highscore 0.2.1 → 0.3.1

data/History.txt

@@ -1,3 +1,7 @@
+== 0.3.1 / 2012-01-21
+
+* refactored and threw away all that array stuff that nobody wants
+
 == 0.2.1 / 2012-01-15
 
 * added parameter multiplier to set a general multiplier for the ranking

data/README.md

@@ -19,8 +19,15 @@ Examples
       set :long_words_threshold, 15
     end
 
-    text.keywords # => Hash
-    text.keywords.top(50) # => Array
+    # get all keywords
+    text.keywords.rank => Array
+
+    # get the top 50 keywords
+    text.keywords.top(50).each do |keyword|
+      keyword.text   # => keyword text
+      keyword.weight # => rank weight (float)
+    end
+
 
 Requirements
 ------------

data/highscore.gemspec

@@ -0,0 +1,34 @@
+# -*- encoding: utf-8 -*-
+
+Gem::Specification.new do |s|
+  s.name = "highscore"
+  s.version = "0.3.0"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
+  s.authors = ["Dominik Liebler"]
+  s.date = "2012-01-21"
+  s.description = "Rank keywords in long texts."
+  s.email = "liebler.dominik@googlemail.com"
+  s.executables = ["highscore"]
+  s.extra_rdoc_files = ["History.txt", "bin/highscore"]
+  s.files = [".gitignore", "History.txt", "README.md", "Rakefile", "bin/highscore", "lib/highscore.rb", "lib/highscore/content.rb", "lib/highscore/keyword.rb", "lib/highscore/keywords.rb", "test/highscore/test_content.rb", "test/highscore/test_keyword.rb", "test/highscore/test_keywords.rb", "test/test_highscore.rb", "version.txt"]
+  s.homepage = "http://thewebdev.de"
+  s.rdoc_options = ["--main", "README.md"]
+  s.require_paths = ["lib"]
+  s.rubyforge_project = "highscore"
+  s.rubygems_version = "1.8.11"
+  s.summary = "Rank keywords in long texts."
+  s.test_files = ["test/highscore/test_content.rb", "test/highscore/test_keyword.rb", "test/highscore/test_keywords.rb", "test/test_highscore.rb"]
+
+  if s.respond_to? :specification_version then
+    s.specification_version = 3
+
+    if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
+      s.add_development_dependency(%q<bones>, [">= 3.7.3"])
+    else
+      s.add_dependency(%q<bones>, [">= 3.7.3"])
+    end
+  else
+    s.add_dependency(%q<bones>, [">= 3.7.3"])
+  end
+end

data/lib/highscore/content.rb

@@ -34,36 +34,23 @@ module Highscore
     #   keywords -> Keywords
     #
     def keywords
-      keywords = Keywords.new(0)
+      keywords = Keywords.new
 
-      find_keywords.each do |k|
+      Keywords.find_keywords(@content).each do |text|
         weight = @emphasis[:multiplier]
 
-        if k.length >= @emphasis[:long_words_threshold]
+        if text.length >= @emphasis[:long_words_threshold]
           weight *= @emphasis[:long_words]
         end
 
-        if k[0] == k[0].upcase
+        if text[0] == text[0].upcase
           weight *= @emphasis[:upper_case]
         end
 
-        keywords[k] += weight
+        keywords << Highscore::Keyword.new(text, weight)
       end
 
       keywords
     end
-
-    private
-
-    # find keywords in the content and rate them
-    #
-    def find_keywords
-      keywords = @content.scan(/\w+/)
-      keywords.delete_if do |x|
-        x.match(/^[\d]+(\.[\d]+){0,1}$/) or x.length <= 2
-      end
-
-      keywords.sort
-    end
   end
 end

data/lib/highscore/keyword.rb

@@ -0,0 +1,19 @@
+module Highscore
+
+  # keywords read from the content
+  #
+  class Keyword
+    attr_accessor :weight, :text
+
+    # init a keyword
+    def initialize(text, weight)
+      @text = text
+      @weight = weight.to_f
+    end
+
+    # sort keywords
+    def <=>(other)
+      other.weight <=> @weight
+    end
+  end
+end

