»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
AlexDaniel Juerd: yea. If they want to keep opening issues, then perhaps they should also write code to resolve some of them 00:00
Juerd They are beautifully labeled, though.
00:00 cdg joined
AlexDaniel I'm running out of colors :) 00:01
Juerd Doesn't matter, it looks very professional
00:02 Cabanossi left 00:03 Cabanossi joined, wamba joined 00:04 cdg_ left 00:05 cdg left
Geth whateverable: 9a8b269474 | (Aleks-Daniel Jakimenko-Aleksejev)++ | t/lib/Testable.pm6
Add missing mkdir

In case it does not exist yet (e.g. fresh install).		 00:17
whateverable: d5fc45d357 | (Aleks-Daniel Jakimenko-Aleksejev)++ | t/lib/Testable.pm6
Use consistent logfile names

We use filenames like “unicodable_2017-09-19_0117.log” for regular log files, so let's use a similar format for other files (this format may be LTA, but consistency first).
AlexDaniel Juerd: well, if you like it… maybe you can help? 😍 00:22
Zoffix AlexDaniel: do you remember what you meant here by "Bools aren't tested"? rt.perl.org/Ticket/Display.html?id...xn-1453512 00:24
AlexDaniel Juerd: I can give you ssh access to the machine with bots and do a quick tutorial on how to hack them
Zoffix: okay… hmmm… I think I see 00:26
Zoffix: I was talking about things like a => True and a => False
whether this comment makes sense or not I'm not sure
but that's what I meant 00:27
Zoffix ok
AlexDaniel m: say (a=>False,3).perl
camelia (:!a, 3)
AlexDaniel m: say (a=>True,3).perl
camelia (:a, 3)
AlexDaniel m: say (:!a,3).perl
camelia (:!a, 3)
AlexDaniel m: say (:a,3).perl
camelia (:a, 3)
00:28 kyan joined
AlexDaniel Zoffix: I guess this includes :a and :!a which is what the OP is about 00:28
Juerd AlexDaniel: Thanks for the offer. Unfortunately I can't. 00:29
I'm afraid basically all I get around to, regarding Perl 6, is reading the weeklies.
kyan Zoffix: Sorry to bother you again, but I was wondering, when you mentioned earlier that '`$fail === "f"` is better written as `$fail eq "f"`', I noticed that the suggested version gives a warning if $fail is not defined (it's an optional parameter). Should I test and conditionally explicitly set it to "" if it's unset, and then use eq? 00:30
Zoffix kyan: I'd just set the default to an empty string: sub foo ($fail = "") {} 00:32
kyan: basically `===` is an object identity test, while `eq` tests stringy equality
kyan Zoffix: Ah, cool, thanks!
Zoffix kyan: will $fail only ever have one value? If so, I'd just have it as a bool named parameter
kyan (different copies of the same string are the same object?) 00:33
Zoffix m: sub foo (:$fail) { if $fail { say "has fail" } else { say "no fail" }; foo :fail; foo
camelia 5===SORRY!5=== Error while compiling <tmp>
Missing block
at <tmp>:1
------> 3} else { say "no fail" }; foo :fail; foo7⏏5<EOL>
expecting any of:
statement end
statement modifier
statement modifier loop
Zoffix m: sub foo (:$fail) { if $fail { say "has fail" } else { say "no fail" } }; foo :fail; foo
camelia has fail
no fail
kyan ...That's a good idea I hadn't thought of. The bash "everything is a string" philosophy is thoroughly embedded in me, it seems
Thanks! :)
Zoffix kyan: yeah, they are, but the intent of the test is less clear with `===` because all ideomatic code would use `eq` for stringy equality 00:34
m: say "foo" === "foobar".substr: 0, 3 00:35
camelia True
kyan cool.
Juerd Most of the time when you get 42, and you wanted "42", you just want to continue because it'll stringify as "42" just fine.
00:36 geekosaur left, pilne joined
kyan If I declare the param as Str $foo, it will get mad at the caller if it gives it a number, though, right? 00:36
Zoffix kyan: right 00:37
Juerd Yes, but you could declare it as Str() $foo, and it'll happily coerce things for you
Whether you'd want to do that depends on the use case. If you're just going to display the string, you probably don't really care about type safety. If you're going to build JSON from it, you might. 00:38
kyan Mm, I see. Cool, thanks!
00:43 geekosaur joined
AlexDaniel Juerd: That's OK, thanks anyway :) 00:43
00:43 troys is now known as troys_
AlexDaniel Juerd: FWIW I'd love to see you during the squashathon (in two weeks). Nowadays I don't have lots of time too, but squashathons are fun 00:47
Juerd: (pssst, there's free virtual pizza) 00:48
00:58 skids left, cognominal left 00:59 cognominal joined 01:02 mcmillhj left, skids joined 01:21 geekosaur left 01:29 geekosaur joined 01:35 itaipu joined 01:48 ilbot3 left 01:49 eliasr left 01:52 astj left 01:53 astj joined, rngoodn joined, astj left, astj joined 01:54 ilbot3 joined, ChanServ sets mode: +v ilbot3 02:00 astj left 02:01 astj joined 02:02 Cabanossi left 02:03 Cabanossi joined 02:04 astj left
skids .tell Zoffix fail.rakudo.party seems to list some "deleted" tickets, e.g. 131808 02:05
yoleaux skids: I'll pass your message to Zoffix.
02:06 astj joined 02:08 astj left 02:09 astj joined 02:11 astj left, astj joined
AlexDaniel .tell MasterDuke I think you were interested in this stuff: github.com/rakudo/rakudo/pull/1168 02:14
yoleaux AlexDaniel: I'll pass your message to MasterDuke.
02:29 AlexDani` joined 02:31 AlexDaniel left 02:32 astj left 02:33 zakharyas joined 02:43 cpage_ left, geekosaur left 02:44 cpage_ joined, geekosaur joined 02:45 rngoodn left
skids r-j: class A { has str $.x; method BUILD(:$!x) { } }; say A.new(:x<foo>).x 02:47
camelia Error while reading '/home/camelia/p6eval-token': No such file or directory at /home/camelia/rakudo-j-inst/bin/eval-client.pl line 10.
