newscrapi 0.0.11

data/test/test_scrapper.rb

@@ -0,0 +1,257 @@
+require 'helper'
+require 'mocha'
+
+class TestScrapper < Test::Unit::TestCase
+
+  Newscrapi::Scrapper.default_config_file = nil
+
+  context "on settings without sanitization tags" do
+    setup do
+      @scrapper = Newscrapi::Scrapper.new
+      @scrapper.instance_eval do
+        content_mapping do
+          url_pattern /.*/
+          content_at '//div[@id="itext_content"]'
+        end
+      end
+      content = File.open("#{File.dirname(__FILE__)}/test_pages/pretty.html").read 
+      stringio = StringIO.new(content)
+      Kernel.expects(:open).returns(stringio)
+    end
+    should 'not sanitize' do
+      assert !@scrapper.scrap_content('http://www.pretty.url/fsdsd').nil?
+    end
+  end
+
+  context "on common settings" do
+    setup do
+      @scrapper = Newscrapi::Scrapper.new
+      @scrapper.instance_eval do
+        content_mapping do
+          url_pattern /^http:\/\/www\.pretty\.url/
+          content_at '//div[@id="failing_content"]'
+          content_at '//div[@id="itext_content"]'
+        end
+
+        content_mapping do
+          url_pattern /^http:\/\/www\.twopatterns\.url/
+          content_at '//div[@id="failing_content"]'
+          content_at '//div[@id="itext_content"]'
+        end
+
+        content_mapping do
+          url_pattern /^http:\/\/www\.twopatterns\.url/
+          content_at '//div[@id="itext_second_content"]'
+        end
+
+        content_mapping do
+          url_pattern /^http:\/\/www\.skipper\.url/
+        end
+
+        content_mapping do
+          url_pattern /^http:\/\/www\.skipper\.url/
+          content_at '//div[@id="never_should_be_here"]'
+        end
+
+        sanitize_tags do
+          {:elements => ['p','br', 'b', 'em', 'i', 'strong', 'u', 'a', 'h1', 'h2', 'h3', 'li', 'ol', 'ul'], \
+                       :attributes => { 'a' => ['href'] }}
+        end
+      end
+    end
+
+    should "identify the correct content mapper" do
+      content_mapper = @scrapper.matching_content_mapper('http://www.pretty.url/fsdsd')
+      assert !content_mapper.nil?
+      assert_equal /^http:\/\/www\.pretty\.url/, content_mapper.url_pattern_regexp
+    end
+
+    context "for known sources with expected content scrapping" do
+      setup do
+        pretty_content = File.open("#{File.dirname(__FILE__)}/test_pages/pretty.html").read 
+        stringio = StringIO.new(pretty_content)
+        Kernel.expects(:open).returns(stringio)
+        @entry_content = @scrapper.scrap_content('http://www.pretty.url/fsdsd')
+      end
+      should("identify the content") do
+        assert_match(%r{<p><strong>This is a strong text</strong></p>}, @entry_content)
+      end
+    end
+
+    context "for known pages with unexpected content scrapping" do
+      setup do
+        ugly_content = File.open("#{File.dirname(__FILE__)}/test_pages/ugly.html").read
+        stringio = StringIO.new(ugly_content)
+        Kernel.expects(:open).returns(stringio)
+        @entry_content = @scrapper.scrap_content('http://www.pretty.url/hsdae')
+      end
+      should("return nil") { assert_nil @entry_content }
+    end
+
+    context "for unknown pages" do
+      setup { @entry_content = @scrapper.scrap_content('http://www.unknown.url/hsdae') }
+      should("return nil") { assert_nil @entry_content }
+    end
+
+    context "multiple matching url patterns" do
+      setup do
+        twocontent = File.open("#{File.dirname(__FILE__)}/test_pages/twocontent.html").read
+        stringio = StringIO.new(twocontent)
+        Kernel.expects(:open).with('http://www.twopatterns.url').returns(stringio)
+        @entry_content = @scrapper.scrap_content('http://www.twopatterns.url')
+      end
+      should "match the first content" do
+        assert_equal 'The first one is matched', @entry_content
+      end
+    end
+
+    context "skipper patterns" do
+      setup do
+        Kernel.expects(:open).with('http://www.skipper.url/fdgsw').never
+        @entry_content = @scrapper.scrap_content('http://www.skipper.url/fdgsw')
+      end
+      should("not match enything") { assert_nil @entry_content }
+    end
+
+    context "on already downloaded document" do
+      setup do
+        pretty_content = File.open("#{File.dirname(__FILE__)}/test_pages/pretty.html").read 
+        Kernel.expects(:open).never
+        @scrapped_content = @scrapper.scrap_content('http://www.pretty.url/hsdae',
+                                                    :use_page => pretty_content)
+      end
+      should "scrap from the provided full page" do
+        assert_match(%r{<p><strong>This is a strong text</strong></p>}, @scrapped_content)
+      end
+    end
+
+    context "on scrapping with feedzirra" do
+      setup do
+        require 'newscrapi/feedzirra'
+        require 'sax-machine'
+        require 'feedzirra/parser/rss_entry'
+        require 'feedzirra/parser/atom_entry'
+      end
+
+      context "feed entry with not parsable remote content, but with feed content set" do
+        setup do
+          @feed_entries = [ Feedzirra::Parser::RSSEntry.new, Feedzirra::Parser::AtomEntry.new ]
+          @feed_entries.each do |feed_entry|
+            feed_entry.url = 'http://www.unknown.url/wedhsf'
+            feed_entry.content = 'Pretty well written content is this.'
