jaro_winkler 1.2.1 → 1.2.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 4a1b9c70518b0e53cf56495a67e2ffe90962b25f
-  data.tar.gz: 407eee20cb14e8e2b3fde69ad200547b34b27ebe
+  metadata.gz: 7a1229af20c91c12d0f4aeb8fb13dd57cc617344
+  data.tar.gz: 919b78c919c60d15bdeacdfc84bd6e13952e6181
 SHA512:
-  metadata.gz: 0ec1165eac3e38cbac6462d3920467b3bdf2e08957367f150b51553710dd147ffd3ec78774bd74408867a433bf47ca7e5a810750aaddc2bc92aa3cb7d8dc1d31
-  data.tar.gz: 918f8cc2603b09f6a23e8df6bbdd76d3ea295988f20a57000b9382d84f8e97f7d7205d3992d2b48fa1820e3b01bcb4de8473a0d9f1a39f4735efeac441c26d71
+  metadata.gz: 56011763a68a13b9b04973a33b3575c768f81c6257d42284e99bcdf94c66896493cf68fad4e33338081b251fb5d1d93f0f6815e82eee7a8e33b0e97d5707826a
+  data.tar.gz: 79e38669e7d5d4a7582af83a76bee06b49e3b34304708ac5a0d5c8be661dbfc4196a81157b636d8ed4266645af4fcc671b6afcde506588fa3af56d8178961b92

data/README.md

@@ -2,6 +2,10 @@
 
 It's a implementation of [Jaro-Winkler distance](http://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance) algorithm, it uses C extension and will fallback to pure Ruby version in JRuby.
 
+**Windows Issue**
+
+It will fallabck to pure Ruby implementation on Windows since it can't be compiled currently. (ref [#1](https://github.com/tonytonyjan/jaro_winkler/issues/1))
+
 # Installation
 
 ```
@@ -20,8 +24,8 @@ JaroWinkler.distance "MARTHA", "MARHTA", weight: 0.2
 # => 0.9778
 
 # Force the strategy
-JaroWinkler.c_distance "MARTHA", "MARHTA"
-JaroWinkler.r_distance "MARTHA", "MARHTA"
+JaroWinkler.c_distance "MARTHA", "MARHTA" # C extension
+JaroWinkler.r_distance "MARTHA", "MARHTA" # Pure Ruby
 ```
 
 **Both implementations support UTF-8 string.**
@@ -32,7 +36,7 @@ Name        | Type    | Default | Note
 ----------- | ------  | ------- | ------------------------------------------------------------------------------------------------------------
 case_match  | boolean | false   | All upper case characters are converted to lower case prior to the comparison.
 weight      | number  | 0.1     | A constant scaling factor for how much the score is adjusted upwards for having common prefixes.
-threshold   | number  | 0.7     | The prefix bonus is only added when the compared strings have a Jaro distance above a this.
+threshold   | number  | 0.7     | The prefix bonus is only added when the compared strings have a Jaro distance above the threshold.
 
 # Why This?
 

data/benchmark/native.rb

@@ -1,6 +1,8 @@
 require 'benchmark'
 require 'jaro_winkler'
 require 'fuzzystringmatch'
+require 'hotwater'
+require 'amatch'
 ary = [['al', 'al'], ['martha', 'marhta'], ['jones', 'johnson'], ['abcvwxyz', 'cabvwxyz'], ['dwayne', 'duane'], ['dixon', 'dicksonx'], ['fvie', 'ten']]
 
 n = 100000
@@ -13,8 +15,18 @@ Benchmark.bmbm do |x|
     jarow = FuzzyStringMatch::JaroWinkler.create(:native)
     n.times{ ary.each{ |str1, str2| jarow.getDistance(str1, str2) } }
   end
+
+  x.report 'hotwater' do
+    n.times{ ary.each{ |str1, str2| Hotwater.jaro_winkler_distance(str1, str2) } }
+  end
+
+  x.report 'amatch' do
+    n.times{ ary.each{ |str1, str2| Amatch::Jaro.new(str1).match(str2) } }
+  end
 end
 
 #                        user     system      total        real
-# jaro_winkler       0.380000   0.000000   0.380000 (  0.386071)
-# fuzzystringmatch   0.140000   0.000000   0.140000 (  0.138053)
+# jaro_winkler       0.420000   0.000000   0.420000 (  0.426742)
+# fuzzystringmatch   0.160000   0.000000   0.160000 (  0.160146)
+# hotwater           0.300000   0.000000   0.300000 (  0.297350)
+# amatch             0.980000   0.010000   0.990000 (  0.982874)

data/ext/jaro_winkler/distance.c

@@ -55,10 +55,10 @@ double c_distance(char *s1, char *s2, Option *opt){
   }
   int window_size = ary_2_len / 2 - 1;
   if(window_size < 0) window_size = 0;
-  double matches = 0.0;
+  double matches     = 0.0;
   int transpositions = 0;
   int previous_index = -1;
-  int max_index = ary_2_len - 1;
+  int max_index      = ary_2_len - 1;
   for(int i = 0; i < ary_1_len; i++){
     int left  = i - window_size;
     int right = i + window_size;

data/jaro_winkler.gemspec

@@ -1,6 +1,7 @@
 # coding: utf-8
 lib = File.expand_path('../lib', __FILE__)
 $LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require 'jaro_winkler/fallback'
 require 'jaro_winkler/version'
 
