gscraper 0.1.7 → 0.2.0

data/lib/gscraper/licenses.rb

@@ -1,3 +1,25 @@
+#
+#--
+# GScraper - A web-scraping interface to various Google Services.
+#
+# Copyright (c) 2007-2008 Hal Brodigan (postmodern.mod3 at gmail.com)
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+#++
+#
+
 module GScraper
   module Licenses
     # Any desired license

data/lib/gscraper/page.rb

@@ -0,0 +1,64 @@
+#
+#--
+# GScraper - A web-scraping interface to various Google Services.
+#
+# Copyright (c) 2007-2008 Hal Brodigan (postmodern.mod3 at gmail.com)
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+#++
+#
+
+module GScraper
+  class Page < Array
+
+    #
+    # Creates a new Page object with the given _elements_. If a _block_
+    # is given, it will be passed the newly created Page object.
+    #
+    def initialize(elements=[],&block)
+      super(elements)
+
+      block.call(self) if block
+    end
+
+    #
+    # Returns a mapped Array of the elements within the Page using the
+    # given _block_. If the _block_ is not given, the page will be
+    # returned.
+    #
+    #   page.map # => Page
+    #
+    #   page.map { |element| element.field } # => [...]
+    #
+    def map(&block)
+      return self unless block
+
+      mapped = []
+
+      each { |element| mapped << block.call(element) }
+      return mapped
+    end
+
+    #
+    # Selects the elements within the Page which match the given _block_.
+    #
+    #   page.select { |element| element.field =~ /ruby/i }
+    #
+    def select(&block)
+      self.class.new(super(&block))
+    end
+
+  end
+end

data/lib/gscraper/search.rb

@@ -1 +1,25 @@
+#
+#--
+# GScraper - A web-scraping interface to various Google Services.
+#
+# Copyright (c) 2007-2008 Hal Brodigan (postmodern.mod3 at gmail.com)
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+#++
+#
+
+require 'gscraper/search/web_query'
+require 'gscraper/search/ajax_query'
 require 'gscraper/search/search'