+          end
+          Kernel.expects(:open).with('http://www.unknown.url/wedhsf').never
+        end
+        should("return the original feed content") do
+          @feed_entries.each do |feed_entry|
+            assert_equal 'Pretty well written content is this.', feed_entry.scrap_content(@scrapper)
+            feed_entry.scrap_content!(@scrapper)
+            assert_equal 'Pretty well written content is this.', feed_entry.content
+          end
+        end
+      end
+
+      context "on feed entry with url and scrapping with full_page" do
+        setup do
+          @feed_entries = [ Feedzirra::Parser::RSSEntry.new, Feedzirra::Parser::AtomEntry.new ]
+          @feed_entries.each do |feed_entry|
+            feed_entry.url = 'http://www.pretty.url/wedhsf'
+          end
+          @pretty_content = File.open("#{File.dirname(__FILE__)}/test_pages/pretty.html").read
+          Kernel.expects(:open).never
+        end
+        should("return the original feed content") do
+          @feed_entries.each do |feed_entry|
+            assert_match(%r{<p><strong>This is a strong text</strong></p>},
+              feed_entry.scrap_content(@scrapper, :use_page => @pretty_content))
+            feed_entry.scrap_content!(@scrapper, :use_page => @pretty_content)
+            assert_match(%r{<p><strong>This is a strong text</strong></p>}, feed_entry.content)
+          end
+        end
+      end
+    end
+
+    context "on failing scrapping" do
+      setup do
+        Kernel.expects(:open).raises(Exception, 'something failed')
+        @exception_handle_flag = nil
+        @scrapper.rescue_scrapping do |exception, url|
+          @exception_handle_flag = exception.message
+          @exception_url = url
+        end
+      end
+      should "catch the exception and handle it" do
+        assert_nil @scrapper.scrap_content('http://www.pretty.url')
+        assert_equal 'something failed', @exception_handle_flag
+        assert_equal 'http://www.pretty.url', @exception_url
+      end
+    end
+
+    context "on missing url matcher" do
+      setup do
+        Kernel.expects(:open).never
+        @missing_url_matcher_flag = nil
+        @scrapper.missing_url_matcher do |url|
+          @missing_url_matcher_flag = url
+        end
+        @scrapper.scrap_content('http://missing.url.matcher')
+      end
+      should "call the handler block" do
+        assert_equal 'http://missing.url.matcher', @missing_url_matcher_flag
+      end
+    end
+
+    context "on matching content mapper finding empty content" do
+      setup do
+        twocontent = File.open("#{File.dirname(__FILE__)}/test_pages/pretty_missing_content.html").read
+        stringio = StringIO.new(twocontent)
+        Kernel.expects(:open).with('http://www.pretty.url').returns(stringio)
+        @missing_content_flag = 0
+        @scrapper.missing_content do |url|
+          @missing_content_flag += 1
+        end
+        @scrapper.scrap_content('http://www.pretty.url')
+      end
+      should "call the handler block only once" do
+        assert_equal 1, @missing_content_flag
+      end
+    end
+  end
+
+  context "on setting default content scrapper" do
+    setup { @scrapper = Newscrapi::Scrapper.create_new_default }
+    should "set the default to the recently created" do
+      assert_equal @scrapper, Newscrapi::Scrapper.default
+    end
+    context "when changing default content scrapper" do
+      setup { @new_scrapper = Newscrapi::Scrapper.new.set_as_default }
+      should "change the default to the new content scrapper" do
+        assert_equal @new_scrapper, Newscrapi::Scrapper.default
+      end
+    end
+
+    context "for feed entry" do
+      setup do
+        @feed_entry = Feedzirra::Parser::RSSEntry.new
+        @feed_entry.url = 'http://www.unknown.url/gerhe'
+        @feed_entry.content = 'We should get this.'
+      end
+      should("scrap content by the default scrapper") do
+        assert_equal 'We should get this.', @feed_entry.scrap_content
+      end
+    end
+  end
+
+  context "on giving standard package of settings" do
+    setup do
+      @scrapper = Newscrapi::Scrapper.new
+      @scrapper.report_to_stderr
+    end
+    should "have the reporters set" do
+      assert !@scrapper.scrapping_exception_handler_block.nil?
+      assert !@scrapper.missing_url_matcher_handler_block.nil?
+      assert !@scrapper.missing_content_handler_block.nil?
