1 | // License: GPL. For details, see LICENSE file.
|
---|
2 | package org.openstreetmap.josm.tools;
|
---|
3 |
|
---|
4 | /*
|
---|
5 | * The Alphanum Algorithm is an improved sorting algorithm for strings
|
---|
6 | * containing numbers. Instead of sorting numbers in ASCII order like a standard
|
---|
7 | * sort, this algorithm sorts numbers in numeric order.
|
---|
8 | *
|
---|
9 | * The Alphanum Algorithm is discussed at http://www.DaveKoelle.com
|
---|
10 | *
|
---|
11 | * Released under the MIT License - https://opensource.org/licenses/MIT
|
---|
12 | *
|
---|
13 | * Copyright 2007-2017 David Koelle
|
---|
14 | *
|
---|
15 | * Permission is hereby granted, free of charge, to any person obtaining
|
---|
16 | * a copy of this software and associated documentation files (the "Software"),
|
---|
17 | * to deal in the Software without restriction, including without limitation
|
---|
18 | * the rights to use, copy, modify, merge, publish, distribute, sublicense,
|
---|
19 | * and/or sell copies of the Software, and to permit persons to whom the
|
---|
20 | * Software is furnished to do so, subject to the following conditions:
|
---|
21 | *
|
---|
22 | * The above copyright notice and this permission notice shall be included
|
---|
23 | * in all copies or substantial portions of the Software.
|
---|
24 | *
|
---|
25 | * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
---|
26 | * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
---|
27 | * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
|
---|
28 | * IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
|
---|
29 | * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
|
---|
30 | * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
|
---|
31 | * USE OR OTHER DEALINGS IN THE SOFTWARE.
|
---|
32 | */
|
---|
33 | import java.io.Serializable;
|
---|
34 | import java.text.Collator;
|
---|
35 | import java.util.Comparator;
|
---|
36 |
|
---|
37 | /**
|
---|
38 | * The Alphanum Algorithm is an improved sorting algorithm for strings
|
---|
39 | * containing numbers: Instead of sorting numbers in ASCII order like a standard
|
---|
40 | * sort, this algorithm sorts numbers in numeric order.
|
---|
41 | *
|
---|
42 | * The Alphanum Algorithm is discussed at http://www.DaveKoelle.com
|
---|
43 | *
|
---|
44 | * This is an updated version with enhancements made by Daniel Migowski, Andre
|
---|
45 | * Bogus, David Koelle and others.
|
---|
46 | *
|
---|
47 | */
|
---|
48 | public final class AlphanumComparator implements Comparator<String>, Serializable {
|
---|
49 |
|
---|
50 | private static final long serialVersionUID = 1L;
|
---|
51 |
|
---|
52 | private static final AlphanumComparator INSTANCE = new AlphanumComparator();
|
---|
53 |
|
---|
54 | /**
|
---|
55 | * Replies the unique instance.
|
---|
56 | * @return the unique instance
|
---|
57 | */
|
---|
58 | public static AlphanumComparator getInstance() {
|
---|
59 | return INSTANCE;
|
---|
60 | }
|
---|
61 |
|
---|
62 | /**
|
---|
63 | * Constructs a new Alphanum Comparator.
|
---|
64 | */
|
---|
65 | private AlphanumComparator() {
|
---|
66 | }
|
---|
67 |
|
---|
68 | /**
|
---|
69 | * Returns an alphanum chunk.
|
---|
70 | * Length of string is passed in for improved efficiency (only need to calculate it once).
