| 25 Jul 2024 | |||
| greenfork | Basically I want to do something like this Surface::Fire.draw | 11:24 | |
| I think this is the closest code paste.sr.ht/~greenfork/f10f0f0670f...cf3b6b3e76 | 11:25 | ||
| I also posted on stackoverflow this question, will answer it myself after figuring out the optimal solution I think stackoverflow.com/questions/787930...or-an-enum | |||
| lizmat | that works, so what's wrong with it? | 11:26 | |
| greenfork | Nothing wrong, I just don't like the way it is written | 11:27 | |
| My other classes are like class A { method draw { ... } }, so they are all called like $a.draw, I wanted to have a similar interface for the enum | 11:28 | ||
| But I think that it requires wrapping it in a class | |||
| [Coke] | you can pun a class into that role. | 12:57 | |
| m: role DrawSurface { method draw($a) { say $a } }; my $b = DrawSurface.new; $b.a("hi") | 12:58 | ||
| camelia | No such method 'a' for invocant of type 'DrawSurface' in block <unit> at <tmp> line 1 |
||
| [Coke] | m: role DrawSurface { method draw($a) { say $a } }; my $b = DrawSurface.new; $b.draw("hi") | 12:59 | |
| camelia | hi | ||
| [Coke] | so if you want an instance var, you can still make one, even if it's a role not a class. | ||
| antononcube | I think we can also make a class that does the role Enumeration? (I am not in front a computer, I am just reading the enum documentation on my iPad.) | 13:05 | |
| [Coke] | Yes, you can do it manually too of course. | 13:41 | |
| m: role DrawSurface { method draw($a) { say $a } }; class C does DrawSurface {}; my $b = C.new; $b.draw("hi") | 13:42 | ||
| camelia | hi | ||
| [Coke] | you basically get an anonymous C in the first example. | ||
| greenfork: hope that helps. | 13:54 | ||
| greenfork | [Coke]: this is interesting, thank you for tagggin me! | 13:56 | |
| xinming | Hi, Is it possible that we use PSGI in Cro? | 18:57 | |
| jdv | why? | 19:00 | |
| it was talked about a bunch years ago, psgi and raku that is. | 19:01 | ||
| what is justification? | |||
| antononcube | Is "PSGI" in Raku going to be called "RSGI"? | 19:02 | |
| jdv | afaik, the general psgi/raku thing went nowhere | 19:03 | |
| xinming | Ok, I just check my old test utility written in perl, I have TODO to say upgrade to PSGI, which reminds me about PSGI thing in raku support. :-) | 19:40 | |
| antononcube | @ximming Do it, ximming! Do it! | 20:08 | |
| 26 Jul 2024 | |||
| patrickb | I seem to recall there was a (now unmaintained PSGI stack for Raku. | 05:13 | |
| antononcube | Maybe someone should do a design or motivational document for "Raku SGI" similar to this one: metacpan.org/dist/PSGI/view/PSGI/FAQ.pod | 12:41 | |
| ugexe | raku.land/cpan:HANENKAMP/Smack was an experiment with something more raku-specific | 12:47 | |
| antononcube | 👍 | 12:55 | |
| Xliff_ | \o | 13:55 | |
| Good morning! I'm trying to port some Javascript that wants to use a Map with object keys. I thought this would work, but it doesn't... | |||
| m: class A { }; my $c = Hash[A, Mu].new; my $aa := A.new; my $bb = A.new; $c{$aa} = 1; $c{$bb} = 2; $c.gist.say; $c.keys.map( *.^name ).gist.say | |||
| camelia | Type check failed in assignment to ; expected A but got Int (1) in block <unit> at <tmp> line 1 |
||
| Xliff_ | m: class A { }; my $c = Hash[A, Mu].new; my $aa = A.new; my $bb = A.new; $c{$aa} = 1; $c{$bb} = 2; $c.gist.say; $c.keys.map( *.^name ).gist.say | ||
| camelia | Type check failed in assignment to ; expected A but got Int (1) in block <unit> at <tmp> line 1 |
||
| Xliff_ | I'm a little concerned about the "assignment to ;". Can someone help? | 13:56 | |
| Odd... this does! | 13:58 | ||
| m: class A { has $.a }; my $c = Hash[Mu, Mu].new; my $aa := A.new( a => 42 ); my $bb = A.new( a => 23 ); $c{$aa} = 1; $c{$bb} = 2; $c.gist.say; $c.keys.map( *.a ).gist.say | |||
| camelia | {A.new(a => 42) => 1, A.new(a => 23) => 2} (23 42) |
||
| ugexe | m: class A { }; my %c{A}; my $aa := A.new; my $bb = A.new; %c{$aa} = 1; %c{$bb} = 2; %c.gist.say; %c.keys.map( *.^name ).gist.say | 16:27 | |
| camelia | {A.new => 2, A.new => 1} (A A) |
||
| ugexe | m: class A { }; my $c = Hash[Mu, A].new; my $aa := A.new; my $bb = A.new; $c{$aa} = 1; $c{$bb} = 2; $c.gist.say; $c.keys.map( *.^name ).gist.say; $c{1} = 1 | 16:28 | |
| camelia | {A.new => 2, A.new => 1} Type check failed in binding to parameter 'key'; expected A but got Int (1) in block <unit> at <tmp> line 1 (A A) |
||
| ugexe | m: raku -e 'class A { }; my $c = Hash[Mu, A].new; my $aa := A.new; my $bb = A.new; $c{$aa} = 1; $c{$bb} = 2; $c.gist.say; $c.keys.map( *.^name ).gist.say | 16:29 | |
| camelia | ===SORRY!=== Error while compiling <tmp> Two terms in a row at <tmp>:1 ------> raku -e⏏ 'class A { }; my $c = Hash[Mu, A].new; expecting any of: infix infix stopper postfix statement … |
||
| ugexe | m: class A { }; my $c = Hash[Mu, A].new; my $aa := A.new; my $bb = A.new; $c{$aa} = 1; $c{$bb} = 2; $c.gist.say; $c.keys.map( *.^name ).gist.say | ||
| camelia | {A.new => 2, A.new => 1} (A A) |
||
| ugexe | anyway you can do `my Hash %c{A}` or you need to reverse the parameters to Hash i.e. Hash[Mu, A] instead of Hash[A, Mu] | ||
| Xliff_ | ugexe: Thanks... so it has to be '$' sigil'd and Hash[<val>, <key>] ? | 17:11 | |
| ugexe | my %c{A}; %c{$a} = 1; or my $c = Hash[valType, keyType].new | 18:24 | |