02:47 Cabanossi left 02:48 Cabanossi joined 02:57 noganex joined 03:00 noganex_ left
MasterDuke AlexDani`: thanks, i hadn't seen that 03:07
yoleaux 02:14Z <AlexDaniel> MasterDuke: I think you were interested in this stuff: github.com/rakudo/rakudo/pull/1168
03:07 AlexDani` is now known as AlexDaniel, rngoodn joined 03:08 skids left 03:13 itaipu left 03:14 piojo joined, piojo_ left 03:20 itaipu joined 03:22 pilne left 03:27 itaipu left 03:30 Cabanossi left 03:32 Cabanossi joined 03:34 astj joined 03:38 troys_ is now known as troys 03:48 tardisx joined, astj left 03:49 astj joined, astj left, astj joined 03:59 khw left 04:04 aborazmeh joined, BenGoldberg left 04:05 aborazmeh left, aborazmeh joined 04:10 rngoodn left 04:13 rngoodn joined 04:23 cdg joined 04:28 AlexDaniel left, cdg left 04:43 troys is now known as troys_ 04:44 astj left, astj joined 04:45 astj left, astj joined 04:46 Cabanossi left 04:47 astj left, Cabanossi joined 04:55 llfourn joined 04:57 mr-fooba_ joined 04:59 astj joined, mr-foobar left 05:00 troys_ is now known as troys 05:14 knobo joined 05:19 ufobat joined 05:24 vendethiel- joined 05:25 Ven`` left 05:35 kyan left 05:37 troys left 05:39 rngoodn left 05:43 wamba left 05:45 Cabanossi left 05:47 Cabanossi joined 05:53 cpage_ left 05:59 andrzejku left 06:02 domidumont joined 06:08 domidumont left, domidumont joined 06:15 nadim joined
grondilu m: for <foo>.Mix.pairs X <bar>.Mix.pairs -> ($a, $b) { say "$a, $b" } 06:18
camelia Too few positionals passed to ''; expected 2 arguments but got 0 in sub-signature
in block <unit> at <tmp> line 1
06:18 eliv joined
grondilu ^not sure what I'm doing wrong here 06:18
m: for <foo>.Mix.pairs X <bar>.Mix.pairs -> $a, $b { say "$a, $b" }
camelia Too few positionals passed; expected 2 arguments but got 1
in block <unit> at <tmp> line 1
06:18 eliv left 06:20 andrzejku joined 06:31 darutoko joined 06:35 wamba joined 06:39 cdg joined 06:41 eliv joined 06:42 eliv left
grondilu m: for :a X :b -> ($i, $j) { say "$i $j" } 06:43
camelia Too few positionals passed to ''; expected 2 arguments but got 0 in sub-signature
in block <unit> at <tmp> line 1
06:43 robertle left
grondilu m: for :a X :b -> $i, $j { say "$i $j" } 06:43
camelia Too few positionals passed; expected 2 arguments but got 1
in block <unit> at <tmp> line 1
grondilu wth
m: for <a> X <b> -> ($i, $j) { say "$i $j" } 06:44
camelia a b
06:44 cdg left
moritz I guess X doesn't expect named arguments, but rather positional ones 06:44
grondilu m: for (:a,) X (:b,) -> $i, $j { say "$i $j" }
camelia Too few positionals passed; expected 2 arguments but got 1
in block <unit> at <tmp> line 1
grondilu m: for (:a,) X (:b,) -> ($i, $j) { say "$i $j" }
camelia Too few positionals passed to ''; expected 2 arguments but got 0 in sub-signature
in block <unit> at <tmp> line 1
06:45 aborazmeh left
grondilu m: for [:a] X [:b] -> ($i, $j) { say "$i $j" } 06:45
camelia Too few positionals passed to ''; expected 2 arguments but got 0 in sub-signature
in block <unit> at <tmp> line 1
grondilu jeez
m: for $:a X $:b -> ($i, $j) { say "$i $j" } 06:46
camelia 5===SORRY!5=== Error while compiling <tmp>
Cannot use placeholder parameter $:a outside of a sub or block
at <tmp>:1
------> 3for $:a7⏏5 X $:b -> ($i, $j) { say "$i $j" }
grondilu meh
06:46 cpage_ joined 06:47 ken2 joined 06:48 eliv joined 06:51 araujo left 06:52 piojo left 06:53 araujo joined, araujo left, araujo joined 06:54 raschipi joined, nadim left 07:03 setty1 joined 07:12 llfourn_ joined 07:15 llfourn left 07:24 rindolf joined 07:33 ponbiki left 07:34 ponbiki joined 07:46 Cabanossi left 07:47 Cabanossi joined 07:56 Ven`` joined, vendethiel- left 08:01 imcsk8 left, imcsk8 joined 08:04 mithaldu_ left, leont joined, Praise left, jantore left, mithaldu_ joined, tony-o left 08:05 jantore joined, awwaiid left, wamba left, huf left, Praise joined, Praise left, Praise joined, wamba joined 08:06 mtj_ left, gfldex left, huf joined, gfldex joined, xinming left, tokik left, tony-o joined, awwaiid joined 08:07 ChoHag left, raschipi left 08:08 tokik joined, aindilis left, ChoHag joined, jcallen_ joined, jcallen left 08:11 aindilis joined 08:14 xinming joined 08:20 mtj_ joined 08:24 vendethiel- joined 08:25 Ven`` left, fluca1978 joined
fluca1978 I have a doubt about "is rw" class member access: class Foo { has $.bar is rw; } 08:26
tyil has `set()` from Bailador changed?
fluca1978 Foo.new.bar = 'Hello'; is this an lvalue method call or a variable access? 08:27
moritz Foo.new.bar returns a so-called "container"
and = assigns a value to it
which only works if method bar returns a writable container 08:28
(and container sits between variable (which is a syntactic element) and the value)
fluca1978 moritz: so it is like a container is a reference to the variable within the object instance, something alike? 08:29
moritz fluca1978: it's the other way round 08:30
a variable contains a reference to a container
08:30 leont left
moritz and when you assign to a variable, you update the value that's stored in the container 08:30
fluca1978 moritz: therefore "bar" points (allow me the term) to a container, and each time Foo::bar is called such container is returned. the variable therefore is a kind of refence itself, right? 08:31
moritz fluca1978: yes 08:32
(with the small caveat that the method isn't Foo::bar; that would be a subroutine)
ufobat tyil, nope, do you have any problems? 08:33
fluca1978 moritz: is this true only for classes variables or for any variable in Perl 6?
