knnball 0.0.6 → 0.1.0

data/Rakefile

@@ -1,6 +1,8 @@
 require 'rake/testtask'
 
-Rake::TestTask.new do |tt|
-  tt.test_files = FileList['test/specs/*_spec.rb', 'test/units/*_test.rb']
+Rake::TestTask.new(:test) do |tt|
+  tt.test_files = FileList['test/spec/*_spec.rb', 'test/units/*_test.rb']
   tt.verbose = true
 end
+
+task :default => :test

data/knnball.gemspec

@@ -4,8 +4,8 @@ Gem::Specification.new do |s|
   s.rubygems_version = '1.3.5'
 
   s.name = 'knnball'
-  s.version = '0.0.6'
-  s.date = '2011-09-09'
+  s.version = '0.1.0'
+  s.date = '2012-05-28'
   s.rubyforge_project = 'knnball'
 
   s.summary = "Multi-dimensional nearest neighbor search"
@@ -31,15 +31,15 @@ lib/knnball/ball.rb
 lib/knnball/stat.rb
 lib/knnball/kdtree.rb
 lib/knnball/result_set.rb
-test/specs/ball_spec.rb
-test/specs/data.json
-test/specs/kdtree_spec.rb
-test/specs/knnball_spec.rb
-test/specs/result_set_spec.rb
-test/specs/spec_helpers.rb
+test/spec/ball_spec.rb
+test/spec/data.json
+test/spec/kdtree_spec.rb
+test/spec/knnball_spec.rb
+test/spec/result_set_spec.rb
+test/spec/spec_helpers.rb
 test/units/stat_test.rb
 ]
   # = MANIFEST =
 
-  s.test_files = s.files.select { |path| path =~ /^test\/units\/\.*_test\.rb|test\/specs\/\.*_spec\.rb/ }
+  s.test_files = s.files.select { |path| path =~ /^test\/units\/\.*_test\.rb|test\/spec\/\.*_spec\.rb/ }
 end

data/lib/knnball.rb

@@ -5,8 +5,6 @@
 # knnball is freely distributable under the terms of an MIT license.
 # See LICENSE or http://www.opensource.org/licenses/mit-license.php.
 
-
-
 # This module is used as namespace for every elements of
 # the knnball library.
 module KnnBall
@@ -58,7 +56,7 @@ module KnnBall
     actual_dimension = (max_dimension == actual_dimension ? 1 : actual_dimension)
     
     ball.left = generate(data[0..(median_idx-1)], max_dimension, actual_dimension) if median_idx > 0
-    ball.right = generate(data[(median_idx+1)..-1], max_dimension, actual_dimension) if median_idx < (data.count)
+    ball.right = generate(data[(median_idx+1)..-1], max_dimension, actual_dimension) if median_idx < (data.count - 1)
     return ball
   end
   

data/lib/knnball/stat.rb

@@ -53,8 +53,13 @@ module KnnBall
       return left
     end
     
+    # Retrieve the median index of an array.
+    # array.count == 0 -> 0
+    # array.count == 1 -> 0
+    # array.count == 2 -> 1
+    # array.count == 3 -> 1
     def self.median_index(data)
-      (data.size % 2 == 0) ? (data.size - 1)/ 2 : data.size / 2
+      (data.size % 2 == 0) ? (data.size / 2) : (data.size - 1) / 2
     end
   end
 end

data/test/{specs → spec}/ball_spec.rb

@@ -5,12 +5,9 @@
 # knnball is freely distributable under the terms of an MIT license.
 # See LICENSE or http://www.opensource.org/licenses/mit-license.php.
 
-
-
 require 'minitest/autorun'
 require 'knnball'
 
-
 module KnnBall
   
   describe Ball do

data/test/{specs → spec}/knnball_spec.rb

@@ -49,7 +49,7 @@ describe KnnBall do
           {:id => 6, :point => [13]},
           {:id => 7, :point => [21]},
           {:id => 8, :point => [34]}
-        ]).root.value.must_equal({:id => 4, :point => [5]})
+        ]).root.value.must_equal({:id => 5, :point => [8]})
     end
   end
   
@@ -92,7 +92,7 @@ describe KnnBall do
           msgs << "For #{p}, #{nn_result} retrieved amongs those 2 first results #{tree.nearest(p[:point], :limit => 2)}"
         end
       end
-      must_be_empty msgs
+      msgs.must_be_empty
     end
     
     it "is more efficient than the brute force approach" do
@@ -166,7 +166,7 @@ describe KnnBall do
         nn_result = @index.nearest(p[:point], :limit => 10)
         (nn_result.map{|r| r[:point]}).must_equal(brute_force_result.map{|r| r[:point]})
       end
-      must_be_empty(msgs)
+      msgs.must_be_empty
     end
   end
 end

data/test/{specs → spec}/result_set_spec.rb

@@ -150,8 +150,8 @@ module KnnBall
       end
       
       it "accept smaller value" do
-        assert @rs.eligible? -2
-        assert @rs.eligible? 4
+        @rs.eligible?(-2).must_equal true
+        @rs.eligible?(4).must_equal true
       end
     end
   end

data/test/units/stat_test.rb

@@ -5,16 +5,17 @@
 # knnball is freely distributable under the terms of an MIT license.
 # See LICENSE or http://www.opensource.org/licenses/mit-license.php.
 
-
-
-require 'minitest/autorun'
+require 'test/unit'
 require 'knnball'
 
 module KnnBall
-  class StatTest < MiniTest::Unit::TestCase
+  class StatTest < Test::Unit::TestCase
     def test_median_index
+      assert_equal(0, Stat.median_index([]))
+      assert_equal(0, Stat.median_index([1]))
+      assert_equal(1, Stat.median_index([1] * 2))
       assert_equal(1, Stat.median_index([1] * 3))
-      assert_equal(1, Stat.median_index([1] * 4))
+      assert_equal(2, Stat.median_index([1] * 4))
     end
     
     def test_pivot

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: knnball
 version: !ruby/object:Gem::Version
-  version: 0.0.6
+  version: 0.1.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,8 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2011-09-09 00:00:00.000000000 +02:00
-default_executable: 
+date: 2012-05-28 00:00:00.000000000 Z
 dependencies: []
 description: Implements K-Nearest Neighbor algorithm using a KDTree in Ruby. Usefull
   for sorting geolocation or any other multi-dimensional data.
@@ -30,14 +29,13 @@ files:
 - lib/knnball/stat.rb
 - lib/knnball/kdtree.rb
 - lib/knnball/result_set.rb
-- test/specs/ball_spec.rb
-- test/specs/data.json
-- test/specs/kdtree_spec.rb
-- test/specs/knnball_spec.rb
-- test/specs/result_set_spec.rb
-- test/specs/spec_helpers.rb
+- test/spec/ball_spec.rb
+- test/spec/data.json
+- test/spec/kdtree_spec.rb
+- test/spec/knnball_spec.rb
+- test/spec/result_set_spec.rb
+- test/spec/spec_helpers.rb
 - test/units/stat_test.rb
-has_rdoc: true
 homepage: http://github.com/oliamb/knnball
 licenses: []
 post_install_message: 
@@ -59,7 +57,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: knnball
-rubygems_version: 1.6.2
+rubygems_version: 1.8.24
 signing_key: 
 specification_version: 1
 summary: Multi-dimensional nearest neighbor search