data/lib/gscraper/search/ajax_query.rb

@@ -0,0 +1,176 @@
+#
+#--
+# GScraper - A web-scraping interface to various Google Services.
+#
+# Copyright (c) 2007-2008 Hal Brodigan (postmodern.mod3 at gmail.com)
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+#++
+#
+
+require 'gscraper/search/result'
+require 'gscraper/search/page'
+require 'gscraper/search/query'
+require 'gscraper/extensions/uri'
+require 'gscraper/has_pages'
+require 'gscraper/gscraper'
+
+require 'json'
+
+module GScraper
+  module Search
+    class AJAXQuery < Query
+
+      include HasPages
+
+      # Maximum results per-page
+      RESULTS_PER_PAGE = 8
+
+      # AJAX API host
+      API_HOST = 'www.google.com'
+
+      # AJAX API URL
+      API_URL = "http://#{API_HOST}/uds/GwebSearch?callback=google.search.WebSearch.RawCompletion&context=0&lstkp=0&rsz=large"
+
+      # Default language
+      DEFAULT_LANGUAGE = 'en'
+
+      # Default signature
+      DEFAULT_SIG = '582c1116317355adf613a6a843f19ece'
+
+      # Default key
+      DEFAULT_KEY = 'notsupplied'
+
+      # Default version
+      DEFAULT_VERSION = '1.0'
+
+      # The search language
+      attr_accessor :language
+
+      # The search signature
+      attr_accessor :sig
+
+      # The search key
+      attr_accessor :key
+
+      # The API version
+      attr_accessor :version
+
+      #
+      # Creates a new AJAXQuery with the given _options_. If a _block_ is
+      # given it will be passed the newly created AJAXQuery object.
+      #
+      # _options_ may contain the following keys:
+      # <tt>:language</tt>:: The search language. Defaults to <tt>:en</tt>.
+      # <tt>:sig</tt>:: The search signature. Defaults to
+      #                 +582c1116317355adf613a6a843f19ece+.
+      # <tt>:key</tt>:: The search key. Defaults to <tt>:notsupplied</tt>.
+      # <tt>:version</tt>:: The desired API version. Defaults to
+      #                     <tt>1.0</tt>.
+      #
+      def initialize(options={},&block)
+        @agent = GScraper.web_agent(options)
+
+        @language = (options[:language] || DEFAULT_LANGUAGE)
+
+        @sig = (options[:sig] || DEFAULT_SIG)
+        @key = (options[:key] || DEFAULT_KEY)
+        @version = (options[:version] || DEFAULT_VERSION)
+
+        super(options,&block)
+      end
+
+      #
+      # Creates a new AJAXQuery object from the specified URL. If a block is
+      # given, it will be passed the newly created AJAXQuery object.
+      #
+      def self.from_url(url,options={},&block)
+        url = URI(url.to_s)
+
+        options[:language] = url.query_params['hl']
+        options[:query] = url.query_params['q']
+
+        options[:sig] = url.query_params['sig']
+        options[:key] = url.query_params['key']
+        options[:version] = url.query_params['v']
+
+        return self.new(options,&block)
+      end
+
+      #
+      # Returns +RESULTS_PER_PAGE+.
+      #
+      def results_per_page
+        RESULTS_PER_PAGE
+      end
+
+      #
+      # Returns the URL that represents the query.
+      #
+      def search_url
+        search_url = URI(API_URL)
+
+        search_url.query_params['hl'] = @language
+        search_url.query_params['gss'] = '.com'
+        search_url.query_params['q'] = expression
+        search_url.query_params['sig'] = @sig
+        search_url.query_params['key'] = @key
+        search_url.query_params['v'] = @version
+
+        return search_url
+      end
+
+      #
+      # Returns the URL that represents the query of a specific
+      # _page_index_.
+      #
+      def page_url(page_index)
+        url = search_url
+
+        if page_index > 1
+          url.query_params['start'] = result_offset_of(page_index)
+        end
+
+        return url
+      end
+
+      #
+      # Returns a Page object containing Result objects at the specified
+      # _page_index_.
+      #
+      def page(page_index)
+        Page.new do |new_page|
+          body = @agent.get(page_url(page_index)).body
+          hash = JSON.parse(body.scan(/\{.*\}/).first)
+
+          rank_offset = result_offset_of(page_index)
+
+          if (hash.kind_of?(Hash) && hash['results'])
+            hash['results'].each_with_index do |result,index|
+              rank = rank_offset + (index + 1)
+              title = Hpricot(result['title']).inner_text
+              url = result['unescapedUrl']
+              summary = Hpricot(result['content']).inner_text
+              cached_url = result['cacheUrl']
+
+              new_page << Result.new(rank,title,url,summary,cached_url)
+            end
+          end
+        end
+      end
+
+    end
+  end
+end

data/lib/gscraper/search/page.rb

@@ -1,42 +1,31 @@
+#
+#--
+# GScraper - A web-scraping interface to various Google Services.
+#
+# Copyright (c) 2007-2008 Hal Brodigan (postmodern.mod3 at gmail.com)
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+#++
+#
+
 require 'gscraper/search/result'
+require 'gscraper/page'
 
 module GScraper
   module Search
-    class Page < Array
-
-      #
-      # Creates a new Page object with the given _results_.
-      #
-      def initialize(results=[])
-        super(results)
-      end
-
-      #
-      # Returns a mapped Array of the results within the Page using the
-      # given _block_. If the _block_ is not given, the page will be
-      # returned.
-      #
-      #   page.map # => Page
-      #
-      #   page.map { |result| result.url } # => [...]
-      #
-      def map(&block)
-        return self unless block
-
-        mapped = []
-
-        each { |result| mapped << block.call(result) }
-        return mapped
-      end
-
-      #
-      # Selects the results within the Page which match the given _block_.
-      #
-      #   page.select { |result| result.title =~ /ruby/i }
-      #
-      def select(&block)
-        Page.new(super(&block))
-      end
+    class Page < GScraper::Page
 
