001/* 002 * Licensed to the Apache Software Foundation (ASF) under one or more 003 * contributor license agreements. See the NOTICE file distributed with 004 * this work for additional information regarding copyright ownership. 005 * The ASF licenses this file to You under the Apache License, Version 2.0 006 * (the "License"); you may not use this file except in compliance with 007 * the License. You may obtain a copy of the License at 008 * 009 * https://www.apache.org/licenses/LICENSE-2.0 010 * 011 * Unless required by applicable law or agreed to in writing, software 012 * distributed under the License is distributed on an "AS IS" BASIS, 013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 014 * See the License for the specific language governing permissions and 015 * limitations under the License. 016 */ 017 018package org.apache.commons.codec.language; 019 020import org.apache.commons.codec.EncoderException; 021import org.apache.commons.codec.StringEncoder; 022 023/** 024 * Encodes a string into a Soundex value. Soundex is an encoding used to relate similar names, but can also be used as a 025 * general purpose scheme to find word with similar phonemes. 026 * 027 * <p>This class is thread-safe. 028 * Although not strictly immutable, the mutable fields are not actually used.</p> 029 */ 030public class Soundex implements StringEncoder { 031 032 /** 033 * The marker character used to indicate a silent (ignored) character. 034 * These are ignored except when they appear as the first character. 035 * <p> 036 * Note: The {@link #US_ENGLISH_MAPPING_STRING} does not use this mechanism 037 * because changing it might break existing code. Mappings that don't contain 038 * a silent marker code are treated as though H and W are silent. 039 * </p> 040 * <p> 041 * To override this, use the {@link #Soundex(String, boolean)} constructor. 042 * </p> 043 * 044 * @since 1.11 045 */ 046 public static final char SILENT_MARKER = '-'; 047 048 /** 049 * This is a default mapping of the 26 letters used in US English. A value of {@code 0} for a letter position 050 * means do not encode, but treat as a separator when it occurs between consonants with the same code. 051 * <p> 052 * (This constant is provided as both an implementation convenience and to allow Javadoc to pick 053 * up the value for the constant values page.) 054 * </p> 055 * <p> 056 * <strong>Note that letters H and W are treated specially.</strong> 057 * They are ignored (after the first letter) and don't act as separators 058 * between consonants with the same code. 059 * </p> 060 */ 061 public static final String US_ENGLISH_MAPPING_STRING = "01230120022455012623010202"; 062 063 /** 064 * This is a default mapping of the 26 letters used in US English. A value of {@code 0} for a letter position 065 * means do not encode. 066 * 067 * @see Soundex#Soundex(char[]) 068 */ 069 private static final char[] US_ENGLISH_MAPPING = US_ENGLISH_MAPPING_STRING.toCharArray(); 070 071 /** 072 * An instance of Soundex using the US_ENGLISH_MAPPING mapping. 073 * This treats H and W as silent letters. 074 * Apart from when they appear as the first letter, they are ignored. 075 * They don't act as separators between duplicate codes. 076 * 077 * @see #US_ENGLISH_MAPPING_STRING 078 */ 079 public static final Soundex US_ENGLISH = new Soundex(); 080 081 /** 082 * An instance of Soundex using the Simplified Soundex mapping, as described here: 083 * http://west-penwith.org.uk/misc/soundex.htm 084 * <p> 085 * This treats H and W the same as vowels (AEIOUY). 086 * Such letters aren't encoded (after the first), but they do 087 * act as separators when dropping duplicate codes. 088 * The mapping is otherwise the same as for {@link #US_ENGLISH} 089 * </p> 090 * 091 * @since 1.11 092 */ 093 public static final Soundex US_ENGLISH_SIMPLIFIED = new Soundex(US_ENGLISH_MAPPING_STRING, false); 094 095 /** 096 * An instance of Soundex using the mapping as per the Genealogy site: 097 * http://www.genealogy.com/articles/research/00000060.html 098 * <p> 099 * This treats vowels (AEIOUY), H and W as silent letters. 