realizing while frantically writing on the proceedings for my macros talk: 10:24
AST objects are not *like* closures. they *are* closures.
they represent runnable code. they refer to variables outside of themselves.
the more clearly I see this, the more convinced I am that we should not allow AST.new() in user code. it does not make any sense. 10:25
tadzik less evil available 10:26
moritz I think a good way to manipulate ASTs is to clone them, and change some stuff while cloning
masak the only thing that would make sense is something like AST.new(:outer<here>). but we have no first-class way of talking about lexpads.
yeah, cloning is fine.
anything which preserves the 'outer' attribute. 10:27
splicing an AST and plucking out a small bit of it should be fine too.
in fact, that's probably the primitive operation here, since that's what macros do all the time.
moving AST fragments around and inserting them in each other. 10:28
adding AST nodes to an existing AST is fine, too. 10:30
10:32 jdrab left
moritz we do have a way to talk about lexpads though 10:34
MY:: for example
OUTER::, CALLER:: etc.
masak hm, maybe I'm being to strict here.
maybe "no outer" just means that the AST is not a closure.
and maybe synthetic ASTs could be inserted into bigger ones in such a way that they "inherit" the $!outer of their mother AST. 10:37
though how hygienic that would be I don't know.
10:38 gongyiliao left
jnthn masak: (MOP talk) I guess there'll be new stuff in it, but didn't we see a version of that talk in London? :) 10:41
jnthn checks the schedule to see when he is talking
eek, I'm on when MST is 10:42
But I saw that talk in France. :)
Also my talk is on day 1 :)
masak jnthn: troo. but by the same logic, I've seen my talk twice :P 10:53
jnthn Oh, *that's* why you always have a mirror on the speakers desk... 10:54
masak :P 10:55
masak just took a self portrait with his phone, and feels kinda bad about that now :P
10:56 jdrab joined
masak jdrab: hi. 10:57
oh, that was a rejoin :)
jdrab masak: hi :) 10:58
masak :)
jdrab: welcome. haven't seen you around before, methinks.
jnthn Hm, .sk :) 10:59
Dobry den, jdrab :)
jdrab nope i wasn't here for a long time.. hmm ..never :D
masak vitajte, jdrab . 11:00
jdrab jnthn: hi 11:02
11:04 jdrab left 11:07 crab2313 left 11:09 JimmyZ joined
JimmyZ good evening 11:11
11:13 fridim_ joined
masak JimmyZ: 晚上好 11:15
JimmyZ masak: 下午好
masak .oO( why are we using each other's time zones? ) :P 11:16
JimmyZ .oO(out of courtesy? o_O) 11:17
tadzik just UGT I guess
masak heh, TIL: UGT. 11:19
www.geekality.net/2010/05/14/univer...-time-ugt/
I will so adopt this immediately. 11:20
11:23 HUUB0118408 left
masak for my macro proceedings paper, I've identified *five* stages of macro: parsing of the macro, parsing of the call, running of the macro, inserting of the AST, and running of the inserted code. 11:23
though to be fair, an ordinary sub call has four of those stages. 11:24
TimToady: I think macros should be powerful enough to define classes with parameterized names. do you agree? how will we do that? 11:35
11:38 tokuhiro_ joined
moritz
.oO( universal parsing fallback mechanism to replace any part of a grammar rule with a template )
 11:39
Perl6::Metamodel::ClassHOW.new(:$name)
daxim act.yapc.eu/ye2012/schedule?day=2012-08-20 11:52
jnthn r: constant foo = "lolclass"; class ::(foo) { }; say lolclass 11:55
p6eval rakudo 34e8d4: OUTPUT«===SORRY!===␤multi_dispatch_over_lexical_candidates was unable to find a candidate list␤»
jnthn o.O
star: constant foo = "lolclass"; class ::(foo) { }; say lolclass
p6eval star 2012.07: OUTPUT«Capture[0xa7c9890]()␤»
jnthn Hmm.
That should work.
jnthn leaves masak to do the honors
11:58 fridim_ left
moritz that... should work? really? 12:08
arnsholt Heehee. "Now, instead of spending time figuring out what time of day is it for every member of the channel, we spend time explaining newcomers benefits of UGT." 12:12
12:12 fhelmberger left
jnthn moritz: Did work at one point, even, iirc. 12:13
moritz I don't remember even seeing that working 12:14
masak submits rakudobug 12:18
jnthn: if that worked, it would be a really good start. 12:20
12:33 SamuraiJack joined 12:43 JimmyZ left
moritz would that also work for subs? 12:44
r: constant name = 'foo'; sub ::(name) () { 42 }; say foo()
p6eval rakudo 34e8d4: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&foo' called (line 1)␤»
jnthn r: constant name = 'foo'; say (sub ::(name) () { 42 }).name 12:45
p6eval rakudo 34e8d4: OUTPUT«::(name)␤»
jnthn fel!
masak submits rakudobug
is there any way to call a sub called '::(name)'? :) 12:47
mhasch has just found the yapc::eu schedule. 4-way parallel sessions, geez. 12:49
jnthn ::('::(name)')()
uh 12:50
::('^::(name)')() # more likely
masak r: constant name = 'foo'; sub ::(name) () { 42 }; say ::('^::(name)') 12:51
p6eval rakudo 34e8d4: OUTPUT«Symbol '^::(name)' not found␤ in method gist at src/gen/CORE.setting:9659␤ in sub say at src/gen/CORE.setting:7046␤ in block at /tmp/u60wVQicIq:1␤␤»
masak r: constant name = 'foo'; sub ::(name) () { 42 }; say ::('^::(name)')()
p6eval rakudo 34e8d4: OUTPUT«Symbol '^::(name)' not found␤ in method <anon> at src/gen/CORE.setting:9664␤ in <anon> at src/gen/Metamodel.pm:2304␤ in any find_method_fallback at src/gen/Metamodel.pm:2302␤ in any find_method at src/gen/Metamodel.pm:843␤ in <anon> at src/gen/BOOTSTRAP.pm:824…
masak r: constant name = 'foo'; sub ::(name) () { 42 }; say ::('::(name)')()
p6eval rakudo 34e8d4: OUTPUT«Start of substr out of range. Is: 1, should be in 0..0␤ in method Str at src/gen/CORE.setting:9658␤ in method Stringy at src/gen/CORE.setting:726␤ in sub infix:<eq> at src/gen/CORE.setting:1038␤ in sub INDIRECT_NAME_LOOKUP at src/gen/CORE.setting:11643␤ in blo…
masak oh wow.
jnthn: what does the '^' mean?
r: say ::('::(name)')()
p6eval rakudo 34e8d4: OUTPUT«Start of substr out of range. Is: 1, should be in 0..0␤ in method Str at src/gen/CORE.setting:9658␤ in method Stringy at src/gen/CORE.setting:726␤ in sub infix:<eq> at src/gen/CORE.setting:1038␤ in sub INDIRECT_NAME_LOOKUP at src/gen/CORE.setting:11643␤ in blo…
masak r: ::('::(name)') 12:52
p6eval rakudo 34e8d4: OUTPUT«Start of substr out of range. Is: 1, should be in 0..0␤ in method Str at src/gen/CORE.setting:9658␤ in method Stringy at src/gen/CORE.setting:726␤ in sub infix:<eq> at src/gen/CORE.setting:1038␤ in sub INDIRECT_NAME_LOOKUP at src/gen/CORE.setting:11643␤ in blo…
masak r: ::('::()')
p6eval rakudo 34e8d4: OUTPUT«Start of substr out of range. Is: 1, should be in 0..0␤ in method Str at src/gen/CORE.setting:9658␤ in method Stringy at src/gen/CORE.setting:726␤ in sub infix:<eq> at src/gen/CORE.setting:1038␤ in sub INDIRECT_NAME_LOOKUP at src/gen/CORE.setting:11643␤ in blo…
masak r: ::('::')
p6eval rakudo 34e8d4: OUTPUT«Start of substr out of range. Is: 1, should be in 0..0␤ in method Str at src/gen/CORE.setting:9658␤ in method Stringy at src/gen/CORE.setting:726␤ in sub infix:<eq> at src/gen/CORE.setting:1038␤ in sub INDIRECT_NAME_LOOKUP at src/gen/CORE.setting:11643␤ in blo…
masak r: ::('')
p6eval rakudo 34e8d4: OUTPUT«Start of substr out of range. Is: 1, should be in 0..0␤ in method Str at src/gen/CORE.setting:9658␤ in method Stringy at src/gen/CORE.setting:726␤ in sub infix:<eq> at src/gen/CORE.setting:1038␤ in method exists at src/gen/CORE.setting:6497␤ in sub INDIRECT_NA…
masak submits rakudobug
jnthn er, I meant &, not ^
masak hah! 12:53
also, "phew" :)
jnthn oh, duh
mhasch masak++: boiled down to 6 characters, neat.
jnthn ::('...') pays attention to ::s
masak r: constant name = 'foo'; sub ::(name) () { 42 }; say ::('&::(name)')()
p6eval rakudo 34e8d4: OUTPUT«Symbol '&::(name)' not found␤ in method <anon> at src/gen/CORE.setting:9664␤ in <anon> at src/gen/Metamodel.pm:2304␤ in any find_method_fallback at src/gen/Metamodel.pm:2302␤ in any find_method at src/gen/Metamodel.pm:843␤ in <anon> at src/gen/BOOTSTRAP.pm:824…
jnthn So it will try to split on them
I wanted ::{'&::(name)'}()
masak mhasch: thanks, but this is really nothing out of the ordinary :) golfing is standard procedure. 12:54
r: constant name = 'foo'; sub ::(name) () { 42 }; say ::{'&::(name)'}()
p6eval rakudo 34e8d4: OUTPUT«42␤»
masak \o/
jnthn But yes, the error for giving something wrong to interpolation is LTA there.
masak this all goes into the ticket. 12:55
jnthn If you're going to file that substr thing it is a separate ticket.
(to implemeting indirect method name support)
masak that's the one I'm filing. 12:57
I already filed a separate 'class ::(const)' one and a separate 'sub ::(const)' one.
today is a good day for rakudobugs :)
mhasch them little buggers seem to grow like mushrooms after a summer rain :-) 12:58
masak no. they were there all the time :)
jnthn like flowers in the dessert, they only grow when the rain comes :) 12:59
uh, desert 13:01
mhasch they might have been there, only hidden from masak's laser beam. I wonder what that mars robot will stirr up.
right now I could use a clone of me or two. Threading support in Parrot, e.q. is scheduled in a parallel track to Asynchronous Programming. 13:06
All about parallelism, figures.
Good thing they intend to do video recordings. 13:07
masak they do? neat. 13:08
13:11 xinming_ joined 13:25 skids joined 13:33 atrodo joined 13:37 mucker joined 13:40 bluescreen10 joined
tadzik they did that in Riga too :/ 13:48
13:53 Woodi left, orafu joined
tadzik mhasch: I just complained to The Guys about it, and it "seems fixable" 13:54
13:58 kaare_ joined 13:59 drbean joined
felher r: gist.github.com/3295244 14:05
p6eval rakudo 34e8d4: OUTPUT«sub say-it() { ... }␤»
felher r: gist.github.com/3295275 14:06
p6eval rakudo 34e8d4: OUTPUT«hi␤33␤»
felher The only difference between those two gists is the ';' at the end of line one?
Is that a bug?
14:07 kaleem left, crab2313 joined
felher &brb 14:08
masak felher: rt.perl.org/rt3/Ticket/Display.html?id=76432 14:09
colomon n: gist.github.com/3295244 14:16
p6eval niecza v19-15-g051783d: OUTPUT«sub say-it() { ... }␤»
colomon n: gist.github.com/3295275
p6eval niecza v19-15-g051783d: OUTPUT«sub say-it() { ... }␤»
masak std: repeat { say "OH HAI" } 14:18
p6eval std f43a358: OUTPUT«===SORRY!===␤Undeclared routine:␤ 'repeat' used at line 1␤Check failed␤FAILED 00:00 41m␤»
masak hm.
jnthn did you mean loop? 14:19
felher masak: oh, thanks. Sorry for the noise :)
masak felher: don't be sorry. it's a real problem.
jnthn: no, I meant to check what happens if I do 'repeat' without the 'until' or 'while' after the block.
std: repeat { say "OH HAI" } while True 14:20
p6eval std f43a358: OUTPUT«ok 00:00 41m␤»
masak I'm surprised it parses as a listop when the 'while' is not there.
