-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.
00:31 Limbic_Region_ is now known as Limbic_Region
Alias_ autrijus: ping? 04:27
svnbot6 r7957 | luqui++ | Changed "is lazy" (which is used nowhere) to reflect the semantics 10:00
r7957 | luqui++ | I recently proposed on p6l, hoping that that's the way it ends up.
r7957 | luqui++ | Well... almost; it doesn't cache the value. Maybe that's a good thing
r7957 | luqui++ | (how else would you implement infix:<for>?).
luqui ?eval sub infix:<myfor> ($exp is lazy, @list) { for @list { $exp() } } say "hi" for 1..10 10:01
10:01 evalbot_7956 is now known as evalbot_7957
evalbot_7957 OUTPUT[hi hi hi hi hi hi hi hi hi hi ] undef 10:01
luqui hmm, still need to handle $_ somehow... 10:02
?eval sub bind (*@list, *&code) { @list.map:&code } bind 1,2,3 { $_, $_+1 } 10:04
evalbot_7957 Error: unexpected ":" expecting term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or "}"
luqui ?eval sub bind (*@list, *&code) { @list.map:&code } bind 1,2,3, { $_, $_+1 }
evalbot_7957 Error: unexpected ":" expecting term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or "}"
luqui ?eval sub bind (*@list, *&code) { @list.map(&code) } bind 1,2,3, { $_, $_+1 }
evalbot_7957 Error: Slurpy array followed by slurpy scalar: MkParam {isInvocant = False, isOptional = False, isNamed = False, isLValue = True, isWritable = False, isLazy = False, paramName = "&code", paramContext = CxtSlurpy (mkType "Code"), paramDefault = Noop}
luqui hmm
?eval sub bind (*@list) { my &code = @list.pop; @list.map(&code) } bind 1,2,3, { $_, $_+1 } 10:05
evalbot_7957 Error: cannot cast from VCode (MkCode {isMulti = False, subName = "<anon>", subType = SubBlock, subEnv = Just (MkEnv {envContext = CxtSlurpy (mkType "Any"), envLValue = False, envLexical = MkPad (padToList [("$?1",[(<ref>,<ref>)]),("$_",[(<ref>,<ref>)]),("$__evalbot_print",[(<ref>,<ref>)]),("$code",[(<ref>,<ref>)]),("$lang",[(<ref>,<ref>)]),("&?BLOCK_EXIT",[(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>)]),("&bre
luqui ?eval sub bind (*@list) { my &code = @list.pop; @list.map:{ &code($_) } } bind 1,2,3, { $_, $_+1 }
evalbot_7957 Error: cannot cast from VCode (MkCode {isMulti = False, subName = "<anon>", subType = SubBlock, subEnv = Just (MkEnv {envContext = CxtSlurpy (mkType "Any"), envLValue = False, envLexical = MkPad (padToList [("$?1",[(<ref>,<ref>)]),("$_",[(<ref>,<ref>)]),("$__evalbot_print",[(<ref>,<ref>)]),("$code",[(<ref>,<ref>)]),("$lang",[(<ref>,<ref>)]),("&?BLOCK_EXIT",[(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>)]),("&bre
luqui ?eval sub bind (*@list) { my $code = @list.pop; @list.map:{ $code($_) } } bind 1,2,3, { $_, $_+1 } 10:06
evalbot_7957 Error: cannot cast from VCode (MkCode {isMulti = False, subName = "<anon>", subType = SubBlock, subEnv = Just (MkEnv {envContext = CxtSlurpy (mkType "Any"), envLValue = False, envLexical = MkPad (padToList [("$?1",[(<ref>,<ref>)]),("$_",[(<ref>,<ref>)]),("$__evalbot_print",[(<ref>,<ref>)]),("$code",[(<ref>,<ref>)]),("$lang",[(<ref>,<ref>)]),("&?BLOCK_EXIT",[(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>),(<ref>,<ref>)]),("&bre
luqui grr
x86 whoa 10:15
perl6 has real objects? heh
where you can call a method on a variable using . ? 10:16
luqui x86 what do you mean "where"?
oh
that wasn't a question
x86 ?eval my $test = "hello"; $test.print;
evalbot_7957 OUTPUT[hello] bool::true
luqui it was a followup
x86 whoa, cool!
luqui new to the perl6 world eh?
