ralert 0.2.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: db68c2b3754d20707792ffde3ef60b33a18bbec3
+  data.tar.gz: af6df9818eaa6513b5393becf25dd7a87d226324
+SHA512:
+  metadata.gz: 06e3243385de2ced3f3051cf067a0fa47f4826abbd6dc182f30e95cb8b9747b866610304cd5747ea0effbb2294fb890def0f88332a2e7f0bbe455312ef1207e7
+  data.tar.gz: 643b0b4d7efd401c33b94fd9cc36f374dc0fbf660cb0dd502114738bc315cc8c01b11c0e1b067e22eda334e5a03a34ff264e3af2ae426bb8bd25217fed0c663b

data/lib/ralert.rb

@@ -0,0 +1,161 @@
+#!/usr/bin/env ruby
+#encoding: UTF-8
+
+require 'result'
+require 'search-options'
+
+require 'rubygems'
+require 'nokogiri'
+require 'open-uri'
+
+class Ralert
+  attr_accessor :results, :next_page
+  @page
+
+  def initialize(query, options = nil)
+    @results = Array.new
+
+    # Take care of spaces and other special
+    # characters.
+    query = transform_query(query) if query.index(" ")
+    
+    # Instantiante a new search options object if
+    # the user hasn't provided any when calling
+    # the class
+    options = SearchOptions.new unless !options.nil?
+
+    uri = construct_uri(query, options)
+    @results = perform_search(uri)
+  end
+
+  ##
+  # Takes the query string and a SearchOptions object
+  # and constructs a new search query.
+  #
+  def construct_uri(query, options)
+    base_uri = "https://google.com/search?q="
+    
+    if !options.literal.nil?
+    	query = "\"" + query + "\"" 
+    end
+
+    if !options.date_range.nil?
+      query += "&tbs=qdr:#{options.date_range}"
+    else
+      query += "&tbs=qdr:w"
+    end
+
+    if options.sort_by.nil?
+      query += ",sbd:1"
+    end
+    
+    if !options.safe.nil?
+      query += "&safe=on"
+    else
+      query += "&safe=off"
+    end
+
+    if options.mode.nil?
+      query += "&tbm=nws"
+    end
+
+    return base_uri + query
+  end
+
+  ##
+  # Given a URI, performs a request and scans
+  # the resulting page with the Nokogiri parser.
+  #
+  def perform_search(uri)
+    html = open(uri)
+    @page = Nokogiri::HTML(html.read)
+    @page.encoding = 'utf-8'
+
+    parse_results
+  end
+  
+  ##
+  # Parses the page resulting from the search query
+  # and returns the search items found in a result 
+  # array.
+  #
+  def parse_results
+    cur_results = Array.new
+
+    @page.search('li.g').each do |item|
+      cur_results << node_from_item(item)
+    end
+    
+    update_next_page unless next_page_missing
+    @results += cur_results
+    
+    return cur_results
+  end
+  
+  ##
+  # Takes an HTML li block which represents a search 
+  # result and extracts all the information from it
+  # like: a title, a link, a (relative) date and the
+  # articles source.
+  #
+  def node_from_item(item)
+    result_node = Result.new
+    title = '' 
+
+    link = item.at('h3.r a')
+    meta = item.search('div.slp span.f').inner_html
+    
+    link.children.each do |c|
+      title += c
+    end
+
+    result_node.title  = title
+    result_node.source = meta.split('-')[0]
+    result_node.date = meta.split('-')[1]
+    result_node.link = link['href'].gsub!(/\/url\?q\=/, '').gsub!(/\&sa\=.*/,'')
+
+    return result_node
+  end
+
+  ##
+  # Checks if this is the last of the search result pages
+  # available.
+  #
+  def next_page_missing
+    return @page.at_css("table#nav tr td.b:last-child").at_css("a").nil?
+  end
+  
+  ##
+  # Updates the @next_page instance variable to point
+  # to the next search result page.
+  #
+  def update_next_page
+    next_uri = @page.at_css("table#nav tr td.b:last-child").at_css("a")['href']
+    @next_page = "http://www.google.com" + next_uri
+  end
+
+  ##
+  # Performs the search-parse-update routine on the next 
+  # page of search results if available.
+  #
+  def next_results(page_number = 1)
+    page_number.times.each do
+      !next_page_missing ? perform_search(@next_page) : break
+    end
+
+    return @results
+  end
+
+  ##
+  # Takes a text query and substitutes spaces for plus signs
+  # as the google search engine expects to be fed with.
+  #
+  def transform_query(q)
+    return q.gsub!(/\s/, '+')
+  end
+  
+  def each(&blk)
+    @results.each(&blk)
+  end
+
+end

data/lib/result.rb

@@ -0,0 +1,22 @@
+class Result 
+    attr_accessor :link, :title, :source, :date
+    def initialize(link='', title='', source = '', date = '')
+	@link = link
+	@title = title
+	@source = source
+	@date = date
+    end
+
+    def ==(other)
+        self.class === other and
+        other.author == @link and
+        other.title == @title and 
+	other.source = @source and 
+	other.date == @date
+    end
+
+    alias eql? ==
+    def hash
+        @link.hash ^ @title.hash ^ @source.hash ^ @date.hash # XOR
+    end
+end

data/lib/search-options.rb

@@ -0,0 +1,12 @@
+class SearchOptions
+  attr_accessor :literal, :sort_by, :date_range, :safe, :mode
+
+  def initialize(literal = nil, sort_by = nil, date_range = nil, safe = nil, mode = nil)
+    @literal = literal
+    @sort_by = sort_by
+    @date_range = date_range
+    @safe = safe
+    @mode = mode
+  end
+
+end

metadata

@@ -0,0 +1,47 @@
+--- !ruby/object:Gem::Specification
+name: ralert
+version: !ruby/object:Gem::Version
+  version: 0.2.0
+platform: ruby
+authors:
+- Spyros Livathinos
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2013-08-22 00:00:00.000000000 Z
+dependencies: []
+description: Queries the Google search engine and returns an array of Result objects.
+email: livathinos.spyros@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/ralert.rb
+- lib/result.rb
+- lib/search-options.rb
+homepage: http://thinkcactus.com
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.0.3
+signing_key: 
+specification_version: 4
+summary: Ralert is a simple Ruby gem for parsing Google News search queries programmatically.
+test_files: []
+has_rdoc: