The London Perl and Raku Workshop takes place on 26th Oct 2024. If your company depends on Perl, please consider sponsoring and/or attending.
 / 
Function-Parameters-0.06_01
River stage three • 48 direct dependents • 188 total dependents
/ Function::Parameters

NAME

Function::Parameters - subroutine definitions with parameter lists

SYNOPSIS

 use Function::Parameters;
 
 # simple function
 fun foo($bar, $baz) {
   return $bar + $baz;
 }
 
 # function with prototype
 fun mymap($fun, @args) :(&@) {
   my @res;
   for (@args) {
     push @res, $fun->($_);
   }
   @res
 }
 
 print "$_\n" for mymap { $_ * 2 } 1 .. 4;
 
 # method with implicit $self
 method set_name($name) {
   $self->{name} = $name;
 }

 # function with default arguments
 fun search($haystack, $needle = qr/^(?!)/, $offset = 0) {
   ...
 }

 # method with default arguments
 method skip($amount = 1) {
   $self->{position} += $amount;
 }
 # use different keywords
 use Function::Parameters {
   proc => 'function',
   meth => 'method',
 };
 
 my $f = proc ($x) { $x * 2 };
 meth get_age() {
   return $self->{age};
 }

DESCRIPTION

This module lets you use parameter lists in your subroutines. Thanks to PL_keyword_plugin it works without source filters.

WARNING: This is my first attempt at writing XS code and I have almost no experience with perl's internals. So while this module might appear to work, it could also conceivably make your programs segfault. Consider this module alpha quality.

Basic stuff

To use this new functionality, you have to use fun instead of sub - sub continues to work as before. The syntax is almost the same as for sub, but after the subroutine name (or directly after fun if you're writing an anonymous sub) you can write a parameter list in parentheses. This list consists of comma-separated variables.

The effect of fun foo($bar, $baz) { is as if you'd written sub foo { my ($bar, $baz) = @_; , i.e. the parameter list is simply copied into my and initialized from @_.

In addition you can use method, which understands the same syntax as fun but automatically creates a $self variable for you. So by writing method foo($bar, $baz) { you get the same effect as sub foo { my $self = shift; my ($bar, $baz) = @_; .

Customizing the generated keywords

You can customize the names of the keywords injected into your scope. To do that you pass a reference to a hash mapping keywords to types in the import list:

 use Function::Parameters {
   KEYWORD1 => TYPE1,
   KEYWORD2 => TYPE2,
   ...
 };

Or more concretely:

 use Function::Parameters { proc => 'function', meth => 'method' }; # -or-
 use Function::Parameters { proc => 'function' }; # -or-
 use Function::Parameters { meth => 'method' }; # etc.

The first line creates two keywords, proc and meth (for defining functions and methods, respectively). The last two lines only create one keyword. Generally the hash keys (keywords) can be any identifiers you want while the values (types) have to be either a hash reference (see below) or 'function', 'method', 'classmethod', 'function_strict', 'method_strict', or 'classmethod_strict'. The main difference between 'function' and 'method' is that 'method's automatically shift their first argument into $self ('classmethod's are similar but shift into $class).

The following shortcuts are available:

 use Function::Parameters;
    # is equivalent to #
 use Function::Parameters { fun => 'function', method => 'method' };

The following shortcuts are deprecated and may be removed from a future version of this module:

 # DEPRECATED
 use Function::Parameters 'foo';
   # is equivalent to #
 use Function::Parameters { 'foo' => 'function' };
 # DEPRECATED
 use Function::Parameters 'foo', 'bar';
   # is equivalent to #
 use Function::Parameters { 'foo' => 'function', 'bar' => 'method' };

That is, if you want to pass arguments to Function::Parameters, use a hashref, not a list of strings.

You can customize the properties of the generated keywords even more by passing a hashref instead of a string. This hash can have the following keys:

name

Valid values: optional (default), required (all uses of this keyword must specify a function name), and prohibited (all uses of this keyword must not specify a function name). This means a name => 'prohibited' keyword can only be used for defining anonymous functions.

shift

Valid values: strings that look like a scalar variable. Any function created by this keyword will automatically shift its first argument into a local variable whose name is specified here.

attributes, attrs

Valid values: strings that are valid source code for attributes. Any value specified here will be inserted as a subroutine attribute in the generated code. Thus:

 use Function::Parameters { sub_l => { attributes => ':lvalue' } };
 sub_l foo() {
   ...
 }

turns into

 sub foo :lvalue {
   ...
 }

It is recommended that you use attributes in new code but attrs is also accepted for now.

