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.
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.
16 Redistribution and use in source and binary forms, with or without
17 modification, are permitted provided that the following conditions are
19 1. Redistributions of source code must retain the above copyright
20 notice, the list of contributors, this list of conditions, and the
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.
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/"
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.
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).
55 package org.egothor.stemmer;
57 import java.io.DataInput;
58 import java.io.DataOutput;
59 import java.io.IOException;
60 import java.util.ArrayList;
61 import java.util.List;
64 * The MultiTrie is a Trie of Tries. It stores words and their associated patch
65 * commands. The MultiTrie handles patch commmands individually (each command by
68 public class MultiTrie extends Trie {
70 final String EOM_NODE = "" + EOM;
72 List<Trie> tries = new ArrayList<Trie>();
77 * Constructor for the MultiTrie object.
79 * @param is the input stream
80 * @exception IOException if an I/O error occurs
82 public MultiTrie(DataInput is) throws IOException {
84 forward = is.readBoolean();
86 for (int i = is.readInt(); i > 0; i--) {
87 tries.add(new Trie(is));
92 * Constructor for the MultiTrie object
94 * @param forward set to <tt>true</tt> if the elements should be read left to
97 public MultiTrie(boolean forward) {
102 * Return the element that is stored in a cell associated with the given key.
104 * @param key the key to the cell holding the desired element
105 * @return the element
108 public CharSequence getFully(CharSequence key) {
109 StringBuilder result = new StringBuilder(tries.size() * 2);
110 for (int i = 0; i < tries.size(); i++) {
111 CharSequence r = tries.get(i).getFully(key);
112 if (r == null || (r.length() == 1 && r.charAt(0) == EOM)) {
121 * Return the element that is stored as last on a path belonging to the given
124 * @param key the key associated with the desired element
125 * @return the element that is stored as last on a path
128 public CharSequence getLastOnPath(CharSequence key) {
129 StringBuilder result = new StringBuilder(tries.size() * 2);
130 for (int i = 0; i < tries.size(); i++) {
131 CharSequence r = tries.get(i).getLastOnPath(key);
132 if (r == null || (r.length() == 1 && r.charAt(0) == EOM)) {
141 * Write this data structure to the given output stream.
143 * @param os the output stream
144 * @exception IOException if an I/O error occurs
147 public void store(DataOutput os) throws IOException {
148 os.writeBoolean(forward);
150 os.writeInt(tries.size());
151 for (Trie trie : tries)
156 * Add an element to this structure consisting of the given key and patch
159 * This method will return without executing if the <tt>cmd</tt>
160 * parameter's length is 0.
163 * @param cmd the patch command
166 public void add(CharSequence key, CharSequence cmd) {
167 if (cmd.length() == 0) {
170 int levels = cmd.length() / BY;
171 while (levels >= tries.size()) {
172 tries.add(new Trie(forward));
174 for (int i = 0; i < levels; i++) {
175 tries.get(i).add(key, cmd.subSequence(BY * i, BY * i + BY));
177 tries.get(levels).add(key, EOM_NODE);
181 * Remove empty rows from the given Trie and return the newly reduced Trie.
183 * @param by the Trie to reduce
184 * @return the newly reduced Trie
187 public Trie reduce(Reduce by) {
188 List<Trie> h = new ArrayList<Trie>();
189 for (Trie trie : tries)
190 h.add(trie.reduce(by));
192 MultiTrie m = new MultiTrie(forward);
198 * Print the given prefix and the position(s) in the Trie where it appears.
200 * @param prefix the desired prefix
203 public void printInfo(CharSequence prefix) {
205 for (Trie trie : tries)
206 trie.printInfo(prefix + "[" + (++c) + "] ");