x86 like java and so forth
yeah lol 10:17
all i know is that it uses parrot ;)
and pugs
what does the : operator do?
luqui it appends an "adverb" to a method call
x86 adverb? 10:18
luqui it was introduced to get rid of those ugly parens in: @list.map({ ... })
er, long story
x86 ah i see
makes sense
now can you still do the oldschool way like print $variable ? 10:19
luqui turns out that that adverb block is the same block that is given to the "for" function (which you can now define yourself)
certainly
x86 or do you have to do $variable.print ?
luqui ?eval print "hello, world"
evalbot_7957 OUTPUT[hello, world] bool::true
x86 hehe i hope perl6 maintains perl5 compatibility for a while ;)
luqui hehe, nope
x86 no?
luqui the point of the project was to break perl5 compatibility 10:20
because perl5 is full
x86 there is no perl5.pm?
"full" ?
luqui there's no more syntax we can add to it
x86 why?
luqui just about every string of characters already has a valid interpretation :-)
x86 so how does perl6 fix that? 10:21
luqui oh, but you mean the ability to use eg. perl5 modules in perl6 programs?
x86 right
luqui oh yeah, we're working on that
you can do it in pugs (with caveats)
x86 "we're" ?
luqui use perl5:DBI;
design team. Nicholas Clark mostly
x86 working with larry wall? 10:22
luqui uh huh
x86 you know him?
luqui uh huh :-)
x86 in real life?
luqui visited the japanese garden with him and Gloria
x86 gloria?
luqui his wife
x86 japanese garden?
luqui in portland, during OSCON
x86 cool :)
i hear he hates IRC :P 10:23
is that true?
luqui yeah, he says he wastes enough time on email
x86 hehe
prolly true
so what's his dayjob?
luqui he finally got hired somewhere 10:24
I don't remember though
some sort of consulting I think
he was employed only by TPF grants for a while...
play with perl 6. you'll like it (well, probably not without learning some of it you won't, but hey) 10:27
wolverian how I love 1000+kB/s connections 10:35
luqui somebody who knows the pugs internals: ping 10:43
tewk ll 10:52
luqui SS
x86 tt 10:57
tewk So how is the object stuff in per6 going? and compiling to parrot? 10:59
luqui no idea
IIRC, according to autrijus's journal, compiling to parrot is working nicely 11:00
svnbot6 r7958 | luqui++ | r717@jabberwock: luqui | 2005-11-16 06:01:01 +0000 13:04
r7958 | luqui++ | Made the name of an operator being defined visible within the sub. This
r7958 | luqui++ | is important for unary ops and operator definitions.
r7958 | luqui++ | Also made it possible to take new operator names (say, @@) as parameters,
r7958 | luqui++ | not already-parsing ones.
luqui ?eval product 1..10 13:11
13:11 evalbot_7957 is now known as evalbot_7958
evalbot_7958 Error: No compatible subroutine found: "&product" 13:11
luqui ?eval [*] 1..10
evalbot_7958 3628800
luqui ?eval sub postfix:<!> ($x) { [*] 1..$x } 3!
evalbot_7958 6
luqui ?eval sub postfix:<!> ($x) { [*] 1..$x } 4!
evalbot_7958 24
luqui ?eval sub postfix:<!> ($x) { $x ?? $x * ($x-1)! !! 1 } 4! 13:12
evalbot_7958 Error: unexpected "!" expecting term postfix, operator, ">>xx<<", ">>+&<<", ">>+<<<", ">>+><<", ">>~&<<", ">>~<<<", ">>~><<", ">>*<<", ">>/<<", ">>%<<", ">>x<<", "\187xx\171", "\187+&\171", "\187+<\171", "\187+>\171", "\187~&\171", "\187~<\171", "\187~>\171", "\187*\171", "\187/\171", "\187%\171", "\187x\171", "xx", "+&", "+<", "+>", "~&", "~<", "~>", "*", "/", "%", "x" or "!!"
luqui ?eval sub postfix:<!> ($x) { $x ?? $x * (($x-1)!) !! 1 } 4!
evalbot_7958 Error: unexpected "!" expecting term postfix, operator or ")"
luqui ?eval sub postfix:<!> ($x) { [*] 1..$x } 5!
evalbot_7958 120
luqui ?eval sub postfix:<!> ($x) { [*] 1..$x } (5+1)!
evalbot_7958 720
luqui ?eval sub postfix:<!> ($x) { if $x == 0 { 1 } else { $x * (($x-1)!) } } 13:13
evalbot_7958 Error: unexpected "i" expecting ";", statements or "}" reserved word
wolverian that looks rather broken :)
luqui ?eval sub postfix:<@@> ($x, $y) { if $y == 0 { $x } else { $y @@ $x % $y } } 6 @@ 9 13:15
evalbot_7958 Error: unexpected "i" expecting ";", statements or "}" reserved word
luqui sonofa!
