Wiki_Category_Extractor 0.0.1

data/lib/Wiki_Category_Extractor.rb

@@ -0,0 +1,40 @@
+class Wiki_Category_Extractor
+def self.extract(a,pagelimit)
+@pages = {}
+@categories = []
+@cat_links = []
+
+@categories << a
+@cat_links << "http://en.wikipedia.org/wiki/Category:" + a
+
+while @categories != []
+	@doc = Nokogiri::HTML(open(@cat_links[0]))
+	@x = @doc.css("#mw-subcategories")
+	j = 0
+	while j < @x.css("a").length
+		@categories << @x.css("a")[j].text
+		@cat_links << "http://en.wikipedia.org" + @x.css("a")[j].attributes["href"].value
+		j = j + 1
+	end
+	
+	@y = @doc.css("#mw-pages")
+	j = 0
+	while j < @y.css("a").length
+		if @y.css("a")[j].text != ""
+		@pages[@y.css("a")[j].text] = "http://en.wikipedia.org" + @y.css("a")[j].attributes["href"].value
+		end
+		j = j + 1
+	end
+	
+	if @pages.length > pagelimit
+	break
+	end
+
+	puts @categories[0]
+	@categories.delete_at(0)
+	@cat_links.delete_at(0)	
+
+end
+	return @pages
+end
+end

metadata

@@ -0,0 +1,45 @@
+--- !ruby/object:Gem::Specification
+name: Wiki_Category_Extractor
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+  prerelease: 
+platform: ruby
+authors:
+- Prateek Papriwal
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2013-12-20 00:00:00.000000000 Z
+dependencies: []
+description: Category Extractor
+email: papriwalprateek@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/Wiki_Category_Extractor.rb
+homepage: http://rubygems.org/gems/Wiki_Category_Extractor
+licenses: []
+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: 1.8.11
+signing_key: 
+specification_version: 3
+summary: ! '!!!'
+test_files: []