default_arguments

Valid values: booleans. This property is on by default, so you have to pass default_arguments => 0 to turn it off. If it is disabled, using = in a parameter list causes a syntax error. Otherwise it lets you specify default arguments directly in the parameter list:

 fun foo($x, $y = 42, $z = []) {
   ...
 }

turns into

 sub foo {
   my ($x, $y, $z) = @_;
   $y = 42 if @_ < 2;
   $z = [] if @_ < 3;
   ...
 }

You can even refer to previous parameters in the same parameter list:

 print fun ($x, $y = $x + 1) { "$x and $y" }->(9);  # "9 and 10"

This also works with the implicit first parameter of methods:

 method scale($factor = $self->default_factor) {
   $self->{amount} *= $factor;
 }
check_argument_count

Valid values: booleans. This property is off by default. If it is enabled, the generated code will include checks to make sure the number of passed arguments is correct (and otherwise throw an exception via Carp::croak):

  fun foo($x, $y = 42, $z = []) {
    ...
  }

turns into

 sub foo {
   Carp::croak "Not enough arguments for fun foo" if @_ < 1;
   Carp::croak "Too many arguments for fun foo" if @_ > 3;
   my ($x, $y, $z) = @_;
   $y = 42 if @_ < 2;
   $z = [] if @_ < 3;
   ...
 }

Plain 'function' is equivalent to:

 {
   name => 'optional',
   default_arguments => 1,
   check_argument_count => 0,
 }

(These are all default values so 'function' is also equivalent to {}.)

'function_strict' is like 'function' but with check_argument_count => 1.

'method' is equivalent to:

 {
   name => 'optional',
   default_arguments => 1,
   check_argument_count => 0,
   attributes => ':method',
   shift => '$self',
 }

'method_strict' is like 'method' but with check_argument_count => 1.

'classmethod' is equivalent to:

 {
   name => 'optional',
   default_arguments => 1,
   check_argument_count => 0,
   attributes => ':method',
   shift => '$class',
 }

'classmethod_strict' is like 'classmethod' but with check_argument_count => 1.

Syntax and generated code

Normally, Perl subroutines are not in scope in their own body, meaning the parser doesn't know the name foo or its prototype while processing the body of sub foo ($) { foo $bar[1], $bar[0]; }, parsing it as $bar->foo([1], $bar[0]). Yes. You can add parens to change the interpretation of this code, but foo($bar[1], $bar[0]) will only trigger a foo() called too early to check prototype warning. This module attempts to fix all of this by adding a subroutine declaration before the function body, so the parser knows the name (and possibly prototype) while it processes the body. Thus fun foo($x) :($) { $x } really turns into sub foo ($) { sub foo ($); my ($x) = @_; $x }.

If you need subroutine attributes, you can put them after the parameter list with their usual syntax.

Syntactically, these new parameter lists live in the spot normally occupied by prototypes. However, you can include a prototype by specifying it as the first attribute (this is syntactically unambiguous because normal attributes have to start with a letter while a prototype starts with ().

As an example, the following declaration uses every available feature (subroutine name, parameter list, default arguments, prototype, default attributes, attributes, argument count checks, and implicit $self):

 method foo($x, $y, $z = sqrt 5) :($$$;$) :lvalue :Banana(2 + 2) {
   ...
 }

And here's what it turns into:

 sub foo ($$$;$) :method :lvalue :Banana(2 + 2) {
   sub foo ($$$;$);
   Carp::croak "Not enough arguments for method foo" if @_ < 2;
   Carp::croak "Too many arguments for method foo" if @_ > 4;
   my $self = shift;
   my ($x, $y, $z) = @_;
   $z = sqrt 5 if @_ < 3;
   ...
 }

Another example:

 my $coderef = fun ($p, $q) :(;$$)
   :lvalue
   :Gazebo((>:O)) {
   ...
 };

And the generated code:

 my $coderef = sub (;$$) :lvalue :Gazebo((>:O)) {
   # vvv   only if check_argument_count is enabled    vvv
   Carp::croak "Not enough arguments for fun (anon)" if @_ < 2;
   Carp::croak "Too many arguments for fun (anon)" if @_ > 2;
   # ^^^                                              ^^^
   my ($p, $q) = @_;
   ...
 };

Wrapping Function::Parameters

If you want to wrap Function::Parameters, you just have to call its import method. It always applies to the file that is currently being parsed and its effects are lexical (i.e. it works like warnings or strict).

 package Some::Wrapper;
 use Function::Parameters ();
 sub import {
   Function::Parameters->import;
   # or Function::Parameters->import(@custom_import_args);
 }

AUTHOR

Lukas Mai, <l.mai at web.de>

COPYRIGHT & LICENSE

Copyright 2010, 2011, 2012 Lukas Mai.

This program is free software; you can redistribute it and/or modify it under the terms of either: the GNU General Public License as published by the Free Software Foundation; or the Artistic License.

See http://dev.perl.org/licenses/ for more information.