I thought I just fixed that
?eval sub infix:<@@> ($x, $y) { if $y == 0 { $x } else { $y @@ $x % $y } } 6 @@ 9
evalbot_7958 \3
luqui oh, I did, but apparently just for infix 13:16
?eval sub postfix:<!> ($x) { say $x; ($x-1)! }
evalbot_7958 Error: unexpected "!" expecting term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or "}"
luqui ?eval sub postfix:<``> ($x) { say $x; ``($x-1) }
evalbot_7958 Error: unexpected "`" expecting ";", statements or "}"
luqui ?eval sub prefix:<``> ($x) { say $x; ``($x-1) }
evalbot_7958 Error: unexpected "`" expecting ";", statements or "}" 13:17
luqui geez
?eval sub postfixfix:<``> ($x) { say $x; ($x-1)`` }
evalbot_7958 Error: unexpected ":" expecting word character, "::", bare trait, subroutine parameters, trait or block
luqui heh
?eval sub postfix:<``> ($x) { say $x; ($x-1)`` }
evalbot_7958 Error: unexpected "`" expecting term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or "}"
luqui ?eval sub postfix:<@@> ($x) { say $x; ($x-1)@@ }
evalbot_7958 Error: unexpected "@" expecting term postfix, operator, postfix conditional, postfix loop, postfix iteration, ";" or "}"
luqui okay fine, it doesn't work
13:49 r0nny_ is now known as r0nny
r0nny yo 13:50
robkinyon autrijus: ping 15:22
liz6 seen autrijus 15:23
jabbot liz6: autrijus was seen 2 days 9 hours 16 minutes 2 seconds ago
robkinyon gack
anyone else here good w/svk?
I can't seem to get SVN::Core installed, even though I've recompiled svn w/swig bindings 15:24
broquaint You might want to try clkao, robkinyon. 15:25
robkinyon seen clkao
jabbot robkinyon: clkao was seen 4 hours 50 minutes 58 seconds ago
robkinyon grr
clkao: ping 15:26
(for whenever)
clkao hi 15:27
robkinyon clkao: I installed swig, then compiled svn, but SVN::Core doesn't seem to be installed 15:31
what am I doing wrong?
clkao did you do make check-swig-pl and install-swig-pl ? 15:32
robkinyon no ... was that documented somewhere?
I installed swig off of CPAN, so I though it would install the Perl bindings
or is that for SVN? 15:33
clkao yes, last line of subversion's INSTALL 15:34
robkinyon clkao: thanks 15:35
pawel ping 16:03
pmurias does anyone know how to make ghc emit C--? 16:27
ajs_ C--? 16:31
As in the GTK bindings for C++? 16:32
SamB pmurias: hmm. 16:33
ajs_: thats GTK--
theorbtwo C-- is a language closer to portable assembler then C.
www.cminusminus.org 16:34
SamB pmurias: you'd need to hack it, it looks like
pmurias: or at least use a more recent version of GHC than 6.4
pmurias: it shouldn't be too hard though 16:35
svnbot6 r7959 | luqui++ | Ugh. Rolled back my two most recent changes, as they cause 16:41
r7959 | luqui++ | the Precompiled Prelude generation to hang. Don't have time
r7959 | luqui++ | to "fix" it right now, so I'll just leave it in a working state.
pmurias Is it possible and how do you manipulate haskell data structures from perl6? 16:46
Would a i have to inline haskell code? 16:49
SamB pmurias: its not even really possible from C... 16:50
pmurias are there any parts of pugs which are untested? I could write some tests if they are needed. 17:16
xinming pmurias: There are many, But you have to find it yourself. ;-) 17:18
Alias_ First, write a Devel::Cover equivalent
theorbtwo You might want to start by looking at hat: www.cs.york.ac.uk/fp/hat/ 17:25
r0nny yo 17:31
is there a exclusive exteonsion for perl6 modules ?