       #
       # Selects the results using the specified _block_.
@@ -160,7 +149,7 @@ module GScraper
       #   page.cached_urls # => [...]
       #
       def cached_urls
-        map { |result| result.cached_url }
+        map { |result| result.cached_url }.compact
       end
 
       #
@@ -170,7 +159,7 @@ module GScraper
       #   page.cached_pages # => [...]
       #
       def cached_pages
-        map { |result| result.cached_page }
+        map { |result| result.cached_page }.compact
       end
 
       #
@@ -180,17 +169,7 @@ module GScraper
       #   page.similar_urls # => [...]
       #
       def similar_urls
-        map { |result| result.similar_url }
-      end
-
-      #
-      # Returns an Array containing the similar Queries of the results
-      # within the Page.
-      #
-      #   page.similar_queries # => [...]
-      #
-      def similar_queries
-        map { |result| result.similar_query }
+        map { |result| result.similar_url }.compact
       end
 
       #
@@ -263,20 +242,6 @@ module GScraper
         similar_urls.each(&block)
       end
 
-      #
-      # Iterates over each result's similar Query within the Page, passing
-      # each to the given _block_.
-      #
-      #   each_similar_query do |q|
-      #     q.first_page do |page|
-      #       puts page.urls.join("\n")
-      #     end
-      #   end
-      #
-      def each_similar_query(&block)
-        similar_queries.each(&block)
-      end
-
       #
       # Returns the ranks of the results that match the specified _block_.
       #
@@ -345,16 +310,6 @@ module GScraper
         results_with(&block).similar_urls
       end
 
-      #
-      # Returns the similar Queries of the results that match the
-      # specified _block_.
-      #
-      #   page.similar_queries_of { |result| result.title =~ /hackety/ }
-      #
-      def similar_queries_of(&block)
-        results_with(&block).similar_queries
-      end
-
     end
   end
 end

data/lib/gscraper/search/query.rb

@@ -1,10 +1,33 @@
+#
+#--
+# GScraper - A web-scraping interface to various Google Services.
+#
+# Copyright (c) 2007-2008 Hal Brodigan (postmodern.mod3 at gmail.com)
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+#++
+#
+
 require 'gscraper/search/result'
 require 'gscraper/search/page'
 require 'gscraper/sponsored_ad'
 require 'gscraper/sponsored_links'
 require 'gscraper/extensions/uri'
+require 'gscraper/has_pages'
 require 'gscraper/licenses'
-require 'gscraper/web_agent'
+require 'gscraper/gscraper'
 
 require 'hpricot'
 
@@ -12,20 +35,6 @@ module GScraper
   module Search
     class Query
 
-      include WebAgent
-
-      # Search host
-      SEARCH_HOST = 'www.google.com'
-
-      # Search URL
-      SEARCH_URL = "http://#{SEARCH_HOST}/search"
-
-      # Default results per-page
-      RESULTS_PER_PAGE = 10
-
-      # Results per-page
-      attr_accessor :results_per_page
-
       # Search query
       attr_accessor :query
 
@@ -71,67 +80,14 @@ module GScraper
       # Search for results with-out the words
       attr_accessor :without_words
 
-      # Search for results written in the language
-      attr_accessor :language
-
-      # Search for results from the region
-      attr_accessor :region
-
-      # Search for results in the format
-      attr_accessor :in_format
-
-      # Search for results not in the format
-      attr_accessor :not_in_format
-
-      # Search for results within the past day
-      attr_accessor :within_past_day
-
-      # Search for results within the past week
-      attr_accessor :within_past_week
-
-      # Search for results within the past months
-      attr_accessor :within_past_months
-
-      # Search for results within the past year
-      attr_accessor :within_past_year
-
       # Search for results containing numbers between the range
       attr_accessor :numeric_range
 