 Gem::Specification.new do |spec|
@@ -8,7 +9,7 @@ Gem::Specification.new do |spec|
   spec.version       = JaroWinkler::VERSION
   spec.authors       = ["Jian Weihang"]
   spec.email         = ["tonytonyjan@gmail.com"]
-  spec.extensions    = ["ext/jaro_winkler/extconf.rb"] unless RUBY_PLATFORM == 'java'
+  spec.extensions    = ["ext/jaro_winkler/extconf.rb"] unless JaroWinkler.fallback?
   spec.summary       = %q{Pure Ruby implementation of Jaro-Winkler distance algorithm.}
   spec.description   = %q{Pure Ruby implementation of Jaro-Winkler distance algorithm.}
   spec.homepage      = "https://github.com/tonytonyjan/jaro_winkler"

data/lib/jaro_winkler.rb

@@ -1,4 +1,5 @@
-require 'jaro_winkler/jaro_winkler.so' unless RUBY_PLATFORM == 'java'
+require 'jaro_winkler/fallback'
+require 'jaro_winkler/jaro_winkler.so' unless JaroWinkler.fallback?
 module JaroWinkler
   module_function
   def jaro_distance s1, s2
@@ -8,12 +9,12 @@ module JaroWinkler
       s1, s2 = s2, s1
       length1, length2 = length2, length1
     end
-    window_size      = (length2 / 2) - 1
-    window_size = 0 if window_size < 0
-    matches          = 0.0
-    transpositions   = 0
-    previous_index   = -1
-    max_index = length2 - 1
+    window_size    = (length2 / 2) - 1
+    window_size    = 0 if window_size < 0
+    matches        = 0.0
+    transpositions = 0
+    previous_index = -1
+    max_index      = length2 - 1
     s1.chars.each_with_index do |c1, i|
       left      = i - window_size
       right     = i + window_size
@@ -46,9 +47,9 @@ module JaroWinkler
     options = {weight: 0.1, threshold: 0.7, case_match: false}.merge options
     weight, threshold, case_match = options[:weight], options[:threshold], options[:case_match]
     raise 'Scaling factor should not exceed 0.25, otherwise the distance can become larger than 1' if weight > 0.25
-    s1, s2 = s1.downcase, s2.downcase if case_match
-    distance = jaro_distance(s1, s2)
-    prefix = 0
+    s1, s2     = s1.downcase, s2.downcase if case_match
+    distance   = jaro_distance(s1, s2)
+    prefix     = 0
     max_length = [4, s1.length, s2.length].min
     s1[0, max_length].chars.each_with_index do |c1, i|
       c1 == s2[i] ? prefix += 1 : break
@@ -56,7 +57,7 @@ module JaroWinkler
     distance < threshold ? distance : distance + ((prefix * weight) * (1 - distance))
   end
 
-  if RUBY_PLATFORM == 'java'
+  if JaroWinkler.fallback?
     alias :distance :r_distance
     alias :c_distance :r_distance
     module_function :distance, :c_distance

data/lib/jaro_winkler/fallback.rb

@@ -0,0 +1,7 @@
+module JaroWinkler
+  module_function
+  def fallback?
+    # TODO: Make windows compilable.
+    RUBY_PLATFORM == 'java' || RUBY_PLATFORM == /cygwin|mswin|mingw|bccwin|wince|emx/
+  end
+end

data/lib/jaro_winkler/version.rb

@@ -1,3 +1,3 @@
 module JaroWinkler
-  VERSION = "1.2.1"
+  VERSION = "1.2.2"
 end

data/spec/jaro_winkler_spec.rb

@@ -25,28 +25,24 @@ describe JaroWinkler do
 
   it 'works' do
     @ary.each do |s1, s2, ans|
-      expect(distance(s1, s2)).to be_within(0.0001).of(ans)
-    end
-  end
-
-  it 'supports C extension' do
-    @ary.each do |s1, s2, ans|
+      expect(r_distance(s1, s2)).to be_within(0.0001).of(ans)
       expect(c_distance(s1, s2)).to be_within(0.0001).of(ans)
     end
   end
 
   it 'works with UTF-8' do
-    expect(distance('變形金剛4:絕跡重生', '變形金剛4: 絕跡重生')).to eq c_distance('0123456789', '01234x56789')
+    expect(c_distance('變形金剛4:絕跡重生', '變形金剛4: 絕跡重生')).to eq c_distance('0123456789', '01234x56789')
   end
 
   it 'can ignore case' do
-    expect(distance('MARTHA', 'marhta', case_match: true)).to be_within(0.0001).of(0.9611)
+    expect(r_distance('MARTHA', 'marhta', case_match: true)).to be_within(0.0001).of(0.9611)
     expect(c_distance('MARTHA', 'marhta', case_match: true)).to be_within(0.0001).of(0.9611)
   end
 
   it 'can set weight' do
-    expect(distance('MARTHA', 'MARHTA', weight: 0.2)).to be_within(0.0001).of(0.9778)
+    expect(r_distance('MARTHA', 'MARHTA', weight: 0.2)).to be_within(0.0001).of(0.9778)
     expect(c_distance('MARTHA', 'MARHTA', weight: 0.2)).to be_within(0.0001).of(0.9778)
-    expect{ distance('MARTHA', 'MARHTA', weight: 0.26) }.to raise_error
+    expect{ r_distance('MARTHA', 'MARHTA', weight: 0.26) }.to raise_error
+    expect{ c_distance('MARTHA', 'MARHTA', weight: 0.26) }.to raise_error
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: jaro_winkler
 version: !ruby/object:Gem::Version
-  version: 1.2.1
+  version: 1.2.2
 platform: ruby
 authors:
 - Jian Weihang
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-09-07 00:00:00.000000000 Z
+date: 2014-09-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -76,6 +76,7 @@ files:
 - ext/jaro_winkler/strcmp95.c
 - jaro_winkler.gemspec
 - lib/jaro_winkler.rb
+- lib/jaro_winkler/fallback.rb
 - lib/jaro_winkler/version.rb
 - spec/jaro_winkler_spec.rb
 - spec/spec_helper.rb