want to change the vim autosyntax stuff
autrijus .p6 I think 17:33
oh you mean modules
no, probably not
and .p6 is just transitional anyway
but modules usually begin with "module" or "class" 17:34
or "role"
geoffb autrijus!!
autrijus greetings!
geoffb Good Day!
autrijus I survived! (was abducted by another book after Unicode Demystified)
but it's worth the abduction. 17:35
geoffb oh, which book?
autrijus Code Complete 2nd Edition (cc2e.com)
I...wish I had this book 10 years ago ;)
autrijus is tempted to add it to READTHEM 17:36
geoffb I was wondering if that was worth picking up . . .
autrijus oh, totally totally worth it.
Alias_ Unicode Demystified? 17:37
Must be a VERY big book
autrijus Alias_: it is.
but Code Complete is bigger
and more relevant, anyway
geoffb I tend to pick books by going to a bookstore, finding likely titles, paging through them and reading random bits. If there's something new in there, I pick it up. It may have been bad luck, but CC failed that test last time.
ntgrl _Code Complete_ is definately a good one :) lots of lovely experimental results
autrijus CC is really not for random access :) 17:38
geoffb OK, fair enough.
>> reading-pile
autrijus it's almost completely sequential and would only make sense that way
imvvho
geoffb nodnod 17:39
Alias_ ugh... I hate when you are getting distracted from your distractions to fix code that is need by ... etc...
The worst bit is when you fix the very most distracted thing and then forget how you got to where you are
geoffb Alias_, you have just discovered the joy of yak shaving 17:40
Alias_ it has a name?
geoffb yup
autrijus and reading CC is definitely yakshaving for perl6/pugs :) 17:41
Alias_ yep, I'm most definitely yak shaving
Actually, I managed to stop myself from refactoring Template Toolkit 17:42
geoffb oh my
good save
Alias_ Which would have been where I got a bit too close to the Yak's ass for comfort
geoffb lol
Alias_ codegen is just one big excuse for yak shaving
"... hmm... but I could solve this in the generic case by ..." 17:43
"... and I'd never have to solve this problem in any application ever again if..."
50% of my modules are probably due to it too
geoffb Go ahead, Alias_, be the "unreasonable man" -- you know you want to
heh
ntgrl Is there a module yet for doing nicely indented/prettified output? 17:44
theorbtwo Data::Dump::Streamer?
Alias_ This time around it's "you know, with a Template filter I'd never have to work out which forms needed enctype="multipart/form-data"
geoffb Nice -- the specs page on the C-- site is 403
ntgrl theorbtwo: umm, but I'm not outputting perl code...
geoffb ntgrl, are you asking about Perl 5? Or Perl 6? 17:45
ntgrl geoffb: either. I was thinking something like Text.PrettyPrint.HughesPJ
geoffb ah
theorbtwo Oh. 17:46
theorbtwo suddenly notices that this isn't #london.pm
Alias_ aha! I finally remember where I was 17:48
theorbtwo Ugh! The current version of hat in debian is from 2003!
autrijus ...and wouldn't work with GHC anyway 17:49
geoffb $time, ingy 17:50
theorbtwo wonders if he wants current-release or CVS.
Well, since 2.04 doesn't even get past it's make system, let's try CVS. 17:51
autrijus ingy: yo 17:52
Alias_: no matter what I did on rex.com it always bails with "internal system error"
what's wrong with that site?
theorbtwo Whee: detects that I don't have hmake, and tries to use it anyway.
Alias_ It's COMPLETELY fucked
I wrote a huge rant to the company 17:53
theorbtwo This does not bode well to the usability of the code.
Alias_ autrijus: You pretty much need to change browsers
I think it's the sessions getting screwed
autrijus wtf :(
Alias_ So maybe flush cookies or something
I needed three browsers to get through
autrijus so what worked?
Alias_ They all work, you just can't date leave the most direct line 17:54
dare
So flush everything, then search, select, book, pay immediately
Don't let the session get stale
It's that bad
autrijus I see.
yup, that's it. 17:55
geoffb ponders what to do with day: write overdue O'Reilly blog, hack gjbforth, hack PIGGE, fall asleep with eyes open and hands at kb . . . .
Alias_ I'm going to ring in the morning
autrijus how very retarded.