jnthn Hmmmm. 14:21
Surely it doesn't backtrack there...
masak right.
jnthn std: repeat { my $x; say "OH HAI" }
p6eval std f43a358: OUTPUT«===SORRY!===␤Undeclared routine:␤ 'repeat' used at line 1␤Potential difficulties:␤ $x is declared but not used at /tmp/uVJxqUWmAt line 1:␤------> repeat { my $x⏏; say "OH HAI" }␤Check failed␤FAILED 00:00 42m␤»
jnthn r: repeat { say "OH HAI" } # curious 14:22
p6eval rakudo 34e8d4: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&repeat' called (line 1)␤»
PerlJam That could make for some interesting code.
jnthn r: repeat { BEGIN { say "here" }; say "OH HAI" } # curious 14:23
p6eval rakudo 34e8d4: OUTPUT«here␤here␤===SORRY!===␤CHECK FAILED:␤Undefined routine '&repeat' called (line 1)␤»
PerlJam r: sub repeat { $^a.() }; repeat { say "hi" }
jnthn Wow!
p6eval rakudo 34e8d4: OUTPUT«hi␤»
PerlJam r: sub repeat { $^a.() }; repeat { say "hi" }; my $i = 0; repeat { say "foo" } while $i++ < 3;
p6eval rakudo 34e8d4: OUTPUT«hi␤foo␤foo␤foo␤foo␤»
PerlJam good for obfuscation I guess
masak was gonna say.
I feel 'repeat' is a badly chosen name for that sub :P 14:24
jnthn I think the parser is gonna need to commit after seeing repeat, fwiw
masak submits stdbug 14:25
masak submits rakudobug
n: repeat { say "OH HAI" }
p6eval niecza v19-15-g051783d: OUTPUT«===SORRY!===␤␤Undeclared routine:␤ 'repeat' used at line 1␤␤Unhandled exception: Check failed␤␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 1402 (die @ 5) ␤ at /home/p6eval/niecza/src/STD.pm6 line 1147 (P6.comp_unit @ 37) ␤ at /h…
masak submits nieczabug
jnthn will wait for an STD fix and steal that
14:29 SatoAmbush joined
moritz r: for <a b c> <-> { .say } 14:29
p6eval rakudo 34e8d4: OUTPUT«===SORRY!===␤Missing block␤at /tmp/giT_v0kyy6:1␤»
moritz back in the days people said that LTM was missing to parse RW lambdas properly 14:30
and now we have LTM
so, what's missing? :-)
jnthn I dunno, check out the grammar and try a --parse-trace or something :)
14:30 SatoAmbush left
jnthn b: for <a b c> <-> { .say } 14:31
p6eval b 922500: OUTPUT«===SORRY!===␤Missing block at line 22, near ""␤»
jnthn Hm, same error.
std: for <a b c> <-> { .say } 14:32
p6eval std f43a358: OUTPUT«ok 00:00 42m␤»
PerlJam Weird. I thought rakudo parsed <-> at one time.
(parsed it, but treated it as -> ) 14:33
GlitchMr I'm confused. 14:34
Why this isn't syntax error?
Is it some weird Perl 6 feature?
PerlJam
.oO( What Perl 6 features *aren't* weird? ;-)
 14:35
GlitchMr What is <-> anyways?
masak GlitchMr: it's some weird Perl 6 feature.
PerlJam GlitchMr: it makes the parameters rw 14:36
GlitchMr Oh, ok
Why <a b c> would be read write?
PerlJam GlitchMr: -> $a, $b { ... } # $a and $b are the parameters 14:38
masak r: for <a b c> -> { say 1 }; say "alive" 14:39
p6eval rakudo 34e8d4: OUTPUT«Too many positional parameters passed; got 1 but expected 0␤ in block at /tmp/clZxY8Y8n4:1␤␤»
masak thought so 14:40
jnthn Urgh, somebody looked at one of my Perl 6 scripts at $dayjob today, didn't recognize the language (it's a .Net shop), and asked, "Is that PHP?!" 14:42
tadzik eww
well, dollar signs
arnsholt Sad panda, indeed
brrt .... :-o
hugme: hug jnthn
hugme hugs jnthn
jnthn Happily, they were fairly impressed with what the script actually produced. :) 14:43
14:43 alester joined
flussence this week I'm slowly becoming convinced dhcpd.conf's syntax is worse than PHP... 14:44
pmichaud before LTM, the problem with <-> was that it would parse as a less-than operator, followed by a -> term.
14:44 wtw left
pmichaud one could do various things to the grammar to try to prevent it, but that just pushed the problems elsewhere. 14:45
GlitchMr Actually, I think that PHP is better than Malborge
Malbolge* 14:46
pmichaud anyway, with LTM in place, I think we just need some tweaks to the grammar to get it to parse properly. 14:47
PerlJam Hmm. Line 606 of Grammar.pm says: token lambda { '->' | '<->' } 14:48
(that's probably why I thought it already parsed <-> )
14:49 Chillance joined
GlitchMr gist.github.com/3295609 14:50
I have problems with compiling Rakudo
(or rather, NQP)
masak someone else did not so long ago. 14:55
14:56 sftp joined 15:02 jeffreykegler joined 15:08 Chillance left
jnthn klockan decommute \o/ 15:08
masak \o/ 15:10
sjohnson masak.snuggle 15:13
masak <3 15:14
15:20 Chillance joined 15:22 jeffreykegler left
masak smiles while re-reading strangelyconsistent.org/blog/how-pe...lls-itself for the Nth time 15:25
15:28 bbkr joined
bbkr perl6: my $x is readonly = 3; $x = 6; 15:28
p6eval rakudo 34e8d4: ( no output )
..niecza v19-15-g051783d: OUTPUT«===SORRY!===␤␤Trait readonly not available on variables at /tmp/JIBaQU1ih_ line 1:␤------> my $x is readonly ⏏= 3; $x = 6;␤␤Unhandled exception: Check failed␤␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 1402 (di…
bbkr std: my $x is readonly = 3; $x = 6;
p6eval std f43a358: OUTPUT«ok 00:00 42m␤»
tadzik r: my $zebra is striped; 15:31
p6eval rakudo 34e8d4: ( no output )
tadzik std: my $zebra is striped;
p6eval std f43a358: OUTPUT«ok 00:00 41m␤»
15:32 tokuhiro_ left
masak b: my $x is readonly = 3; $x = 6; say "alive" 15:32
p6eval b 922500: OUTPUT«Cannot modify readonly value␤ in '&infix:<=>' at line 1:src/metamodel/RoleToInstanceApplier.nqp␤ in main program body at line 22:/tmp/dvjZcbMChs␤» 15:33
masak we used to have it.
