NAME

Web::Machine::Util - General Utility module

VERSION

version 0.17

SYNOPSIS

  use Web::Machine::Util;

DESCRIPTION

This is just a basic utility module used internally by Web::Machine. There are no real user serviceable parts in here.

FUNCTIONS

"first"

This is imported from List::Util and passed on here for export.

"pair_key"

"pair_value"

These two functions are used for fetching the key and value out of a pair in the Web::Machine internals. We represent a pair simply as a HASH ref with one key.

"inflate_headers( $request )"

This will call "inflate" on an instance of HTTP::Headers::ActionPack.

"create_header( @args )"

This will call "create" on an instance of HTTP::Headers::ActionPack.

"create_date( $date_string | $time_peice )"

Given either a $date_string or an instance of Time::Piece, this will inflate it into a HTTP::Headers::ActionPack::DateHeader object, suitable for use in the FSM.

"bind_path( $path_spec, $path )"

Given a $path_spec (described below) and a $path, this will either bind the path to the spec and return and array of bound values, or it will return nothing. Returning nothing indicates that no match was found. Additionally, if this function is called in scalar context, and there is only one match, it will return that item. Otherwise it will return the array as normal. This all makes it easy to use the following idiom:

  if ( my $id = bind_path( '/:id', $request->path_info ) ) {
      # handle the case with an ID here
  }
  else {
      # handle other cases here
  }
    

The $path_spec follows a pretty standard convention. Literal path parts must match corresponding literal. Variable path parts are prefixed by a colon and are captured for returning later, if a question mark (?) prefixes the colon, that element will be considered optional. And lastly the "splat" operator ("*") is supported and causes all the rest of the path segments to be returned. Below are a few examples of this:

  spec                  path             result
  ------------------------------------------------------------
  /test/:foo/:bar       /test/1/2        ( 1, 2 )
  /test/:foo/:bar       /test/1/         undef #failure-case
  /test/*               /test/1/2/3      ( 1, 2, 3 )
  /user/:id/:action     /user/1/edit     ( 1, 'edit' )
  /?:id                 /201             ( 201 )
  /?:id                 /                ( )
    

This function is kept deliberately simple and it is expected that the user will use "my" in the array form to assign multiple variables, like this:

  my ( $foo, $bar ) = bind_path( '/test/:foo/:bar', $path );
    

In the future we might add a "bind_path_hash" function which captures the variable names as well, but to be honest, if you feel you need that, you likely want one of the many excellent path dispatching modules available on CPAN.

NOTE: Some care should be taken when using path specs in which the only things are either optional parameters (prefixed with "?:") or the "splat" operator ("*") as they can return empty arrays, which in certain contexts can look like match failure. In these cases you can test the match in scalar context to verify, a match failure will be "undef" whereas a match success (in which nothing was matched) will return 0 (indicating an array with zero size).

SUPPORT

bugs may be submitted through <https://github.com/houseabsolute/webmachine-perl/issues>.

AUTHORS

• Stevan Little <stevan@cpan.org>
• Dave Rolsky <autarch@urth.org>

COPYRIGHT AND LICENCE

This software is copyright (c) 2016 by Infinity Interactive, Inc.

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