org.elasticsearch.index.analysis.phonetic.Nysiis Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of elasticsearch-analysis-phonetic
Show all versions of elasticsearch-analysis-phonetic
The Phonetic Analysis plugin integrates phonetic token filter analysis with elasticsearch.
/*
* Licensed to Elasticsearch under one or more contributor
* license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright
* ownership. Elasticsearch licenses this file to you under
* the Apache License, Version 2.0 (the "License"); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package org.elasticsearch.index.analysis.phonetic;
import org.apache.commons.codec.EncoderException;
import org.apache.commons.codec.StringEncoder;
import java.util.regex.Pattern;
/**
*
* Taken from commons-codec trunk (unreleased yet)
*
* Encodes a string into a NYSIIS value. NYSIIS is an encoding used to relate
* similar names, but can also be used as a general purpose scheme to find word
* with similar phonemes.
*
* NYSIIS features an accuracy increase of 2.7% over the traditional Soundex
* algorithm.
*
* Algorithm description:
*
* 1. Transcode first characters of name
* 1a. MAC -> MCC
* 1b. KN -> NN
* 1c. K -> C
* 1d. PH -> FF
* 1e. PF -> FF
* 1f. SCH -> SSS
* 2. Transcode last characters of name
* 2a. EE, IE -> Y
* 2b. DT,RT,RD,NT,ND -> D
* 3. First character of key = first character of name
* 4. Transcode remaining characters by following these rules, incrementing by one character each time
* 4a. EV -> AF else A,E,I,O,U -> A
* 4b. Q -> G
* 4c. Z -> S
* 4d. M -> N
* 4e. KN -> N else K -> C
* 4f. SCH -> SSS
* 4g. PH -> FF
* 4h. H -> If previous or next is nonvowel, previous
* 4i. W -> If previous is vowel, previous
* 4j. Add current to key if current != last key character
* 5. If last character is S, remove it
* 6. If last characters are AY, replace with Y
* 7. If last character is A, remove it
* 8. Collapse all strings of repeated characters
* 9. Add original first character of name as first character of key
*
*
* @see NYSIIS on Wikipedia
* @see NYSIIS on dropby.com
*
*/
public class Nysiis implements StringEncoder {
private static final char[] CHARS_A = new char[]{'A'};
private static final char[] CHARS_AF = new char[]{'A', 'F'};
private static final char[] CHARS_C = new char[]{'C'};
private static final char[] CHARS_FF = new char[]{'F', 'F'};
private static final char[] CHARS_G = new char[]{'G'};
private static final char[] CHARS_N = new char[]{'N'};
private static final char[] CHARS_NN = new char[]{'N', 'N'};
private static final char[] CHARS_S = new char[]{'S'};
private static final char[] CHARS_SSS = new char[]{'S', 'S', 'S'};
private static final Pattern PAT_MAC = Pattern.compile("^MAC");
private static final Pattern PAT_KN = Pattern.compile("^KN");
private static final Pattern PAT_K = Pattern.compile("^K");
private static final Pattern PAT_PH_PF = Pattern.compile("^(PH|PF)");
private static final Pattern PAT_SCH = Pattern.compile("^SCH");
private static final Pattern PAT_EE_IE = Pattern.compile("(EE|IE)$");
private static final Pattern PAT_DT_ETC = Pattern.compile("(DT|RT|RD|NT|ND)$");
private static final char SPACE = ' ';
private static final int TRUE_LENGTH = 6;
/**
* Tests if the given character is a vowel.
*
* @param c the character to test
* @return {@code true} if the character is a vowel, {@code false} otherwise
*/
private static boolean isVowel(final char c) {
return c == 'A' || c == 'E' || c == 'I' || c == 'O' || c == 'U';
}
/**
* Transcodes the remaining parts of the String. The method operates on a
* sliding window, looking at 4 characters at a time: [i-1, i, i+1, i+2].
*
* @param prev the previous character
* @param curr the current character
* @param next the next character
* @param aNext the after next character
* @return a transcoded array of characters, starting from the current
* position
*/
private static char[] transcodeRemaining(final char prev, final char curr, final char next, final char aNext) {
// 1. EV -> AF
if (curr == 'E' && next == 'V') {
return CHARS_AF;
}
// A, E, I, O, U -> A
if (isVowel(curr)) {
return CHARS_A;
}
// 2. Q -> G, Z -> S, M -> N
if (curr == 'Q') {
return CHARS_G;
} else if (curr == 'Z') {
return CHARS_S;
} else if (curr == 'M') {
return CHARS_N;
}
// 3. KN -> NN else K -> C
if (curr == 'K') {
if (next == 'N') {
return CHARS_NN;
} else {
return CHARS_C;
}
}
// 4. SCH -> SSS
if (curr == 'S' && next == 'C' && aNext == 'H') {
return CHARS_SSS;
}
// PH -> FF
if (curr == 'P' && next == 'H') {
return CHARS_FF;
}
// 5. H -> If previous or next is a non vowel, previous.
if (curr == 'H' && (!isVowel(prev) || !isVowel(next))) {
return new char[]{prev};
}
// 6. W -> If previous is vowel, previous.
if (curr == 'W' && isVowel(prev)) {
return new char[]{prev};
}
return new char[]{curr};
}
/**
* Indicates the strict mode.