100 * Such letters are ignored (after the first) and do not 101 * act as separators when dropping duplicate codes. 102 * </p> 103 * <p> 104 * The codes for consonants are otherwise the same as for 105 * {@link #US_ENGLISH_MAPPING_STRING} and {@link #US_ENGLISH_SIMPLIFIED} 106 * </p> 107 * 108 * @since 1.11 109 */ 110 public static final Soundex US_ENGLISH_GENEALOGY = new Soundex("-123-12--22455-12623-1-2-2"); 111 // ABCDEFGHIJKLMNOPQRSTUVWXYZ 112 113 /** 114 * The maximum length of a Soundex code - Soundex codes are only four characters by definition. 115 * 116 * @deprecated This feature is not needed since the encoding size must be constant. Will be removed in 2.0. 117 */ 118 @Deprecated 119 private int maxLength = 4; 120 121 /** 122 * Every letter of the alphabet is "mapped" to a numerical value. This char array holds the values to which each 123 * letter is mapped. This implementation contains a default map for US_ENGLISH 124 */ 125 private final char[] soundexMapping; 126 127 /** 128 * Should H and W be treated specially? 129 * <p> 130 * In versions of the code prior to 1.11, 131 * the code always treated H and W as silent (ignored) letters. 132 * If this field is false, H and W are no longer special-cased. 133 * </p> 134 */ 135 private final boolean specialCaseHW; 136 137 /** 138 * Creates an instance using US_ENGLISH_MAPPING. 139 * 140 * @see Soundex#Soundex(char[]) 141 * @see Soundex#US_ENGLISH_MAPPING_STRING 142 */ 143 public Soundex() { 144 this.soundexMapping = US_ENGLISH_MAPPING; 145 this.specialCaseHW = true; 146 } 147 148 /** 149 * Creates a Soundex instance using the given mapping. This constructor can be used to provide an internationalized 150 * mapping for a non-Western character set. 151 * <p> 152 * Every letter of the alphabet is "mapped" to a numerical value. This char array holds the values to which each 153 * letter is mapped. This implementation contains a default map for US_ENGLISH 154 * </p> 155 * <p> 156 * If the mapping contains an instance of {@link #SILENT_MARKER} then H and W are not given special treatment. 157 * </p> 158 * 159 * @param mapping 160 * Mapping array to use when finding the corresponding code for a given character. 161 */ 162 public Soundex(final char[] mapping) { 163 this.soundexMapping = mapping.clone(); 164 this.specialCaseHW = !hasMarker(this.soundexMapping); 165 } 166 167 /** 168 * Creates a refined Soundex instance using a custom mapping. This constructor can be used to customize the mapping, 169 * and/or possibly provide an internationalized mapping for a non-Western character set. 170 * <p> 171 * If the mapping contains an instance of {@link #SILENT_MARKER} then H and W are not given special treatment. 172 * </p> 173 * 174 * @param mapping 175 * Mapping string to use when finding the corresponding code for a given character. 176 * @since 1.4 177 */ 178 public Soundex(final String mapping) { 179 this.soundexMapping = mapping.toCharArray(); 180 this.specialCaseHW = !hasMarker(this.soundexMapping); 181 } 182 183 /** 184 * Creates a refined Soundex instance using a custom mapping. This constructor can be used to customize the mapping, 185 * and/or possibly provide an internationalized mapping for a non-Western character set. 186 * 187 * @param mapping 188 * Mapping string to use when finding the corresponding code for a given character. 189 * @param specialCaseHW if true, then 190 * @since 1.11 191 */ 192 public Soundex(final String mapping, final boolean specialCaseHW) { 193 this.soundexMapping = mapping.toCharArray(); 194 this.specialCaseHW = specialCaseHW; 195 } 196 197 /** 198 * Encodes the Strings and returns the number of characters in the two encoded Strings that are the same. This return value ranges from 0 through 4: 0 199 * indicates little or no similarity, and 4 indicates strong similarity or identical values. 200 * 201 * @param s1 A String that will be encoded and compared. 202 * @param s2 A String that will be encoded and compared. 203 * @return The number of characters in the two encoded Strings that are the same from 0 to 4. 