bbkr yes, there is ticket rt.perl.org/rt3/Ticket/Display.html?id=71356 about it and I was curious if it's still a bug 15:34
masak ok. 15:35
bbkr was this trait renamed or just dropped from spec? 15:37
masak S06:2290 15:38
bbkr so the answer is: regression 15:40
std: class A { has $!b is readonly = "foo";} 15:42
p6eval std f43a358: OUTPUT«ok 00:00 42m␤»
15:42 SamuraiJack left
masak std: class A { has $!b is rhubarb = "foo"; } 15:42
p6eval std f43a358: OUTPUT«ok 00:00 42m␤»
masak std: class A { has $!b is slinky_toy = "foo"; } 15:43
p6eval std f43a358: OUTPUT«ok 00:00 42m␤»
masak std: class A { has $!b is Mr_President = "foo"; }
p6eval std f43a358: OUTPUT«ok 00:00 42m␤»
bbkr ok, I get it :)
masak :)
15:45 ingy joined, jeffreykegler joined
bbkr is there a way to check later in the code if $b had slinky_toy trait applied? 15:46
brrt bbkr: i goes with does? 15:47
bbkr what do you mean? 15:48
moritz traits are just subroutines with funny names that get executed at compile time 15:52
if they mix in a role, you can check if an object does the role
if they do nothing, then you can't check for the trait 15:53
currently it seems that traits on attributes are not yet implemented, with 'is rw' just being special-cased 15:54
jnthn Traits on attributes are implemented. 15:56
It's traits on variables that are not in Rakudo.
moritz oh, that was std, not rakudo
never mind then 15:57
bbkr thanks for explanation 15:59
16:01 brrt left
moritz r: role Confused { }; multi trait_mod:<is>(&f, :$incomprehensible!) { &f does Confused }; sub lk is incomprehensible { }; say &lk ~~ Confused 16:04
p6eval rakudo 34e8d4: OUTPUT«True␤»
bbkr awesome, just what I need to introduce notifications in JSON::RPC without introducing variables of special meaning to method params! 16:06
masak bbkr: please show this use case when you have written it. I'm curious. 16:09
bbkr sure 16:12
16:19 benabik left 16:20 benabik joined
masak is suddenly reminded of www.nntp.perl.org/group/perl.perl6....18764.html where it was attempted to think up new, better names for push/pop/unshift/shift 16:20
some things just run too deep to be changed, methinks.
16:20 flussence joined 16:31 SamuraiJack joined, SamuraiJack left 16:32 [particle]1 joined, birdwindupbird left, birdwindupbird joined 16:33 kaare_ left, buubot_backup left, thelazydeveloper left
moritz p6l at its best 16:33
16:33 Colby` joined, Colby` is now known as Pleiades` 16:34 charsbar_____ joined
bbkr "pull" and "put" were considered? well, "shift" is much better, it suggests that elements have to be shifted to make space for new one. unshift sounds horrible due to lack of natural opposition word to shift . but "pull" and "put" do not give any information about which side of list they operate on. 16:34
16:34 kaare_ joined, buubot_backup joined, rhr_ joined, pjcj joined, ranguard joined, PerlJam joined, Tene joined, SamuraiJack joined
TimToady unshift is correctly huffmanized :) 16:38
phenny TimToady: 08:53Z <moritz> tell TimToady www.wall.org/~larry/perl.html is a 404
16:39 birdwindupbird left
geekosaur agreed about huffmanization 16:39
moritz push-to-front, pop-from-front
:-)
masak &unshift is often the most expensive one in terms of what must be done memory-wise. 16:40
TimToady unshift is often a code smell
masak anyway, it is too late now. pop/push/shift/unshift are all in JavaScript, so they will never go away.
TimToady well, one could have said the same for P5 regexes 16:41
and it is true that they'll never go away...
masak moritz: re 'p6l at its best'. in some sense I'm very grateful that discussion took place. and p6l is the natural place for it.
if Perl 6 didn't have that trait of rethinking everything that needs rethinking, I probably wouldn't find the language as interesting. I probably wouldn't be here. 16:42
sometimes it just happens that after rethinking something, you decide not to change anything ;)
TimToady: right. P5 regexes will never go away. the best they can hope for is completely supplanting sed and awk and vim and Java and PHP regexes ;) 16:43
but they won't do that either. 16:44
the best P6 regexes can hope for is to be favorably compared to P5 regexes.
16:44 PerlPilot joined
masak (even though, technically, they are clearly better) 16:44
16:45 kaare_ left, buubot_backup left, rhr_ left, pjcj left, ranguard left, PerlJam left, Tene left, kaare_ joined, buubot_backup joined, rhr_ joined, pjcj joined, ranguard joined, Tene joined, bbkr left
moritz note that while niecza and rakudo have good-ish grammar engines, they are rather bad regex engines 16:45
mostly missing lots of optimizations
masak aye, point. 16:46
16:52 benabik left, SamuraiJack left 16:55 Chillance left
sorear good * #perl6 16:58
diakopter hi 16:59
moritz good (UGT) morning sorear, diakopter 17:01
diakopter heh 17:05
17:06 PerlPilot is now known as PerlJam, benabik joined
masak good (UGT) morning, good fellows on the west coast of the world. 17:07
sorear huh, I thought this was only the west coast of the USA! 17:08
o/ masak
I live on the east coast of the Northern Pacific...