data/lib/highscore/keywords.rb

@@ -1,7 +1,32 @@
+# external
+require 'digest/sha1'
+
 module Highscore
+
   # keywords that were found in content
   #
-  class Keywords < Hash
+  class Keywords
+    include Enumerable
+
+    # find keywords in a piece of content
+    def self.find_keywords content
+      keywords = content.scan(/\w+/)
+      keywords.delete_if do |x|
+        x.match(/^[\d]+(\.[\d]+){0,1}$/) or x.length <= 2
+      end
+
+      keywords.delete_if do |key, value|
+        %w{the and that post add not see about using some something under our comments comment run you want for will file are with end new this use all but can your just get very data blog format out first they posts second}.include? key.downcase
+      end
+
+      keywords.sort
+    end
+
+    # init a new keyword collection
+    #
+    def initialize
+      @keywords = {}
+    end
 
     # ranks the keywords and removes keywords that have a low ranking
     # or are blacklisted
@@ -10,50 +35,44 @@ module Highscore
     #   rank -> array
     #
     def rank
-      filter
-      sort_it
+      sort
     end
 
     # get the top n keywords
     #
     def top n = 10
-      filter
       rank[0..(n - 1)]
     end
 
-    # sorts the keywords and returns a array of arrays
-    #
-    # :call-seq:
-    #   sort_it -> array
+    # add new keywords
     #
-    def sort_it
-      sort {|x,y| y[1] <=> x[1]}
+    def <<(keyword)
+      key = Digest::SHA1.hexdigest(keyword.text)
+
+      if @keywords.has_key?(key)
+        @keywords[key].weight += keyword.weight
+      else
+        @keywords[key] = keyword
+      end
     end
 
-    private
+    # sort
+    def sort
+      sorted = @keywords.sort {|a,b| a[1] <=> b[1] }
 
-    # filter out unwanted results
-    #
-    def filter
-      run_blacklist
-      filter_low
+      # convert Array from sort back to Array of Keyword objects
+      sorted.collect {|x| x[1]}
     end
 
-    # filter low ranked keywords
+    # Enumerable
     #
-    def filter_low
-      delete_if do |key, value|
-        value <= 0
-      end
+    def each &block
+      @keywords.each {|keyword| keyword.call(keyword)}
     end
 
-    # remove blacklisted keywords
-    #
-    def run_blacklist
-      # FIXME: add more keywords!
-      delete_if do |key, value|
-        %w{the and that post add not see about using some something under our comments comment run you want for will file are with end new this use all but can your just get very data blog format out first they posts second}.include? key.downcase
-      end
+    # number of Keywords given
+    def length
+      @keywords.length
     end
   end
 end

data/test/highscore/test_content.rb

@@ -15,4 +15,11 @@ class TestContent < Test::Unit::TestCase
   def test_keywords
     assert_instance_of(Highscore::Keywords, @content.keywords)
   end
+
+  def test_multiple_keywords
+    content = 'Ruby Ruby Ruby and so forth ...'
+
+    content = Highscore::Content.new content
+    assert_equal 2, content.keywords.length
+  end
 end

data/test/highscore/test_keyword.rb

@@ -0,0 +1,31 @@
+$:.unshift(File.join(File.dirname(__FILE__), %w{.. .. lib highscore}))
+require 'keyword'
+require "test/unit"
+
+class TestKeyword < Test::Unit::TestCase
+  def setup
+    @keyword = Highscore::Keyword.new('Ruby', 2)
+  end
+
+  def test_init
+
+    # don't allow 'empty' keywords
+    assert_raise(ArgumentError) do
+      Highscore::Keyword.new
+    end
+  end
+
+  def test_text
+    assert_equal 'Ruby', @keyword.text
+
+    @keyword.text = 'Foobar'
+    assert_equal 'Foobar', @keyword.text
+  end
+
+  def test_weight
+    assert_equal 2, @keyword.weight
+
+    @keyword.weight = 10.123
+    assert_equal 10.123, @keyword.weight
+  end
+end

