How to remove a multi method in Raku - raku

How can I augment a class with a multi-method that will be called before the already defined one ?
I am trying to enable negative subscript: #arr[-1] like in this article but without changing the source.
So I augment Array with:
augment class Array {
proto method AT-POS(Array:D: Int:D $i where <0 ) {
say "AT-POS called";
my $pos = -1;
my $ix = $pos + self.elems;
return self.AT-POS($ix);
}
};
But as stated in the doc
Please note that adding a multi candidate that differs only
in its named parameters will add that candidate behind the already defined one
and as such it won't be picked by the dispatcher.
So my multi is never called:
say .signature for #arr.^method_table{'AT-POS'}.candidates ;
(Any:U \SELF: int \pos, *%_)
(Any:U \SELF: Int:D \pos, *%_)
(Any:U: Num:D \pos, *%_)
(Any:U: Any:D \pos, *%_)
(Any:D: int \pos, *%_)
(Any:D: Int:D \pos, *%_)
(Any:D: Num:D \pos, *%_)
(Any:D: Any:D \pos, *%_)
($: Any:U \pos, *%_)
(Any:D: \one, \two, *%_)
(Any:D: \one, \two, \three, *%_)
(Any:D: **#indices, *%_)
(List:D: int $pos, *%_)
(List:D: Int:D $pos, *%_)
(Array:D: int $pos, *%_)
(Array:D: Int:D $pos, *%_) # Their
(Array: $a, *%_)
(Array:D: Int:D $pos, *%_) # My
I want my method to be called before their. How can can I modify the dispatcher?

Named parameters don't come into the matter; there aren't any here. The problem is that instead of adding a multi candidate that is more specific, the code in the question instead tries to replace the proto. If instead a multi candidate is added like this:
use MONKEY-TYPING;
augment class Array {
multi method AT-POS(Array:D: Int:D $i where $i < 0) {
nextwith($i + self.elems)
}
}
Then, due to the presence of the where clause, it will be considered before the usual AT-POS candidate without one. Since the standard candidate still applies too, nextwith can be used to defer to it. Using the above augment, the program:
my #arr = 1, 2, 3;
my $idx = -1;
say #arr[$idx];
Will output 3.
The usual caveats about augment apply, and since every array index operation will pay this cost, expect a significant slowdown.

Related

How to flatten or stringify an object (esp. Match)?

How do we flatten or stringify Match (or else) object to be string data type (esp. in multitude ie. as array elements)? e.g.
'foobar' ~~ m{ (foo) };
say $0.WHAT;
my $foo = $0;
say $foo.WHAT
(Match)
(Match)
How to end up with (Str)?
~ is the Str contextualizer:
'foobar' ~~ m{ (foo) };
say ~$0
will directly coerce it to a Str. You can use that if you have many matches, i. e.:
'foobar' ~~ m{ (f)(o)(o) };
say $/.map: ~*; # (f o o)
Just treat the objects as if they were strings.
If you apply a string operation to a value/object Raku will almost always just automatically coerce it to a string.
String operations include functions such as print and put, operators such as infix eq and ~ concatenation, methods such as .starts-with or .chop, interpolation such as "A string containing a $variable", and dedicated coercers such as .Str and Str(...).
A Match object contains an overall match. Any "children" (sub-matches) are just captures of substrings of that overall match. So there's no need to flatten anything because you can just deal with the single overall match.
A list of Match objects is a list. And a list is itself an object. If you apply a string operation to a list, you get the elements of the list stringified with a space between each element.
So:
'foobar' ~~ m{ (f) (o) (o) };
put $/; # foo
put $/ eq 'foo'; # True
put $/ ~ 'bar'; # foobar
put $/ .chop; # fo
put "[$/]"; # [foo]
put $/ .Str; # foo
my Str() $foo = $/;
say $foo.WHAT; # (Str)
put 'foofoo' ~~ m:g{ (f) (o) (o) }; # foo foo
The constructor for Str takes any Cool value as argument, including a regex Match object.
'foobar' ~~ m{ (foo) };
say $0.WHAT; # (Match)
say $0.Str.WHAT; # (Str)

How to pass pointer to a container from function?

