You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@commons.apache.org by br...@apache.org on 2015/03/01 13:14:31 UTC
svn commit: r941823 [8/21] - in
/websites/production/commons/content/sandbox/commons-text: ./ apidocs/
apidocs/org/apache/commons/text/diff/
apidocs/org/apache/commons/text/diff/class-use/
apidocs/org/apache/commons/text/similarity/ apidocs/org/apache/...
Modified: websites/production/commons/content/sandbox/commons-text/apidocs/src-html/org/apache/commons/text/similarity/JaroWrinklerDistance.html
==============================================================================
--- websites/production/commons/content/sandbox/commons-text/apidocs/src-html/org/apache/commons/text/similarity/JaroWrinklerDistance.html (original)
+++ websites/production/commons/content/sandbox/commons-text/apidocs/src-html/org/apache/commons/text/similarity/JaroWrinklerDistance.html Sun Mar 1 12:14:29 2015
@@ -25,23 +25,23 @@
<span class="sourceLineNo">017</span>package org.apache.commons.text.similarity;<a name="line.17"></a>
<span class="sourceLineNo">018</span><a name="line.18"></a>
<span class="sourceLineNo">019</span>/**<a name="line.19"></a>
-<span class="sourceLineNo">020</span> * <p><a name="line.20"></a>
-<span class="sourceLineNo">021</span> * The Jaro measure is the weighted sum of percentage of matched characters<a name="line.21"></a>
-<span class="sourceLineNo">022</span> * from each file and transposed characters. Winkler increased this measure<a name="line.22"></a>
-<span class="sourceLineNo">023</span> * for matching initial characters.<a name="line.23"></a>
-<span class="sourceLineNo">024</span> * </p><a name="line.24"></a>
-<span class="sourceLineNo">025</span> *<a name="line.25"></a>
-<span class="sourceLineNo">026</span> * <p><a name="line.26"></a>
-<span class="sourceLineNo">027</span> * This implementation is based on the Jaro Winkler similarity algorithm<a name="line.27"></a>
-<span class="sourceLineNo">028</span> * from <a href="http://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance"><a name="line.28"></a>
-<span class="sourceLineNo">029</span> * http://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance</a>.<a name="line.29"></a>
-<span class="sourceLineNo">030</span> * </p><a name="line.30"></a>
-<span class="sourceLineNo">031</span> *<a name="line.31"></a>
-<span class="sourceLineNo">032</span> * <p><a name="line.32"></a>
-<span class="sourceLineNo">033</span> * This code has been adapted from Apache Commons Lang 3.3.<a name="line.33"></a>
-<span class="sourceLineNo">034</span> * </p><a name="line.34"></a>
-<span class="sourceLineNo">035</span> *<a name="line.35"></a>
-<span class="sourceLineNo">036</span> * @since 1.0<a name="line.36"></a>
+<span class="sourceLineNo">020</span> * A similarity algorithm indicating the percentage of matched characters between two character sequences.<a name="line.20"></a>
+<span class="sourceLineNo">021</span> *<a name="line.21"></a>
+<span class="sourceLineNo">022</span> * <p><a name="line.22"></a>
+<span class="sourceLineNo">023</span> * The Jaro measure is the weighted sum of percentage of matched characters<a name="line.23"></a>
+<span class="sourceLineNo">024</span> * from each file and transposed characters. Winkler increased this measure<a name="line.24"></a>
+<span class="sourceLineNo">025</span> * for matching initial characters.<a name="line.25"></a>
+<span class="sourceLineNo">026</span> * </p><a name="line.26"></a>
+<span class="sourceLineNo">027</span> *<a name="line.27"></a>
+<span class="sourceLineNo">028</span> * <p><a name="line.28"></a>
+<span class="sourceLineNo">029</span> * This implementation is based on the Jaro Winkler similarity algorithm<a name="line.29"></a>
+<span class="sourceLineNo">030</span> * from <a href="http://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance"><a name="line.30"></a>
+<span class="sourceLineNo">031</span> * http://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance</a>.<a name="line.31"></a>
+<span class="sourceLineNo">032</span> * </p><a name="line.32"></a>
+<span class="sourceLineNo">033</span> *<a name="line.33"></a>
+<span class="sourceLineNo">034</span> * <p><a name="line.34"></a>
+<span class="sourceLineNo">035</span> * This code has been adapted from Apache Commons Lang 3.3.<a name="line.35"></a>
+<span class="sourceLineNo">036</span> * </p><a name="line.36"></a>
<span class="sourceLineNo">037</span> */<a name="line.37"></a>
<span class="sourceLineNo">038</span>public class JaroWrinklerDistance implements StringMetric<Double> {<a name="line.38"></a>
<span class="sourceLineNo">039</span><a name="line.39"></a>
@@ -51,334 +51,321 @@
<span class="sourceLineNo">043</span> public static final int INDEX_NOT_FOUND = -1;<a name="line.43"></a>
<span class="sourceLineNo">044</span><a name="line.44"></a>
<span class="sourceLineNo">045</span> /**<a name="line.45"></a>
-<span class="sourceLineNo">046</span> * <p><a name="line.46"></a>
-<span class="sourceLineNo">047</span> * Find the Jaro Winkler Distance which indicates the similarity score<a name="line.47"></a>
-<span class="sourceLineNo">048</span> * between two Strings.<a name="line.48"></a>
-<span class="sourceLineNo">049</span> * </p><a name="line.49"></a>
-<span class="sourceLineNo">050</span> *<a name="line.50"></a>
-<span class="sourceLineNo">051</span> * <pre><a name="line.51"></a>
-<span class="sourceLineNo">052</span> * StringUtils.getJaroWinklerDistance(null, null) = IllegalArgumentException<a name="line.52"></a>
-<span class="sourceLineNo">053</span> * StringUtils.getJaroWinklerDistance("","") = 0.0<a name="line.53"></a>
-<span class="sourceLineNo">054</span> * StringUtils.getJaroWinklerDistance("","a") = 0.0<a name="line.54"></a>
-<span class="sourceLineNo">055</span> * StringUtils.getJaroWinklerDistance("aaapppp", "") = 0.0<a name="line.55"></a>
-<span class="sourceLineNo">056</span> * StringUtils.getJaroWinklerDistance("frog", "fog") = 0.93<a name="line.56"></a>
-<span class="sourceLineNo">057</span> * StringUtils.getJaroWinklerDistance("fly", "ant") = 0.0<a name="line.57"></a>
-<span class="sourceLineNo">058</span> * StringUtils.getJaroWinklerDistance("elephant", "hippo") = 0.44<a name="line.58"></a>
-<span class="sourceLineNo">059</span> * StringUtils.getJaroWinklerDistance("hippo", "elephant") = 0.44<a name="line.59"></a>
-<span class="sourceLineNo">060</span> * StringUtils.getJaroWinklerDistance("hippo", "zzzzzzzz") = 0.0<a name="line.60"></a>
-<span class="sourceLineNo">061</span> * StringUtils.getJaroWinklerDistance("hello", "hallo") = 0.88<a name="line.61"></a>
-<span class="sourceLineNo">062</span> * StringUtils.getJaroWinklerDistance("ABC Corporation", "ABC Corp") = 0.91<a name="line.62"></a>
-<span class="sourceLineNo">063</span> * StringUtils.getJaroWinklerDistance("D N H Enterprises Inc", "D &amp; H Enterprises, Inc.") = 0.93<a name="line.63"></a>
-<span class="sourceLineNo">064</span> * StringUtils.getJaroWinklerDistance("My Gym Children's Fitness Center", "My Gym. Childrens Fitness") = 0.94<a name="line.64"></a>
-<span class="sourceLineNo">065</span> * StringUtils.getJaroWinklerDistance("PENNSYLVANIA", "PENNCISYLVNIA") = 0.9<a name="line.65"></a>
-<span class="sourceLineNo">066</span> * </pre><a name="line.66"></a>
-<span class="sourceLineNo">067</span> *<a name="line.67"></a>
-<span class="sourceLineNo">068</span> * @param left the first String, must not be null<a name="line.68"></a>
-<span class="sourceLineNo">069</span> * @param right the second String, must not be null<a name="line.69"></a>
-<span class="sourceLineNo">070</span> * @return result distance<a name="line.70"></a>
-<span class="sourceLineNo">071</span> * @throws IllegalArgumentException if either String input {@code null}<a name="line.71"></a>
-<span class="sourceLineNo">072</span> */<a name="line.72"></a>
-<span class="sourceLineNo">073</span> @Override<a name="line.73"></a>
-<span class="sourceLineNo">074</span> public Double compare(CharSequence left, CharSequence right) {<a name="line.74"></a>
-<span class="sourceLineNo">075</span> final double DEFAULT_SCALING_FACTOR = 0.1;<a name="line.75"></a>
-<span class="sourceLineNo">076</span><a name="line.76"></a>
-<span class="sourceLineNo">077</span> if (left == null || right == null) {<a name="line.77"></a>