masak :P 17:09
flussence
.oO( should probably stay away from the west coast of the world, you might fall off )
moritz Universal greeting time is Universal :-)
flussence: no worries, we've never had any first-hand reports from people who fell off 17:10
17:12 jaldhar left, [particle]1 is now known as [particle]
sorear we do, however, have the geographic region of West Antarctica. 17:13
diakopter also North Antarctica
benabik North Antarctica? You mean the coast? 17:14
masak moritz: oh, sure -- just because we never heard back from anyone who fell off, that means there must be no edge of the world, right? with that kind of reasoning you could probably prove that the Earth revolves around the sun rather than vice versa! 17:16
moritz masak: only if one way or another might produce victims :-) 17:17
masak victims who never write back. 17:18
moritz correct
17:19 crab2313 left 17:21 thou joined
[Coke] aloha, ugt? 17:29
moritz aloha: ugt is Univeral Greeting TIme
aloha: ugt? 17:30
seems that feature is disabled
can't say I'll weep for it
TimToady masak: re irclog.perlgeek.de/perl6/2012-08-08#i_5884009 I'd say, no, macros should be *more* powerful than that :)
the use of ::() is a code smell that you're trying to treat a grammar element as a term when it isn't :)
benabik moritz: seems disabled here, but works in #parrot 17:31
TimToady s/you're/jnthn is/
masak TimToady: so, you agree. good. how will we do that? 17:33
17:34 BooK joined, mucker_ joined
masak that is, what's the syntax for saying "define a class with the name provided as a parameter to the macro"? 17:34
17:34 nebuchadnezzar joined
TimToady we talked about that some time ago; we need better ways to parse, and better ways to stitch in non-term quasi-unquotes 17:34
masak what's the type of such a parameter? Str?
yeah. we need that.
17:35 mucker left
TimToady ¤ and all that 17:35
masak right.
I brought this up because it seems to me this is a more pressing problem than the ones we/you tried to solve in that discussion.
TimToady to me it's the same problem I was trying to solve then
masak that is, I felt at the time, and still feel, that we attacked too much of the problem at once and got unsolvable consequences that we weren't prepared for. 17:36
17:38 sjn joined 17:40 mucker_ left
TimToady and that is different from the rest of Perl 6 how? :) 17:40
17:42 fhelmberger joined 17:44 Telgalizer joined
masak point taken. 17:44
sorear masak: fwiw, niecza uses circular buffers to implement lists so unshfit and push are exactly equal in speed
masak still, we're missing something here, and I'm suggesting we might discovering by not focusing on the bigger picture.
sorear: ooh, nice.
17:44 Shozan joined
sorear moritz: i've spent a little bit of time thinking about the requirements of regexes versus grammars ... 17:46
it seems to me that one of the most important things to do is to recognize freestanding regexes and add an internal .*? to them 17:47
diakopter masak: also, unshift==push speed in Pm's qlist in NQP (and thus rakudo, I think)
masak huh.
sorear with *handwaving* to handle the case of .match( pos ) 17:48
once the .*? is inside the regex the existing regex optimization machinery can be deployed to optimize .*? camel into <.BM('camel')> 17:49
TimToady some fbm after .*? would not be out of order
grammars of course do very little .*?
masak puts on his TODO list to compose a list of possible optimizations for regex/grammar engines 17:50
17:50 jeffreykegler left
sorear another thing to look into is to recognize feature-poor regexes and turn them into data objects / use specialized code for ACCEPTS 17:50
the most obvious case being something like /foo/
TimToady and even when it does something like that in the nibbler, it's looking for single-char matches, so BM need not apply
sorear $text ~~ /foo/, IndexRegex.new('foo').ACCEPTS($text) 17:51
TimToady sorear: more generally, not collecting regex match data that we can determine won't be sued
or used, even :)
sorear statically determining that seems extremely difficult, so I'd prefer not to try 17:52
TimToady well, but we could change the default on something like .caps if it turns out to be expensive
we should be alert for sawampersandish penalties 17:53
sorear I am thinking that the main point of PL_sawampersand is that the original string has to be copied into offside storage? 17:54
particularly $', because that part wouldn't otherwise be touched at all
TimToady well, and that it applied globally
17:54 phenny joined 17:55 fhelmberger left, charsbar_____ left, felher left
sorear $` and $& have to be scanned anyway to make the regex work 17:55
TimToady since we could not determine which regex set it
sorear anyway the regex feature I am a liiitle concerned about is .from' 17:57
GlitchMr eval 'a' =~ /a/; ${'&'}
buubot_backup GlitchMr: No output.
sorear or is that .pos
GlitchMr eval 'a' =~ /a/; ${&}
buubot_backup GlitchMr: a
17:58 charsbar_____ joined, felher joined, kcwu joined, moritz joined, smash joined, Gesh joined, pochi joined, yeltzooo7 joined
GlitchMr It's probably known bug, I know :) 17:58
sorear because it interferes a bit with replacing regexes with DFAs
PerlJam masak: I'm more interested in: what's the syntax for saying "define the sub with the name and signature provided as parameters to the macro"
sorear another thing which troubles me is requiring ropes to get decent append performance in p6
diakopter sorear: a regex could start out not not creating matches, but when it is requested, the match is run again to get the match, and then matches are collected for that regex for all subsequent runs
sorear diakopter: doesn't work for regexes with embedded code 17:59
diakopter k, so don't do it for them :)
s/not not/not/ 18:00
masak PerlJam: yes, that's important too.
tadzik seems that we're the "Dont's", by whoever's definition: 140stitches.com/2012/8/2/nerds-its-...shion-game 18:03
masak GlitchMr: even if it's not known... this is #perl6.
PerlJam tadzik: uh ... wow
sorear working as designed. 18:04
masak tadzik: I find that post slightly offensive.
flussence ah, another site that breaks hilariously with noscript... 18:05
(or is it looking as intended?)
masak tadzik: besides, I think you look very trendy in that t-shirt. it suits you!
tadzik: how... how did you find that post? 18:06
jnthn :/
tadzik is formulating a response
masak: hacker news
masak hah!
tadzik actually a friend of mine
18:06 mhasch joined
tadzik "hey, is that you? Found on HN" 18:06
me: "haha, yes, that's me :D... wait, wtf" 18:07
masak "The guys who you can't have a conversation with without getting lost in the first 10 seconds." -- the post isn't made better by generalizations such as this. :(
PerlJam decrying "hackathon shirts" with images of guys taken *from a hackathon* is really wrong too. 18:08
masak oh, right. I forgot to say that.
there's a time and place for everything.
PerlJam s:2nd/hackathon/hackathon-like-activity/
masak it was *a hackathon*.
in Oslo.
though I'm not sure what effect showing up looking like a clothes model would have, either. maybe none. 18:09
PerlJam maybe next time someone should show up in a tux with a tophat and cane :) 18:10
masak maybe people would think it'd be weird in that setting. maybe they would unconsciously pay more attention to this obviously-alpha geek. I dunno.
