Gecko [ v99win.today ]

Name	Size	Permission
AllTypes.pod	7.84 KB	-rw-r--r--
Coercions.pod	12.81 KB	-rw-r--r--
Contributing.pod	2.86 KB	-rw-r--r--
Installation.pod	4.75 KB	-rw-r--r--
Libraries.pod	14.02 KB	-rw-r--r--
NonOO.pod	3.34 KB	-rw-r--r--
Optimization.pod	8.76 KB	-rw-r--r--
Params.pod	8.12 KB	-rw-r--r--
Policies.pod	4.05 KB	-rw-r--r--
UsingWithClassTiny.pod	4.16 KB	-rw-r--r--
UsingWithMoo.pod	25.38 KB	-rw-r--r--
UsingWithMoo2.pod	10.63 KB	-rw-r--r--
UsingWithMoo3.pod	9.26 KB	-rw-r--r--
UsingWithMoose.pod	6.11 KB	-rw-r--r--
UsingWithMouse.pod	6.53 KB	-rw-r--r--
UsingWithOther.pod	4.32 KB	-rw-r--r--
UsingWithTestMore.pod	2.58 KB	-rw-r--r--

Code Editor : Params.pod

=pod

=encoding utf-8

=head1 NAME

Type::Tiny::Manual::Params - advanced information on Type::Params
  
=head1 MANUAL

To get started with Type::Params, please read
L<Type::Tiny::Manual::UsingWithMoo> which will cover a lot of the basics,
even if you're not using Moo.

=head2 C<validate> and C<validate_named>

The generally recommended way of using Type::Params is this:

sub mysub {
    state $check = compile( SIGNATURE );
    my @args = $check->( @_ );
  }

But it is possible to do it in one call:

sub mysub {
    my @args = validate( \@_, SIGNATURE );
  }

There is also a C<validate_named> function which acts as a counterpart for
C<compile_named>.

This will generally be slower and less efficient than using C<compile> first
because Type::Tiny can do a lot of optimizations in that first stage to make
the second stage a lot faster. (And the results of C<compile> get stored in
the C<state> variable so that only has to happen once.)

There is rarely a reason to use C<validate> and C<validate_named>, but
they exist if you want them.

=head2 C<multisig>

Multisig allows you to allow multiple ways of calling a sub.

sub repeat_string {
   state $check = multisig(
     compile(
       Int,
       Str,
     ),
     compile_named(
       { named_to_list => 1 },
       count  => Int,
       string => Str,
     ),
   );
   
   my ($count, $string) = $check->(@_);
   return $string x $count;
 }
 
 repeat_string(            "Hello",          42  );    # works
 repeat_string(  string => "Hello", count => 42  );    # works
 repeat_string({ string => "Hello", count => 42 });    # works
 repeat_string( qr/hiya/ );                            # dies

It combines multiple checks and tries each until one works.

=head2 C<wrap_subs> and C<wrap_methods>

C<wrap_subs> turns the C<compile> idea inside out.

Instead of this:

sub foobar {
   state $check = compile(Int, Str);
   my ($foo, $bar) = @_;
   ...;
 }

You do this:

sub foobar {
   my ($foo, $bar) = @_;
   ...;
 }
 wrap_subs foobar => [ Int, Str ];

Or this:

sub foobar {
   my ($foo, $bar) = @_;
   ...;
 }
 wrap_subs foobar => compile( Int, Str );

=head2 Mixed Named and Positional Parameters

This can be faked using positional parameters and a slurpy dictionary.

state $check = compile(
   Int,
   slurpy Dict[
     foo => Int,
     bar => Optional[Int],
     baz => Optional[Int],
   ],
 );
 
 @_ = (42, foo => 21);                 # ok
 @_ = (42, foo => 21, bar  => 84);     # ok
 @_ = (42, foo => 21, bar  => 10.5);   # not ok
 @_ = (42, foo => 21, quux => 84);     # not ok

From Type::Params 1.009_002, C<head> and C<tail> options are accepted, which
provide another option for mixed named and positional arguments:

state $check = compile_named(
   { head => [ Int ] },
   foo => Int,
   bar => Optional[Int],
   baz => Optional[Int],
   ],
 );

The C<head> is shifted off C<< @_ >> before C<< @_ >> is considered as a hash.
The C<tail> is popped off C<< @_ >> before C<< @_ >> is considered as a hash.

=head2 Proper Signatures

Don't you wish your subs could look like this?

sub set_name (Object $self, Str $name) {
    $self->{name} = $name;
  }

Well; here are a few solutions for sub signatures that work with
L<Type::Tiny>...