-<span class="sourceLineNo">078</span> throw new IllegalArgumentException("Strings must not be null");<a name="line.78"></a>
-<span class="sourceLineNo">079</span> }<a name="line.79"></a>
-<span class="sourceLineNo">080</span><a name="line.80"></a>
-<span class="sourceLineNo">081</span> final double jaro = score(left, right);<a name="line.81"></a>
-<span class="sourceLineNo">082</span> final int cl = commonPrefixLength(left, right);<a name="line.82"></a>
-<span class="sourceLineNo">083</span> final double matchScore = Math.round((jaro + (DEFAULT_SCALING_FACTOR<a name="line.83"></a>
-<span class="sourceLineNo">084</span> * cl * (1.0 - jaro))) *100.0)/100.0;<a name="line.84"></a>
-<span class="sourceLineNo">085</span><a name="line.85"></a>
-<span class="sourceLineNo">086</span> return matchScore;<a name="line.86"></a>
-<span class="sourceLineNo">087</span> }<a name="line.87"></a>
-<span class="sourceLineNo">088</span><a name="line.88"></a>
-<span class="sourceLineNo">089</span> // TODO: we can move these methods to a Util class, keep them here,<a name="line.89"></a>
-<span class="sourceLineNo">090</span> // create a common abstract class, shade lang-3.3...<a name="line.90"></a>
-<span class="sourceLineNo">091</span><a name="line.91"></a>
-<span class="sourceLineNo">092</span> /**<a name="line.92"></a>
-<span class="sourceLineNo">093</span> * Calculates the number of characters from the beginning of the strings<a name="line.93"></a>
-<span class="sourceLineNo">094</span> * that match exactly one-to-one, up to a maximum of four (4) characters.<a name="line.94"></a>
-<span class="sourceLineNo">095</span> *<a name="line.95"></a>
-<span class="sourceLineNo">096</span> * @param first The first string.<a name="line.96"></a>
-<span class="sourceLineNo">097</span> * @param second The second string.<a name="line.97"></a>
-<span class="sourceLineNo">098</span> * @return A number between 0 and 4.<a name="line.98"></a>
-<span class="sourceLineNo">099</span> */<a name="line.99"></a>
-<span class="sourceLineNo">100</span> private static int commonPrefixLength(final CharSequence first,<a name="line.100"></a>
-<span class="sourceLineNo">101</span> final CharSequence second) {<a name="line.101"></a>
-<span class="sourceLineNo">102</span> final int result = getCommonPrefix(first.toString(), second.toString())<a name="line.102"></a>
-<span class="sourceLineNo">103</span> .length();<a name="line.103"></a>
-<span class="sourceLineNo">104</span><a name="line.104"></a>
-<span class="sourceLineNo">105</span> // Limit the result to 4.<a name="line.105"></a>
-<span class="sourceLineNo">106</span> return result > 4 ? 4 : result;<a name="line.106"></a>
-<span class="sourceLineNo">107</span> }<a name="line.107"></a>
-<span class="sourceLineNo">108</span><a name="line.108"></a>
-<span class="sourceLineNo">109</span> /**<a name="line.109"></a>
-<span class="sourceLineNo">110</span> * <p><a name="line.110"></a>
-<span class="sourceLineNo">111</span> * Compares all Strings in an array and returns the initial sequence of<a name="line.111"></a>
-<span class="sourceLineNo">112</span> * characters that is common to all of them.<a name="line.112"></a>
-<span class="sourceLineNo">113</span> * </p><a name="line.113"></a>
-<span class="sourceLineNo">114</span> *<a name="line.114"></a>
-<span class="sourceLineNo">115</span> * <p><a name="line.115"></a>
-<span class="sourceLineNo">116</span> * For example,<a name="line.116"></a>
-<span class="sourceLineNo">117</span> * <code>getCommonPrefix(new String[] {"i am a machine", "i am a robot"}) -&gt; "i am a "</code><a name="line.117"></a>
-<span class="sourceLineNo">118</span> * </p><a name="line.118"></a>
-<span class="sourceLineNo">119</span> *<a name="line.119"></a>
-<span class="sourceLineNo">120</span> * <pre><a name="line.120"></a>
-<span class="sourceLineNo">121</span> * StringUtils.getCommonPrefix(null) = ""<a name="line.121"></a>
-<span class="sourceLineNo">122</span> * StringUtils.getCommonPrefix(new String[] {}) = ""<a name="line.122"></a>
-<span class="sourceLineNo">123</span> * StringUtils.getCommonPrefix(new String[] {"abc"}) = "abc"<a name="line.123"></a>
-<span class="sourceLineNo">124</span> * StringUtils.getCommonPrefix(new String[] {null, null}) = ""<a name="line.124"></a>
-<span class="sourceLineNo">125</span> * StringUtils.getCommonPrefix(new String[] {"", ""}) = ""<a name="line.125"></a>
-<span class="sourceLineNo">126</span> * StringUtils.getCommonPrefix(new String[] {"", null}) = ""<a name="line.126"></a>
-<span class="sourceLineNo">127</span> * StringUtils.getCommonPrefix(new String[] {"abc", null, null}) = ""<a name="line.127"></a>
-<span class="sourceLineNo">128</span> * StringUtils.getCommonPrefix(new String[] {null, null, "abc"}) = ""<a name="line.128"></a>
-<span class="sourceLineNo">129</span> * StringUtils.getCommonPrefix(new String[] {"", "abc"}) = ""<a name="line.129"></a>
-<span class="sourceLineNo">130</span> * StringUtils.getCommonPrefix(new String[] {"abc", ""}) = ""<a name="line.130"></a>
-<span class="sourceLineNo">131</span> * StringUtils.getCommonPrefix(new String[] {"abc", "abc"}) = "abc"<a name="line.131"></a>
-<span class="sourceLineNo">132</span> * StringUtils.getCommonPrefix(new String[] {"abc", "a"}) = "a"<a name="line.132"></a>
-<span class="sourceLineNo">133</span> * StringUtils.getCommonPrefix(new String[] {"ab", "abxyz"}) = "ab"<a name="line.133"></a>
-<span class="sourceLineNo">134</span> * StringUtils.getCommonPrefix(new String[] {"abcde", "abxyz"}) = "ab"<a name="line.134"></a>
-<span class="sourceLineNo">135</span> * StringUtils.getCommonPrefix(new String[] {"abcde", "xyz"}) = ""<a name="line.135"></a>
-<span class="sourceLineNo">136</span> * StringUtils.getCommonPrefix(new String[] {"xyz", "abcde"}) = ""<a name="line.136"></a>
-<span class="sourceLineNo">137</span> * StringUtils.getCommonPrefix(new String[] {"i am a machine", "i am a robot"}) = "i am a "<a name="line.137"></a>
-<span class="sourceLineNo">138</span> * </pre><a name="line.138"></a>
-<span class="sourceLineNo">139</span> *<a name="line.139"></a>
-<span class="sourceLineNo">140</span> * @param strs array of String objects, entries may be null<a name="line.140"></a>
-<span class="sourceLineNo">141</span> * @return the initial sequence of characters that are common to all Strings<a name="line.141"></a>
-<span class="sourceLineNo">142</span> * in the array; empty String if the array is null, the elements are<a name="line.142"></a>
-<span class="sourceLineNo">143</span> * all null or if there is no common prefix.<a name="line.143"></a>
-<span class="sourceLineNo">144</span> * @since 2.4<a name="line.144"></a>
-<span class="sourceLineNo">145</span> */<a name="line.145"></a>
-<span class="sourceLineNo">146</span> public static String getCommonPrefix(final String... strs) {<a name="line.146"></a>
-<span class="sourceLineNo">147</span> if (strs == null || strs.length == 0) {<a name="line.147"></a>
-<span class="sourceLineNo">148</span> return "";<a name="line.148"></a>
-<span class="sourceLineNo">149</span> }<a name="line.149"></a>
-<span class="sourceLineNo">150</span> final int smallestIndexOfDiff = indexOfDifference(strs);<a name="line.150"></a>
-<span class="sourceLineNo">151</span> if (smallestIndexOfDiff == INDEX_NOT_FOUND) {<a name="line.151"></a>
-<span class="sourceLineNo">152</span> // all strings were identical<a name="line.152"></a>
-<span class="sourceLineNo">153</span> if (strs[0] == null) {<a name="line.153"></a>
-<span class="sourceLineNo">154</span> return "";<a name="line.154"></a>
-<span class="sourceLineNo">155</span> }<a name="line.155"></a>
-<span class="sourceLineNo">156</span> return strs[0];<a name="line.156"></a>
-<span class="sourceLineNo">157</span> } else if (smallestIndexOfDiff == 0) {<a name="line.157"></a>
-<span class="sourceLineNo">158</span> // there were no common initial characters<a name="line.158"></a>
-<span class="sourceLineNo">159</span> return "";<a name="line.159"></a>
-<span class="sourceLineNo">160</span> } else {<a name="line.160"></a>
-<span class="sourceLineNo">161</span> // we found a common initial character sequence<a name="line.161"></a>