-      # Search for results where the query ocurrs within the area
-      attr_accessor :occurrs_within
-
-      # Search for results inside the domain
-      attr_accessor :inside_domain
-
-      # Search for results outside the domain
-      attr_accessor :outside_domain
-
-      # Search for results which have the rights
-      attr_accessor :rights
-
-      # Filter the search results
-      attr_accessor :filtered
-
-      # Search for results similar to the page
-      attr_accessor :similar_to
-
-      # Search for results linking to the page
-      attr_accessor :links_to
-
       #
       # Creates a new Query object from the given search options. If a
-      # block is given, it will be passed the newly created query object.
-      #
-      #   Query.new(:query => 'ruby', :with_words => 'sow rspec')
-      #
-      #   Query.new(:exact_phrase => 'fluent interfaces') do |q|
-      #     q.within_past_week = true
-      #   end
+      # block is given, it will be passed the newly created Query object.
       #
       def initialize(options={},&block)
-        @results_per_page = (options[:results_per_page] || RESULTS_PER_PAGE)
-
         @query = options[:query]
 
         @link = options[:link]
@@ -151,175 +107,34 @@ module GScraper
         @with_words = options[:with_words]
         @without_words = options[:without_words]
 
-        @language = options[:language]
-        @region = options[:region]
-        @in_format = options[:in_format]
-        @not_in_format = options[:not_in_format]
-
-        if options[:within_past_day]
-          @within_past_day = options[:within_past_day]
-          @within_past_week = false
-          @within_past_months = false
-          @within_past_year = false
-        elsif options[:within_past_week]
-          @within_past_day = false
-          @within_past_week = options[:within_past_week]
-          @within_past_months = false
-          @within_past_year = false
-        elsif options[:within_past_months]
-          @within_past_day = false
-          @within_past_week = false
-          @within_past_months = options[:within_past_months]
-          @within_past_year = false
-        elsif options[:within_past_year]
-          @within_past_day = false
-          @within_past_week = false
-          @within_past_months = false
-          @within_past_year = options[:within_past_year]
-        else
-          @within_past_day = false
-          @within_past_week = false
-          @within_past_months = false
-          @within_past_year = false
-        end
-
         @numeric_range = options[:numeric_range]
-        @occurrs_within = options[:occurrs_within]
-        @inside_domain = options[:inside_domain]
-        @outside_domain = options[:outside_domain]
-        @rights = options[:rights]
-        @filtered = options[:filtered]
-
-        @similar_to = options[:similar_to]
-        @links_to = options[:links_to]
 