Alias_: so, conf ends 7th @ melbourne. are you going to osdc, and how/when are you flying back if so?
Alias_ I'm going, I haven't booked flights yet 17:56
autrijus I need to give a talk on 12th in sydney, no other engagements
Alias_ hmm
Let me see if anyone does Ballina to Melbourne direct
That might be a better option
stevan hey autrijus :) 17:57
Alias_ but I don't think they do
autrijus anyway, the plan is to fly back with you whenever your flight is
Alias_ ok
I'll let you know the flight to book
autrijus stevan: hey! CC2e totally r0x0rz
Alias_: good, because it will just be two roundtrips for me
sydney<->melbourne and then sydney<->lismore
stevan hasn't read that, but has been told to by many people 17:58
autrijus maybe more reasonable than sydney->melbounre->ballina->lismore->sydney
stevan is currently reading The Stars My Destination, by (Alfred Bester)++
Alias_ oh, we'd drive from ballina
half an hour
oh wait, I see
So you get into Syd and someone is paying Syd<->Mel 17:59
But they haven't booked specific flights?
stevan autrijus: I had/am-having a crazy idea about introducing SML-like module signatures into the Perl6::Core:: 18:00
then the metamodel can program to the signatures,.. and the core:: then can be easily swapped out 18:01
autrijus stevan: good, been thinking about the same
Alias_ SML?
stevan and instead of wrapping the metamodel,... we can do something which maps Roles to module sigs
autrijus sexp is a bit too dyn
stevan Alias_: Standard ML
autrijus: sexp is mostly just for fun :)
autrijus well, luqui's idea has always been that functor (module sigs) becomes theories 18:02
so that may be that "something"
stevan yeah it maps nicely to theories :)
in fact.,.. theory.pod suddenly made much more sense when I re-read the Functor chapter in my ML book 18:03
autrijus right, because I believe it originated from Scala and Scala took heavily from there 18:05
Alias_ autrijus: OK, there is no combination of flights that can get us from Melbourne to Lismore on the night of the 7th 18:08
autrijus: So whoever is hosting you, plan to stay the night of the 7th
autrijus ack 18:09
so we fly 8th and I fly again on 13th
Alias_ we fly the 8th... I have no idea what you do on the 13th :) 18:10
autrijus er wait
I mean 11th ;)
Alias_ where and when is your Sydney commitment
specifically
autrijus unsw, 12 morning
Alias_ ouch 18:11
hmm
autrijus ponders if Alias_ can also fly to unsw for more fun 18:12
Alias_ Maybe, but probably not
I'm a bit overcommitted at the moment
clkao giggles at autrijus
autrijus k... just tell me the times then
Alias_ It would be nice to go back to unsw though
ok then, so looking at flight times, we have to take rex on the Lis<->Syd legs 18:13
Nothing else can match the schedule
Which means flight ZL338 departing Sydney 3:10pm 18:14
on the 8th 18:15
This app is SO fucked
autrijus okie, ack.
Alias_ and for you, return flight ZL319 departing 8:55am on the 11th 18:17
which gives you time to grab a taxi to UNSW
with about half an hour spare :)
autrijus that... is a bit mean
Alias_ oh this is FUCKED 18:18
autrijus it's 12th morning
Alias_ what?
autrijus not 12 noon morning
misparse, sorry for bogus input
(take this to /msg) ?
sorry for the flooding :) 18:20
stevan autrijus: ping 19:33
autrijus pong
stevan hey,.. got time to chat about the module signature idea/ 19:34
autrijus I'm very sorry but no, it's already way past bedtime :/
stevan ah,.. ok,.. :)
autrijus I'm back at an unfortunate time zone ;)
3:34am here
stevan 2:33 pm, here :)
autrijus almost maximally wrong, sigh 19:35
stevan ok,.. so maybe tomorrow morning my time/evening your time?
or maybe talk late night tonight for me,,.. early morning tomorrow for you?
whenever you have some time basically :)
autrijus write it down first, then? 19:36
stevan good idea,,.. i will do that
autrijus I'll have plenty of time starting tomorrow
but net connectivity might be limited
stevan ok
obra where you headed?
autrijus I'm discovering this Reality thing isn't that bad after all and am coding things up in hot springs etc.