|
---|
71 | * @param s string
|
---|
72 | * @param slength string length
|
---|
73 | * @param marker position
|
---|
74 | * @return alphanum chunk found at given position
|
---|
75 | */
|
---|
76 | private static String getChunk(String s, int slength, int marker) {
|
---|
77 | final int startMarker = marker;
|
---|
78 | char c = s.charAt(marker);
|
---|
79 | marker++;
|
---|
80 | if (Character.isDigit(c)) {
|
---|
81 | while (marker < slength) {
|
---|
82 | c = s.charAt(marker);
|
---|
83 | if (!Character.isDigit(c)) {
|
---|
84 | break;
|
---|
85 | }
|
---|
86 | marker++;
|
---|
87 | }
|
---|
88 | } else {
|
---|
89 | while (marker < slength) {
|
---|
90 | c = s.charAt(marker);
|
---|
91 | if (Character.isDigit(c)) {
|
---|
92 | break;
|
---|
93 | }
|
---|
94 | marker++;
|
---|
95 | }
|
---|
96 | }
|
---|
97 | return s.substring(startMarker, marker);
|
---|
98 | }
|
---|
99 |
|
---|
100 | /**
|
---|
101 | * Check if a string is ASCII only
|
---|
102 | * @param string The string to check
|
---|
103 | * @param stringLength The length of the string (for performance reasons)
|
---|
104 | * @return {@code true} if the string only contains ascii characters
|
---|
105 | */
|
---|
106 | private static boolean isAscii(String string, int stringLength) {
|
---|
107 | for (int i = 0; i < stringLength; i++) {
|
---|
108 | char c = string.charAt(i);
|
---|
109 | if (c >= 128) {
|
---|
110 | return false;
|
---|
111 | }
|
---|
112 | }
|
---|
113 | return true;
|
---|
114 | }
|
---|
115 |
|
---|
116 | /**
|
---|
117 | * Compare two string chunks
|
---|
118 | * @param thisChunk The first chunk to compare
|
---|
119 | * @param thisChunkLength The length of the first chunk (for performance reasons)
|
---|
120 | * @param thatChunk The second chunk to compare
|
---|
121 | * @param thatChunkLength The length of the second chunk (for performance reasons)
|
---|
122 | * @return The {@link Comparator} result
|
---|
123 | */
|
---|
124 | private static int compareChunk(String thisChunk, int thisChunkLength, String thatChunk, int thatChunkLength) {
|
---|
125 | int result;
|
---|
126 | if (Character.isDigit(thisChunk.charAt(0)) && Character.isDigit(thatChunk.charAt(0))) {
|
---|
127 | // Simple chunk comparison by length.
|
---|
128 | result = thisChunkLength - thatChunkLength;
|
---|
129 | // If equal, the first different number counts
|
---|
130 | if (result == 0) {
|
---|
131 | for (int i = 0; i < thisChunkLength; i++) {
|
---|
132 | result = thisChunk.charAt(i) - thatChunk.charAt(i);
|
---|
133 | if (result != 0) {
|
---|
134 | return result;
|
---|
135 | }
|
---|
136 | }
|
---|
137 | }
|
---|
138 | } else {
|
---|
139 | // Check if both chunks are ascii only
|
---|
140 | if (isAscii(thisChunk, thisChunkLength) && isAscii(thatChunk, thatChunkLength)) {
|
---|
141 | return thisChunk.compareTo(thatChunk);
|
---|
142 | }
|
---|
143 | // Instantiate the collator
|
---|
144 | Collator compareOperator = Collator.getInstance();
|
---|
145 | // Compare regardless of accented letters
|
---|
146 | compareOperator.setStrength(Collator.SECONDARY);
|
---|
147 | result = compareOperator.compare(thisChunk, thatChunk);
|
---|
148 | }
|
---|
149 | return result;
|
---|
150 | }
|
---|
151 |
|
---|
152 | @Override
|
---|
153 | public int compare(String s1, String s2) {
|
---|
154 | if (s1 == null && s2 == null) {
|
---|
155 | return 0;
|
---|
156 | } else if (s1 == null) {
|
---|
157 | return -1;
|
---|
158 | } else if (s2 == null) {
|
---|
159 | return 1;
|
---|
160 | }
|
---|
161 |
|
---|
162 | int thisMarker = 0;
|
---|
163 | int thatMarker = 0;
|
---|
164 | int s1Length = s1.length();
|
---|
165 | int s2Length = s2.length();
|
---|
166 |
|
---|
167 | while (thisMarker < s1Length && thatMarker < s2Length) {
|
---|
168 | final String thisChunk = getChunk(s1, s1Length, thisMarker);
|
---|
169 | final int thisChunkLength = thisChunk.length();
|
---|
170 | thisMarker += thisChunkLength;
|
---|
171 |
|
---|
172 | String thatChunk = getChunk(s2, s2Length, thatMarker);
|
---|
173 | final int thatChunkLength = thatChunk.length();
|
---|
174 | thatMarker += thatChunkLength;
|
---|
175 |
|
---|
176 | // If both chunks contain numeric characters, sort them numerically
|
---|
177 | int result = compareChunk(thisChunk, thisChunkLength, thatChunk, thatChunkLength);
|
---|
178 |
|
---|
179 | if (result != 0) {
|
---|
180 | return result;
|
---|
181 | }
|
---|
182 | }
|
---|
183 |
|
---|
184 | return s1Length - s2Length;
|
---|
185 | }
|
---|
186 | }
|
---|