baiduserp 2.0.2 → 2.0.3

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 925a7067129b157540753444a3e54c78676d2a9f
+  data.tar.gz: 908b484a96fbbed83429804b09d50ff1b7828ff5
+SHA512:
+  metadata.gz: 0c7fb5fd7f6fa734934522cfe738f0725be40c4f1504f0c87fa0e1d08bff5fd8d8c19580701a2dd17d014f02df9b2b850aef752f535a614d4fd62ba7044b001d
+  data.tar.gz: 793333893a7e9c51f936d21a861066dfa47237349976b1bdee59a1046c4641e059b829530d5dbaffc51322cf3b2a035c1e71b8323e8b0cfecc7803695a892a54

data/lib/baiduserp/parser.rb

@@ -6,7 +6,7 @@ require 'baiduserp/helper'
 
 module Baiduserp
   class Parser
-    Dir[File.expand_path('../parser/*.rb', __FILE__)].each{|f| require f}
+    Dir[File.expand_path('../../parsers/*.rb', __FILE__)].each{|f| require f}
 
     def parse(html)
       @file = Hash.new
@@ -15,9 +15,10 @@ module Baiduserp
       @file[:html] = html
       @file[:doc] = Nokogiri::HTML(html)
 
-      self.class.constants.each do |m|
-        #puts m
-        eval "@serp[:#{m.downcase}] = #{m}.parse @file"
+      self.methods.each do |m|
+        next unless m =~ /^_parse_/
+        #p m
+        @serp[m.to_s.sub('_parse_','').to_sym] = self.send m,@file
         #p @serp.keys
       end
 

data/lib/baiduserp/version.rb

@@ -1,3 +1,3 @@
 module Baiduserp
-  VERSION = "2.0.2"
+  VERSION = "2.0.3"
 end

data/lib/{baiduserp/parser → parsers}/ads_right.rb

@@ -1,5 +1,5 @@
-module Baiduserp::Parser::Ads_Right
-  def self.parse(file)
+class Baiduserp::Parser
+  def _parse_ads_right(file)
     result = []
     file[:doc].search('div.EC_im').each do |div|
       r = {}

data/lib/{baiduserp/parser → parsers}/ads_top.rb

@@ -1,5 +1,5 @@
-module Baiduserp::Parser::Ads_Top
-  def self.parse(file)
+class Baiduserp::Parser
+  def _parse_ads_top(file)
     result = []
     file[:doc].search('div.ec_pp_f').each do |div|
       id = div['id'].to_i

data/lib/parsers/pinpaizhuanqu.rb

@@ -0,0 +1,5 @@
+class Baiduserp::Parser
+  def _parse_pinpaizhuanqu(file)
+    file[:html].include? 'bs.baidu.com/adcoup-mat'
+  end
+end

data/lib/{baiduserp/parser → parsers}/ranks.rb

@@ -1,13 +1,17 @@
-module Baiduserp::Parser::Ranks
-  def self.parse(file)
+class Baiduserp::Parser
+  def _parse_ranks(file)
     result = []
     file[:doc].search("//table").each do |table|
+      next if table.nil?
       id = table['id'].to_i
       next unless id > 0
       r = {:rank => id}
 
-      url = table.search('h3/a').first['href']
-      url = Baiduserp::Client.get(url).headers['location'] if url.include?('http://www.baidu.com/link?')
+      url = table.search('h3/a').first
+      unless url.nil?
+        url = url['href'] 
+        url = Baiduserp::Client.get(url).headers['location'] if url.include?('http://www.baidu.com/link?')
+      end
       r[:url] = url
 
       r[:title] = Baiduserp::Helper.get_content_safe(table.search('h3'))

data/lib/{baiduserp/parser → parsers}/related_keywords.rb

@@ -1,5 +1,5 @@
-module Baiduserp::Parser::Related_Keywords
-  def self.parse(file)
+class Baiduserp::Parser
+  def _parse_related_keywords(file)
     result = []
     file[:doc].search('div[@id="rs"]').each do |rs|
       rs.css('a').each do |link|

data/lib/{baiduserp/parser → parsers}/result_num.rb

@@ -1,7 +1,7 @@
 # coding: utf-8
 
-module Baiduserp::Parser::Result_Num
-  def self.parse(file)
+class Baiduserp::Parser
+  def _parse_result_num(file)
     html = file[:html]
     str = html.scan(/找到相关结果(.*)个/).join
     str = str.gsub('约','')