PerlJam: I'll consider it. that sounds like fun. 18:11
tadzik "Awaiting moderation". We'll see
masak .oO( Awaiting admiration )
tadzik gist.github.com/3297168 in case it'll not come to sb's liking
PerlJam or, if anyone happens to be bulging with muscles (Damian is the only hacker that immediately comes to mind though), they could walk around shirtless with a sign that reads "I'm too sexy for my shirt" 18:12
masak tadzik: angry comment. though I don't think it's over-the-top. 18:13
jnthn That was...angry?
It was quite contained, imho :)
masak it wasn't *Internet* angry, mind.
jnthn Well, quite.
masak but clearly expressing dislike and dissatisfaction.
jnthn Heh. "It's holding you back from getting the girl" - well, if it filters out ones like the author of that post, I'm very glad of it. :-)
PerlJam Hmm. 18:14
masak well, what's sad about it is that she's clearly out to help.
she's just being a bit of an asshat about it.
PerlJam Lots of presumption though
lots and lots and lotgs of presumption
masak and cliches and generalization.
not a good start.
tadzik masak: I don't fell non-angry when being used as a Bad Example, jokingly or not 18:15
PerlJam I wonder if she polled the people in those pictures, how many are actually either married or in a relationship?
masak tadzik: quite.
diakopter I'm kindof confused why any of you is offended
PerlJam (or how many aren't because they don't want to be)
masak diakopter: ahahahah. I was waiting for this! :) 18:16
diakopter PerlJam: also, how many of them want the Guy instead of the Girl
PerlJam diakopter: exactly
diakopter the commenter who's in the photo says he's honored (not offended) 18:17
and commends the blog for an interesting idea
18:17 Telgalizer left
masak oh, there's a lot of comments there already. 18:17
PerlJam diakopter: that's just because he agrees with the underlying premise :) 18:18
masak I agree with the underlying premise to some extent too.
I just think the delivery was tone-deaf.
18:18 REPLeffect joined
diakopter PerlJam: I don't understand what's wrong with the underlying premise.. 18:18
masak "And as some people have noted, trying to 1) insult your target demographic rather than try to raise them up and 2) using pretty low-level tricks like, "It's holding you back from getting the girl" on a crowd who you yourself acknowledge are a pretty bright bunch isn't exactly tactical." +1 18:19
PerlJam well ... since my picture isn't up there, I get to point and laugh at you guys for having no "fashion sense" ;->
diakopter I really don't see how it's insulting
I don't see how that's a low-level trick
she's just stating an opinion.
not a lure 18:20
implying... "if you care about getting the girl, I'm here to tell you this is holding you back, if you weren't already aware"
where "the girl" means "a much larger pool of non-disgusted females" 18:21
masak diakopter: how would you dress to a hackathon, ooc?
flussence the tone of it sounds a lot like the anecdote here: www.reddit.com/r/linux/comments/xiy...re/c5mzgil 18:22
awwaiid not enough of the "if you care" preamble
diakopter same as I wear to work and wear nearly every other place other than work, loose jeans and polo with white undershirt, leather sandals with white socks.
sirrobert n
diakopter masak: it's kindof my uniform 18:23
note: I have a Girl 18:24
masak diakopter: I'm a geek. I have a girlfriend. I have a burgeoning fashion sense. improving it *interests* me. the tone of this post had an off-putting effect on me, by making assumptions that I don't recognize and don't like.
18:24 nebuchadnezzar left
masak I'm not sure if "offended" is the right word. I didn't feel happier reading that. 18:25
diakopter hm. perhaps the difference here is how much we care about the inaccuracy of a random blogger's generalizations.
masak isn't it a fundamental rule of any writing (or speaking) that it's a bad idea to insult one's audience?
TimToady hackers are just too Lazy to try to please everyone, and smart enough to realize that pleasing a small subset is usually going to be adequate, long term 18:26
masak "no offense, but you people suck" -- I've never seen this have a net positive effect.
rjbs The small subset containing one element, $Spouse.
TimToady but I think the actual implied insult is that hackers aren't smart enough to worry about fashion while they're worrying about important things :)
awwaiid pft. clearly the "no offense" part cancels out _anything_ that follows!
diakopter masak: again, I really don't see how anything she says is insulting
or says "you suck" 18:27
masak diakopter: she says "you guys need help".
diakopter so?
how is that insulting? 18:28
awwaiid goes back to hacking
masak that's on the same size of 0 as "you suck".
side*
diakopter everybody needs help.
masak stops arguing
TimToady the intent was to insult, in my opinion
sirrobert She's not claiming helping "nerds" address a problem they know they have. She's not even helping show them a problem they don't know they have. 18:29
she's simply asserting the problem, which doesn't work towards a solution, it simply espresses opposition =) 18:30
expresses
diakopter she promises to offer solutions in forthcoming posts, it seems 18:31
sirrobert yeah, but if we don't agree there's a problem (and what the problem is) then we're not interested in her solutions
TimToady solutions of the form, "think less about hacking and more about fashion"
sirrobert so it does her no good =)
masak I agree about the underlying premise. fashion isn't a prime concern at a hackathon. 18:32
diakopter I don't think it would mean spending more time thinking about fashion.
sirrobert so, I have this: HTTP::Server::Simple.new(8080).run; 18:33
diakopter she's not talking about attending hackathons
sirrobert if I go to it in my browser, it works fine =)
if I hit refresh twice fast, it works fine
if I hit refresh 5+ times fast, it crashes silently.
how do I troubleshoot that? =)
(I'm also open to an alternative to HTTP::Server::Simple) 18:34
18:36 gv joined
masak TimToady: 'Foo' in 'class Foo' is a <longname> according to STD.pm6. unquotes are terms, so something is missing. how would a solution that allowed 'Foo' to be parameterized look? 18:40
18:41 nebuchadnezzar joined
moritz add an | <indirect> branch to token longname 18:42
TimToady well, ¤longname() or some such on the interpolation end, or a generic ¤() on interpolation and a longname parser attached to the parameter
18:43 sudokode joined
TimToady is parsed(/<longname>/) ain't so far off 18:43
moritz and have some sensible syntax for <indirect>
TimToady I fail to see how doctoring the grammar helps the macro know how to parse the parameter 18:44
maybe you're solving a different part of the problem from me
moritz I might be 18:45
TimToady but it seems like identifying the rule to parse the parameter with would be close to adequate
where the default is <term>, more or less 18:46
18:46 Vlavv joined
diakopter twitter.com/jordynclee/status/2330...8346207233 18:46
she replies to nearly every comment in that ycombinator thread
masak TimToady: thank you. I'll put that in my brain and let it go through some cycles to see what comes out. 18:49
TimToady the fashionistas always wage war on utilitarian garb, nothing new here...
my wife thinks there's a time and place for both, and I like someone who thinks that way :) 18:50
masak I'm considering the possibility that going to a hackathon in a conf t-shirt *is* a fashion statement.
or going to a conf in a conf t-shirt. 18:51
moritz was that picture from the oslo hackathon published under some creative commons license? 18:52
masak moritz: howcaniexplainthis.blogspot.se/2012...tures.html 18:53
I don't see a license.