moritz fluca1978: for all variables 08:34
tyil ufobat: set() doesnt seem to actually set anything in my latest script, yet it works as expected in an older project of mine
if I `set("host", "127.1")` it still hosts on 0.0.0.0
fluca1978 moritz:thanks a lot 08:35
tyil ufobat: also, trying to set `config.host = "127.1"` results in undeclared routine 08:36
(yes, I have `use Bailador`) 08:37
ufobat [~/.workspace/p6/Bailador] - [Fr Sep 22, 10:37]
└─[$] <git:(dev*)> perl6 -Ilib -e 'use Bailador; set("host", "127.1.2.3"); dd config.host'
Str $!host = "127.1.2.3"
do you have a older version installed? i tested it against "dev" 08:38
i also tested it against 0.0.12
08:40 cdg joined, robertle joined
tyil I though I was on 0.0.12 too, zef upgrade Bailador seems to fail with "o such method 'dist' for invocant of type 'Any'. Did you mean any of these?" 08:42
ufobat --verbose, maybe? 08:43
08:44 cdg left
tyil that doesnt give more output than before 08:45
just some suggestions and a backtrace
ufobat does the backtrace help? i dont know what's wrong :( 08:46
tyil idk either
I'll try a zef install --force Bailador see if that helps anything 08:47
===> Fetching [FAIL]: Bailador:ver('0.0.12') from git://github.com/Bailador/Bailador.git
welp
ufobat i dont know how verions and github work, Bailador is also on cpan. on my box zef takes it from there 08:50
===> Found: Bailador:ver<0.0.12> [via Zef::Repository::Ecosystems<cpan>]
08:50 fluca1978 left
tyil ugh, I'll take a look at it again at a later date 08:58
09:13 lowbro joined, lowbro left, lowbro joined 09:25 piojo joined 09:28 gfldex_work joined 09:31 eliasr joined 09:32 gfldex_work left 09:44 wamba left 09:46 Cabanossi left 09:47 Cabanossi joined 09:52 dustinm` left 09:56 wamba joined 09:57 dustinm` joined 09:59 margeas joined 10:01 TEttinger left 10:07 zakharyas left, zakharyas joined 10:08 astj left 10:11 zakharyas left
Zoffix . 10:17
yoleaux 02:05Z <skids> Zoffix: fail.rakudo.party seems to list some "deleted" tickets, e.g. 131808
Zoffix yes it does
10:19 hkaimenas joined
Zoffix RT's API doesn't report deletions as status changes, so unless you fetch the thousand+ of tickets (slow as hell), you keep the deleted 10:19
(and yes, there's a way to toss deletions by, say, fetching the 1000+ tickets every 24 hours, but I haven't had the time to code that: github.com/zoffixznet/r6 ) 10:25
10:36 jonas1 joined
mspo did 09 come out? 10:40
rakudo.org has stalled in blog posts 10:41
oh maybe not 10:42
Zoffix mspo: yes, but compiler releases are only announced on the compiler mailing list. There's no .09 Rakudo Star 10:43
mspo oh there's a moar 09.1
Zoffix huggable: debs
huggable Zoffix, CentOS, Debian, Fedora and Ubuntu Rakudo packages: github.com/nxadm/rakudo-pkg/releases
Zoffix huggable: star 10:51
huggable Zoffix, Estimated Rakudo Star releases for 2017: .01, .04, .07 & .10
11:00 eliv left 11:01 itaipu joined 11:03 eliv joined 11:14 skids joined 11:18 hkaimenas left 11:20 skids left 11:21 wamba left 11:23 piojo left 11:26 rgrau joined 11:31 Cabanossi left 11:32 Cabanossi joined 11:34 cdg joined 11:39 cdg left 11:48 vendethiel- left, llfourn_ left 11:52 ShalokShalom_ joined 11:53 domidumont left 11:54 domidumont joined 11:55 ShalokShalom left 11:59 ken2 left 12:01 ken2 joined 12:07 eliv left, eliv joined 12:10 eliv left 12:11 eliv joined 12:12 piojo joined 12:24 rngoodn joined 12:25 wamba joined 12:30 Cabanossi left 12:32 Cabanossi joined 12:35 n0tjack joined
n0tjack We're having a mini idiom showcase in our language nerd channel at work ... what's the prettiest and most idiomatic way to express the volume of a N-ball in Perl6? That is, pi^(n/2) / gamma(n/2 - 1) where gamma is the generalized factorial? 12:37
DrForr Well, Gamma(1/2-n)=(pi**1/2)/((n**-1/2)*(n!) and of course Gamma(1/2)=pi**(1/2) 12:42
12:45 ken2 left, ken1 joined
TimToady see also rosettacode.org/wiki/Gamma_function#Perl_6 12:45
n0tjack oh, nice
just what I needed, thanks guys
TimToady: what's with the constants in that function? Just for faster convergence? Why not a closed form function? 12:48
TimToady rakudo doesn't always do things quite at the right time for that yet 12:50
12:51 eliv left, ken1 is now known as thunktone 12:52 reu left
n0tjack does p6 have a polynomial application function which takes a series of coefficients and produces the sum of the product of those coefficients with successively greater powers of a value, or should I roll my own with [+] and Z* ? 12:53
12:53 wamba left
DrForr Maybe polymod does what you're after? 12:54
TimToady that's the other direction 12:55
timotimo so polyproduct :)
perlpilot There's rosettacode.org/wiki/Horner%27s_rul...ion#Perl_6
TimToady m: say :10[4,3,2,1] 12:57
camelia 4321
TimToady maybe that, or the function under it
12:57 cdg joined
n0tjack whoa, [o] is cool 12:59
never thought of that
perlpilot Maybe someone needs to write a short-ish "Perl 6 for Mathematicians" or something 13:00
:-)
TimToady I guess the underlying function as the un-pretty name of &UNBASE_BRACKET
*has
13:02 skids joined, ShalokShalom_ is now known as ShalokShalom 13:04 cdg_ joined, HoboWithAShotgun joined
HoboWithAShotgun good afternoon you lovely people 13:05
Zoffix \o 13:06
n0tjack perlpilot: I think a Perl6 module for generating OEIS sequences, including recipes and idioms for common patterns is a worthwhile exercise 13:07
13:08 cdg left
n0tjack I may never get around to it 13:08
but it's a neat idea
13:09 geekosaur left
HoboWithAShotgun the docs say this would be nyi 13:10
m: my &s = sub (Int $a where * < 0) { say $a; }; &s(-1);
camelia -1
timotimo what part of it?
HoboWithAShotgun docs.perl6.org/type/Callable 13:11
the anonymous signature
or closure with signature
Zoffix HoboWithAShotgun: the one with `my &f:(Int)` parts?
HoboWithAShotgun yup
Zoffix m: my &s:(Int) = sub (Int{ say $a; }; &s(-1); 13:12
camelia 5===SORRY!5=== Error while compiling <tmp>
You can't adverb &s
at <tmp>:1
------> 3my &s:(Int)7⏏5 = sub (Int{ say $a; }; &s(-1);
13:12 mcmillhj joined
HoboWithAShotgun nevermind i overlooked the :(Int) part 13:12
:-/
13:16 mr-foobar joined 13:17 mr-fooba_ left
Geth doc: 3ecad59a15 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Type/IO/Socket/INET.pod6
Use constants for sock family
 13:17
13:19 geekosaur joined
Zoffix
.oO( 🎜 🎝 We—Are—Faaaamilyyyy )
 13:20
DrForr Hopefully the package name isn't ClothesDryer. 13:21
Zoffix huh, PF_INET isn't a constant, it's an enum, yet they're constants here: github.com/rakudo/rakudo/blob/nom/....pm#L3-L18 13:22
Zoffix bets there's a duplicate set of enums somewhere
Yup :( github.com/rakudo/rakudo/blob/nom/...pm#L11-L17 13:23
eater those are also wrong D:
13:27 gregf_ left
HoboWithAShotgun i just learned that say 2.&g.&f; and say f( g(3) ); is equivalent, but why? i mean the functions clearly are not methods of some Int 13:27
2 and 2, not 2 and 3 13:28
Zoffix HoboWithAShotgun: $foo.&g($bar) is equivalent to g($foo, $bar)
moritz it's just a syntactic construct to invoke a subroutine 13:29
Zoffix (more or less; there seem to be differences that mess with the Failures; there's a ticket for it)
HoboWithAShotgun ah, i see.
m: class foo { sub bar( $self ) { $self.baz }; sub baz { say "hi" } }; foo.new.bar; 13:30
camelia No such method 'bar' for invocant of type 'foo'
in block <unit> at <tmp> line 1
HoboWithAShotgun allright, but assuming the dot would resolve then this would work 13:31
Zoffix HoboWithAShotgun: subs are lexical by default. Also, you're missing the `&` 13:32
m: class foo { our sub bar( $self ) { $self.baz }; sub baz { say "hi" } }; foo.new.&bar;
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
bar used at line 1. Did you mean 'VAR', 'bag'?