-<span class="sourceLineNo">162</span> return strs[0].substring(0, smallestIndexOfDiff);<a name="line.162"></a>
-<span class="sourceLineNo">163</span> }<a name="line.163"></a>
-<span class="sourceLineNo">164</span> }<a name="line.164"></a>
-<span class="sourceLineNo">165</span><a name="line.165"></a>
-<span class="sourceLineNo">166</span> /**<a name="line.166"></a>
-<span class="sourceLineNo">167</span> * This method returns the Jaro-Winkler score for string matching.<a name="line.167"></a>
-<span class="sourceLineNo">168</span> *<a name="line.168"></a>
-<span class="sourceLineNo">169</span> * @param first the first string to be matched<a name="line.169"></a>
-<span class="sourceLineNo">170</span> * @param second the second string to be machted<a name="line.170"></a>
-<span class="sourceLineNo">171</span> * @return matching score without scaling factor impact<a name="line.171"></a>
-<span class="sourceLineNo">172</span> */<a name="line.172"></a>
-<span class="sourceLineNo">173</span> protected static double score(final CharSequence first,<a name="line.173"></a>
-<span class="sourceLineNo">174</span> final CharSequence second) {<a name="line.174"></a>
-<span class="sourceLineNo">175</span> String shorter;<a name="line.175"></a>
-<span class="sourceLineNo">176</span> String longer;<a name="line.176"></a>
-<span class="sourceLineNo">177</span><a name="line.177"></a>
-<span class="sourceLineNo">178</span> // Determine which String is longer.<a name="line.178"></a>
-<span class="sourceLineNo">179</span> if (first.length() > second.length()) {<a name="line.179"></a>
-<span class="sourceLineNo">180</span> longer = first.toString().toLowerCase();<a name="line.180"></a>
-<span class="sourceLineNo">181</span> shorter = second.toString().toLowerCase();<a name="line.181"></a>
-<span class="sourceLineNo">182</span> } else {<a name="line.182"></a>
-<span class="sourceLineNo">183</span> longer = second.toString().toLowerCase();<a name="line.183"></a>
-<span class="sourceLineNo">184</span> shorter = first.toString().toLowerCase();<a name="line.184"></a>
-<span class="sourceLineNo">185</span> }<a name="line.185"></a>
-<span class="sourceLineNo">186</span><a name="line.186"></a>
-<span class="sourceLineNo">187</span> // Calculate the half length() distance of the shorter String.<a name="line.187"></a>
-<span class="sourceLineNo">188</span> final int halflength = shorter.length() / 2 + 1;<a name="line.188"></a>
-<span class="sourceLineNo">189</span><a name="line.189"></a>
-<span class="sourceLineNo">190</span> // Find the set of matching characters between the shorter and longer<a name="line.190"></a>
-<span class="sourceLineNo">191</span> // strings. Note that<a name="line.191"></a>
-<span class="sourceLineNo">192</span> // the set of matching characters may be different depending on the<a name="line.192"></a>
-<span class="sourceLineNo">193</span> // order of the strings.<a name="line.193"></a>
-<span class="sourceLineNo">194</span> final String m1 = getSetOfMatchingCharacterWithin(shorter, longer,<a name="line.194"></a>
-<span class="sourceLineNo">195</span> halflength);<a name="line.195"></a>
-<span class="sourceLineNo">196</span> final String m2 = getSetOfMatchingCharacterWithin(longer, shorter,<a name="line.196"></a>
-<span class="sourceLineNo">197</span> halflength);<a name="line.197"></a>
-<span class="sourceLineNo">198</span><a name="line.198"></a>
-<span class="sourceLineNo">199</span> // If one or both of the sets of common characters is empty, then<a name="line.199"></a>
-<span class="sourceLineNo">200</span> // there is no similarity between the two strings.<a name="line.200"></a>
-<span class="sourceLineNo">201</span> if (m1.length() == 0 || m2.length() == 0) {<a name="line.201"></a>
-<span class="sourceLineNo">202</span> return 0.0;<a name="line.202"></a>
-<span class="sourceLineNo">203</span> }<a name="line.203"></a>
-<span class="sourceLineNo">204</span><a name="line.204"></a>
-<span class="sourceLineNo">205</span> // If the set of common characters is not the same size, then<a name="line.205"></a>
-<span class="sourceLineNo">206</span> // there is no similarity between the two strings, either.<a name="line.206"></a>
-<span class="sourceLineNo">207</span> if (m1.length() != m2.length()) {<a name="line.207"></a>
-<span class="sourceLineNo">208</span> return 0.0;<a name="line.208"></a>
-<span class="sourceLineNo">209</span> }<a name="line.209"></a>
-<span class="sourceLineNo">210</span><a name="line.210"></a>
-<span class="sourceLineNo">211</span> // Calculate the number of transposition between the two sets<a name="line.211"></a>
-<span class="sourceLineNo">212</span> // of common characters.<a name="line.212"></a>
-<span class="sourceLineNo">213</span> final int transpositions = transpositions(m1, m2);<a name="line.213"></a>
-<span class="sourceLineNo">214</span><a name="line.214"></a>
-<span class="sourceLineNo">215</span> // Calculate the distance.<a name="line.215"></a>
-<span class="sourceLineNo">216</span> final double dist = (m1.length() / ((double) shorter.length())<a name="line.216"></a>
-<span class="sourceLineNo">217</span> + m2.length() / ((double) longer.length()) + (m1.length() - transpositions)<a name="line.217"></a>
-<span class="sourceLineNo">218</span> / ((double) m1.length())) / 3.0;<a name="line.218"></a>
-<span class="sourceLineNo">219</span> return dist;<a name="line.219"></a>
-<span class="sourceLineNo">220</span> }<a name="line.220"></a>
-<span class="sourceLineNo">221</span><a name="line.221"></a>
-<span class="sourceLineNo">222</span> /**<a name="line.222"></a>
-<span class="sourceLineNo">223</span> * Calculates the number of transposition between two strings.<a name="line.223"></a>
-<span class="sourceLineNo">224</span> *<a name="line.224"></a>
-<span class="sourceLineNo">225</span> * @param first The first string.<a name="line.225"></a>
-<span class="sourceLineNo">226</span> * @param second The second string.<a name="line.226"></a>
-<span class="sourceLineNo">227</span> * @return The number of transposition between the two strings.<a name="line.227"></a>
-<span class="sourceLineNo">228</span> */<a name="line.228"></a>
-<span class="sourceLineNo">229</span> protected static int transpositions(final CharSequence first,<a name="line.229"></a>
-<span class="sourceLineNo">230</span> final CharSequence second) {<a name="line.230"></a>
-<span class="sourceLineNo">231</span> int transpositions = 0;<a name="line.231"></a>
-<span class="sourceLineNo">232</span> for (int i = 0; i < first.length(); i++) {<a name="line.232"></a>
-<span class="sourceLineNo">233</span> if (first.charAt(i) != second.charAt(i)) {<a name="line.233"></a>
-<span class="sourceLineNo">234</span> transpositions++;<a name="line.234"></a>
-<span class="sourceLineNo">235</span> }<a name="line.235"></a>
-<span class="sourceLineNo">236</span> }<a name="line.236"></a>
-<span class="sourceLineNo">237</span> return transpositions / 2;<a name="line.237"></a>
-<span class="sourceLineNo">238</span> }<a name="line.238"></a>
-<span class="sourceLineNo">239</span><a name="line.239"></a>
-<span class="sourceLineNo">240</span> /**<a name="line.240"></a>
-<span class="sourceLineNo">241</span> * <p><a name="line.241"></a>
-<span class="sourceLineNo">242</span> * Compares all CharSequences in an array and returns the index at which the<a name="line.242"></a>
-<span class="sourceLineNo">243</span> * CharSequences begin to differ.<a name="line.243"></a>
-<span class="sourceLineNo">244</span> * </p><a name="line.244"></a>
-<span class="sourceLineNo">245</span> *<a name="line.245"></a>
-<span class="sourceLineNo">246</span> * <p><a name="line.246"></a>
-<span class="sourceLineNo">247</span> * For example,<a name="line.247"></a>
-<span class="sourceLineNo">248</span> * <code>indexOfDifference(new String[] {"i am a machine", "i am a robot"}) -&gt; 7</code><a name="line.248"></a>
-<span class="sourceLineNo">249</span> * </p><a name="line.249"></a>
-<span class="sourceLineNo">250</span> *<a name="line.250"></a>
-<span class="sourceLineNo">251</span> * <pre><a name="line.251"></a>
-<span class="sourceLineNo">252</span> * StringUtils.indexOfDifference(null) = -1<a name="line.252"></a>
-<span class="sourceLineNo">253</span> * StringUtils.indexOfDifference(new String[] {}) = -1<a name="line.253"></a>