         block.call(self) if block
       end
 
       #
-      # Creates a new Query object from the specified URL. If a block is
-      # given, it will be passed the newly created Query object.
-      #
-      #   Query.from_url('http://www.google.com/search?q=ruby+zen)
+      # Returns the query expression.
       #
-      #   Query.from_url('http://www.google.com/search?q=ruby') do |q|
-      #     q.within_last_month = true
-      #     q.occurrs_within = :title
-      #   end
-      #
-      def self.from_url(url,options={},&block)
-        url = URI.parse(url)
-
-        options[:results_per_page] = url.query_params['num']
-
-        options[:query] = url.query_params['as_q']
-        options[:exact_phrase] = url.query_params['as_epq']
-        options[:with_words] = url.query_params['as_oq']
-        options[:without_words] = url.query_params['as_eq']
-
-        options[:language] = url.query_params['lr']
-        options[:region] = url.query_params['cr']
-
-        case url.query_params['as_ft']
-        when 'i'
-          options[:in_format] = url.query_params['as_filetype']
-        when 'e'
-          options[:not_in_format] = url.query_params['as_filetype']
-        end
-
-        case url.query_params['as_qdr']
-        when 'd'
-          options[:within_past_day] = true
-        when 'w'
-          options[:within_past_week] = true
-        when 'm'
-          options[:within_past_months] = 1
-        when 'm2'
-          options[:within_past_months] = 2
-        when 'm3'
-          options[:within_past_months] = 3
-        when 'm6'
-          options[:within_past_months] = 6
-        when 'y'
-          options[:within_past_year] = true
-        end
-
-        if (url.query_params['as_nlo'] || url.query_params['as_nhi'])
-          options[:numeric_range] = Range.new(url.query_params['as_nlo'].to_i,url.query_params['as_nhi'].to_i)
-        end
-
-        case url.query_params['as_occt']
-        when 'title'
-          options[:occurrs_within] = :title
-        when 'body'
-          options[:occurrs_within] = :body
-        when 'url'
-          options[:occurrs_within] = :url
-        when 'links'
-          options[:occurrs_within] = :links
-        end
-
-        case url.query_params['as_dt']
-        when 'i'
-          options[:inside_domain] = url.query_params['as_sitesearch']
-        when 'e'
-          options[:outside_domain] = url.query_params['as_sitesearch']
-        end
-
-        case url.query_params['as_rights']
-        when '(cc_publicdomain|cc_attribute|cc_sharealike|cc_noncommercial|cc_nonderived)'
-          options[:rights] = Licenses::CC_BY_NC_ND
-        when '(cc_publicdomain|cc_attribute|cc_sharealike|cc_nonderived).-(cc_noncommercial)'
-          options[:rights] = Licenses::CC_BY_SA
-        when '(cc_publicdomain|cc_attribute|cc_sharealike|cc_noncommercial).-(cc_nonderived)'
-          options[:rights] = Licenses::CC_BY_NC
-        when '(cc_publicdomain|cc_attribute|cc_sharealike).-(cc_noncommercial|cc_nonderived)'
-          options[:rights] = Licenses::CC_BY
-        end
-
-        if url.query_params[:safe]=='active'
-          options[:filtered] = true
-        end
-
-        if url.query_params['as_rq']
-          options[:similar_to] = url.query_params['as_rq']
-        elsif url.query_params['as_lq']
-          options[:links_to] = url.query_params['as_lq']
-        end
-
-        return self.new(options,&block)
-      end
-
-      #
-      # Returns the URL that represents the query.
-      #
-      def search_url
-        url = URI(SEARCH_URL)
-        query_expr = []
-
-        set_param = lambda { |param,value|
-          url.query_params[param.to_s] = value if value
-        }
+      def expression
+        expr = []
 
         append_modifier = lambda { |name|
           modifier = instance_variable_get("@#{name}")
 
-          query_expr << "#{name}:#{modifier}" if modifier
+          expr << "#{name}:#{modifier}" if modifier
         }
 
-        join_ops = lambda { |name|
+        append_options = lambda { |name|
           ops = instance_variable_get("@#{name}")
 
           if ops.kind_of?(Array)
-            query_expr << "#{name}:#{ops.join(' ')}"
+            expr << "#{name}:#{ops.join(' ')}"
           elsif ops
-            query_expr << "#{name}:#{ops}"
+            expr << "#{name}:#{ops}"
           end
         }
 
-        set_param.call('num',@results_per_page)
-
-        query_expr << @query if @query
+        expr << @query if @query
 
         append_modifier.call(:link)
         append_modifier.call(:related)
@@ -327,256 +142,30 @@ module GScraper
         append_modifier.call(:site)
         append_modifier.call(:filetype)
 
-        join_ops.call(:allintitle)
+        append_options.call(:allintitle)
         append_modifier.call(:intitle)
-        join_ops.call(:allinurl)
+        append_options.call(:allinurl)
         append_modifier.call(:inurl)
-        join_ops.call(:allintext)
+        append_options.call(:allintext)
         append_modifier.call(:intext)
 