Zoffix hmmm
oh
Well, not oh, but I see `&` is also missing on baz... Weird it's saying bar is undeclared tho 13:33
13:33 reu joined
Zoffix m: class foo { sub bar( $self ) is export { $self.&baz }; sub baz ($) { say "hi" } }; import foo; foo.new.&bar; 13:34
camelia hi
Zoffix hm, there's something I don't understand about symbols :/
HoboWithAShotgun dont bother. this was just to see wether i got that first argument thing right in my head
Zoffix I still want to know what I missing :)
m: class { our sub foo { 42.say } }; foo 13:35
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
foo used at line 1
Zoffix I thought `our` would make `foo` available where I'm calling it
m: class Foo { our sub foo { 42.say } }; Foo::foo 13:36
camelia 42
Zoffix There's the missing piece :D
moritz m: class { our sub foo { 42.say }.WHO
camelia 5===SORRY!5=== Error while compiling <tmp>
Missing block
at <tmp>:1
------> 3class { our sub foo { 42.say }.WHO7⏏5<EOL>
expecting any of:
statement end
statement modifier
statement modifier loop
moritz m: class { our sub foo { 42.say }}.WHO
camelia ( no output )
moritz m: class { our sub foo { 42.say }}.WHO<&foo>
Zoffix m: class Foo { our sub bar( $self ) { $self.&baz }; sub baz ($) { say "hi" } }; Foo.new.&Foo::bar;
camelia ( no output )
hi
moritz m: class { our sub foo { 42.say }}.WHO<&foo>()
camelia 42
Zoffix :)
moritz that's how you can call an our-sub from an anonymous package
Zoffix Neat 13:37
moritz though why anybody would put an our-sub into an anon class is anybody's guess :-)
HoboWithAShotgun nevertheless it is covered
which probably tells us something 13:38
moritz that jnthn++ designed the types and package table system really well? :-) 13:39
HoboWithAShotgun this pony tailed late hippy, yeah
i love his talks
13:41 mephinet joined, piojo left
HoboWithAShotgun when he talks, there is this hint of pride in his voice, yet he is too humble to show it :-) 13:41
fully i mean
13:42 mr-fooba_ joined, piojo joined 13:43 mr-foobar left
HoboWithAShotgun speaking of talks, did you guys watch damians "three little words"? 13:44
13:45 Cabanossi left 13:47 Cabanossi joined 13:59 HoboWithAShotgun left 14:01 thunktone left 14:03 thunktone joined 14:21 itaylor57 left, HoboWithAShotgun joined, itaylor57 joined 14:28 nekomune joined 14:30 geekosaur left 14:31 Cabanossi left, lowbro left 14:32 khw joined, wander4096 joined, Cabanossi joined
wander4096 can i just compile p6-code but not execute it? 14:35
like when works on java, it is separated into `javac` and `java`
i want to do `rakudo-c`, not the whole `perl6 ...` 14:36
14:37 geekosaur joined
DrForr You might be able to use --target= in order to just do the precompile. 14:40
I don't see any other options in the main perl6 binary, though there's also a -o that might aid in things. 14:41
wander4096 i try this command `perl6 main.p6 --target=jvm`
it just runs the code
14:42 piojo left
wander4096 no files look like bytecode file generated 14:42
DrForr Right, --target accepts a compilation stage, probably not the VM to compile against.
jnthn I'd have to be --target=jar --output=SomeFile.jar I guess
Though I don't know if it will produce something that will run directly 14:43
wander4096 it seems that -o option has no effect 14:44
14:44 piojo joined
wander4096 run `perl6 main.p6 -o=main` but file `main` wasn't generated 14:45
DrForr That may not be the intent of the -o option, I was simply reading the output of 'perl6 -h'. 14:46
14:46 HoboWithAShotgun left 15:01 Cabanossi left 15:02 Cabanossi joined, cdg_ left 15:03 Sgeo_ joined, cdg joined 15:05 Sgeo left 15:06 cdg_ joined
nekomune i'm having trouble figuring out how i can pass an array as an argument to a function expecting a NativeCall Pointer argument. 15:06
is there a way I can reference this array through a pointer? 15:07
15:08 cdg left 15:09 troys joined 15:12 wander4096 left 15:13 mcmillhj left
ugexe wander4096: its `perl6-j --target=jar --output=SomeFile.jar main.p6`. but you still need a perl6 entry point to run it i believe 15:14
before you were running your main.p6 script and passing the script itself the -o=main 15:17
Zoffix nekomune: don't know anything about NativeCall, but I see a section on Arrays in the docs. Perhaps that'll be of help: docs.perl6.org/language/nativecall#Arrays 15:37
15:37 mcmillhj joined
nekomune Zoffix: If I use NativeCall's CArray instead, I would still get something like: expected NativeCall::Types::Pointer but got Array ($[NativeCall::Types::...) 15:40
Zoffix No idea, sorry.
nekomune np, i'll still try to figure this out. 15:41
15:42 Aaronepower joined
nekomune I've seen docs.perl6.org/language/nativecall...f_Pointers as well, but it only goes into dealing with pointers returned from a native call. 15:42
there's even a helper function 'nativecast'
but not much about the other way around...
15:43 gregf_ joined
Aaronepower Hello, is there a way to use Grammars? Like call another grammar as part of a rule? 15:43
re use*
15:47 eliv joined 15:48 geekosaur left, cdg_ left 15:49 domidumont left, cdg_ joined, geekosaur joined, cdg_ left
perlpilot m: grammar A { token foo { "foo" } }; grammar B { token TOP { <A::foo> } }; say B.parse("foo"); 15:49
camelia ｢foo｣
A::foo => ｢foo｣
15:49 cdg_ joined 15:50 cdg__ joined, geekosaur left
Zoffix They're just classes. All the same benefits apply. You can even mix in roles 15:50
m: role A { token foo { "foo" } }; grammar B does A { token TOP { <foo> } }; say B.parse("foo"); 15:51
camelia ｢foo｣
foo => ｢foo｣
15:51 geekosaur joined
Aaronepower Sorry I think I wasn't clear. I don't want to use methods from another grammar I want to call the whole grammar. 15:52
m: grammar A { token foo { "foo" } }; grammar B { token TOP { <A> } }; say B.parse("foo");
camelia No such method 'A' for invocant of type 'B'
in regex TOP at <tmp> line 1
in block <unit> at <tmp> line 1
 15:53
15:53 jonas1 left, cdg_ left
Zoffix Aaronepower: just use the entry token instead of <A> (so, it'd be <A::TOP> for common case) 15:54
perlpilot m: grammar A { token TOP { "foo" } }; grammar B { token TOP { <A::TOP> } }; say B.parse("foo");
camelia ｢foo｣
A::TOP => ｢foo｣
Zoffix .parse just uses TOP as default; it's changeable 15:55
m: grammar B { token meows { "foo" } }; say B.parse: "foo", :rule<meows>;
camelia ｢foo｣
perlpilot Aaronepower: so ... why do you want to 'call the whole grammar' from within a rule of another grammar?
