RTM `printplanner' scraper

rtm-printplanner-scraper.rb

#! /usr/bin/env ruby

require 'rubygems'
require 'nokogiri'

#
# RTM `printplanner' scraper
#

#TARGET = 'http://www.rememberthemilk.com/printplanner/USER/'
TARGET = 'rtm-weekly.html'

#
# Return list of list
#
# [{h1.inner_text => [ li, li, li, li, ...]},
#  {h1.inner_text => [ li, li, li, li, ...]},
#  ...
# ]
#
# each li
#
# li == { 'name'  => li's text node,
#         'list'  => list,
#         'limit' => due
#        }
#
Nokogiri( open( TARGET ).read 
          ).search( '//h1[following-sibling::ul]' ).map { |h1|
  li = h1.search( './following-sibling::ul[1]/li' )
  if ( li.size > 0 )
    {
      h1.inner_text =>
      li.map { |e|
        list, limit = e.search( '.tasklist'
                                  ).inner_text.sub( /\A\(/, ''
                                                    ).sub( /\)\z/, ''
                                                           ).split( /,/ )
        {
          'name'   => e.children.map { |n|
            if ( n.node_name == 'text' )
              n
            else
              nil
            end
          }.join.strip,
          'list'  => list.strip,
          'limit' => limit.strip
        }
      }
    }
  else
    nil
  end
}.compact