data/test/highscore/test_keywords.rb

@@ -1,14 +1,15 @@
 $:.unshift(File.join(File.dirname(__FILE__), %w{.. .. lib highscore}))
-require "keywords"
+require 'keywords'
+require 'keyword'
 require "test/unit"
 
 class TestKeywords < Test::Unit::TestCase
   def setup
     @keywords = Highscore::Keywords.new
-    @keywords['Ruby'] = 2
-    @keywords['Sinatra'] = 3
-    @keywords['Highscore'] = 1
-    @keywords['the'] = 10
+    @keywords << Highscore::Keyword.new('Ruby', 2)
+    @keywords << Highscore::Keyword.new('Sinatra', 3)
+    @keywords << Highscore::Keyword.new('Highscore', 1)
+    @keywords << Highscore::Keyword.new('the', 10)
   end
 
   def test_init
@@ -20,10 +21,16 @@ class TestKeywords < Test::Unit::TestCase
 
     ranked = @keywords.rank
 
-    assert_instance_of(Array, ranked)
+    ranked_texts = []
+    ranked.each do |keyword|
+      assert(keyword.instance_of?(Highscore::Keyword),
+             "keywords must be instances of Highscore::Keyword, #{keyword.class} given")
+      ranked_texts << keyword.text
+    end
 
-    should_rank = [['Sinatra', 3], ['Ruby', 2], ['Highscore', 1]]
-    assert_equal should_rank, ranked
+    should_rank = %w{the Sinatra Ruby Highscore}
+
+    assert_equal should_rank, ranked_texts
   end
 
   def test_rank_empty
@@ -31,18 +38,13 @@ class TestKeywords < Test::Unit::TestCase
   end
 
   def test_top
-    assert_equal [['Sinatra', 3]], @keywords.top(1)
+    top = @keywords.top(1)
+
+    assert_equal('the', top[0].text)
+    assert_equal(10.0, top[0].weight)
   end
 
   def test_top_empty
     assert_equal [], Highscore::Keywords.new.top(0)
   end
-
-  def test_sort
-    keywords = Highscore::Keywords.new
-    keywords['Test'] = 1
-    keywords['Foobar'] = 2
-
-    assert_equal [['Foobar', 2], ['Test', 1]], keywords.sort_it
-  end
 end

data/version.txt

@@ -1 +1 @@
-0.2.1
+0.3.1

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: highscore
 version: !ruby/object:Gem::Version
-  version: 0.2.1
+  version: 0.3.1
   prerelease: 
 platform: ruby
 authors:
@@ -9,11 +9,11 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-01-18 00:00:00.000000000 Z
+date: 2012-01-21 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bones
-  requirement: &70306539324720 !ruby/object:Gem::Requirement
+  requirement: &70191947101320 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -21,7 +21,7 @@ dependencies:
         version: 3.7.3
   type: :development
   prerelease: false
-  version_requirements: *70306539324720
+  version_requirements: *70191947101320
 description: Rank keywords in long texts.
 email: liebler.dominik@googlemail.com
 executables:
@@ -36,10 +36,13 @@ files:
 - README.md
 - Rakefile
 - bin/highscore
+- highscore.gemspec
 - lib/highscore.rb
 - lib/highscore/content.rb
+- lib/highscore/keyword.rb
 - lib/highscore/keywords.rb
 - test/highscore/test_content.rb
+- test/highscore/test_keyword.rb
 - test/highscore/test_keywords.rb
 - test/test_highscore.rb
 - version.txt
@@ -71,5 +74,6 @@ specification_version: 3
 summary: Rank keywords in long texts.
 test_files:
 - test/highscore/test_content.rb
+- test/highscore/test_keyword.rb
 - test/highscore/test_keywords.rb
 - test/test_highscore.rb