Aaronepower perlpilot: Because I want to call A on it's own as well. 15:56
15:59 astj joined, troys is now known as troys_ 16:00 Cabanossi left 16:01 geekosaur left, nadim joined, robertle left 16:02 Cabanossi joined 16:03 astj left 16:04 eliv left
Aaronepower Is there a way to concat hashs? Like `{foo => 1} + {bar => 2} = {foo => 1, bar => 2}`? 16:07
16:07 geekosaur joined
perlpilot Aaronepower: what do you want to happen with same-named keys ? 16:08
Aaronepower perlpilot: Can't happen in my case. 16:09
As in there will never be a conflict.
perlpilot m: my %h = foo => 24; my %g = bar => 42; say %h.push(%g); 16:10
camelia {bar => 42, foo => 24}
perlpilot m: my %h = foo => 24, bar => 87; my %g = bar => 42; say %h.push(%g);
camelia {bar => [87 42], foo => 24}
perlpilot (note what happens if there are dups)
16:12 roguelazer left 16:14 HoboWithAShotgun joined
perlpilot Aaronepower: I suppose I should have mentioned the simpler: %x = %h, %g; # :-) 16:14
16:14 roguelazer joined
perlpilot m: my %h = foo => 24, bar => 87; my %g = bar => 42; my %x = %h, %g; say %h 16:14
camelia {bar => 87, foo => 24}
perlpilot m: my %h = foo => 24, bar => 87; my %g = bar => 42; my %x = %h, %g; say %x
camelia {bar => 42, foo => 24}
hythm Hello, I'm trying to use IoC module in another module, but I get Undeclared routine when I' try to use IoC routine. Here is a gist, What I'm doing wrong? gist.github.com/hythm7/aabc37aa540...834380e86f 16:16
16:17 troys_ is now known as troys
perlpilot hythm: A::IoC doesn't automatically export routines from IoC; 16:19
Aaronepower Zoffix: How would I call the Actions object on that grammar?
hythm perlpilot: is there a way I can tell A::I0C to export them, or it has to be exported in IoC module at first? 16:20
s/0/o/
perlpilot hythm: there's docs.perl6.org/language/modules#in...sub_EXPORT 16:21
16:21 piojo left, Sgeo_ left 16:22 jameslenz joined
hythm Thanks perlpilot, checking that 16:22
perlpilot hythm: you'll probably have to change lib-A-IoC.pm to no longer use `unit`. 16:27
16:28 Aaronepower left, Aaronepower joined
perlpilot hythm: something like: use IoC; sub EXPORT { %(IoC::EXPORT::DEFAULT::); } module A::IoC { sub setup-ioc is export {} } 16:28
perlpilot is having a sloooow day today
16:29 thunktone left 16:30 thunktone joined
Aaronepower How do you call The Actions object on Grammar inside another grammar? `grammar Foo { token TOP { foo } } class FooActions { method TOP($/) { 1 } } grammar Bar {token TOP { <Foo::TOP> } } class BarActions { method TOP { make $<Foo::TOP>.made } }` 16:31
hythm perlpilot: That worked great, thanks.
16:31 domidumont joined 16:32 ufobat left
grondilu m: say Inf ~~ Num; say Inf ~~ Int 16:33
camelia True
False
grondilu :/
Aaronepower How could have it so `Bar.parse("foo", actions => BarActions).made` returns 1?
grondilu: Inf is a Float no?
grondilu wanted to extend an integer index to infinity with Inf, but learnt that Inf is not an Int
Aaronepower: apparently
16:34 astronavt joined
grondilu I suppose it's the IEEE-754 (IIRC) floating point value or something 16:34
Aaronepower grondilu: An Int can only go up to 2147483647.
grondilu I don't think so 16:35
no in Perl 6
*not
an Int is not an int
Zoffix Inf is a Num
Yeah, we follow IEEE 754-2008 (with some parts yet to be stuffed into 6.d language release)
perlpilot Aaronepower: I don't think that you do. At least not without calling Foo.parse(…) You'd have to have the Bar grammar slurp up enough of its input to feed to Foo though.
Zoffix m: say 0/0 16:36
camelia Attempt to divide by zero using div
in block <unit> at <tmp> line 1
grondilu m: say 1/0
camelia Attempt to divide 1 by zero using div
in block <unit> at <tmp> line 1
Zoffix m: .Num.say for 1/0, -1/0, 0/0
camelia Inf
-Inf
NaN
Zoffix You need Num view to get IEEE semantics
Aaronepower perlpilot: Is there no way to pass $<Foo::TOP> to FooActions?
grondilu Kind of annoying I can't have a version of Inf that is Int :/ 16:37
Aaronepower perlpilot: Like in BarActions
grondilu it would have been elegant in my code
16:38 mcmillhj left
grondilu m: say Inf.Int 16:38
camelia Cannot coerce Inf to an Int
in block <unit> at <tmp> line 1
grondilu :P
Aaronepower grondilu: It wouldn't really make sense though, as Int isn't supposed to represent entire spectrum of numbers. 16:39
Zoffix Yeah. There might be something borne out of it in 6.* release, but last we thought of making Inf an Int and a Num it sounded like a bad idea
grondilu Aaronepower: it could though, couldn't it?