*/
private final boolean strict;
/**
* Creates an instance of the {@link Nysiis} encoder with strict mode
* (original form), i.e. encoded strings have a maximum length of 6.
*/
public Nysiis() {
this(true);
}
/**
* Create an instance of the {@link Nysiis} encoder with the specified
* strict mode:
*
* - {@code true}: encoded strings have a maximum length of 6
- {@code false}:
* encoded strings may have arbitrary length
*
* @param strict the strict mode
*/
public Nysiis(final boolean strict) {
this.strict = strict;
}
/**
* Encodes an Object using the NYSIIS algorithm. This method is provided in
* order to satisfy the requirements of the Encoder interface, and will
* throw an {@link EncoderException} if the supplied object is not of type
* {@link String}.
*
* @param obj Object to encode
* @return An object (or a {@link String}) containing the NYSIIS code which
* corresponds to the given String.
* @throws EncoderException if the parameter supplied is not of a {@link String}
* @throws IllegalArgumentException if a character is not mapped
*/
@Override
public Object encode(Object obj) throws EncoderException {
if (!(obj instanceof String)) {
throw new EncoderException("Parameter supplied to Nysiis encode is not of type java.lang.String");
}
return this.nysiis((String) obj);
}
/**
* Encodes a String using the NYSIIS algorithm.
*
* @param str A String object to encode
* @return A Nysiis code corresponding to the String supplied
* @throws IllegalArgumentException if a character is not mapped
*/
@Override
public String encode(String str) {
return this.nysiis(str);
}
/**
* Indicates the strict mode for this {@link Nysiis} encoder.
*
* @return {@code true} if the encoder is configured for strict mode, {@code false}
* otherwise
*/
public boolean isStrict() {
return this.strict;
}
/**
* Retrieves the NYSIIS code for a given String object.
*
* @param str String to encode using the NYSIIS algorithm
* @return A NYSIIS code for the String supplied
*/
public String nysiis(String str) {
if (str == null) {
return null;
}
// Use the same clean rules as Soundex
str = clean(str);
if (str.length() == 0) {
return str;
}
// Translate first characters of name:
// MAC -> MCC, KN -> NN, K -> C, PH | PF -> FF, SCH -> SSS
str = PAT_MAC.matcher(str).replaceFirst("MCC");
str = PAT_KN.matcher(str).replaceFirst("NN");
str = PAT_K.matcher(str).replaceFirst("C");
str = PAT_PH_PF.matcher(str).replaceFirst("FF");
str = PAT_SCH.matcher(str).replaceFirst("SSS");
// Translate last characters of name:
// EE -> Y, IE -> Y, DT | RT | RD | NT | ND -> D
str = PAT_EE_IE.matcher(str).replaceFirst("Y");
str = PAT_DT_ETC.matcher(str).replaceFirst("D");
// First character of key = first character of name.
StringBuffer key = new StringBuffer(str.length());
key.append(str.charAt(0));
// Transcode remaining characters, incrementing by one character each time
final char[] chars = str.toCharArray();
final int len = chars.length;
for (int i = 1; i < len; i++) {
final char next = i < len - 1 ? chars[i + 1] : SPACE;
final char aNext = i < len - 2 ? chars[i + 2] : SPACE;
final char[] transcoded = transcodeRemaining(chars[i - 1], chars[i], next, aNext);
System.arraycopy(transcoded, 0, chars, i, transcoded.length);
// only append the current char to the key if it is different from the last one
if (chars[i] != chars[i - 1]) {
key.append(chars[i]);
}
}
if (key.length() > 1) {
char lastChar = key.charAt(key.length() - 1);
// If last character is S, remove it.
if (lastChar == 'S') {
key.deleteCharAt(key.length() - 1);
lastChar = key.charAt(key.length() - 1);
}
if (key.length() > 2) {
final char last2Char = key.charAt(key.length() - 2);
// If last characters are AY, replace with Y.
if (last2Char == 'A' && lastChar == 'Y') {
key.deleteCharAt(key.length() - 2);
}
}
// If last character is A, remove it.
if (lastChar == 'A') {
key.deleteCharAt(key.length() - 1);
}
}
final String string = key.toString();
return this.isStrict() ? string.substring(0, Math.min(TRUE_LENGTH, string.length())) : string;
}
static String clean(String str) {
if (str == null || str.length() == 0) {
return str;
}
int len = str.length();
char[] chars = new char[len];
int count = 0;
for (int i = 0; i < len; i++) {
if (Character.isLetter(str.charAt(i))) {
chars[count++] = str.charAt(i);
}
}
if (count == len) {
return str.toUpperCase(java.util.Locale.ENGLISH);
}
return new String(chars, 0, count).toUpperCase(java.util.Locale.ENGLISH);
}
}