TimToady and part of the (unspoken) subtext of putting that picture there is not just about clothing; it's really about the pose, or lack thereof, which reinforces the insult that "hackers just don't care enough about how they look to pose for a picture properly" 18:54
gee, they should've hired a profession photographer for the photoshoot 18:55
masak :) 18:56
TimToady or maybe it's "hackers don't care enough to take 100 pictures and pick the best one" 18:57
sirrobert I'd be interested in software that could pick the best one based on parameters =) 18:58
18:59 birdwindupbird joined
masak clearly this blogger feels a need to address geeks but doesn't quite know how to do it. she needs help. :) 19:01
sirrobert heh
19:03 GlitchMr left
moritz well, she can help me by leaving me alone :-) 19:08
I know how to dress up, if that's what I want
I just don't want, 99% of all days
sirrobert That's the problem with the problem =)
It's a problem *for her*, not for you/me/us
moritz I rather think it's a business for her, not a problem 19:09
sirrobert well, businesses are about solving problems
but it has to be your customer's problem from your customer's perspective
if it's his problem from your perspective, your problem from yours/his, there's no sale =) 19:10
diakopter don't read her blog if you don't want to :P "something is wrong on the ....... "
moritz no, businesses are about making money
solving problems is just one way to do that 19:11
sirrobert I don't see any modules for handling socket IO. Is there a good help resource for it right now?
moritz IO::Socket::INET is built in
masak diakopter: it just happens to be a blog post with a picture with people I know on it labeled as "Don't".
moritz happens to be one of those people
sirrobert I was defining "problem" as "a need sufficient to part with one's money willingly for a given addressing of the need" 19:12
diakopter has realized that 10 minutes ago
sirrobert moritz: thanks (re: Socket)
masak diakopter: "hey, don't take it personal. it's not like she's talking about *you*. oh wait."
diakopter (right)
19:13 daxim left
colomon finally looks at the article and realizes he is at this very moment wearing one of the t-shirts pictured... 19:15
19:25 alester left
moritz programmers.stackexchange.com/users/login has "why OpenID?" box, but I can't find the option to log in with OpenID 19:30
am I stupid, or are they?
flussence you have to turn JS *off* to log in with it :D
huf O_o 19:31
moritz wtf.
huf websites can be so <3
moritz flussence++
diakopter moritz: click "show more login options"
jnthn wtf, I thought flussence was kidding! 19:32
Wow...
huf apparently he was, but i took him for serious :)
sirrobert my script is exiting (unexpectedly) with an exit code of 141. Is there a list of codes somewhere?
jnthn Oh!
diakopter moritz: (the openid icon is the first on the left)
jnthn I took the <moritz> flussence++ as "that solved my problem" :D
moritz yes, turning off JS did solve my problem 19:33
huf well, it works better without js, apparently ;)))
moritz hense flussence++
*hence
huf one less click before you go
jnthn Oh, hah :)
diakopter strange you didn't see the "show more login options" link/button
s/you didn't see/it didn't show you/
it shows in every browser I've tried 19:34
sorear sirrobert: unix shells will map signal N to exit code 128+N 19:35
sirrobert ah 19:36
sorear sirrobert: subtract 128 and look in the output of kill -l
sirrobert sorear: thanks =)
sorear HOWEVER, this is not necessarily the source
because the exit() system call takes values from 0 to 255
sirrobert the code makes sense in context 19:37
(now I have to figure out what to do with it =)
sorear n: say 141-128
p6eval niecza v19-15-g051783d: OUTPUT«13␤»
sirrobert SIGPIPE
from HTTP::Server::Simple crashing
huf i think exit actually takes an int, but "portable" values are between 0 and 126 inclusive 19:38
or something horrible like that
sorear depends on which standard you ask 19:40
ansi C, I think it takes an int but the only portable values are EXIT_SUCCESS and EXIT_FAILURE 19:41
and 0, which is transparently mapped to EXIT_SUCCESS
e.g. on VMS exit values are a full 32-bits, but the native EXIT_SUCCESS is not 0, it's like 0xC000 or something
huf yeah, the 0-126 is for unixlikes and mostly takes the shell's mangling into account 19:42
sorear when I talk about 8-bit exit() values I am referring only to traditional unix behavior
exit(127) is not the best idea because the shell calls exit(127) if exec fails
huf yes, but if you're gonna run it from shell, 127 and 128+n are out
yeah
sorear 192 is probably safe though, I've never seen a unix with more than 64 signals :D 19:43
huf :)
19:45 pmurias joined
sorear at the level of 'union wait' there are 256+3*NSIG distinguishable statuses but for some reason I rarely see that in shells 19:46
(granted, WIFSTOPPED would pretty rarely come up in a shell situation...) 19:47
geekosaur hm? ctrl-z 19:54
I'd think it more common than the other case it comes up (debuggers)
sorear d'oh 19:57
20:05 birdwindupbird left, tokuhiro_ joined 20:06 SamuraiJack joined
dalek rl6-roast-data: 9e0aac1 | coke++ | p (2 files):
today
 20:08
20:15 benabik left 20:16 benabik joined 20:20 tokuhiro_ left
TimToady ./viv -e 'repeat { ... };' 20:22
===SORRY!===
"repeat" is missing its "while" or "until" at (eval) line 1:
can't check in till I fix the other stuff though 20:23
masak \o/
TimToady nap & 20:25
20:25 birdwindupbird joined
masak good UGT night, TimToady. 20:25
PerlJam predicts that will be shortened to just "good ugt" (as both a greeting and a farewell) in common usage 20:27
20:27 renormalist joined 20:29 spider-mario joined
moritz \UGT 20:29
20:30 bruges left 20:31 bruges joined 20:32 alvis joined
[Coke] gugt. 20:34
gug.