=head3 Zydeco

L<Zydeco> is a Perl OO syntax toolkit with Type::Tiny support baked in
throughout.

package MyApp {
    use Zydeco;
    
    class Person {
      has name ( type => Str );
      
      method rename (Str $new_name) {
        printf("%s will now be called %s\n", $self->name, $new_name);
        $self->name($new_name);
      }
      
      coerce from Str via {
        $class->new(name => $_)
      }
    }
    
    class Company {
      has owner ( type => 'Person' );
    }
  }
  
  my $acme = MyApp->new_company(owner => "Robert");
  $acme->owner->rename("Bob");

=head3 Kavorka

L<Kavorka> is a sub signatures implementation written to natively use
L<Type::Utils>' C<dwim_type> for type constraints, and take advantage
of Type::Tiny's features such as inlining, and coercions.

method set_name (Str $name) {
    $self->{name} = $name;
  }

Kavorka's signatures provide a lot more flexibility, and slightly more
speed than Type::Params. (The speed comes from inlining almost all type
checks into the body of the sub being declared.)

Kavorka also includes support for type checking of the returned value.

Kavorka can also be used as part of L<Moops>, a larger framework for
object oriented programming in Perl.

=head3 Function::Parameters

Function::Parameters offers support for Type::Tiny and MooseX::Types.

use Types::Standard qw( Str );
  use Function::Parameters;
  
  method set_name (Str $name) {
      $self->{name} = $name;
  }

=head3 Attribute::Contract

Both Kavorka and Function::Parameters require a relatively recent
version of Perl. L<Attribute::Contract> supports older versions by
using a lot less magic.

You want Attribute::Contract 0.03 or above.

use Attribute::Contract -types => [qw/Object Str/];
  
  sub set_name :ContractRequires(Object, Str) {
      my ($self, $name) = @_;
      $self->{name} = $name;
  }

Attribute::Contract also includes support for type checking of the
returned value.

=head2 Type::Params versus X

=head3 Params::Validate

L<Type::Params> is not really a drop-in replacement for L<Params::Validate>;
the API differs far too much to claim that. Yet it performs a similar task,
so it makes sense to compare them.

=over

=item *

Type::Params will tend to be faster if you've got a sub which is called
repeatedly, but may be a little slower than Params::Validate for subs that
are only called a few times. This is because it does a bunch of work the
first time your sub is called to make subsequent calls a lot faster.

=item *

Params::Validate doesn't appear to have a particularly natural way of
validating a mix of positional and named parameters.

=item *

Type::Utils allows you to coerce parameters. For example, if you expect
a L<Path::Tiny> object, you could coerce it from a string.

=item *

If you are primarily writing object-oriented code, using Moose or similar,
and you are using Type::Tiny type constraints for your attributes, then
using Type::Params allows you to use the same constraints for method calls.

=item *

Type::Params comes bundled with Types::Standard, which provides a much
richer vocabulary of types than the type validation constants that come
with Params::Validate. For example, Types::Standard provides constraints
like C<< ArrayRef[Int] >> (an arrayref of integers), while the closest from
Params::Validate is C<< ARRAYREF >>, which you'd need to supplement with
additional callbacks if you wanted to check that the arrayref contained
integers.

Whatsmore, Type::Params doesn't just work with Types::Standard, but also
any other Type::Tiny type constraints.

=back

=head3 Params::ValidationCompiler

L<Params::ValidationCompiler> does basically the same thing as
L<Type::Params>.

=over

=item *

Params::ValidationCompiler and Type::Params are likely to perform fairly
similarly. In most cases, recent versions of Type::Params seem to be
I<slightly> faster, but except in very trivial cases, you're unlikely to
notice the speed difference. Speed probably shouldn't be a factor when
choosing between them.

=item *

Type::Params's syntax is more compact:

state $check = compile(Object, Optional[Int], slurpy ArrayRef);

Versus:

state $check = validation_for(
      params => [
         { type => Object },
         { type => Int,      optional => 1 },
         { type => ArrayRef, slurpy => 1 },
      ],
   );

=item *

L<Params::ValidationCompiler> probably has slightly better exceptions.

=back

=head1 NEXT STEPS

Here's your next step:

=over

=item * L<Type::Tiny::Manual::NonOO>

Type::Tiny in non-object-oriented code.

=back

=head1 AUTHOR

Toby Inkster E<lt>tobyink@cpan.orgE<gt>.

=head1 COPYRIGHT AND LICENCE

This is free software; you can redistribute it and/or modify it under
the same terms as the Perl 5 programming language system itself.

=head1 DISCLAIMER OF WARRANTIES

THIS PACKAGE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR IMPLIED
WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.

=cut

${this.title}

Code Editor : Params.pod