Zoffix There are lots of places where an arg could be Int number OR an * (Inf) and just because of that extra condition you can't use a native int (well, without multi dispatch) 16:40
Well, not can't, but it's extra hassle and extra lost CPU cycles
Aaronepower grondilu: In the same way that an Int could hold floating point numbers. Int implies a lot about how the number works and how to add and sub from it. 16:41
grondilu I suppose I may create a subset of Real 16:42
m: subset Intish of Real where { $_ ~~ Int or $_ ~~ Inf }
camelia ( no output )
perlpilot Aaronepower: If you were using roles, you could compose the Foo role into the Bar grammar and FooActions role into the BarActions class.
grondilu m: subset Intish of Real where { $_ ~~ Int or $_ ~~ Inf }; say Inf ~~ Intish
camelia True
grondilu is not too happy about the name "Intish" though 16:43
perlpilot Aaronepower: and to use Foo stand-alone you'd have to make a FooOnly grammar and a FooActionsOnly class that are empty except for composing the roles
thunktone Surely you could call it "Integer" if you want? 16:44
Aaronepower perlpilot: But Is there no way of saying something like `FooActions.new($<Foo::TOP>)` in the BarActions method?
grondilu m: say Inf ~~ Real # just checking
camelia True
grondilu thunktone: yeah, "Integer" should do
Zoffix m: say Int.Range 16:45
camelia -Inf^..^Inf
16:46 robertle joined
Zoffix m: say int.Range 16:46
camelia -9223372036854775808..9223372036854775807
thunktone So Int goes up to infinity minus one? 16:47
perlpilot Aaronepower: I can't think of one, but my brain power isn't very reliable today for some reason :)
16:47 AlexDaniel joined
Zoffix thunktone: no, it goes up to infinity without including infinity itself 16:48
m: say Inf - 1; # "minus one" is still infinity
16:48 AlexDaniel left
camelia Inf 16:48
Zoffix m: say eager 1.3..^2.5; # the ^ means exclude the endpoint, not minus one, even for non-Inf 16:49
camelia (1.3 2.3)
Zoffix You know what would be a good use for that .Bridge method? It takes an arg and returns appropriate matching arg 16:50
m: class InfI is Num is Int { method Bridge { 42 } }; say 0e0 + InfI
camelia 42
Zoffix m: class InfI is Num is Int { method Bridge { 42 } }; say 0 + InfI
camelia 42
Zoffix Like here, it'd receive 0 or 0e0 and return appropriate value
16:50 AlexDaniel joined 16:52 lichtkind joined
grondilu m: subset Int∞ of Real where { $_ ~~ Int or $_ ~~ Inf } 16:52
camelia 5===SORRY!5=== Error while compiling <tmp>
Bogus postfix
at <tmp>:1
------> 3subset Int7⏏5∞ of Real where { $_ ~~ Int or $_ ~~ Inf
expecting any of:
infix
infix stopper
postfix
statement end…
grondilu thought Int∞ would be an allowed identifier for some reason
Zoffix m: sub infix:<+> { &CORE::infix:<+>($^a, $^b.Bridge: $a) }; class InfI is Num is Int { method Bridge($v) { $v ~~ Int ?? int.Range.max !! Inf } }; say 0e0 + InfI 16:53
camelia Inf
Zoffix m: sub infix:<+> { &CORE::infix:<+>($^a, $^b.Bridge: $a) }; class InfI is Num is Int { method Bridge($v) { $v ~~ Int ?? int.Range.max !! Inf } }; say 0 + InfI
camelia 9223372036854775807
Zoffix Like that
∞ is not a valid identifier
u: ∞
unicodable6 Zoffix, U+221E INFINITY [Sm] (∞)
Zoffix Need a Lu Ll Nd - or ' 16:54
grondilu ol
*ok
wasn't it used as an alias for Inf, though?
m: say ∞
camelia Inf
Zoffix Yeah, it's just converted to Inf by the grammar. It's not an idenfier-based term 16:55
grondilu ok
wasn't there a junction-style form for types? 16:56
Zoffix m: my \term:<Int∞> = subset :: of Real where { $_ ~~ Int or $_ ~~ Inf }; say ∞ ~~ Int∞; say 42 ~~ Int∞; say 0e0 ~~ Int∞
camelia True
True
False
grondilu m: say 13 ~~ Int|Inf
camelia True
grondilu m: say Inf ~~ Int|Inf
camelia True
grondilu m: say pi ~~ Int|Inf
camelia False
Zoffix m: my \term:<Int∞> = subset :: of Real where Int|Inf; say ∞ ~~ Int∞; say 42 ~~ Int∞; say 0e0 ~~ Int∞
camelia True
True
False
grondilu m: my Int|Inf $
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed my
at <tmp>:1
------> 3my Int7⏏5|Inf $
Zoffix m: my $ where Int|Inf 16:57
camelia ( no output )
grondilu that could do
16:57 AlexDani` joined
grondilu that's actually pretty much as simple as I could have hoped 16:58
16:58 mcmillhj joined 16:59 AlexDaniel left
grondilu m: role Foo[Int $sign where -1|1] {}; class :: does Foo[1] {}.new 17:04
camelia ( no output )
grondilu m: role Foo[Int $sign where -1|1] {}; say class :: does Foo[1] {}.new 17:05
camelia <anon|79109168>.new
grondilu hum, can't reproduce an error here
m: role Foo[Int $sign where -1|1] {}; say class :: does Foo[pi] {}.new
camelia 5===SORRY!5=== Error while compiling <tmp>
No appropriate parametric role variant available for 'Foo'
at <tmp>:1
grondilu nvm 17:06
17:07 AlexDani` left 17:11 leont joined
Zoffix m: my \term:<Int∞> = subset :: of Real where { $_ ~~ Int or $_ ~~ Inf }; sub (Int∞) {}(42) # can't actually use a term as a type constraint 17:11
camelia 5===SORRY!5=== Error while compiling <tmp>
A parameter may only have one prefix type constraint
at <tmp>:1
------> 3re { $_ ~~ Int or $_ ~~ Inf }; sub (Int∞7⏏5) {}(42) # can't actually use a term as
Zoffix m: my \term:<Int∞> = * ~~ Inf|Int; say ∞ ~~ Int∞; say 42 ~~ Int∞; say 0e0 ~~ Int∞ # and for this usecase it could well be just a Callabler 17:12
camelia True
True
False
Aaronepower What's the perl way to convert a hash to named arguments? 17:22
perl6*
jnthn |
Zoffix m: my %h = :42a, :70b; { dd [ $:a, $:b ] }( |%h ) 17:23
camelia [42, 70]
17:24 dogbert11 joined
Zoffix m: grammar A { token TOP { "A" } }; class A-A { method TOP($/) { make 42 } }; class B-B { method TOP($_) { .make: "Meow {.<A::TOP>.made}" } }; grammar B { token TOP { {$/.set_actions: A-A } <A::TOP> { $/.set_actions: B-B } "B" } }.parse(:actions(B-B), "AB").made.say 17:27
camelia Meow 42
Zoffix Aaronepower: ^ .set_actions is not part of the language, but the above seems to do what you want
Aaronepower Zoffix: thanks! 17:28
17:29 troys is now known as troys_ 17:30 a3r0 left 17:32 wamba joined 17:33 aindilis left 17:34 aindilis joined 17:36 Sgeo joined
HoboWithAShotgun m: my $s = supply { for 1 .. 3 { sleep(2); emit($_); } }; react { whenever $s -> $x { say $x; } } 17:36
camelia 1
2
3
HoboWithAShotgun also on the shell this has no output for 6 seconds, then dumps all values at once 17:37
why isn't the output coming in two second intervals? 17:39
hey jnthn++ i just watched your talk about parallelism and concurreny 17:40
17:40 a3r0 joined
timotimo HoboWithAShotgun: the whenever only starts once the supply block has finished running 17:42
Aaronepower How do you get a value from a hash with a variable? 17:43
Zoffix m: my %h = :42a; my $v = 'a'; say %h{$v}
camelia 42
Aaronepower Zoffix: cheers!