ug.
moritz hands [Coke] a banana
masak hands [Coke] a glass of water
20:42 benabik left
mhasch would have thought [Coke] preferred another beverage 20:45
20:47 MayDaniel left
dalek kudo/nom: 4acaef3 | coke++ | tools/autounfudge.pl:
Allow autounfudge to use --section=1|01
 20:48
20:53 brrt joined 20:55 thou left 20:58 MayDaniel joined
dalek kudo/nom: 038a3c7 | coke++ | tools/autounfudge.pl:
allow override of autounfudge output file
 21:00
21:04 araujo joined, thou joined 21:09 skids left 21:11 obra joined 21:14 kaare_ left 21:19 brrt left 21:21 SamuraiJack left 21:22 atrodo left 21:25 birdwindupbird left, moukeddar joined, benabik joined 21:27 moukeddar left 21:34 spider-mario_ joined, spider-mario left 21:35 spider-mario_ is now known as spider-mario 21:38 gv left
masak managed to send in both proceedings articles before the deadline 21:43
21:44 Chillance joined
tadzik good job :) 21:44
21:46 jeffreykegler joined
masak I will definitely write proceedings from now on. they made me rethink the exposition of the talks. 21:47
21:48 whiteknight joined, whiteknight is now known as Guest39929
jnthn sent his in too 21:52
21:52 supernovus joined
supernovus A quick question. Say I have a BUILD submethod, that needs to populate object attributes based on a list of fields and a set of data for those fields. How would you get a pointer to the private attribute storage for said fields based on a variable name? I know for attributes marked 'rw' it's a simple self."$fieldname"() = $value; however, having to have all fields marked 'rw' may not always 22:00
be desirable? If this is not possible, don't worry about it, I'll just force the use of 'rw' attributes in this particular case.
jnthn Can you not write a method with attributive parameters and map your list of fields etc into a call to that? 22:02
I mean like a method setup(:$!foo, :$!bar) { ... } 22:04
And then if you have a hash of the ones you want to set you self.setup(|%that_hash)
22:04 fgomez joined
supernovus Hmm, I will look at that. I was hoping to find a way of doing it so that the person making the child class didn't have to do anything other than define their list of @.fields, and the rest would be done by the parent class in the background. I will play with it some more I think. 22:08
22:08 MayDaniel left
jnthn Oh 22:09
Then you're into the real of meta-programming
Use the MOP.
*realm
.^attributes gets you the attributes
TimToady you can reimplement Ruby if you want :) 22:10
jnthn And the Attribute objects provide a way to get/set the attributes.
TimToady: :P
22:10 bluescreen10 left 22:11 pmurias left 22:16 jeffreykegler left
masak awwww i.chzbgr.com/completestore/12/8/5/...u1baw2.png 22:18
tadzik masak: awesome :) 22:19
22:20 jeffreykegler joined
masak I knew you would like it, tadzik :) 22:20
tadzik it has Hugs and Bison. At the same time! \o/
masak jnthn++ found it :) 22:21
supernovus jnthn: thanks, the MOP method will work great. 22:23
22:28 skids joined 22:31 spider-mario left
sirrobert supernovus: I'm playing with WWW::App (thanks =) but having a little trouble setting it up with lighttpd. have you used it? 22:40
or anyone else, for that matter ;)
supernovus sirrobert: Yes, I am using it with lighttpd at the moment (in testing of WWW::App::Easy, an MVC framework for building rapid web apps.)
sirrobert cool. I've never used scgi or psgi before. I don't quite understand how to set up lighttpd for it. Can you point me to an example config or something? 22:41
I'm trying to configure to use with scgi right now 22:42
fwiw, I'm using the example app (scgi.p6) from your SCGI module 22:43
github
supernovus sirrobert: make sure you have enabled the scgi plugin for lighttpd: server.modules += ( "mod_scgi" )
sirrobert did that
let me pastebin you my conf... one sec 22:44
crud, I had just scrapped it to start over =)
supernovus I will pastebin you the one I'm using
sirrobert cool, thanks
supernovus pastebin.com/E63Weapt 22:47
sirrobert thanks, working on integrating that 22:48
might bug you once or twice more if you're still around =)
supernovus No problem. lighttpd + SCGI + WWW::App makes a pretty good web framework. Once WWW::App::Easy is finished, building dynamic web apps in Perl 6 will be super easy! 22:49
sirrobert It looks promising already. my company is building a virtual appliance and I'm working on setting this up as the api front-end initially 22:50
(well, api/docs) 22:51
supernovus++: ah, great. That got it going for me. Thanks =) 22:54
as I'm working, I'll see if my team or I can send some stuff upstream while we work 22:56
[Coke] just listened to blip.tv/timbunce/devel-nytprof-v4-o...7-3932242, where tim says "I'm kind of distracted with perl6 stuff at the moment." 22:58
22:59 gongyiliao joined 23:01 gongyiliao left
diakopter o_O 23:04
23:05 gongyiliao joined 23:15 benabik left 23:17 awwaiid left, jeffreykegler left 23:18 benabik joined, jeffreykegler joined
[Coke] does remember some DBIish things going on, but doesn't think it landed. 23:24
23:30 awwaiid joined
diakopter [Coke]: maybe there was a secret project at the time 23:36
(that would've never appeared, or at least is taking a very long time) 23:37
23:37 awwaiid left 23:38 jeffreykegler left, awwaiid joined
masak good nugt, #perl6. 23:38
sirrobert supernovus++ (for WWW::App and SCGI)
masak: wave 23:39
23:39 thou left 23:43 awwaiid left 23:44 awwaiid joined 23:52 jeffreykegler joined 23:54 raiph joined 23:55 Chillance left
raiph hello #perl6 # showing emily, 8, how irc chat works 23:56
tadzik hello raiph and emily :) 23:57
raiph hello
23:57 benabik left, benabik joined
raiph perl6: say 1 23:58
p6eval rakudo 038a3c, niecza v19-15-g051783d: OUTPUT«1␤»

Powered by Cro and the Raku® Programming Language.

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

Thank you!