-        unless query_expr.empty?
-          url.query_params['as_q'] = query_expr.join(' ')
-        end
-
-        set_param.call('as_epq',@exact_phrase)
-        set_param.call('as_oq',@with_words)
-        set_param.call('as_eq',@without_words)
-
-        set_param.call('lr',@language)
-        set_param.call('cr',@region)
-
-        if @in_format
-          url.query_params['as_ft'] = 'i'
-          url.query_params['as_filtetype'] = @in_format
-        elsif @not_in_format
-          url.query_params['as_ft'] = 'e'
-          url.query_params['as_filtetype'] = @not_in_format
-        end
-
-        if @within_past_day
-          url.query_params['as_qdr'] = 'd'
-        elsif @within_past_week
-          url.query_params['as_qdr'] = 'w'
-        elsif @within_past_months
-          case @within_past_months
-          when 1
-            url.query_params['as_qdr'] = 'm'
-          when 2
-            url.query_params['as_qdr'] = 'm2'
-          when 3
-            url.query_params['as_qdr'] = 'm3'
-          when 6
-            url.query_params['as_qdr'] = 'm6'
-          end
-        elsif @within_past_year
-          url.query_params['as_qdr'] = 'y'
-        end
-
-        if @numeric_range
-          url.query_params['as_nlo'] = @numeric_range.begin
-          url.query_params['as_nhi'] = @numeric_range.end
-        end
-
-        case @occurrs_within
-        when :title, 'title'
-          url.query_params['as_occt'] = 'title'
-        when :body, 'body'
-          url.query_params['as_occt'] = 'body'
-        when :url, 'url'
-          url.query_params['as_occt'] = 'url'
-        when :links, 'links'
-          url.query_params['as_occt'] = 'links'
+        if @exact_phrase
+          expr << "\"#{@exact_phrase}\""
         end
 
-        if @inside_domain
-          url.query_params['as_dt'] = 'i'
-          url.query_params['as_sitesearch'] = @inside_domain
-        elsif @outside_domain
-          url.query_params['as_dt'] = 'e'
-          url.query_params['as_sitesearch'] = @outside_domain
+        if @with_words.kind_of?(Array)
+          expr << @with_words.join(' OR ')
         end
-
-        case @rights
-        when Licenses::CC_BY_NC_ND
-          url.query_params['as_rights'] = '(cc_publicdomain|cc_attribute|cc_sharealike|cc_noncommercial|cc_nonderived)'
-        when Licenses::CC_BY_SA
-          url.query_params['as_rights'] = '(cc_publicdomain|cc_attribute|cc_sharealike|cc_nonderived).-(cc_noncommercial)'
-        when Licenses::CC_BY_ND
-          url.query_params['as_rights'] = '(cc_publicdomain|cc_attribute|cc_sharealike|cc_noncommercial).-(cc_nonderived)'
-        when Licenses::CC_BY
-          url.query_params['as_rights'] = '(cc_publicdomain|cc_attribute|cc_sharealike).-(cc_noncommercial|cc_nonderived)'
+        
+        if @without_words.kind_of?(Array)
+          expr << @without_words.map { |word| "-#{word}" }.join(' ')
         end
 
-        url.query_params['safe'] = true if @filtered
-
-        if @similar_to
-          url.query_params['as_rq'] = @similar_to
-        elsif @links_to
-          url.query_params['as_lq'] = @links_to
+        if @numeric_range.kind_of?(Range)
+          expr << "#{@numeric_range.begin}..#{@numeric_range.end}"
         end
 