204 * @see SoundexUtils#difference(StringEncoder,String,String) 205 * @see <a href="https://msdn.microsoft.com/library/default.asp?url=/library/en-us/tsqlref/ts_de-dz_8co5.asp"> MS T-SQL DIFFERENCE</a> 206 * 207 * @throws EncoderException if an error occurs encoding one of the strings. 208 * @since 1.3 209 */ 210 public int difference(final String s1, final String s2) throws EncoderException { 211 return SoundexUtils.difference(this, s1, s2); 212 } 213 214 /** 215 * Encodes an Object using the Soundex algorithm. This method is provided in order to satisfy the requirements of the Encoder interface, and will throw an 216 * EncoderException if the supplied object is not of type {@link String}. 217 * 218 * @param obj Object to encode. 219 * @return An object (or type {@link String}) containing the Soundex code which corresponds to the String supplied. 220 * @throws EncoderException if the parameter supplied is not of type {@link String}. 221 * @throws IllegalArgumentException if a character is not mapped. 222 */ 223 @Override 224 public Object encode(final Object obj) throws EncoderException { 225 if (!(obj instanceof String)) { 226 throw new EncoderException("Parameter supplied to Soundex encode is not of type java.lang.String"); 227 } 228 return soundex((String) obj); 229 } 230 231 /** 232 * Encodes a String using the Soundex algorithm. 233 * 234 * @param str A String object to encode. 235 * @return A Soundex code corresponding to the String supplied. 236 * @throws IllegalArgumentException if a character is not mapped. 237 */ 238 @Override 239 public String encode(final String str) { 240 return soundex(str); 241 } 242 243 /** 244 * Returns the maxLength. Standard Soundex 245 * 246 * @deprecated This feature is not needed since the encoding size must be constant. Will be removed in 2.0. 247 * @return the maxLength. 248 */ 249 @Deprecated 250 public int getMaxLength() { 251 return this.maxLength; 252 } 253 254 private boolean hasMarker(final char[] mapping) { 255 for (final char ch : mapping) { 256 if (ch == SILENT_MARKER) { 257 return true; 258 } 259 } 260 return false; 261 } 262 263 /** 264 * Maps the given upper-case character to its Soundex code. 265 * 266 * @param ch 267 * An upper-case character. 268 * @return A Soundex code. 269 * @throws IllegalArgumentException 270 * Thrown if {@code ch} is not mapped. 271 */ 272 private char map(final char ch) { 273 final int index = ch - 'A'; 274 if (index < 0 || index >= this.soundexMapping.length) { 275 throw new IllegalArgumentException("The character is not mapped: " + ch + " (index=" + index + ")"); 276 } 277 return this.soundexMapping[index]; 278 } 279 280 /** 281 * Sets the maxLength. 282 * 283 * @deprecated This feature is not needed since the encoding size must be constant. Will be removed in 2.0. 284 * @param maxLength 285 * The maxLength to set. 286 */ 287 @Deprecated 288 public void setMaxLength(final int maxLength) { 289 this.maxLength = maxLength; 290 } 291 292 /** 293 * Retrieves the Soundex code for a given String object. 294 * 295 * @param str String to encode using the Soundex algorithm. 296 * @return A Soundex code for the String supplied. 297 * @throws IllegalArgumentException if a character is not mapped. 298 */ 299 public String soundex(String str) { 300 if (str == null) { 301 return null; 302 } 303 str = SoundexUtils.clean(str); 304 if (str.isEmpty()) { 305 return str; 306 } 307 final char[] out = { '0', '0', '0', '0' }; 308 int count = 0; 309 final char first = str.charAt(0); 310 out[count++] = first; 311 char lastDigit = map(first); // previous digit 312 for (int i = 1; i < str.length() && count < out.length; i++) { 313 final char ch = str.charAt(i); 314 if (this.specialCaseHW && (ch == 'H' || ch == 'W')) { // these are ignored completely 315 continue; 316 } 317 final char digit = map(ch); 318 if (digit == SILENT_MARKER) { 319 continue; 320 } 321 if (digit != '0' && digit != lastDigit) { // don't store vowels or repeats 322 out[count++] = digit; 323 } 324 lastDigit = digit; 325 } 326 return new String(out); 327 } 328 329}