-<span class="sourceLineNo">254</span> * StringUtils.indexOfDifference(new String[] {"abc"}) = -1<a name="line.254"></a>
-<span class="sourceLineNo">255</span> * StringUtils.indexOfDifference(new String[] {null, null}) = -1<a name="line.255"></a>
-<span class="sourceLineNo">256</span> * StringUtils.indexOfDifference(new String[] {"", ""}) = -1<a name="line.256"></a>
-<span class="sourceLineNo">257</span> * StringUtils.indexOfDifference(new String[] {"", null}) = 0<a name="line.257"></a>
-<span class="sourceLineNo">258</span> * StringUtils.indexOfDifference(new String[] {"abc", null, null}) = 0<a name="line.258"></a>
-<span class="sourceLineNo">259</span> * StringUtils.indexOfDifference(new String[] {null, null, "abc"}) = 0<a name="line.259"></a>
-<span class="sourceLineNo">260</span> * StringUtils.indexOfDifference(new String[] {"", "abc"}) = 0<a name="line.260"></a>
-<span class="sourceLineNo">261</span> * StringUtils.indexOfDifference(new String[] {"abc", ""}) = 0<a name="line.261"></a>
-<span class="sourceLineNo">262</span> * StringUtils.indexOfDifference(new String[] {"abc", "abc"}) = -1<a name="line.262"></a>
-<span class="sourceLineNo">263</span> * StringUtils.indexOfDifference(new String[] {"abc", "a"}) = 1<a name="line.263"></a>
-<span class="sourceLineNo">264</span> * StringUtils.indexOfDifference(new String[] {"ab", "abxyz"}) = 2<a name="line.264"></a>
-<span class="sourceLineNo">265</span> * StringUtils.indexOfDifference(new String[] {"abcde", "abxyz"}) = 2<a name="line.265"></a>
-<span class="sourceLineNo">266</span> * StringUtils.indexOfDifference(new String[] {"abcde", "xyz"}) = 0<a name="line.266"></a>
-<span class="sourceLineNo">267</span> * StringUtils.indexOfDifference(new String[] {"xyz", "abcde"}) = 0<a name="line.267"></a>
-<span class="sourceLineNo">268</span> * StringUtils.indexOfDifference(new String[] {"i am a machine", "i am a robot"}) = 7<a name="line.268"></a>
-<span class="sourceLineNo">269</span> * </pre><a name="line.269"></a>
-<span class="sourceLineNo">270</span> *<a name="line.270"></a>
-<span class="sourceLineNo">271</span> * @param css array of CharSequences, entries may be null<a name="line.271"></a>
-<span class="sourceLineNo">272</span> * @return the index where the strings begin to differ; -1 if they are all<a name="line.272"></a>
-<span class="sourceLineNo">273</span> * equal<a name="line.273"></a>
-<span class="sourceLineNo">274</span> * @since 2.4<a name="line.274"></a>
-<span class="sourceLineNo">275</span> * @since 3.0 Changed signature from indexOfDifference(String...) to<a name="line.275"></a>
-<span class="sourceLineNo">276</span> * indexOfDifference(CharSequence...)<a name="line.276"></a>
-<span class="sourceLineNo">277</span> */<a name="line.277"></a>
-<span class="sourceLineNo">278</span> protected static int indexOfDifference(final CharSequence... css) {<a name="line.278"></a>
-<span class="sourceLineNo">279</span> if (css == null || css.length <= 1) {<a name="line.279"></a>
-<span class="sourceLineNo">280</span> return INDEX_NOT_FOUND;<a name="line.280"></a>
-<span class="sourceLineNo">281</span> }<a name="line.281"></a>
-<span class="sourceLineNo">282</span> boolean anyStringNull = false;<a name="line.282"></a>
-<span class="sourceLineNo">283</span> boolean allStringsNull = true;<a name="line.283"></a>
-<span class="sourceLineNo">284</span> final int arrayLen = css.length;<a name="line.284"></a>
-<span class="sourceLineNo">285</span> int shortestStrLen = Integer.MAX_VALUE;<a name="line.285"></a>
-<span class="sourceLineNo">286</span> int longestStrLen = 0;<a name="line.286"></a>
-<span class="sourceLineNo">287</span><a name="line.287"></a>
-<span class="sourceLineNo">288</span> // find the min and max string lengths; this avoids checking to make<a name="line.288"></a>
-<span class="sourceLineNo">289</span> // sure we are not exceeding the length of the string each time through<a name="line.289"></a>
-<span class="sourceLineNo">290</span> // the bottom loop.<a name="line.290"></a>
-<span class="sourceLineNo">291</span> for (int i = 0; i < arrayLen; i++) {<a name="line.291"></a>
-<span class="sourceLineNo">292</span> if (css[i] == null) {<a name="line.292"></a>
-<span class="sourceLineNo">293</span> anyStringNull = true;<a name="line.293"></a>
-<span class="sourceLineNo">294</span> shortestStrLen = 0;<a name="line.294"></a>
-<span class="sourceLineNo">295</span> } else {<a name="line.295"></a>
-<span class="sourceLineNo">296</span> allStringsNull = false;<a name="line.296"></a>
-<span class="sourceLineNo">297</span> shortestStrLen = Math.min(css[i].length(), shortestStrLen);<a name="line.297"></a>
-<span class="sourceLineNo">298</span> longestStrLen = Math.max(css[i].length(), longestStrLen);<a name="line.298"></a>
-<span class="sourceLineNo">299</span> }<a name="line.299"></a>
-<span class="sourceLineNo">300</span> }<a name="line.300"></a>
-<span class="sourceLineNo">301</span><a name="line.301"></a>
-<span class="sourceLineNo">302</span> // handle lists containing all nulls or all empty strings<a name="line.302"></a>
-<span class="sourceLineNo">303</span> if (allStringsNull || longestStrLen == 0 && !anyStringNull) {<a name="line.303"></a>
-<span class="sourceLineNo">304</span> return INDEX_NOT_FOUND;<a name="line.304"></a>
-<span class="sourceLineNo">305</span> }<a name="line.305"></a>
-<span class="sourceLineNo">306</span><a name="line.306"></a>
-<span class="sourceLineNo">307</span> // handle lists containing some nulls or some empty strings<a name="line.307"></a>
-<span class="sourceLineNo">308</span> if (shortestStrLen == 0) {<a name="line.308"></a>
-<span class="sourceLineNo">309</span> return 0;<a name="line.309"></a>
-<span class="sourceLineNo">310</span> }<a name="line.310"></a>
-<span class="sourceLineNo">311</span><a name="line.311"></a>
-<span class="sourceLineNo">312</span> // find the position with the first difference across all strings<a name="line.312"></a>
-<span class="sourceLineNo">313</span> int firstDiff = -1;<a name="line.313"></a>
-<span class="sourceLineNo">314</span> for (int stringPos = 0; stringPos < shortestStrLen; stringPos++) {<a name="line.314"></a>
-<span class="sourceLineNo">315</span> final char comparisonChar = css[0].charAt(stringPos);<a name="line.315"></a>
-<span class="sourceLineNo">316</span> for (int arrayPos = 1; arrayPos < arrayLen; arrayPos++) {<a name="line.316"></a>
-<span class="sourceLineNo">317</span> if (css[arrayPos].charAt(stringPos) != comparisonChar) {<a name="line.317"></a>
-<span class="sourceLineNo">318</span> firstDiff = stringPos;<a name="line.318"></a>
-<span class="sourceLineNo">319</span> break;<a name="line.319"></a>
-<span class="sourceLineNo">320</span> }<a name="line.320"></a>
-<span class="sourceLineNo">321</span> }<a name="line.321"></a>
-<span class="sourceLineNo">322</span> if (firstDiff != -1) {<a name="line.322"></a>
-<span class="sourceLineNo">323</span> break;<a name="line.323"></a>
-<span class="sourceLineNo">324</span> }<a name="line.324"></a>
-<span class="sourceLineNo">325</span> }<a name="line.325"></a>
-<span class="sourceLineNo">326</span><a name="line.326"></a>
-<span class="sourceLineNo">327</span> if (firstDiff == -1 && shortestStrLen != longestStrLen) {<a name="line.327"></a>
-<span class="sourceLineNo">328</span> // we compared all of the characters up to the length of the<a name="line.328"></a>
-<span class="sourceLineNo">329</span> // shortest string and didn't find a match, but the string lengths<a name="line.329"></a>
-<span class="sourceLineNo">330</span> // vary, so return the length of the shortest string.<a name="line.330"></a>
-<span class="sourceLineNo">331</span> return shortestStrLen;<a name="line.331"></a>
-<span class="sourceLineNo">332</span> }<a name="line.332"></a>
-<span class="sourceLineNo">333</span> return firstDiff;<a name="line.333"></a>
-<span class="sourceLineNo">334</span> }<a name="line.334"></a>
-<span class="sourceLineNo">335</span><a name="line.335"></a>
-<span class="sourceLineNo">336</span> /**<a name="line.336"></a>
-<span class="sourceLineNo">337</span> * Gets a set of matching characters between two strings.<a name="line.337"></a>
-<span class="sourceLineNo">338</span> *<a name="line.338"></a>