-        return url
-      end
-
-      #
-      # Returns the URL that represents the query at the specific
-      # _page_index_.
-      #
-      def page_url(page_index)
-        url = search_url
-
-        url.query_params['start'] = page_result_offset(page_index)
-        url.query_params['sa'] = 'N'
-
-        return url
-      end
-
-      #
-      # Returns a Page object containing Result objects at the specified
-      # _page_index_. If a _block_ is given, it will be passed the newly
-      # created Page.
-      #
-      def page(page_index,&block)
-        doc = get_page(page_url(page_index))
-
-        new_page = Page.new
-        results = doc.search('//div.g')[0...@results_per_page.to_i]
-
-        results.each_with_index do |result,index|
-          rank = page_result_offset(page_index) + (index + 1)
-          link = result.at('//a.l')
-          title = link.inner_text
-          url = link.get_attribute('href')
-          summary_text = ''
-          cached_url = nil
-          similar_url = nil
-
-          if (content = (result.at('//td.j//font|//td.j/div.sml')))
-            content.children.each do |elem|
-              break if (!(elem.text?) && elem.name=='br')
-
-              summary_text << elem.inner_text
-            end
-            
-            if (cached_link = result.at('nobr/a:first'))
-              cached_url = cached_link.get_attribute('href')
-            end
-
-            if (similar_link = result.at('nobr/a:last'))
-              similar_url = "http://#{SEARCH_HOST}" + similar_link.get_attribute('href')
-            end
-          end
-
-          new_page << Result.new(rank,title,url,summary_text,cached_url,similar_url)
-        end
-
-        block.call(new_page) if block
-        return new_page
-      end
-
-      #
-      # Returns the Results on the first page. If a _block_ is given it
-      # will be passed the newly created Page.
-      #
-      def first_page(&block)
-        page(1,&block)
-      end
-
-      #
-      # Returns the Result at the specified _index_.
-      #
-      def result_at(index)
-        page(result_page_index(index))[page_result_index(index)]
-      end
-
-      #
-      # Returns the first Result on the first_page.
-      #
-      def top_result
-        result_at(1)
-      end
-
-      #
-      # Iterates over the results at the specified _page_index_, passing
-      # each to the given _block_.
-      #
-      #   query.each_on_page(2) do |result|
-      #     puts result.title
-      #   end
-      #
-      def each_on_page(page_index,&block)
-        page(page_index).each(&block)
-      end
-
-      #
-      # Iterates over the results on the first page, passing each to the
-      # given _block_.
-      #
-      #   query.each_on_first_page do |result|
-      #     puts result.url
-      #   end
-      #
-      def each_on_first_page(&block)
-        each_on_page(1,&block)
-      end
-
-      #
-      # Returns a SponsoredLinks object containing SponsoredAd objects of
-      # the query. If a _block_ is given, it will be passed the newly
-      # created Page.
-      #
-      def sponsored_links(&block)
-        doc = get_page(search_url)
-        new_links = SponsoredLinks.new
-
-        # top and side ads
-        doc.search('//a[@id="pa1"]|//a[@id*="an"]').each do |link|
-          title = link.inner_text
-          url = "http://#{SEARCH_HOST}" + link.get_attribute('href')
-
-          new_links << SponsoredAd.new(title,url)
-        end
-
-        block.call(new_links) if block
-        return new_links
-      end
-
-      #
-      # Returns the first sponsored link on the first page of results.
-      #
-      def top_sponsored_link
-        top_sponsored_links.first
-      end
-
-      #
-      # Iterates over the sponsored links on the first page of
-      # results passing each to the specified _block_.
-      #
-      def each_sponsored_link(&block)
-        sponsored_links.each(&block)
-      end
-
-      protected
-
-      #
-      # Returns the rank offset for the specified _page_index_.
-      #
-      def page_result_offset(page_index)
-        (page_index.to_i - 1) * @results_per_page.to_i
-      end
-
-      #
-      # Returns the in-Page index of the _result_index_.
-      #
-      def page_result_index(result_index)
-        (result_index.to_i - 1) % @results_per_page.to_i
-      end
-
-      #
-      # Returns the page index for the specified _result_index_
-      #
-      def result_page_index(result_index)
-        ((result_index.to_i - 1) / @results_per_page.to_i) + 1
+        return expr.join(' ')
       end
 
     end