Zoffix %h<foo> is just a shortcut for %h{'foo'}
%h<foo> is just a shortcut for %h{<foo>} # more accurately, since it does allomorphs 17:44
timotimo and %h<foo bar> is short for %h{'foo', 'bar'}, which is also important
(but including the allomorph thing)
Zoffix m: my %h := :{ <42> => "allomorph", 42 => "Int", "42" => "Str" }; say %h<42>; say %h{42}; say %h{"42"}; say %h{<42>} 17:45
camelia allomorph
Int
Str
allomorph
Aaronepower Is there a way to refer an outer `$_` from an inner for loop? 17:47
Zoffix (don't matter much for regular hashes, I suppose; only for Setties/Baggies/Object hashes)
Aaronepower: rewrite to not need it :) Also, $OUTER::_ or something
m: with 42 { say $OUTER::_ for ^3 } 17:48
camelia 42
42
42
Zoffix m: with 42 { say "$_:$OUTER::_" for ^3 }
camelia 0:42
1:42
2:42
Zoffix m: with 42 { say "$_:$^hehehe" for ^3 } 17:49
camelia 0:42
1:42
2:42
Zoffix ^_^
Aaronepower Zoffix: cheers 17:52
17:53 cdg joined 17:54 leont left 17:55 andrzejku_ joined
thunktone Is there any way to add to the end of the autogenerated "Usage" message without erasing the generated part? 17:56
I can't seem to access $?USAGE at all
Zoffix I think it's NYI 17:57
17:57 cdg__ left
Zoffix hm, actually I see it being set.... hm 17:58
17:58 cdg left
thunktone It is producing output, I just can't access it within the program. 17:58
Zoffix m: use nqp; sub MAIN ($a) { }; sub USAGE { say nqp::getlexcaller(q|$?USAGE|) ~ "moaaaar stuff" } 18:00
camelia Usage:
<tmp> <a> moaaaar stuff
18:00 leont joined
Zoffix thunktone: you could use this in the meantime, though `nqp` is not standard language or supported. I'll poke at it right now and see to it that $?USAGE is available from userland 18:00
18:00 Cabanossi left
thunktone thanks 18:01
18:02 Cabanossi joined 18:06 a3r0 left 18:07 leont left 18:10 leont joined
HoboWithAShotgun how do i get out of a whenever Supply.interval(1) -> $v { } block, so the reacting block returns and the program continues 18:15
i can die, but that can't be it, right?
jnthn done 18:16
yoleaux 17:26Z <dogbert11> jnthn: wrt the failing syntax.t test, got this after several attempts: gist.github.com/dogbert17/d018b0f6...0f734d0dbc
jnthn m: react whenever Supply.interval(0.2) { .say; done if $++ == 5 } 18:17
camelia 0
1
2
3
4
5
jnthn m: react whenever Supply.interval(0.2) { .say; done if $++ == 5 }; say "afterwards"
camelia 0
1
2
3
4
5
afterwards
HoboWithAShotgun yay :)
18:19 leont left
Zoffix So what are "compile time variables"? 18:21
Or more to the point: why is $?USAGE a compile time variable
m: use nqp; sub MAIN ($a) { }; sub USAGE { BEGIN say nqp::getlexcaller(q|$?USAGE|) ~ "moaaaar stuff" } 18:22
camelia Use of uninitialized value of type Any in string context.
Methods .^name, .perl, .gist, or .say can be used to stringify it to something meaningful.
moaaaar stuff
in code at <tmp> line 1
Zoffix I "fixed" it, but it's not compile time and I don't get why we need to pay for generation of usage messages if we don't even know if we'll ever need 'em
18:24 a3r0 joined
Zoffix moves this to #perl6-dev 18:24
18:25 espadrine joined 18:38 nightfrog left, Rawriful joined
Zoffix thunktone: it might take awhile to fix it actually. Some design issues are blocking it currently: github.com/rakudo/rakudo/commit/bbedf18d3c 18:39
18:41 thunktone left, domidumont left 18:43 thunktone joined 18:45 Cabanossi left 18:47 Cabanossi joined 18:48 AlexDaniel joined 19:02 mcmillhj left
weabot is there a way to import a file from the path as with #include "path/to/header.h" in C? 19:06
to just import existing subroutines from it
19:06 mcmillhj joined
Zoffix weabot: try: EVALFILE 'path/to/header.h' 19:07
weabot: nevermind, it runs in its own "scope", so not like #include 19:08
weabot that only returns a single structure from the file doesn't it?
Zoffix As far as I understand, it's like 'path/to/header.h'.IO.slurp.EVAL
weabot hmm 19:09
Zoffix But I just tried it with `sub foo {}` in the file and it doesn't become available in the program
m: 'sub foo { say "hi" }'.EVAL; foo 19:10
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
foo used at line 1
Zoffix m: 'our sub foo { say "hi" }'.EVAL; foo
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
foo used at line 1
19:10 astj joined 19:14 astj left 19:16 pecastro left 19:26 darutoko left 19:27 pecastro joined 19:31 grondilu left 19:39 krunen left, krunen_ is now known as krunen 19:46 TEttinger joined 19:52 Aaronepower left 19:56 eliasr left, astronavt left 20:09 cdg joined 20:10 mcmillhj left 20:17 LanceW joined 20:22 mcmillhj joined 20:23 AlexDaniel left 20:27 mcmillhj left 20:32 andrzejku_ left, lichtkind left 20:35 pecastro left 20:39 mcmillhj joined 20:41 Ven`` joined, pecastro joined 20:42 itaipu left 20:43 mcmillhj left 20:51 LanceW left 20:52 |oLa| left, mcmillhj joined 20:57 mcmillhj left 21:03 Rawriful left 21:05 Rawriful joined 21:07 troys_ is now known as troys 21:09 skids left 21:18 setty1 left, itaipu joined 21:29 perlpilot left 21:30 rgrau left 21:45 Cabanossi left 21:47 Cabanossi joined 21:55 astj joined 21:59 astj left 22:03 zoli__ joined
zoli__ (1..5).say 22:04
(1..5).join
geekosaur "m: "
zoli__ wondering why .say (or .perl) returns 1..5, but join (and reduce...) returns what I expect (12345) 22:05
geekosaur m: (1..5).WHAT.say 22:06
camelia (Range)
zoli__ ah, so it is a range, but just not expanded; thanks
geekosaur join and reduce operate on lists/arrays, so trigger conversion
Zoffix m: say eager 1..5 22:07
camelia (1 2 3 4 5)
zoli__ just starting with perl6; I'll have to use WHAT more
geekosaur m: (1..5).list.say
camelia (1 2 3 4 5)
Zoffix zoli__: use .^name instead. .WHAT returns a type object, while .^name returns the string with the actual name
zoli__ m: (1..5).name 22:08
camelia No such method 'name' for invocant of type 'Range'. Did you mean any of these?