-<span class="sourceLineNo">339</span> * <p><a name="line.339"></a>
-<span class="sourceLineNo">340</span> * Two characters from the first string and the second string are<a name="line.340"></a>
-<span class="sourceLineNo">341</span> * considered matching if the character's respective positions are no<a name="line.341"></a>
-<span class="sourceLineNo">342</span> * farther than the limit value.<a name="line.342"></a>
-<span class="sourceLineNo">343</span> * </p><a name="line.343"></a>
-<span class="sourceLineNo">344</span> *<a name="line.344"></a>
-<span class="sourceLineNo">345</span> * @param first The first string.<a name="line.345"></a>
-<span class="sourceLineNo">346</span> * @param second The second string.<a name="line.346"></a>
-<span class="sourceLineNo">347</span> * @param limit The maximum distance to consider.<a name="line.347"></a>
-<span class="sourceLineNo">348</span> * @return A string contain the set of common characters.<a name="line.348"></a>
-<span class="sourceLineNo">349</span> */<a name="line.349"></a>
-<span class="sourceLineNo">350</span> protected static String getSetOfMatchingCharacterWithin(<a name="line.350"></a>
-<span class="sourceLineNo">351</span> final CharSequence first, final CharSequence second, final int limit) {<a name="line.351"></a>
-<span class="sourceLineNo">352</span> final StringBuilder common = new StringBuilder();<a name="line.352"></a>
-<span class="sourceLineNo">353</span> final StringBuilder copy = new StringBuilder(second);<a name="line.353"></a>
-<span class="sourceLineNo">354</span><a name="line.354"></a>
-<span class="sourceLineNo">355</span> for (int i = 0; i < first.length(); i++) {<a name="line.355"></a>
-<span class="sourceLineNo">356</span> final char ch = first.charAt(i);<a name="line.356"></a>
-<span class="sourceLineNo">357</span> boolean found = false;<a name="line.357"></a>
-<span class="sourceLineNo">358</span><a name="line.358"></a>
-<span class="sourceLineNo">359</span> // See if the character is within the limit positions away from the<a name="line.359"></a>
-<span class="sourceLineNo">360</span> // original position of that character.<a name="line.360"></a>
-<span class="sourceLineNo">361</span> for (int j = Math.max(0, i - limit); !found<a name="line.361"></a>
-<span class="sourceLineNo">362</span> && j < Math.min(i + limit, second.length()); j++) {<a name="line.362"></a>
-<span class="sourceLineNo">363</span> if (copy.charAt(j) == ch) {<a name="line.363"></a>
-<span class="sourceLineNo">364</span> found = true;<a name="line.364"></a>
-<span class="sourceLineNo">365</span> common.append(ch);<a name="line.365"></a>
-<span class="sourceLineNo">366</span> copy.setCharAt(j, '*');<a name="line.366"></a>
-<span class="sourceLineNo">367</span> }<a name="line.367"></a>
-<span class="sourceLineNo">368</span> }<a name="line.368"></a>
-<span class="sourceLineNo">369</span> }<a name="line.369"></a>
-<span class="sourceLineNo">370</span> return common.toString();<a name="line.370"></a>
-<span class="sourceLineNo">371</span> }<a name="line.371"></a>
-<span class="sourceLineNo">372</span><a name="line.372"></a>
-<span class="sourceLineNo">373</span>}<a name="line.373"></a>
+<span class="sourceLineNo">046</span> * Find the Jaro Winkler Distance which indicates the similarity score<a name="line.46"></a>
+<span class="sourceLineNo">047</span> * between two CharSequences.<a name="line.47"></a>
+<span class="sourceLineNo">048</span> *<a name="line.48"></a>
+<span class="sourceLineNo">049</span> * <pre><a name="line.49"></a>
+<span class="sourceLineNo">050</span> * distance.compare(null, null) = IllegalArgumentException<a name="line.50"></a>
+<span class="sourceLineNo">051</span> * distance.compare("","") = 0.0<a name="line.51"></a>
+<span class="sourceLineNo">052</span> * distance.compare("","a") = 0.0<a name="line.52"></a>
+<span class="sourceLineNo">053</span> * distance.compare("aaapppp", "") = 0.0<a name="line.53"></a>
+<span class="sourceLineNo">054</span> * distance.compare("frog", "fog") = 0.93<a name="line.54"></a>
+<span class="sourceLineNo">055</span> * distance.compare("fly", "ant") = 0.0<a name="line.55"></a>
+<span class="sourceLineNo">056</span> * distance.compare("elephant", "hippo") = 0.44<a name="line.56"></a>
+<span class="sourceLineNo">057</span> * distance.compare("hippo", "elephant") = 0.44<a name="line.57"></a>
+<span class="sourceLineNo">058</span> * distance.compare("hippo", "zzzzzzzz") = 0.0<a name="line.58"></a>
+<span class="sourceLineNo">059</span> * distance.compare("hello", "hallo") = 0.88<a name="line.59"></a>
+<span class="sourceLineNo">060</span> * distance.compare("ABC Corporation", "ABC Corp") = 0.91<a name="line.60"></a>
+<span class="sourceLineNo">061</span> * distance.compare("D N H Enterprises Inc", "D &amp; H Enterprises, Inc.") = 0.93<a name="line.61"></a>
+<span class="sourceLineNo">062</span> * distance.compare("My Gym Children's Fitness Center", "My Gym. Childrens Fitness") = 0.94<a name="line.62"></a>
+<span class="sourceLineNo">063</span> * distance.compare("PENNSYLVANIA", "PENNCISYLVNIA") = 0.9<a name="line.63"></a>
+<span class="sourceLineNo">064</span> * </pre><a name="line.64"></a>
+<span class="sourceLineNo">065</span> *<a name="line.65"></a>
+<span class="sourceLineNo">066</span> * @param left the first String, must not be null<a name="line.66"></a>
+<span class="sourceLineNo">067</span> * @param right the second String, must not be null<a name="line.67"></a>
+<span class="sourceLineNo">068</span> * @return result distance<a name="line.68"></a>
+<span class="sourceLineNo">069</span> * @throws IllegalArgumentException if either String input {@code null}<a name="line.69"></a>
+<span class="sourceLineNo">070</span> */<a name="line.70"></a>
+<span class="sourceLineNo">071</span> @Override<a name="line.71"></a>
+<span class="sourceLineNo">072</span> public Double compare(CharSequence left, CharSequence right) {<a name="line.72"></a>
+<span class="sourceLineNo">073</span> final double DEFAULT_SCALING_FACTOR = 0.1;<a name="line.73"></a>
+<span class="sourceLineNo">074</span><a name="line.74"></a>
+<span class="sourceLineNo">075</span> if (left == null || right == null) {<a name="line.75"></a>
+<span class="sourceLineNo">076</span> throw new IllegalArgumentException("Strings must not be null");<a name="line.76"></a>
+<span class="sourceLineNo">077</span> }<a name="line.77"></a>
+<span class="sourceLineNo">078</span><a name="line.78"></a>
+<span class="sourceLineNo">079</span> final double jaro = score(left, right);<a name="line.79"></a>
+<span class="sourceLineNo">080</span> final int cl = commonPrefixLength(left, right);<a name="line.80"></a>
+<span class="sourceLineNo">081</span> final double matchScore = Math.round((jaro + (DEFAULT_SCALING_FACTOR<a name="line.81"></a>
+<span class="sourceLineNo">082</span> * cl * (1.0 - jaro))) *100.0)/100.0;<a name="line.82"></a>
+<span class="sourceLineNo">083</span><a name="line.83"></a>
+<span class="sourceLineNo">084</span> return matchScore;<a name="line.84"></a>
+<span class="sourceLineNo">085</span> }<a name="line.85"></a>
+<span class="sourceLineNo">086</span><a name="line.86"></a>
+<span class="sourceLineNo">087</span> /**<a name="line.87"></a>
+<span class="sourceLineNo">088</span> * Calculates the number of characters from the beginning of the strings<a name="line.88"></a>
+<span class="sourceLineNo">089</span> * that match exactly one-to-one, up to a maximum of four (4) characters.<a name="line.89"></a>
+<span class="sourceLineNo">090</span> *<a name="line.90"></a>
+<span class="sourceLineNo">091</span> * @param first The first string.<a name="line.91"></a>
+<span class="sourceLineNo">092</span> * @param second The second string.<a name="line.92"></a>
+<span class="sourceLineNo">093</span> * @return A number between 0 and 4.<a name="line.93"></a>
+<span class="sourceLineNo">094</span> */<a name="line.94"></a>
+<span class="sourceLineNo">095</span> private static int commonPrefixLength(final CharSequence first,<a name="line.95"></a>
+<span class="sourceLineNo">096</span> final CharSequence second) {<a name="line.96"></a>
+<span class="sourceLineNo">097</span> final int result = getCommonPrefix(first.toString(), second.toString())<a name="line.97"></a>