I can bind containers to new names:
my %h;
my $p := %h{ "a" }{ "b" }{ "c" };
$p = 1;
say %h;
Which outputs expected:
{a => {b => {c => 1}}}
But what if I need to return such pointer from subroutine?
my %h;
sub get-pointer {
my $p := %h{ "a" }{ "b" }{ "c" };
return $p;
};
my $q := get-pointer();
$q = 1;
say %h;
Gives:
Cannot assign to a readonly variable or a value
That thing puzzles me - $p.WHERE and $q.WHERE give the same address, so why is it suddenly read-only?
Nevermind, I had some tunnel-vision moment and wanted aliases to behave as C pointers.
Found it clearly explained here at Raku Documentation.
The sub return will return values, not containers. Those are immutable
To return a mutable container, use return-rw.

What's the equivalent of Rebol apply in Red if any?

Rebol has apply Creating map function in Red language what's the equivalent of Rebol apply in Red if any ?
Currently, there's no native apply in Red. You can write apply on your own:
apply: func [
"Apply a function to a block of arguments"
fn [any-function!] "Function value to apply"
args [block!] "Block of arguments (to quote refinement use QUOTE keyword)"
/local refs vals val
][
refs: copy []
vals: copy []
set-val: [set val skip (append/only vals val)]
parse args [
some [
'quote set-val
| set val refinement! (append refs to word! val)
| set-val
]
]
do compose [(make path! head insert refs 'fn) (vals)]
]
It works bit differently than Rebol's apply (because I don't like Rebol's apply syntax). If you define some function:
f: func [
foo
/bar
baz
][
reduce [foo bar baz]
]
then here's how to use this apply:
>> apply :f [1 /bar 1]
== [1 true 1]
>> apply :f [quote /bar]
== [/bar false none]
>> apply :f [quote /bar /bar 1]
== [/bar true 1]
See http://red.qyz.cz/apply-and-ufcs.html for details.

Scope of nested regexes in Perl 6

Is it possible to define nested regexes in arbitrary sequence?
The following program works as expected:
my regex letter { <[a b]> }
my regex word { <letter> + }
my $string = 'abab';
$string ~~ &word;
put $/; # abab
If I swap the first two lines, compiler produces an error.
Is there a way to override this restriction (without using grammars)?
You can put the regex in a variable you declare up front but later set:
my $letter;
my regex word { <$letter> + }
$letter = regex { <[a b]> }
my $string = 'abab';
$string ~~ &word;
put $/; # abab

Bison Grammar %type and %token

Why is it that I have to use $<nVal>4 explicitly in the below grammar snippet?
I thought the %type <nVal> expr line would remove the need so that I can simply put $4?
Is it not possible to use a different definition for expr so that I can?
%union
{
int nVal;
char *pszVal;
}
%token <nVal> tkNUMBER
%token <pszVal> tkIDENT
%type <nVal> expr
%%
for_statement : tkFOR
tkIDENT { printf( "I:%s\n", $2 ); }
tkEQUALS
expr { printf( "A:%d\n", $<nVal>4 ); } // Why not just $4?
tkTO
expr { printf( "B:%d\n", $<nVal>6 ); } // Why not just $6?
step-statement
list
next-statement;
expr : tkNUMBER { $$ = $1; }
;
Update following rici's answer. This now works a treat:
for_statement : tkFOR
tkIDENT { printf( "I:%s\n", $2 ); }
tkEQUALS
expr { printf( "A:%d\n", $5 /* $<nVal>5 */ ); }
tkTO
expr { printf( "A:%d\n", $8 /* $<nVal>8 */ ); }
step-statement
list
next-statement;
Why is it that I have to use $<nVal>4 explicitly in the below grammar snippet?
Actually, you should use $5 if you want to refer to the expr. $4 is the tkEQUALS, which has no declared type, so any use must be explicitly typed. $3 is the previous midrule action, which has no value since $$ is not assigned in that action.
By the same logic, the second expr is $8; $6 is the second midrule action, which also has no value (and no type).
See the Bison manual:
The mid-rule action itself counts as one of the components of the rule. This makes a difference when there is another action later in the same rule (and usually there is another at the end): you have to count the actions along with the symbols when working out which number n to use in $n.