none
note
race
take

in block <unit> at <tmp> line 1
Zoffix It's .^name not .name
zoli__ m: (1..5).^name
camelia ( no output )
zoli__ m: (1..5).^name.say
camelia Range
zoli__ cool, thanks Zoffix and geekosaur
Zoffix $foo.^bar is short for $foo.HOW.bar($foo); it's a method call on the object's metaobject 22:09
gfldex m: LABEL: for 1..3 {}; say LABEL.WHAT
camelia (Label)
gfldex m: LABEL: for 1..3 {}; say &LABEL.WHAT
camelia Nil
gfldex o.0
what the heck is an &-sigiled Label and why is it Nil? 22:10
Zoffix m: constant foo = 42; say &foo 22:11
camelia Nil
Zoffix mc: constant foo = 42; say &foo
committable6 Zoffix, ¦2015.12: «Nil»
Zoffix Ah, `foo` with a `&` coercer
m: say &42
camelia Nil
Zoffix m: say &&&&&&&&&&&&&&&&&&&&&&&&&&@@@@@@@@@@@@@@$$$$$$$$$$$$$$$$$$$$$$$$$$%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 22:13
camelia ()
Zoffix m: constant $z = 42; say &$z
camelia 42
Zoffix m: constant z = 42; say &z 22:14
camelia Nil
Zoffix inconsistent
m: constant z = 42; say &(z)
camelia 42
22:15 troys is now known as troys_ 22:16 zoli__ left
Zoffix Filed as rt.perl.org/Ticket/Display.html?id=132146 22:18
22:18 itaipu left 22:19 espadrine left
Geth doc: 1382193d15 | (Will "Coke" Coleda)++ | doc/Language/5to6-nutshell.pod6
'Taint there.
 22:25
doc: e5f5dd96bd | (Will "Coke" Coleda)++ | 2 files
whitespace
doc: 2ac4c00de3 | (Will "Coke" Coleda)++ | 2 files
Fix code samples
22:26 Aaronepower joined, troys_ is now known as troys 22:30 mcmillhj joined 22:31 Cabanossi left
Aaronepower How would I with a Grammar get everything up to two certain words preserving those words to be parsed by the next rules? Here's what I tried so far: "Hello World BREAK Foo Bar STOP" code: `grammar Foo { rule TOP { <text> <foo> <bar> } rule foo { 'BREAK' <text> } rule bar { 'STOP' <text> } rule text { .+ <?before 'BREAK'> | <?before 'STOP'>} }` 22:32
22:32 Cabanossi joined 22:34 mcmillhj left 22:36 nightfrog joined, vike left 22:37 cdg left 22:38 vike joined, vike left, cdg joined 22:44 skids joined
timotimo Aaronepower: are you intending for the .+ in the rule text to only be allowed before a "BREAK"? 22:45
because the | puts the .+ and <?before BREAK> on one side of the alteration and only <?before STOP> on the othe
22:45 mcmillhj joined 22:46 AlexDaniel joined
Aaronepower timotimo: No I just want all the items before "BREAK", and allow for BREAK to be used by the grammar for rules after this one. 22:46
22:47 thunktone left 22:49 HoboWithAShotgun left 22:50 mcmillhj left
timotimo right, that should work 22:51
i'd expect you can also use )> before the BREAK | STOP part
will still have to put [ ] around the break/stop part, though 22:52
[Coke] Readline erroring out trying to install on mac, using a readline from macports.
ah, github.com/drforr/perl6-readline/issues/12 22:54
22:55 rindolf left
Aaronepower timotimo: I'm sorry I don't understand. I just that code doesn't work paste.rs/dAF You're saying it is? 22:57
just said*
timotimo: Here's the updated version. paste.rs/spX 22:59
timotimo well, that can't work
it expects a text after STOP
so there has to be more text and another BREAK or STOP there
23:00 robertle left
Aaronepower timotimo: paste.rs/WTc 23:00
timotimo that looks more correct
what i generally find is whitespace stuff is difficult
that often screws me over when i use "rule" 23:01
23:01 vike joined
Aaronepower So what I'm expecting is essentially three <text> rules like `[Hello World]` `[Foo Bar]` `[Real Talk]` 23:01
timotimo i'll grab the code and try it out locally
23:01 mcmillhj joined
timotimo oh 23:02
you have another problem
the text rule matches everything until the last BREAK 23:03
then not problematic, since rule does do backtracking
making it .+? instead of .+ makes it parse for me
Aaronepower timotimo: Can you show me that code as just changing it from `.+` to `.+> 23:04
timotimo hold on, something's not right
Aaronepower `.+?` did not work.
timotimo yes, bear with me for another moment 23:06
23:07 mcmillhj left
timotimo ah, yes 23:08
you're not consuming the last BREAK or STOP
so you successfully match, but you don't match up until the end of the source text
.parse will consider that a failure
if you use .subparse instead, that won't force beginning-to-end match, and thus will succeed
23:09 cdg left
Aaronepower timotimo: Nope, changing it to subparse just gives me `#<failed match>` 23:09
timotimo m: paste.rs/J4Y
camelia 5===SORRY!5=== Error while compiling <tmp>
Confused
at <tmp>:1
------> 3https:7⏏5//paste.rs/J4Y
expecting any of:
colon pair
timotimo aaw
anyway, my working code is at that address
Aaronepower timotimo: Is there no way to have that functionality with parse? 23:11
timotimo sure
just gotta eat up the final BREAK or STOP in the TOP rule
Aaronepower timotimo: So `<text> BREAK <foo> STOP <bar>`? 23:12
timotimo no, more like <text> <foo> <bar> [ BREAK | STOP ]
if you want it to accept strings that don't end in break or stop, you have to also accept $ as an alternative to BREAK or STOP in the text rule 23:13
though if your input data is really this simple, can i interest you in the awesome and versatile tool that is .split?
m: "Hello World BREAK Foo Bar STOP Real Talk".split(["BREAK", "STOP"], :k).perl.say 23:14
camelia ("Hello World ", 0, " Foo Bar ", 1, " Real Talk").Seq
timotimo m: "Hello World BREAK Foo Bar STOP Real Talk".split(["BREAK", "STOP"], :v).perl.say
camelia ("Hello World ", "BREAK", " Foo Bar ", "STOP", " Real Talk").Seq
Aaronepower timotimo: Hang on. You added a `BREAK` to the end. That isn't how my input will look. It won't have BREAK at the end.
timotimo ok, i'll put the necessary modification in and paste agai
npaste.rs/6ES
oops
paste.rs/6ES 23:15
Aaronepower timotimo: Okay, thank you for your time.
timotimo sure 23:19
i don't grammar terribly often, so it was a good opportunity to freshen up
23:25 mcmillhj joined 23:29 mcmillhj left 23:37 troys is now known as troys_ 23:38 wamba left 23:46 Cabanossi left 23:47 Cabanossi joined 23:54 BenGoldberg joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!