+<span class="sourceLineNo">098</span> .length();<a name="line.98"></a>
+<span class="sourceLineNo">099</span><a name="line.99"></a>
+<span class="sourceLineNo">100</span> // Limit the result to 4.<a name="line.100"></a>
+<span class="sourceLineNo">101</span> return result > 4 ? 4 : result;<a name="line.101"></a>
+<span class="sourceLineNo">102</span> }<a name="line.102"></a>
+<span class="sourceLineNo">103</span><a name="line.103"></a>
+<span class="sourceLineNo">104</span> /**<a name="line.104"></a>
+<span class="sourceLineNo">105</span> * Compares all Strings in an array and returns the initial sequence of<a name="line.105"></a>
+<span class="sourceLineNo">106</span> * characters that is common to all of them.<a name="line.106"></a>
+<span class="sourceLineNo">107</span> *<a name="line.107"></a>
+<span class="sourceLineNo">108</span> * <p><a name="line.108"></a>
+<span class="sourceLineNo">109</span> * For example,<a name="line.109"></a>
+<span class="sourceLineNo">110</span> * <code>getCommonPrefix(new String[] {"i am a machine", "i am a robot"}) -&gt; "i am a "</code><a name="line.110"></a>
+<span class="sourceLineNo">111</span> * </p><a name="line.111"></a>
+<span class="sourceLineNo">112</span> *<a name="line.112"></a>
+<span class="sourceLineNo">113</span> * <pre><a name="line.113"></a>
+<span class="sourceLineNo">114</span> * getCommonPrefix(null) = ""<a name="line.114"></a>
+<span class="sourceLineNo">115</span> * getCommonPrefix(new String[] {}) = ""<a name="line.115"></a>
+<span class="sourceLineNo">116</span> * getCommonPrefix(new String[] {"abc"}) = "abc"<a name="line.116"></a>
+<span class="sourceLineNo">117</span> * getCommonPrefix(new String[] {null, null}) = ""<a name="line.117"></a>
+<span class="sourceLineNo">118</span> * getCommonPrefix(new String[] {"", ""}) = ""<a name="line.118"></a>
+<span class="sourceLineNo">119</span> * getCommonPrefix(new String[] {"", null}) = ""<a name="line.119"></a>
+<span class="sourceLineNo">120</span> * getCommonPrefix(new String[] {"abc", null, null}) = ""<a name="line.120"></a>
+<span class="sourceLineNo">121</span> * getCommonPrefix(new String[] {null, null, "abc"}) = ""<a name="line.121"></a>
+<span class="sourceLineNo">122</span> * getCommonPrefix(new String[] {"", "abc"}) = ""<a name="line.122"></a>
+<span class="sourceLineNo">123</span> * getCommonPrefix(new String[] {"abc", ""}) = ""<a name="line.123"></a>
+<span class="sourceLineNo">124</span> * getCommonPrefix(new String[] {"abc", "abc"}) = "abc"<a name="line.124"></a>
+<span class="sourceLineNo">125</span> * getCommonPrefix(new String[] {"abc", "a"}) = "a"<a name="line.125"></a>
+<span class="sourceLineNo">126</span> * getCommonPrefix(new String[] {"ab", "abxyz"}) = "ab"<a name="line.126"></a>
+<span class="sourceLineNo">127</span> * getCommonPrefix(new String[] {"abcde", "abxyz"}) = "ab"<a name="line.127"></a>
+<span class="sourceLineNo">128</span> * getCommonPrefix(new String[] {"abcde", "xyz"}) = ""<a name="line.128"></a>
+<span class="sourceLineNo">129</span> * getCommonPrefix(new String[] {"xyz", "abcde"}) = ""<a name="line.129"></a>
+<span class="sourceLineNo">130</span> * getCommonPrefix(new String[] {"i am a machine", "i am a robot"}) = "i am a "<a name="line.130"></a>
+<span class="sourceLineNo">131</span> * </pre><a name="line.131"></a>
+<span class="sourceLineNo">132</span> *<a name="line.132"></a>
+<span class="sourceLineNo">133</span> * @param strs array of String objects, entries may be null<a name="line.133"></a>
+<span class="sourceLineNo">134</span> * @return the initial sequence of characters that are common to all Strings<a name="line.134"></a>
+<span class="sourceLineNo">135</span> * in the array; empty String if the array is null, the elements are<a name="line.135"></a>
+<span class="sourceLineNo">136</span> * all null or if there is no common prefix.<a name="line.136"></a>
+<span class="sourceLineNo">137</span> */<a name="line.137"></a>
+<span class="sourceLineNo">138</span> public static String getCommonPrefix(final String... strs) {<a name="line.138"></a>
+<span class="sourceLineNo">139</span> if (strs == null || strs.length == 0) {<a name="line.139"></a>
+<span class="sourceLineNo">140</span> return "";<a name="line.140"></a>
+<span class="sourceLineNo">141</span> }<a name="line.141"></a>
+<span class="sourceLineNo">142</span> final int smallestIndexOfDiff = indexOfDifference(strs);<a name="line.142"></a>
+<span class="sourceLineNo">143</span> if (smallestIndexOfDiff == INDEX_NOT_FOUND) {<a name="line.143"></a>
+<span class="sourceLineNo">144</span> // all strings were identical<a name="line.144"></a>
+<span class="sourceLineNo">145</span> if (strs[0] == null) {<a name="line.145"></a>
+<span class="sourceLineNo">146</span> return "";<a name="line.146"></a>
+<span class="sourceLineNo">147</span> }<a name="line.147"></a>
+<span class="sourceLineNo">148</span> return strs[0];<a name="line.148"></a>
+<span class="sourceLineNo">149</span> } else if (smallestIndexOfDiff == 0) {<a name="line.149"></a>
+<span class="sourceLineNo">150</span> // there were no common initial characters<a name="line.150"></a>
+<span class="sourceLineNo">151</span> return "";<a name="line.151"></a>
+<span class="sourceLineNo">152</span> } else {<a name="line.152"></a>
+<span class="sourceLineNo">153</span> // we found a common initial character sequence<a name="line.153"></a>
+<span class="sourceLineNo">154</span> return strs[0].substring(0, smallestIndexOfDiff);<a name="line.154"></a>
+<span class="sourceLineNo">155</span> }<a name="line.155"></a>
+<span class="sourceLineNo">156</span> }<a name="line.156"></a>
+<span class="sourceLineNo">157</span><a name="line.157"></a>
+<span class="sourceLineNo">158</span> /**<a name="line.158"></a>
+<span class="sourceLineNo">159</span> * This method returns the Jaro-Winkler score for string matching.<a name="line.159"></a>
+<span class="sourceLineNo">160</span> *<a name="line.160"></a>
+<span class="sourceLineNo">161</span> * @param first the first string to be matched<a name="line.161"></a>
+<span class="sourceLineNo">162</span> * @param second the second string to be machted<a name="line.162"></a>
+<span class="sourceLineNo">163</span> * @return matching score without scaling factor impact<a name="line.163"></a>
+<span class="sourceLineNo">164</span> */<a name="line.164"></a>
+<span class="sourceLineNo">165</span> protected static double score(final CharSequence first,<a name="line.165"></a>
+<span class="sourceLineNo">166</span> final CharSequence second) {<a name="line.166"></a>
+<span class="sourceLineNo">167</span> String shorter;<a name="line.167"></a>
+<span class="sourceLineNo">168</span> String longer;<a name="line.168"></a>
+<span class="sourceLineNo">169</span><a name="line.169"></a>
+<span class="sourceLineNo">170</span> // Determine which String is longer.<a name="line.170"></a>
+<span class="sourceLineNo">171</span> if (first.length() > second.length()) {<a name="line.171"></a>
+<span class="sourceLineNo">172</span> longer = first.toString().toLowerCase();<a name="line.172"></a>
+<span class="sourceLineNo">173</span> shorter = second.toString().toLowerCase();<a name="line.173"></a>
+<span class="sourceLineNo">174</span> } else {<a name="line.174"></a>
+<span class="sourceLineNo">175</span> longer = second.toString().toLowerCase();<a name="line.175"></a>
+<span class="sourceLineNo">176</span> shorter = first.toString().toLowerCase();<a name="line.176"></a>
+<span class="sourceLineNo">177</span> }<a name="line.177"></a>
+<span class="sourceLineNo">178</span><a name="line.178"></a>
+<span class="sourceLineNo">179</span> // Calculate the half length() distance of the shorter String.<a name="line.179"></a>
+<span class="sourceLineNo">180</span> final int halflength = shorter.length() / 2 + 1;<a name="line.180"></a>
+<span class="sourceLineNo">181</span><a name="line.181"></a>
+<span class="sourceLineNo">182</span> // Find the set of matching characters between the shorter and longer<a name="line.182"></a>
+<span class="sourceLineNo">183</span> // strings. Note that<a name="line.183"></a>
+<span class="sourceLineNo">184</span> // the set of matching characters may be different depending on the<a name="line.184"></a>
