WWW::Scraper::NorthernLight - Scrapes NorthernLight.com

    require WWW::Scraper;
    $search = new WWW::Scraper('NorthernLight');

package WWW::Scraper::NorthernLight;

#####################################################################

use strict;
use vars qw($VERSION @ISA);
@ISA = qw(WWW::Scraper);
$VERSION = sprintf("%d.%02d", q$Revision: 1.0 $ =~ /(\d+)\.(\d+)/);

use Carp ();
use WWW::Scraper(qw(2.27 generic_option addURL trimTags));
use WWW::Scraper::FieldTranslation;

my $scraperRequest = 
   { 
      'type' => 'FORM'       # Type of query generation is 'QUERY'
     ,'formNameOrNumber' => 'powSearch'
     ,'submitButton' => 'search'
      # This is the basic URL on which to build the query.
     ,'url' => 'http://www.northernlight.com/power.html'
      # This is the Scraper attributes => native input fields mapping
      ,'nativeQuery' => 'qr'
      ,'nativeDefaults' => {
                            'qr' => undef
                           }
      ,'fieldTranslations' =>
              {
                  '*' =>
                      {    'skills'    => 'qr'
#                            ,'payrate'   => undef
#                            ,'locations' => new WWW::Scraper::FieldTranslation('NorthernLight', 'Job', 'locations')
                          ,'*'         => '*'
                      }
              }
      # Some more options for the Scraper operation.
     ,'cookies' => 0
   };

my $scraperFrame =
        [ 'HTML', 
           [ 
               #        </b> found <b>10,032,977 items</b>
               [ 'COUNT', 'found\s+<b>([0-9,]+)\s+items?</b>']
              ,[ 'NEXT', 'alt="Next Page"' ]
              ,[ 'BODY', '<!--NLBannerStart-->', '<!--NLResultListEnd-->',
                  [  
                     [ 'HIT*',
                        [  
                           [ 'BODY', '<!--NLResultStart-->', '<!--NLResultEnd-->',
                             [
                               [ 'TR',
                                  [
                                    # <!--  --><!-- <td>&nbsp;</td> -->
                                     [ 'SNIP', '<!--[^>]*?<td>.*?-->',
                                       [
                                         [ 'TD',
                                              [['SPAN', 'number']]
                                         ]
                                        ,[ 'TD', 
                                            [
                                               [ 'A', 'url', 'title' ]
                                              ,['REGEX', '<!--NLResultRelevanceStart-->(\d+)% -', 'relevance']
                                              ,['REGEX', '<!--NLResultRelevanceEnd-->(.*?)&nbsp;', 'source']
                                              ,['REGEX', '</b>(.*?)<br>', 'description']
                                              ,['REGEX', '<!-- Misc Block --><!-- \d+ -->(.*?)<!-- Misc Block --><!-- \d+ -->', 'miscBlock']
                                              ,[ 'TABLE',
                                                 [
                                                   ['TR']
                                                  ,['REGEX', '(\d+)%:', 'secondRelevance']
                                                  ,[ 'A', 'secondUrl', 'secondTitle' ]
                                                 ]
                                               ]
                                              #,['SPAN', 'avail'] #needs better treatment of the <SPAN> at the top of this <TD> for this to work.
                                              ,[ 'AQ', 'more\s+results', 'moreResultsUrl', undef ]
                                              # <!-- Inline Clustering -->
                                            ]
                                         ]
                                       ]
                                     ]
                                  ]
                               ]
                             ]
                           ]
                        ]
                     ] 
                  ]
              ]
           ]
        ];



# Access methods for the structural declarations of this Scraper engine.
sub scraperRequest { $scraperRequest }
sub scraperFrame { $_[0]->SUPER::scraperFrame($scraperFrame); }
sub scraperDetail{ undef }

sub testParameters {
    my ($self) = @_;

    if ( ref $self ) {
        $self->{'isTesting'} = 1;
    }
    
    return { 
             'SKIP' => "NorthernLight's search engine seems to be down these days!?"
            ,'testNativeQuery' => 'search scraper'
            ,'expectedOnePage' => 9
            ,'expectedMultiPage' => 12
            ,'expectedBogusPage' => 0
           };
}

1;