1 /* GENERATED SOURCE. DO NOT MODIFY. */ 2 // © 2016 and later: Unicode, Inc. and others. 3 // License & terms of use: http://www.unicode.org/copyright.html#License 4 /* 5 * Copyright (C) 1996-2011, International Business Machines Corporation and 6 * others. All Rights Reserved. 7 */ 8 package ohos.global.icu.text; 9 import ohos.global.icu.impl.PatternProps; 10 import ohos.global.icu.impl.UCharacterName; 11 import ohos.global.icu.impl.Utility; 12 import ohos.global.icu.lang.UCharacter; 13 14 /** 15 * A transliterator that performs name to character mapping. 16 * @author Alan Liu 17 */ 18 class NameUnicodeTransliterator extends Transliterator { 19 20 static final String _ID = "Name-Any"; 21 22 static final String OPEN_PAT = "\\N~{~"; 23 static final char OPEN_DELIM = '\\'; // first char of OPEN_PAT 24 static final char CLOSE_DELIM = '}'; 25 static final char SPACE = ' '; 26 27 28 /** 29 * System registration hook. 30 */ register()31 static void register() { 32 Transliterator.registerFactory(_ID, new Transliterator.Factory() { 33 @Override 34 public Transliterator getInstance(String ID) { 35 return new NameUnicodeTransliterator(null); 36 } 37 }); 38 } 39 40 /** 41 * Constructs a transliterator. 42 */ NameUnicodeTransliterator(UnicodeFilter filter)43 public NameUnicodeTransliterator(UnicodeFilter filter) { 44 super(_ID, filter); 45 } 46 47 /** 48 * Implements {@link Transliterator#handleTransliterate}. 49 */ 50 @Override handleTransliterate(Replaceable text, Position offsets, boolean isIncremental)51 protected void handleTransliterate(Replaceable text, 52 Position offsets, boolean isIncremental) { 53 54 int maxLen = UCharacterName.INSTANCE.getMaxCharNameLength() + 1; // allow for temporary trailing space 55 56 StringBuffer name = new StringBuffer(maxLen); 57 58 // Get the legal character set 59 UnicodeSet legal = new UnicodeSet(); 60 UCharacterName.INSTANCE.getCharNameCharacters(legal); 61 62 int cursor = offsets.start; 63 int limit = offsets.limit; 64 65 // Modes: 66 // 0 - looking for open delimiter 67 // 1 - after open delimiter 68 int mode = 0; 69 int openPos = -1; // open delim candidate pos 70 71 int c; 72 while (cursor < limit) { 73 c = text.char32At(cursor); 74 75 switch (mode) { 76 case 0: // looking for open delimiter 77 if (c == OPEN_DELIM) { // quick check first 78 openPos = cursor; 79 int i = Utility.parsePattern(OPEN_PAT, text, cursor, limit); 80 if (i >= 0 && i < limit) { 81 mode = 1; 82 name.setLength(0); 83 cursor = i; 84 continue; // *** reprocess char32At(cursor) 85 } 86 } 87 break; 88 89 case 1: // after open delimiter 90 // Look for legal chars. If \s+ is found, convert it 91 // to a single space. If closeDelimiter is found, exit 92 // the loop. If any other character is found, exit the 93 // loop. If the limit is reached, exit the loop. 94 95 // Convert \s+ => SPACE. This assumes there are no 96 // runs of >1 space characters in names. 97 if (PatternProps.isWhiteSpace(c)) { 98 // Ignore leading whitespace 99 if (name.length() > 0 && 100 name.charAt(name.length()-1) != SPACE) { 101 name.append(SPACE); 102 // If we are too long then abort. maxLen includes 103 // temporary trailing space, so use '>'. 104 if (name.length() > maxLen) { 105 mode = 0; 106 } 107 } 108 break; 109 } 110 111 if (c == CLOSE_DELIM) { 112 113 int len = name.length(); 114 115 // Delete trailing space, if any 116 if (len > 0 && 117 name.charAt(len-1) == SPACE) { 118 name.setLength(--len); 119 } 120 121 c = UCharacter.getCharFromExtendedName(name.toString()); 122 if (c != -1) { 123 // Lookup succeeded 124 125 // assert(UTF16.getCharCount(CLOSE_DELIM) == 1); 126 cursor++; // advance over CLOSE_DELIM 127 128 String str = UTF16.valueOf(c); 129 text.replace(openPos, cursor, str); 130 131 // Adjust indices for the change in the length of 132 // the string. Do not assume that str.length() == 133 // 1, in case of surrogates. 134 int delta = cursor - openPos - str.length(); 135 cursor -= delta; 136 limit -= delta; 137 // assert(cursor == openPos + str.length()); 138 } 139 // If the lookup failed, we leave things as-is and 140 // still switch to mode 0 and continue. 141 mode = 0; 142 openPos = -1; // close off candidate 143 continue; // *** reprocess char32At(cursor) 144 } 145 146 if (legal.contains(c)) { 147 UTF16.append(name, c); 148 // If we go past the longest possible name then abort. 149 // maxLen includes temporary trailing space, so use '>='. 150 if (name.length() >= maxLen) { 151 mode = 0; 152 } 153 } 154 155 // Invalid character 156 else { 157 --cursor; // Backup and reprocess this character 158 mode = 0; 159 } 160 161 break; 162 } 163 164 cursor += UTF16.getCharCount(c); 165 } 166 167 offsets.contextLimit += limit - offsets.limit; 168 offsets.limit = limit; 169 // In incremental mode, only advance the cursor up to the last 170 // open delimiter candidate. 171 offsets.start = (isIncremental && openPos >= 0) ? openPos : cursor; 172 } 173 174 /* (non-Javadoc) 175 * @see ohos.global.icu.text.Transliterator#addSourceTargetSet(ohos.global.icu.text.UnicodeSet, ohos.global.icu.text.UnicodeSet, ohos.global.icu.text.UnicodeSet) 176 */ 177 @Override addSourceTargetSet(UnicodeSet inputFilter, UnicodeSet sourceSet, UnicodeSet targetSet)178 public void addSourceTargetSet(UnicodeSet inputFilter, UnicodeSet sourceSet, UnicodeSet targetSet) { 179 UnicodeSet myFilter = getFilterAsUnicodeSet(inputFilter); 180 if (!myFilter.containsAll(UnicodeNameTransliterator.OPEN_DELIM) || !myFilter.contains(CLOSE_DELIM)) { 181 return; // we have to contain both prefix and suffix 182 } 183 UnicodeSet items = new UnicodeSet() 184 .addAll('0', '9') 185 .addAll('A', 'F') 186 .addAll('a', 'z') // for controls 187 .add('<').add('>') // for controls 188 .add('(').add(')') // for controls 189 .add('-') 190 .add(' ') 191 .addAll(UnicodeNameTransliterator.OPEN_DELIM) 192 .add(CLOSE_DELIM); 193 items.retainAll(myFilter); 194 if (items.size() > 0) { 195 sourceSet.addAll(items); 196 // could produce any character 197 targetSet.addAll(0, 0x10FFFF); 198 } 199 } 200 } 201