plucky (3) Sublike::Extended.3pm.gz

NAME

       "Sublike::Extended" - enable extended features when parsing "sub"-like syntax

SYNOPSIS

          use v5.26;
          use Sublike::Extended;
          use experimental 'signatures';

          extended sub greet (:$name = "world") {
             say "Hello, $name";
          }

          greet( name => $ENV{USER} );

       Or, since version 0.29:

          use v5.26;
          use Sublike::Extended 0.29 'sub';
          use experimental 'signatures';

          sub greet (:$name = "world") {
             say "Hello, $name";
          }

          greet( name => $ENV{USER} );

DESCRIPTION

       This module extends the syntax for declaring named or anonymous subroutines using Perl's builtin "sub"
       keyword, or other similar keywords provided by third-party modules, to enable parsing of extra features.

       By default, this module provides a new keyword, "extended", which parses the extra syntax required.
       Optionally since version 0.29, this module can additionally take over the handling of the "sub" keyword
       itself, allowing this extra syntax to be used without the "extended" prefix keyword.  As this ability may
       be surprising to unsuspecting readers, this is not done by default and must be explicitly requested with
       the "sub" import argument:

          use Sublike::Extended 'sub';

       On Perl 5.38 or above, this can also take over handling of the "method" keyword when using "feature
       'class'".

          use Sublike::Extended 'method';

       Currently, the only extended features that are provided are related to the parsing of a subroutine
       signature. Since signatures are only available on Perl version 5.26 or later, this module is unlikely to
       be useful in earlier versions of Perl.

   Named parameters
       Extended subroutines can declare named parameters in the signature, after any positional ones. These take
       the form of a name prefixed by a colon character.  The caller of such a function should pass values for
       these parameters by the usual name-value pair syntax that would be used for passing into a regular hash.
       Within the body of the subroutine the values passed into these are unpacked into regular lexical
       variables.

          sub colour (:$red, :$green, :$blue) {
             ... # $red, $green and $blue are available as regular lexicals
          }

          # argument order at the caller site is not important
          colour(green => 1, blue => 2, red => 3);

       Positional parameters can be placed after optional positional ones, but in order to make use of them the
       caller would have to pass a value for every positional parameter including the optional ones first. This
       is unlikely to be very useful; if you want to have optional parameters and named parameters, use named
       optional ones after any mandatory positional parameters.

       As with positional parameters, they are normally mandatory, but can be made optional by supplying a
       defaulting expression. If the caller fails to pass a value corresponding to an optional parameter, the
       default expression is evaluated and used instead.

          sub f (:$x0, :$x1, :$x2 = 0) { ... }
          # The caller must provide x0 and x1, but x2 is optional

       Since version 0.23 named parameters can be given defaulting expressions with the "//=" or "||="
       operators, meaning their defaults apply also if the caller passed a present-but-undef, or present-but-
       false value.

          sub f (:$x0, :$x1, :$x2 //= 0) { ... }
          # $x2 will be set to 0 even if the caller passes  x2 => undef

       An optional slurpy hash or (since version 0.24) slurpy array is also permitted after all of these. It
       will contain the values of any other name-value pairs given by the caller, after those corresponding to
       named parameters have already been extracted.

          sub g (:$alpha, :$beta, %rest) { ... }

          sub g (:$alpha, :$beta, @rest) { ... }

       In the case of a slurpy array, it will contain every argument value that was not consumed as a named
       parameter pair, in the original order passed by the caller, including any duplicates.

   Parameter Attributes
       Parameters to extended subroutines can use attribute syntax to apply extra attributes to individual
       parameters.

          sub info ($x :Attribute) { ... }

       Any attributes that are available are ones that have been previously registered with XS::Parse::Sublike
       using its XS-level API. The particular behaviour of such an attribute would be defined by whatever module
       provided the attribute.

KEYWORDS

   extended
          extended sub NAME (SIGNATURE...) { BODY... }

          extended sub (SIGNATURE...) { BODY... };

       This prefix keyword enables extra parsing features when handling a "sub" (or other sub-like function
       keyword).

       This keyword can be freely mixed with other "sub"-prefix keywords, such as "async" from
       Future::AsyncAwait

          async extended sub f (:$param) { ... }

       This can also be used with other keywords that provide "sub"-like syntax, such as "method" from
       Object::Pad or the core "use feature 'class'".

          extended method f (:$param) { ... }

TODO

       •   Support defined-or and true-or positional parameters even on versions of Perl before they were
           officially added (v5.38).

AUTHOR

       Paul Evans <leonerd@leonerd.org.uk>