+<span class="sourceLineNo">185</span> // order of the strings.<a name="line.185"></a>
+<span class="sourceLineNo">186</span> final String m1 = getSetOfMatchingCharacterWithin(shorter, longer,<a name="line.186"></a>
+<span class="sourceLineNo">187</span> halflength);<a name="line.187"></a>
+<span class="sourceLineNo">188</span> final String m2 = getSetOfMatchingCharacterWithin(longer, shorter,<a name="line.188"></a>
+<span class="sourceLineNo">189</span> halflength);<a name="line.189"></a>
+<span class="sourceLineNo">190</span><a name="line.190"></a>
+<span class="sourceLineNo">191</span> // If one or both of the sets of common characters is empty, then<a name="line.191"></a>
+<span class="sourceLineNo">192</span> // there is no similarity between the two strings.<a name="line.192"></a>
+<span class="sourceLineNo">193</span> if (m1.length() == 0 || m2.length() == 0) {<a name="line.193"></a>
+<span class="sourceLineNo">194</span> return 0.0;<a name="line.194"></a>
+<span class="sourceLineNo">195</span> }<a name="line.195"></a>
+<span class="sourceLineNo">196</span><a name="line.196"></a>
+<span class="sourceLineNo">197</span> // If the set of common characters is not the same size, then<a name="line.197"></a>
+<span class="sourceLineNo">198</span> // there is no similarity between the two strings, either.<a name="line.198"></a>
+<span class="sourceLineNo">199</span> if (m1.length() != m2.length()) {<a name="line.199"></a>
+<span class="sourceLineNo">200</span> return 0.0;<a name="line.200"></a>
+<span class="sourceLineNo">201</span> }<a name="line.201"></a>
+<span class="sourceLineNo">202</span><a name="line.202"></a>
+<span class="sourceLineNo">203</span> // Calculate the number of transposition between the two sets<a name="line.203"></a>
+<span class="sourceLineNo">204</span> // of common characters.<a name="line.204"></a>
+<span class="sourceLineNo">205</span> final int transpositions = transpositions(m1, m2);<a name="line.205"></a>
+<span class="sourceLineNo">206</span><a name="line.206"></a>
+<span class="sourceLineNo">207</span> // Calculate the distance.<a name="line.207"></a>
+<span class="sourceLineNo">208</span> final double dist = (m1.length() / ((double) shorter.length())<a name="line.208"></a>
+<span class="sourceLineNo">209</span> + m2.length() / ((double) longer.length()) + (m1.length() - transpositions)<a name="line.209"></a>
+<span class="sourceLineNo">210</span> / ((double) m1.length())) / 3.0;<a name="line.210"></a>
+<span class="sourceLineNo">211</span> return dist;<a name="line.211"></a>
+<span class="sourceLineNo">212</span> }<a name="line.212"></a>
+<span class="sourceLineNo">213</span><a name="line.213"></a>
+<span class="sourceLineNo">214</span> /**<a name="line.214"></a>
+<span class="sourceLineNo">215</span> * Calculates the number of transposition between two strings.<a name="line.215"></a>
+<span class="sourceLineNo">216</span> *<a name="line.216"></a>
+<span class="sourceLineNo">217</span> * @param first The first string.<a name="line.217"></a>
+<span class="sourceLineNo">218</span> * @param second The second string.<a name="line.218"></a>
+<span class="sourceLineNo">219</span> * @return The number of transposition between the two strings.<a name="line.219"></a>
+<span class="sourceLineNo">220</span> */<a name="line.220"></a>
+<span class="sourceLineNo">221</span> protected static int transpositions(final CharSequence first,<a name="line.221"></a>
+<span class="sourceLineNo">222</span> final CharSequence second) {<a name="line.222"></a>
+<span class="sourceLineNo">223</span> int transpositions = 0;<a name="line.223"></a>
+<span class="sourceLineNo">224</span> for (int i = 0; i < first.length(); i++) {<a name="line.224"></a>
+<span class="sourceLineNo">225</span> if (first.charAt(i) != second.charAt(i)) {<a name="line.225"></a>
+<span class="sourceLineNo">226</span> transpositions++;<a name="line.226"></a>
+<span class="sourceLineNo">227</span> }<a name="line.227"></a>
+<span class="sourceLineNo">228</span> }<a name="line.228"></a>
+<span class="sourceLineNo">229</span> return transpositions / 2;<a name="line.229"></a>
+<span class="sourceLineNo">230</span> }<a name="line.230"></a>
+<span class="sourceLineNo">231</span><a name="line.231"></a>
+<span class="sourceLineNo">232</span> /**<a name="line.232"></a>
+<span class="sourceLineNo">233</span> * Compares all CharSequences in an array and returns the index at which the<a name="line.233"></a>
+<span class="sourceLineNo">234</span> * CharSequences begin to differ.<a name="line.234"></a>
+<span class="sourceLineNo">235</span> *<a name="line.235"></a>
+<span class="sourceLineNo">236</span> * <p><a name="line.236"></a>
+<span class="sourceLineNo">237</span> * For example,<a name="line.237"></a>
+<span class="sourceLineNo">238</span> * <code>indexOfDifference(new String[] {"i am a machine", "i am a robot"}) -&gt; 7</code><a name="line.238"></a>
+<span class="sourceLineNo">239</span> * </p><a name="line.239"></a>
+<span class="sourceLineNo">240</span> *<a name="line.240"></a>
+<span class="sourceLineNo">241</span> * <pre><a name="line.241"></a>
+<span class="sourceLineNo">242</span> * distance.indexOfDifference(null) = -1<a name="line.242"></a>
+<span class="sourceLineNo">243</span> * distance.indexOfDifference(new String[] {}) = -1<a name="line.243"></a>
+<span class="sourceLineNo">244</span> * distance.indexOfDifference(new String[] {"abc"}) = -1<a name="line.244"></a>
+<span class="sourceLineNo">245</span> * distance.indexOfDifference(new String[] {null, null}) = -1<a name="line.245"></a>
+<span class="sourceLineNo">246</span> * distance.indexOfDifference(new String[] {"", ""}) = -1<a name="line.246"></a>
+<span class="sourceLineNo">247</span> * distance.indexOfDifference(new String[] {"", null}) = 0<a name="line.247"></a>
+<span class="sourceLineNo">248</span> * distance.indexOfDifference(new String[] {"abc", null, null}) = 0<a name="line.248"></a>
+<span class="sourceLineNo">249</span> * distance.indexOfDifference(new String[] {null, null, "abc"}) = 0<a name="line.249"></a>
+<span class="sourceLineNo">250</span> * distance.indexOfDifference(new String[] {"", "abc"}) = 0<a name="line.250"></a>
+<span class="sourceLineNo">251</span> * distance.indexOfDifference(new String[] {"abc", ""}) = 0<a name="line.251"></a>
+<span class="sourceLineNo">252</span> * distance.indexOfDifference(new String[] {"abc", "abc"}) = -1<a name="line.252"></a>
+<span class="sourceLineNo">253</span> * distance.indexOfDifference(new String[] {"abc", "a"}) = 1<a name="line.253"></a>
+<span class="sourceLineNo">254</span> * distance.indexOfDifference(new String[] {"ab", "abxyz"}) = 2<a name="line.254"></a>
+<span class="sourceLineNo">255</span> * distance.indexOfDifference(new String[] {"abcde", "abxyz"}) = 2<a name="line.255"></a>
+<span class="sourceLineNo">256</span> * distance.indexOfDifference(new String[] {"abcde", "xyz"}) = 0<a name="line.256"></a>
+<span class="sourceLineNo">257</span> * distance.indexOfDifference(new String[] {"xyz", "abcde"}) = 0<a name="line.257"></a>
+<span class="sourceLineNo">258</span> * distance.indexOfDifference(new String[] {"i am a machine", "i am a robot"}) = 7<a name="line.258"></a>
+<span class="sourceLineNo">259</span> * </pre><a name="line.259"></a>
+<span class="sourceLineNo">260</span> *<a name="line.260"></a>
+<span class="sourceLineNo">261</span> * @param css array of CharSequences, entries may be null<a name="line.261"></a>
+<span class="sourceLineNo">262</span> * @return the index where the strings begin to differ; -1 if they are all<a name="line.262"></a>
+<span class="sourceLineNo">263</span> * equal<a name="line.263"></a>
+<span class="sourceLineNo">264</span> */<a name="line.264"></a>
+<span class="sourceLineNo">265</span> protected static int indexOfDifference(final CharSequence... css) {<a name="line.265"></a>
+<span class="sourceLineNo">266</span> if (css == null || css.length <= 1) {<a name="line.266"></a>
+<span class="sourceLineNo">267</span> return INDEX_NOT_FOUND;<a name="line.267"></a>
+<span class="sourceLineNo">268</span> }<a name="line.268"></a>
+<span class="sourceLineNo">269</span> boolean anyStringNull = false;<a name="line.269"></a>