+    end
+  end
+end
+

metadata

@@ -0,0 +1,191 @@
+--- !ruby/object:Gem::Specification 
+name: newscrapi
+version: !ruby/object:Gem::Version 
+  hash: 9
+  prerelease: false
+  segments: 
+  - 0
+  - 0
+  - 11
+  version: 0.0.11
+platform: ruby
+authors: 
+- Gyorgy Frivolt
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2010-10-11 00:00:00 +02:00
+default_executable: 
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  name: thoughtbot-shoulda
+  prerelease: false
+  requirement: &id001 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 35
+        segments: 
+        - 2
+        - 10
+        - 2
+        version: 2.10.2
+  type: :development
+  version_requirements: *id001
+- !ruby/object:Gem::Dependency 
+  name: mocha
+  prerelease: false
+  requirement: &id002 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 43
+        segments: 
+        - 0
+        - 9
+        - 8
+        version: 0.9.8
+  type: :development
+  version_requirements: *id002
+- !ruby/object:Gem::Dependency 
+  name: nokogiri
+  prerelease: false
+  requirement: &id003 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 5
+        segments: 
+        - 1
+        - 4
+        - 1
+        version: 1.4.1
+  type: :runtime
+  version_requirements: *id003
+- !ruby/object:Gem::Dependency 
+  name: rchardet
+  prerelease: false
+  requirement: &id004 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 3
+        segments: 
+        - 0
+        version: "0"
+  type: :runtime
+  version_requirements: *id004
+description: If you want to cut only the content of pages, without any other part (like the menu, header, footer, commercials, etc.), you might find this gem very handy. A DSL is also defined for nifty definitions for your screen scrapping and sanitization.
+email: gyorgy.frivolt@gmail.com
+executables: []
+
+extensions: []
+
+extra_rdoc_files: 
+- LICENSE
+- README.rdoc
+files: 
+- .document
+- .gitignore
+- LICENSE
+- README.rdoc
+- Rakefile
+- VERSION
+- config/content_scrapper.rb
+- doc/classes/ContentMapping.html
+- doc/classes/ContentMapping.src/M000001.html
+- doc/classes/ContentMapping.src/M000002.html
+- doc/classes/ContentMapping.src/M000003.html
+- doc/classes/ContentMapping.src/M000004.html
+- doc/classes/ContentMapping.src/M000005.html
+- doc/classes/ContentMapping.src/M000006.html
+- doc/classes/ContentScrapper.html
+- doc/classes/ContentScrapper.src/M000007.html
+- doc/classes/ContentScrapper.src/M000008.html
+- doc/classes/ContentScrapper.src/M000009.html
+- doc/classes/ContentScrapper.src/M000010.html
+- doc/classes/ContentScrapper.src/M000011.html
+- doc/classes/ContentScrapper.src/M000012.html
+- doc/classes/ContentScrapper.src/M000013.html
+- doc/classes/ContentScrapper.src/M000014.html
+- doc/classes/ContentScrapper.src/M000015.html
+- doc/classes/ContentScrapper.src/M000016.html
+- doc/classes/Feedzirra.html
+- doc/classes/Feedzirra/FeedEntryUtilities.html
+- doc/classes/Feedzirra/FeedEntryUtilities.src/M000017.html
+- doc/classes/Feedzirra/FeedEntryUtilities.src/M000018.html
+- doc/created.rid
+- doc/files/lib/content_scrapper/content_mapping_rb.html
+- doc/files/lib/content_scrapper/feedzirra_rb.html
+- doc/files/lib/content_scrapper_rb.html
+- doc/fr_class_index.html
+- doc/fr_file_index.html
+- doc/fr_method_index.html
+- doc/index.html
+- doc/rdoc-style.css
+- lib/newscrapi.rb
+- lib/newscrapi/encoding.rb
+- lib/newscrapi/feedzirra.rb
+- lib/newscrapi/mapping.rb
+- lib/newscrapi/scrapper.rb
+- lib/newscrapi/testing.rb
+- newscrapi.gemspec
+- rails/init.rb
+- test/helper.rb
+- test/test_encoding.rb
+- test/test_mapping.rb
+- test/test_pages.rb
+- test/test_pages/cdata.html
+- test/test_pages/page_without_encoding_meta_tag.html
+- test/test_pages/pretty.html
+- test/test_pages/pretty_missing_content.html
+- test/test_pages/twocontent.html
+- test/test_pages/ugly.html
+- test/test_pages/utf-8_page.html
+- test/test_pages/windows-1250_page.html
+- test/test_scrapper.rb
+has_rdoc: true
+homepage: http://github.com/fifigyuri/newscrapi
+licenses: []
+
+post_install_message: 
+rdoc_options: 
+- --charset=UTF-8
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
+      - 0
+      version: "0"
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
+      - 0
+      version: "0"
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.3.7
+signing_key: 
+specification_version: 3
+summary: Gem for those who want to screen scrap only the content part of web pages, blogs or articles.
+test_files: 
+- test/helper.rb
+- test/test_encoding.rb
+- test/test_mapping.rb
+- test/test_pages.rb
+- test/test_scrapper.rb