obra: random pleasant place nearby
s/place/places/ 19:37
obra nice
liz6 autri: yes, but you don't see me... 19:50
autrijus mumbles something about nickserv and freenode 19:55
liz6 mumbles utters something back
autrijus oh well. I'll mail instead and sleep. *wave* 19:56
liz6 waves back
xinming won't stop waving until autrijus got asleep... 20:01
He might get a sleepless night if I do so. ;-)
r0nny_ *back* 20:11
can anyone here have a better vim file then the actual one ?
s/can/does/ 20:12
PerlJam "better vim file"?
r0nny_ u know - syntax highlighting
PerlJam If the syntax highlighter is borked, fix it!
r0nny_ the perl6 syntax file scared me away 20:13
think i found what broke the module name-highlighting 20:14
Mitar would it be possible to implement haskell under parrot? 20:18
so are there any structural limits for implementing running haskell under parrot?
like that parrot does not support anything what would haskell need? 20:19
i know that everyhing could be simulated, but this would give perfomance penalty
xinming 0,8,24,48,80,120 what's next number please?
Mitar 42
xinming a maths question. :-/
Mitar :-)
it could be any number
ntgrl 0*8, 1*8, 3*8, 6*8, 10*8, etc 20:20
Mitar mathematically
xinming ....
Mitar but you could ask which is the next number the author of the sequence thinks of
xinming and found the relation between 0 1 3 6 10.
ntgrl which is good ol' triangular numbess
r0nny_ im to lazy for finding the rule
ntgrl 1/2*n*(n+1)(8 20:21
4*n*(n+1)
xinming I am poor at maths :'(
PerlJam I think the correct solution here is to write a perl6 program that finds the next number ;-)
ntgrl xinming: you can draw out 0, 1, 3, 6, 10 as little patterns of dots and put those into triangles 20:22
xinming ?eval my @a = [0,8,24,48,80,120]; { $_.say } for @a; my @b = [1..10]; { $_.say } for @b; 20:23
20:23 evalbot_7958 is now known as evalbot_7959
evalbot_7959 undef 20:23
xinming :-P
qwr try to subtract following numbers ;)
Mitar so does anybody knows an answer to my question? 20:25
qwr 120?
PerlJam Mitar: parrot could implement haskell, yes. 20:26
xinming Mitar: It will have one. :-)
Mitar is there some info on that?
xinming qwr: Yes, the answer is 120, But I feel regret for my poor maths in the school.
PerlJam I thought 120 was in the given set? 20:27
ntgrl the next number is 168, qwr
xinming the next is 168
:-/
PerlJam the next numbers would be 168, 224, 288 (I think)
xinming PerlJam: Exactly 20:28
ntgrl then it's 360.
Mitar PerlJam?
PerlJam ntgrl: then 440
xinming ... 20:29
ntgrl and eventually 1002000
PerlJam heh
ntgrl: and at some point it will be 1600080000
There's an interesting pattern there. Not sure if it's "real" or not though 20:30
ntgrl what? that they're approximately the squares? :P
PerlJam Why would anyone want to multiply triangluar numbers by 8 anyway? 20:31
qwr (n^2 - 1) * 8
ntgrl qwr: umm, no
theorbtwo To annoy math students, PJ.
ntgrl 4*(n^2 + n)
xinming ...
xinming wish to go back to school and learn maths. 20:32
qwr that is, when guess by looking some middle numbers ;)
svnbot6 r7960 | lwall++ | Could create -I options without an argument. 22:43
ntgrl *blink* lwall! 22:46
Alias_ blinks too 22:48
ntgrl hmm, not a bad patch as patches go 22:50
22:51 cp5 is now known as samy
ingy seen autrijus 22:54
jabbot ingy: autrijus was seen 2 hours 58 minutes 27 seconds ago
ingy hi samy
long time
samy hey ingy
yeah, how are you?
ingy fine 22:55
samy what are you doing these days
ingy heading to Taiwan for a couple months
liz6 has waved autri to sleep
ingy hi liz6
samy wow
ingy long time
liz6 indeed...... trying to catch up...
in the midst of $work++ 22:56
svnbot6 r7961 | lwall++ | range.t didn't test 'z'..'aa' carry, and was confused about 3..2ish things. 23:19
stevan horray for our new committer lwall :) 23:24
dduncan hey, it had to happen sooner or later 23:29
leo recently nearby at #perl6: 00:19 < svnbot6> r7961 | lwall ...range.t didn't test 'z'..'aa' carry, and was confused about 3..2ish
things.
oops pasted into wrong channel 23:31
dbrock how would I make hyphens synonymous to underscores in identifiers, and does Pugs support this? 23:39
dduncan do you mean that you want to use a hyphen as part of an identifier? 23:40
dbrock yes
I also want to be able to call `foo_bar' by typing `foo-bar' 23:41
dduncan I think you do it like ('foo')
dbrock I don't understand
dduncan at least, package name references can contain variables, such as ...
