Steffen Müller > Class-XSAccessor-1.19 > Class::XSAccessor::Array



Annotate this POD


New  2
Open  5
Stalled  2
View/Report Bugs
Module Version: 1.19   Source  


Class::XSAccessor::Array - Generate fast XS accessors without runtime compilation


  package MyClassUsingArraysAsInternalStorage;
  use Class::XSAccessor::Array
    constructor => 'new',
    getters => {
      get_foo => 0, # 0 is the array index to access
      get_bar => 1,
    setters => {
      set_foo => 0,
      set_bar => 1,
    accessors => { # a mutator
      buz => 2,
    predicates => { # test for definedness
      has_buz => 2,
    lvalue_accessors => { # see below
      baz => 3,
    true => [ 'is_token', 'is_whitespace' ],
    false => [ 'significant' ];
  # The imported methods are implemented in fast XS.
  # normal class code here.

As of version 1.05, some alternative syntax forms are available:

  package MyClass;
  # Options can be passed as a HASH reference if you prefer it,
  # which can also help PerlTidy to flow the statement correctly.
  use Class::XSAccessor {
    getters => {
      get_foo => 0,
      get_bar => 1,


The module implements fast XS accessors both for getting at and setting an object attribute. Additionally, the module supports mutators and simple predicates (has_foo() like tests for definedness of an attributes). The module works only with objects that are implemented as arrays. Using it on hash-based objects is bound to make your life miserable. Refer to Class::XSAccessor for an implementation that works with hash-based objects.

A simple benchmark showed a significant performance advantage over writing accessors in Perl.

Since version 0.10, the module can also generate simple constructors (implemented in XS) for you. Simply supply the constructor => 'constructor_name' option or the constructors => ['new', 'create', 'spawn'] option. These constructors do the equivalent of the following Perl code:

  sub new {
    my $class = shift;
    return bless [], ref($class)||$class;

That means they can be called on objects and classes but will not clone objects entirely. Note that any parameters to new() will be discarded! If there is a better idiom for array-based objects, let me know.

While generally more obscure than hash-based objects, objects using blessed arrays as internal representation are a bit faster as its somewhat faster to access arrays than hashes. Accordingly, this module is slightly faster (~10-15%) than Class::XSAccessor, which works on hash-based objects.

The method names may be fully qualified. In the example of the synopsis, you could have written MyClass::get_foo instead of get_foo. This way, you can install methods in classes other than the current class. See also: The class option below.

Since version 1.01, you can generate extremely simple methods which just return true or false (and always do so). If that seems like a really superfluous thing to you, then think of a large class hierarchy with interfaces such as PPI. This is implemented as the true and false options, see synopsis.


In addition to specifying the types and names of accessors, you can add options which modify behaviour. The options are specified as key/value pairs just as the accessor declaration. Example:

  use Class::XSAccessor::Array
    getters => {
      get_foo => 0,
    replace => 1;

The list of available options is:


Set this to a true value to prevent Class::XSAccessor::Array from complaining about replacing existing subroutines.


Set this to a true value to change the return value of setters and mutators (when called with an argument). If chained is enabled, the setters and accessors/mutators will return the object. Mutators called without an argument still return the value of the associated attribute.

As with the other options, chained affects all methods generated in the same use Class::XSAccessor::Array ... statement.


By default, the accessors are generated in the calling class. Using the class option, you can explicitly specify where the methods are to be generated.


Support for lvalue accessors via the keyword lvalue_accessors was added in version 1.08. At this point, THEY ARE CONSIDERED HIGHLY EXPERIMENTAL. Furthermore, their performance hasn't been benchmarked yet.

The following example demonstrates an lvalue accessor:

  package Address;
  use Class::XSAccessor
    constructor => 'new',
    lvalue_accessors => { zip_code => 0 };
  package main;
  my $address = Address->new(2);
  print $address->zip_code, "\n"; # prints 2
  $address->zip_code = 76135; # <--- This is it!
  print $address->zip_code, "\n"; # prints 76135


Probably wouldn't work if your objects are tied. But that's a strange thing to do anyway.

Scary code exploiting strange XS features.

If you think writing an accessor in XS should be a laughably simple exercise, then please contemplate how you could instantiate a new XS accessor for a new hash key or array index that's only known at run-time. Note that compiling C code at run-time a la Inline::C is a no go.

Threading. With version 1.00, a memory leak has been fixed that would leak a small amount of memory if you loaded Class::XSAccessor-based classes in a subthread that hadn't been loaded in the "main" thread before. If the subthread then terminated, a hash key and an int per associated method used to be lost. Note that this mattered only if classes were only loaded in a sort of throw-away thread.

In the new implementation as of 1.00, the memory will not be released again either in the above situation. But it will be recycled when the same class or a similar class is loaded again in any thread.





Steffen Mueller <>

chocolateboy <>


Copyright (C) 2008, 2009, 2010, 2011, 2012, 2013 by Steffen Mueller

This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself, either Perl version 5.8 or, at your option, any later version of Perl 5 you may have available.

syntax highlighting: