| 1 | // Jomic - a viewer for comic book archives. |
| 2 | // Copyright (C) 2004-2008 Thomas Aglassinger |
| 3 | // |
| 4 | // This program is free software: you can redistribute it and/or modify |
| 5 | // it under the terms of the GNU General Public License as published by |
| 6 | // the Free Software Foundation, either version 3 of the License, or |
| 7 | // (at your option) any later version. |
| 8 | // |
| 9 | // This program is distributed in the hope that it will be useful, |
| 10 | // but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 11 | // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 12 | // GNU General Public License for more details. |
| 13 | // |
| 14 | // You should have received a copy of the GNU General Public License |
| 15 | // along with this program. If not, see <http://www.gnu.org/licenses/>. |
| 16 | package net.sf.jomic.tools; |
| 17 | |
| 18 | import java.util.Comparator; |
| 19 | |
| 20 | import net.sf.wraplog.Logger; |
| 21 | |
| 22 | /** |
| 23 | * Comparator to perform a 'natural order' comparisons of strings. |
| 24 | * |
| 25 | * @author Thomas Aglassinger |
| 26 | */ |
| 27 | public class NaturalOrderComparator implements Comparator |
| 28 | { |
| 29 | private static final int DEFAULT_NUMBER_LENGTH = 16; |
| 30 | private static final char TOKEN_DIGIT = 'd'; |
| 31 | private static final char TOKEN_OTHER = 'o'; |
| 32 | private static final char TOKEN_SPACE = 's'; |
| 33 | |
| 34 | private Logger logger; |
| 35 | |
| 36 | public NaturalOrderComparator() { |
| 37 | logger = Logger.getLogger(NaturalOrderComparator.class); |
| 38 | } |
| 39 | |
| 40 | public int compare(Object o1, Object o2) { |
| 41 | int result = Integer.MAX_VALUE; |
| 42 | String a = o1.toString(); |
| 43 | String b = o2.toString(); |
| 44 | int ia = 0; |
| 45 | int ib = 0; |
| 46 | int aLength = a.length(); |
| 47 | int bLength = b.length(); |
| 48 | |
| 49 | // skip leading white space |
| 50 | while ((ia < aLength) && Character.isWhitespace(a.charAt(ia))) { |
| 51 | ia += 1; |
| 52 | } |
| 53 | while ((ib < bLength) && Character.isWhitespace(b.charAt(ib))) { |
| 54 | ib += 1; |
| 55 | } |
| 56 | |
| 57 | while (result == Integer.MAX_VALUE) { |
| 58 | // did one of the strings reach its end? |
| 59 | if (ia == aLength) { |
| 60 | if (ib == bLength) { |
| 61 | // both strings reached their end |
| 62 | result = 0; |
| 63 | } else { |
| 64 | // a reached its end, b goes on |
| 65 | result = -1; |
| 66 | } |
| 67 | } else if (ib == bLength) { |
| 68 | // b reached its end, a goes on |
| 69 | result = 1; |
| 70 | } else { |
| 71 | char aChar = a.charAt(ia); |
| 72 | char aType = tokenOf(aChar); |
| 73 | char bChar = b.charAt(ib); |
| 74 | char bType = tokenOf(bChar); |
| 75 | |
| 76 | if (aType != bType) { |
| 77 | result = sgn(aChar - bChar); |
| 78 | } else if (aType == TOKEN_SPACE) { |
| 79 | // seek until end of space |
| 80 | while (((ia + 1) < aLength) |
| 81 | && (tokenOf(a.charAt(ia + 1)) == TOKEN_SPACE)) { |
| 82 | ia += 1; |
| 83 | } |
| 84 | while (((ib + 1) < bLength) |
| 85 | && (tokenOf(b.charAt(ib + 1)) == TOKEN_SPACE)) { |
| 86 | ib += 1; |
| 87 | } |
| 88 | } else if (aType == TOKEN_DIGIT) { |
| 89 | StringBuffer aNumber = extractNumber(a, ia); |
| 90 | |
| 91 | if (logger.isDebugEnabled()) { |
| 92 | logger.debug("aNumber=" + aNumber); |
| 93 | } |
| 94 | StringBuffer bNumber = extractNumber(b, ib); |
| 95 | |
| 96 | if (logger.isDebugEnabled()) { |
| 97 | logger.debug("bNumber=" + bNumber); |
| 98 | } |
| 99 | ia += aNumber.length() - 1; |
| 100 | ib += bNumber.length() - 1; |
| 101 | stripLeadingZeros(aNumber); |
| 102 | if (logger.isDebugEnabled()) { |
| 103 | logger.debug("aNumber\\0=" + aNumber); |
| 104 | } |
| 105 | stripLeadingZeros(bNumber); |
| 106 | if (logger.isDebugEnabled()) { |
| 107 | logger.debug("bNumber\\0=" + bNumber); |
| 108 | } |
| 109 | int deltaLength = sgn(aNumber.length() - bNumber.length()); |
| 110 | |
| 111 | if (deltaLength == 0) { |
| 112 | // numbers have same length; compare them digit by digit |
| 113 | int i = 0; |
| 114 | int deltaDigit; |
| 115 | |
| 116 | do { |
| 117 | deltaDigit = aNumber.charAt(i) - bNumber.charAt(i); |
| 118 | if (deltaDigit == 0) { |
| 119 | i += 1; |
| 120 | } else { |
| 121 | // digits differ |
| 122 | result = sgn(deltaDigit); |
| 123 | } |
| 124 | } while ((deltaDigit == 0) && (i < aNumber.length())); |
| 125 | } else { |
| 126 | // numbers differ in length |
| 127 | result = deltaLength; |
| 128 | } |
| 129 | } else if (aType == TOKEN_OTHER) { |
| 130 | int cmp = sgn(aChar - bChar); |
| 131 | |
| 132 | if (cmp != 0) { |
| 133 | result = cmp; |
| 134 | } |
| 135 | } else { |
| 136 | assert false : "type = " + aType; |
| 137 | } |
| 138 | ia += 1; |
| 139 | ib += 1; |
| 140 | } |
| 141 | } |
| 142 | return result; |
| 143 | } |
| 144 | |
| 145 | private StringBuffer extractNumber(String some, int startIndex) { |
| 146 | assert some != null; |
| 147 | assert startIndex < some.length(); |
| 148 | StringBuffer result = new StringBuffer(DEFAULT_NUMBER_LENGTH); |
| 149 | int index = startIndex; |
| 150 | boolean foundNonDigit = false; |
| 151 | |
| 152 | if (logger.isDebugEnabled()) { |
| 153 | logger.debug("extractNumber(\"" + some + "\", " + startIndex + ")"); |
| 154 | } |
| 155 | do { |
| 156 | char c = some.charAt(index); |
| 157 | |
| 158 | if (logger.isDebugEnabled()) { |
| 159 | logger.debug(" " + c); |
| 160 | } |
| 161 | foundNonDigit = !Character.isDigit(c); |
| 162 | if (!foundNonDigit) { |
| 163 | result.append(c); |
| 164 | index += 1; |
| 165 | } |
| 166 | } while (!foundNonDigit && (index < some.length())); |
| 167 | return result; |
| 168 | } |
| 169 | |
| 170 | private int sgn(int x) { |
| 171 | int result; |
| 172 | |
| 173 | if (x > 0) { |
| 174 | result = 1; |
| 175 | } else if (x == 0) { |
| 176 | result = 0; |
| 177 | } else { |
| 178 | result = -1; |
| 179 | } |
| 180 | return result; |
| 181 | } |
| 182 | |
| 183 | private void stripLeadingZeros(StringBuffer some) { |
| 184 | int deleteCount = 0; |
| 185 | |
| 186 | while ((deleteCount < some.length() - 1) |
| 187 | && (some.charAt(deleteCount) == '0')) { |
| 188 | deleteCount += 1; |
| 189 | } |
| 190 | some.delete(0, deleteCount); |
| 191 | } |
| 192 | |
| 193 | private char tokenOf(char c) { |
| 194 | char result; |
| 195 | |
| 196 | if (Character.isDigit(c)) { |
| 197 | result = TOKEN_DIGIT; |
| 198 | } else if (Character.isWhitespace(c)) { |
| 199 | result = TOKEN_SPACE; |
| 200 | } else { |
| 201 | result = TOKEN_OTHER; |
| 202 | } |
| 203 | return result; |
| 204 | } |
| 205 | } |