1 /* 2 Egothor Software License version 1.00 3 Copyright (C) 1997-2004 Leo Galambos. 4 Copyright (C) 2002-2004 "Egothor developers" 5 on behalf of the Egothor Project. 6 All rights reserved. 7 8 This software is copyrighted by the "Egothor developers". If this 9 license applies to a single file or document, the "Egothor developers" 10 are the people or entities mentioned as copyright holders in that file 11 or document. If this license applies to the Egothor project as a 12 whole, the copyright holders are the people or entities mentioned in 13 the file CREDITS. This file can be found in the same location as this 14 license in the distribution. 15 16 Redistribution and use in source and binary forms, with or without 17 modification, are permitted provided that the following conditions are 18 met: 19 1. Redistributions of source code must retain the above copyright 20 notice, the list of contributors, this list of conditions, and the 21 following disclaimer. 22 2. Redistributions in binary form must reproduce the above copyright 23 notice, the list of contributors, this list of conditions, and the 24 disclaimer that follows these conditions in the documentation 25 and/or other materials provided with the distribution. 26 3. The name "Egothor" must not be used to endorse or promote products 27 derived from this software without prior written permission. For 28 written permission, please contact Leo.G@seznam.cz 29 4. Products derived from this software may not be called "Egothor", 30 nor may "Egothor" appear in their name, without prior written 31 permission from Leo.G@seznam.cz. 32 33 In addition, we request that you include in the end-user documentation 34 provided with the redistribution and/or in the software itself an 35 acknowledgement equivalent to the following: 36 "This product includes software developed by the Egothor Project. 37 http://egothor.sf.net/" 38 39 THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED 40 WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF 41 MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 42 IN NO EVENT SHALL THE EGOTHOR PROJECT OR ITS CONTRIBUTORS BE LIABLE 43 FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 44 CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 45 SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR 46 BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, 47 WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE 48 OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN 49 IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 50 51 This software consists of voluntary contributions made by many 52 individuals on behalf of the Egothor Project and was originally 53 created by Leo Galambos (Leo.G@seznam.cz). 54 */ 55 package org.egothor.stemmer; 56 57 import java.io.DataInput; 58 import java.io.DataOutput; 59 import java.io.IOException; 60 import java.io.PrintStream; 61 import java.util.ArrayList; 62 import java.util.List; 63 64 /** 65 * The MultiTrie is a Trie of Tries. It stores words and their associated patch commands. The 66 * MultiTrie handles patch commands individually (each command by itself). 67 */ 68 public class MultiTrie extends Trie { 69 final char EOM = '*'; 70 final String EOM_NODE = "" + EOM; 71 72 List<Trie> tries = new ArrayList<>(); 73 74 int BY = 1; 75 76 /** 77 * Constructor for the MultiTrie object. 78 * 79 * @param is the input stream 80 * @exception IOException if an I/O error occurs 81 */ MultiTrie(DataInput is)82 public MultiTrie(DataInput is) throws IOException { 83 super(false); 84 forward = is.readBoolean(); 85 BY = is.readInt(); 86 for (int i = is.readInt(); i > 0; i--) { 87 tries.add(new Trie(is)); 88 } 89 } 90 91 /** 92 * Constructor for the MultiTrie object 93 * 94 * @param forward set to <code>true</code> if the elements should be read left to right 95 */ MultiTrie(boolean forward)96 public MultiTrie(boolean forward) { 97 super(forward); 98 } 99 100 /** 101 * Return the element that is stored in a cell associated with the given key. 102 * 103 * @param key the key to the cell holding the desired element 104 * @return the element 105 */ 106 @Override getFully(CharSequence key)107 public CharSequence getFully(CharSequence key) { 108 StringBuilder result = new StringBuilder(tries.size() * 2); 109 for (int i = 0; i < tries.size(); i++) { 110 CharSequence r = tries.get(i).getFully(key); 111 if (r == null || (r.length() == 1 && r.charAt(0) == EOM)) { 112 return result; 113 } 114 result.append(r); 115 } 116 return result; 117 } 118 119 /** 120 * Return the element that is stored as last on a path belonging to the given key. 121 * 122 * @param key the key associated with the desired element 123 * @return the element that is stored as last on a path 124 */ 125 @Override getLastOnPath(CharSequence key)126 public CharSequence getLastOnPath(CharSequence key) { 127 StringBuilder result = new StringBuilder(tries.size() * 2); 128 for (int i = 0; i < tries.size(); i++) { 129 CharSequence r = tries.get(i).getLastOnPath(key); 130 if (r == null || (r.length() == 1 && r.charAt(0) == EOM)) { 131 return result; 132 } 133 result.append(r); 134 } 135 return result; 136 } 137 138 /** 139 * Write this data structure to the given output stream. 140 * 141 * @param os the output stream 142 * @exception IOException if an I/O error occurs 143 */ 144 @Override store(DataOutput os)145 public void store(DataOutput os) throws IOException { 146 os.writeBoolean(forward); 147 os.writeInt(BY); 148 os.writeInt(tries.size()); 149 for (Trie trie : tries) trie.store(os); 150 } 151 152 /** 153 * Add an element to this structure consisting of the given key and patch command. 154 * 155 * <p>This method will return without executing if the <code>cmd</code> parameter's length is 0. 156 * 157 * @param key the key 158 * @param cmd the patch command 159 */ 160 @Override add(CharSequence key, CharSequence cmd)161 public void add(CharSequence key, CharSequence cmd) { 162 if (cmd.length() == 0) { 163 return; 164 } 165 int levels = cmd.length() / BY; 166 while (levels >= tries.size()) { 167 tries.add(new Trie(forward)); 168 } 169 for (int i = 0; i < levels; i++) { 170 tries.get(i).add(key, cmd.subSequence(BY * i, BY * i + BY)); 171 } 172 tries.get(levels).add(key, EOM_NODE); 173 } 174 175 /** 176 * Remove empty rows from the given Trie and return the newly reduced Trie. 177 * 178 * @param by the Trie to reduce 179 * @return the newly reduced Trie 180 */ 181 @Override reduce(Reduce by)182 public Trie reduce(Reduce by) { 183 List<Trie> h = new ArrayList<>(); 184 for (Trie trie : tries) h.add(trie.reduce(by)); 185 186 MultiTrie m = new MultiTrie(forward); 187 m.tries = h; 188 return m; 189 } 190 191 /** 192 * Print the given prefix and the position(s) in the Trie where it appears. 193 * 194 * @param prefix the desired prefix 195 */ 196 @Override printInfo(PrintStream out, CharSequence prefix)197 public void printInfo(PrintStream out, CharSequence prefix) { 198 int c = 0; 199 for (Trie trie : tries) trie.printInfo(out, prefix + "[" + (++c) + "] "); 200 } 201 } 202