KickJava   Java API By Example, From Geeks To Geeks.

Java > Open Source Codes > org > enhydra > apache > xerces > utils > StringHasher


1 /*
2  * The Apache Software License, Version 1.1
3  *
4  *
5  * Copyright (c) 1999 The Apache Software Foundation. All rights
6  * reserved.
7  *
8  * Redistribution and use in source and binary forms, with or without
9  * modification, are permitted provided that the following conditions
10  * are met:
11  *
12  * 1. Redistributions of source code must retain the above copyright
13  * notice, this list of conditions and the following disclaimer.
14  *
15  * 2. Redistributions in binary form must reproduce the above copyright
16  * notice, this list of conditions and the following disclaimer in
17  * the documentation and/or other materials provided with the
18  * distribution.
19  *
20  * 3. The end-user documentation included with the redistribution,
21  * if any, must include the following acknowledgment:
22  * "This product includes software developed by the
23  * Apache Software Foundation (http://www.apache.org/)."
24  * Alternately, this acknowledgment may appear in the software itself,
25  * if and wherever such third-party acknowledgments normally appear.
26  *
27  * 4. The names "Xerces" and "Apache Software Foundation" must
28  * not be used to endorse or promote products derived from this
29  * software without prior written permission. For written
30  * permission, please contact apache@apache.org.
31  *
32  * 5. Products derived from this software may not be called "Apache",
33  * nor may "Apache" appear in their name, without prior written
34  * permission of the Apache Software Foundation.
35  *
36  * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
37  * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
38  * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
39  * DISCLAIMED. IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
40  * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
41  * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
42  * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
43  * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
44  * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
45  * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
46  * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
47  * SUCH DAMAGE.
48  * ====================================================================
49  *
50  * This software consists of voluntary contributions made by many
51  * individuals on behalf of the Apache Software Foundation and was
52  * originally based on software copyright (c) 1999, International
53  * Business Machines, Inc., http://www.apache.org. For more
54  * information on the Apache Software Foundation, please see
55  * <http://www.apache.org/>.
56  */
57
58 package org.enhydra.apache.xerces.utils;
59
60 /**
61  * Algorithm used to hash char arrays (strings).
62  *
63  * This class was created after it was discovered that parsing some
64  * documents was unexpectedly slow due to many different strings
65  * hashing to the same 32-bit value using the java.lang.String hash
66  * algorithm.
67  *
68  * The trick seems to be the shift of the top eight bits of the hashcode
69  * back down to the bottom to keep them from being rolled out.
70  *
71  * @version
72  */
73 public final class StringHasher {
74     /**
75      * generate a hashcode for a String
76      *
77      * @param str the String to hash
78      * @param strLength the length of the String to hash
79      * @return hashcode for the String
80      */
81     public static int hashString(String JavaDoc str, int strLength) {
82         int hashcode = 0;
83         for (int i = 0; i < strLength; i++) {
84             int top = hashcode >> 24;
85             hashcode += ((hashcode * 37) + top + ((int)str.charAt(i)));
86         }
87         hashcode = (hashcode & 0x7fffffff);
88         return (hashcode == 0) ? 1 : hashcode;
89     }
90     /**
91      * generate a hashcode for a character array
92      *
93      * @param chars the array to hash
94      * @param offset the offset to start hashing
95      * @param length the length of characters to hash
96      * @return hashcode for the character array
97      */
98     public static int hashChars(char[] chars, int offset, int length) {
99         int hashcode = 0;
100         for (int i = 0; i < length; i++) {
101             int top = hashcode >> 24;
102             hashcode += ((hashcode * 37) + top + ((int)(chars[offset++] & 0xFFFF)));
103         }
104         hashcode = (hashcode & 0x7fffffff);
105         return (hashcode == 0) ? 1 : hashcode;
106     }
107     /**
108      * generate partially completed character hashcode.
109      * this is mean to be iterated over individual characters in order to generate
110      * a full hash value
111      * @see #finishHash(int)
112      *
113      * @param hashcode a partially completed character hashcode
114      * @param ch the character to hash
115      * @return a partially completed character hashcode
116      */
117     public static int hashChar(int hashcode, int ch) {
118         int top = hashcode >> 24;
119         hashcode += ((hashcode * 37) + top + ch);
120         return hashcode;
121     }
122     /**
123      * finish hashing a partically completed character hashcode
124      * @see #hashChar(int,int)
125      *
126      * @param hashcode a partially completed character hashcode
127      * @return a character hashcode
128      */
129     public static int finishHash(int hashcode) {
130         hashcode = (hashcode & 0x7fffffff);
131         return (hashcode == 0) ? 1 : hashcode;
132     }
133 }
134
Popular Tags