SamB dbrock: #lisp called. they want their hyphens back. 23:42
dduncan $text = &::($module_name)::get_text_by_key( $msg_key );
in that case, part of an identifier is in a variable ... I suspect that in this usage the identifier can contain any characters ... not sure though 23:43
my example meant &::('foo')::bar(), for example 23:44
dbrock SamB: hey, lots of people, not only Lisp weenies, use hyphens instead of underscores
dduncan or, Pkg::('foo-bar')()
dbrock dduncan: okay... well, I don't want my code to look like that
SamB dbrock: in what?
dbrock dduncan: although that is cool 23:45
dduncan I don't know if that would work, but I think that is how it would look if it did
dbrock SamB: take CSS, for example
dduncan fyi, my quoted example with $text actually does work
also, the paren notation works when specifying the version part of a module, for example, ... 23:46
luqui SamB, also, it is canonical for BNFs
dduncan use Foo-(0.0.2..0.1.3);
versus
dbrock I don't want my code to look like $text = &('get-text-by-key')($('msg-key'));
dduncan use Foo-0.3.0;
theorbtwo Feel free to warp the parser. 23:47
dduncan maybe the & isn't necessary ...
dbrock I want it to look like $text = get-text-by-key($msg-key);
theorbtwo But note that then you can no longer say time-3600 to mean time()-3600.
There are quite likely other conflicts as well.
dduncan the key thing is that if you have parens inside an identifier, they seem to interpolate any variable
dbrock theorbtwo: yes, obviously I can't do that
theorbtwo dduncan: I think it's only ::().
dduncan perhaps... that's only where I've seen it work 23:48
dbrock I would write "time - 3600" anyway
theorbtwo: how do I warp the parser?
dduncan I think that if we're going to start allowing '-' as part of identifiers, we should allow any characters, and require them to be delimited
this is how SQL works, fyi 23:49
an identifier can contain anything, even spaces, if it is double-quote delimited
dbrock but I don't want to double-quote my identifiers
theorbtwo In present-day pugs, by hacking haskell code. In perl6 as specced... well, don't know, but you can do it without hacking anything but rules.
dduncan then don't use funny characters in them
dbrock dduncan: the hyphen is not a "funny character"; it is roughly as common as the underscore as word-separator in identifiers 23:50
dduncan well, I suppose that if any use of subtraction requires spaces to be around the -, it could work 23:51
but only in that case
dbrock yes, and I think that is desirable
theorbtwo I don't. 23:52
dduncan personally, I wouldn't want to use a - in identifiers, though
theorbtwo It's often a good idea, but sometimes not.
dbrock why do you think it's not a good idea in this case? 23:53
dduncan dbrock, if you want to be able to have hyphens in identifiers, make your case on perl6-language
on that note, while perl 6 is highly about blazing new trails and finding a better way, even if no one else has done something before, what existing languages do you know of that allow hyphens in bareword identifiers? 23:55
luqui likes hyphens
dduncan bareword meaning non-delimited 23:56
luqui dbrock, however, can you give another example to back up your claim that hyphens are about as common as underscores?
dduncan in the context I'm referring to
dbrock luqui: I think the fact that all Lisp-derived languages use hyphens is enough to put it in the same ballpark 23:57
that includes languages like Dylan
luqui dbrock, yeah, but they also use many other characters in identifiers 23:58
',*,?,!,etc.
dbrock luqui: sure, they also allow underscores, but noone uses them
SamB the lexical syntax allows them, but the community would lynch you... 23:59
luqui dbrock, there is one problem I have with allowing hyphens
dbrock signal names in GObject allow both hyphens and underscores and treat them as synonyms --- I think this is the way to go if you want to appeal to Lispers and Unix people equally
luqui but it's not the hyphen in particular that I'm opposed to
it's the possibility of having both hyphens and underscores

Powered by Cro and the Raku® Programming Language.

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

Thank you!