data/lib/parsers/right_hotel.rb

@@ -0,0 +1,12 @@
+class Baiduserp::Parser
+  def _parse_right_hotel(file)
+    rh = file[:doc].search('div[@tpl="right_hotel"]')
+    return nil if rh.nil?
+
+    rh = rh.first
+    return nil if rh.nil?
+    title = Baiduserp::Helper.get_content_safe(rh.search('div.opr-hotel-title'))
+    
+    {:title => title}
+  end
+end

data/lib/parsers/right_personinfo.rb

@@ -0,0 +1,13 @@
+class Baiduserp::Parser
+  def _parse_right_personinfo(file)
+    rp = file[:doc].search('div[@tpl="right_personinfo"]')
+    return nil if rp.nil?
+
+    title = Baiduserp::Helper.get_content_safe rp.search('span.opr-personinfo-subtitle-large')
+    info = Baiduserp::Helper.get_content_safe rp.search('div.opr-personinfo-info')
+    source = Baiduserp::Helper.get_content_safe rp.search('div.opr-personinfo-source a')
+    
+    return nil if title.nil? && info.nil? && source.nil?
+    {:title => title, :info => info, :source => source}
+  end
+end

data/lib/parsers/right_relaperson.rb

@@ -0,0 +1,17 @@
+class Baiduserp::Parser
+  def _parse_right_relaperson(file)
+    relapersons = file[:doc].search('div[@tpl="right_relaperson"]')
+    return nil if relapersons.nil?
+
+    result = []
+    relapersons.each do |rr|
+      title = rr.search('span.opr-relaperson-subtitle-tip').first.content
+      r = []
+      rr.search('p.opr-relaperson-name').each do |p|
+        r << p.content
+      end
+      result << {:title => title, :names => r}
+    end
+    result
+  end
+end

data/lib/parsers/right_weather.rb

@@ -0,0 +1,14 @@
+class Baiduserp::Parser
+  def _parse_right_weather(file)
+    rw = file[:doc].search('div[@tpl="right_weather"]')
+    return nil if rw.nil?
+
+    rw = rw.first
+    return nil if rw.nil?
+
+    title = Baiduserp::Helper.get_content_safe(rw.search('div.opr-weather-title'))
+    week = rw.search('a.opr-weather-week').first['href']
+    
+    {:title => title, :week => week}
+  end
+end

metadata

@@ -1,46 +1,41 @@
 --- !ruby/object:Gem::Specification
 name: baiduserp
 version: !ruby/object:Gem::Version
-  version: 2.0.2
-  prerelease: 
+  version: 2.0.3
 platform: ruby
 authors:
 - MingQian Zhang
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-06-20 00:00:00.000000000 Z
+date: 2013-06-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
   name: httparty
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '>='
       - !ruby/object:Gem::Version
         version: '0'
 description: Parse Baidu SERP result page.
@@ -53,39 +48,42 @@ extra_rdoc_files: []
 files:
 - lib/baiduserp/client.rb
 - lib/baiduserp/helper.rb
-- lib/baiduserp/parser/ads_right.rb
-- lib/baiduserp/parser/ads_top.rb
-- lib/baiduserp/parser/pinpaizhuanqu.rb
-- lib/baiduserp/parser/ranks.rb
-- lib/baiduserp/parser/related_keywords.rb
-- lib/baiduserp/parser/result_num.rb
 - lib/baiduserp/parser.rb
 - lib/baiduserp/version.rb
 - lib/baiduserp.rb
+- lib/parsers/ads_right.rb
+- lib/parsers/ads_top.rb
+- lib/parsers/pinpaizhuanqu.rb
+- lib/parsers/ranks.rb
+- lib/parsers/related_keywords.rb
+- lib/parsers/result_num.rb
+- lib/parsers/right_hotel.rb
+- lib/parsers/right_personinfo.rb
+- lib/parsers/right_relaperson.rb
+- lib/parsers/right_weather.rb
 - bin/baiduserp
 - README.md
 homepage: https://github.com/mqzhang/baiduserp
 licenses: []
+metadata: {}
 post_install_message: 
 rdoc_options: []
 require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
-  none: false
   requirements:
-  - - ! '>='
+  - - '>='
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
-  none: false
   requirements:
-  - - ! '>='
+  - - '>='
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 1.8.25
+rubygems_version: 2.0.0
 signing_key: 
-specification_version: 3
+specification_version: 4
 summary: Baidu SERP
 test_files: []

data/lib/baiduserp/parser/pinpaizhuanqu.rb

@@ -1,5 +0,0 @@
-module Baiduserp::Parser::PinPaiZhuanQu
-  def self.parse(file)
-    file[:html].include? 'bs.baidu.com/adcoup-mat'
-  end
-end