+<span class="sourceLineNo">270</span> boolean allStringsNull = true;<a name="line.270"></a>
+<span class="sourceLineNo">271</span> final int arrayLen = css.length;<a name="line.271"></a>
+<span class="sourceLineNo">272</span> int shortestStrLen = Integer.MAX_VALUE;<a name="line.272"></a>
+<span class="sourceLineNo">273</span> int longestStrLen = 0;<a name="line.273"></a>
+<span class="sourceLineNo">274</span><a name="line.274"></a>
+<span class="sourceLineNo">275</span> // find the min and max string lengths; this avoids checking to make<a name="line.275"></a>
+<span class="sourceLineNo">276</span> // sure we are not exceeding the length of the string each time through<a name="line.276"></a>
+<span class="sourceLineNo">277</span> // the bottom loop.<a name="line.277"></a>
+<span class="sourceLineNo">278</span> for (int i = 0; i < arrayLen; i++) {<a name="line.278"></a>
+<span class="sourceLineNo">279</span> if (css[i] == null) {<a name="line.279"></a>
+<span class="sourceLineNo">280</span> anyStringNull = true;<a name="line.280"></a>
+<span class="sourceLineNo">281</span> shortestStrLen = 0;<a name="line.281"></a>
+<span class="sourceLineNo">282</span> } else {<a name="line.282"></a>
+<span class="sourceLineNo">283</span> allStringsNull = false;<a name="line.283"></a>
+<span class="sourceLineNo">284</span> shortestStrLen = Math.min(css[i].length(), shortestStrLen);<a name="line.284"></a>
+<span class="sourceLineNo">285</span> longestStrLen = Math.max(css[i].length(), longestStrLen);<a name="line.285"></a>
+<span class="sourceLineNo">286</span> }<a name="line.286"></a>
+<span class="sourceLineNo">287</span> }<a name="line.287"></a>
+<span class="sourceLineNo">288</span><a name="line.288"></a>
+<span class="sourceLineNo">289</span> // handle lists containing all nulls or all empty strings<a name="line.289"></a>
+<span class="sourceLineNo">290</span> if (allStringsNull || longestStrLen == 0 && !anyStringNull) {<a name="line.290"></a>
+<span class="sourceLineNo">291</span> return INDEX_NOT_FOUND;<a name="line.291"></a>
+<span class="sourceLineNo">292</span> }<a name="line.292"></a>
+<span class="sourceLineNo">293</span><a name="line.293"></a>
+<span class="sourceLineNo">294</span> // handle lists containing some nulls or some empty strings<a name="line.294"></a>
+<span class="sourceLineNo">295</span> if (shortestStrLen == 0) {<a name="line.295"></a>
+<span class="sourceLineNo">296</span> return 0;<a name="line.296"></a>
+<span class="sourceLineNo">297</span> }<a name="line.297"></a>
+<span class="sourceLineNo">298</span><a name="line.298"></a>
+<span class="sourceLineNo">299</span> // find the position with the first difference across all strings<a name="line.299"></a>
+<span class="sourceLineNo">300</span> int firstDiff = -1;<a name="line.300"></a>
+<span class="sourceLineNo">301</span> for (int stringPos = 0; stringPos < shortestStrLen; stringPos++) {<a name="line.301"></a>
+<span class="sourceLineNo">302</span> final char comparisonChar = css[0].charAt(stringPos);<a name="line.302"></a>
+<span class="sourceLineNo">303</span> for (int arrayPos = 1; arrayPos < arrayLen; arrayPos++) {<a name="line.303"></a>
+<span class="sourceLineNo">304</span> if (css[arrayPos].charAt(stringPos) != comparisonChar) {<a name="line.304"></a>
+<span class="sourceLineNo">305</span> firstDiff = stringPos;<a name="line.305"></a>
+<span class="sourceLineNo">306</span> break;<a name="line.306"></a>
+<span class="sourceLineNo">307</span> }<a name="line.307"></a>
+<span class="sourceLineNo">308</span> }<a name="line.308"></a>
+<span class="sourceLineNo">309</span> if (firstDiff != -1) {<a name="line.309"></a>
+<span class="sourceLineNo">310</span> break;<a name="line.310"></a>
+<span class="sourceLineNo">311</span> }<a name="line.311"></a>
+<span class="sourceLineNo">312</span> }<a name="line.312"></a>
+<span class="sourceLineNo">313</span><a name="line.313"></a>
+<span class="sourceLineNo">314</span> if (firstDiff == -1 && shortestStrLen != longestStrLen) {<a name="line.314"></a>
+<span class="sourceLineNo">315</span> // we compared all of the characters up to the length of the<a name="line.315"></a>
+<span class="sourceLineNo">316</span> // shortest string and didn't find a match, but the string lengths<a name="line.316"></a>
+<span class="sourceLineNo">317</span> // vary, so return the length of the shortest string.<a name="line.317"></a>
+<span class="sourceLineNo">318</span> return shortestStrLen;<a name="line.318"></a>
+<span class="sourceLineNo">319</span> }<a name="line.319"></a>
+<span class="sourceLineNo">320</span> return firstDiff;<a name="line.320"></a>
+<span class="sourceLineNo">321</span> }<a name="line.321"></a>
+<span class="sourceLineNo">322</span><a name="line.322"></a>
+<span class="sourceLineNo">323</span> /**<a name="line.323"></a>
+<span class="sourceLineNo">324</span> * Gets a set of matching characters between two strings.<a name="line.324"></a>
+<span class="sourceLineNo">325</span> *<a name="line.325"></a>
+<span class="sourceLineNo">326</span> * <p><a name="line.326"></a>
+<span class="sourceLineNo">327</span> * Two characters from the first string and the second string are<a name="line.327"></a>
+<span class="sourceLineNo">328</span> * considered matching if the character's respective positions are no<a name="line.328"></a>
+<span class="sourceLineNo">329</span> * farther than the limit value.<a name="line.329"></a>
+<span class="sourceLineNo">330</span> * </p><a name="line.330"></a>
+<span class="sourceLineNo">331</span> *<a name="line.331"></a>
+<span class="sourceLineNo">332</span> * @param first The first string.<a name="line.332"></a>
+<span class="sourceLineNo">333</span> * @param second The second string.<a name="line.333"></a>
+<span class="sourceLineNo">334</span> * @param limit The maximum distance to consider.<a name="line.334"></a>
+<span class="sourceLineNo">335</span> * @return A string contain the set of common characters.<a name="line.335"></a>
+<span class="sourceLineNo">336</span> */<a name="line.336"></a>
+<span class="sourceLineNo">337</span> protected static String getSetOfMatchingCharacterWithin(<a name="line.337"></a>
+<span class="sourceLineNo">338</span> final CharSequence first, final CharSequence second, final int limit) {<a name="line.338"></a>
+<span class="sourceLineNo">339</span> final StringBuilder common = new StringBuilder();<a name="line.339"></a>
+<span class="sourceLineNo">340</span> final StringBuilder copy = new StringBuilder(second);<a name="line.340"></a>
+<span class="sourceLineNo">341</span><a name="line.341"></a>
+<span class="sourceLineNo">342</span> for (int i = 0; i < first.length(); i++) {<a name="line.342"></a>
+<span class="sourceLineNo">343</span> final char ch = first.charAt(i);<a name="line.343"></a>
+<span class="sourceLineNo">344</span> boolean found = false;<a name="line.344"></a>
+<span class="sourceLineNo">345</span><a name="line.345"></a>
+<span class="sourceLineNo">346</span> // See if the character is within the limit positions away from the<a name="line.346"></a>
+<span class="sourceLineNo">347</span> // original position of that character.<a name="line.347"></a>
+<span class="sourceLineNo">348</span> for (int j = Math.max(0, i - limit); !found<a name="line.348"></a>
+<span class="sourceLineNo">349</span> && j < Math.min(i + limit, second.length()); j++) {<a name="line.349"></a>
+<span class="sourceLineNo">350</span> if (copy.charAt(j) == ch) {<a name="line.350"></a>
+<span class="sourceLineNo">351</span> found = true;<a name="line.351"></a>
+<span class="sourceLineNo">352</span> common.append(ch);<a name="line.352"></a>
+<span class="sourceLineNo">353</span> copy.setCharAt(j, '*');<a name="line.353"></a>
+<span class="sourceLineNo">354</span> }<a name="line.354"></a>
+<span class="sourceLineNo">355</span> }<a name="line.355"></a>
+<span class="sourceLineNo">356</span> }<a name="line.356"></a>
+<span class="sourceLineNo">357</span> return common.toString();<a name="line.357"></a>
+<span class="sourceLineNo">358</span> }<a name="line.358"></a>
+<span class="sourceLineNo">359</span><a name="line.359"></a>
+<span class="sourceLineNo">360</span>}<a name="line.360"></a>
@@ -442,4 +429,4 @@
</pre>
</div>
</body>
-</html>
+</html>
\ No newline at end of file