J. Duke | 319a3b9 | 2007-12-01 00:00:00 +0000 | [diff] [blame^] | 1 | /* |
| 2 | * Portions Copyright 2005 Sun Microsystems, Inc. All Rights Reserved. |
| 3 | * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. |
| 4 | * |
| 5 | * This code is free software; you can redistribute it and/or modify it |
| 6 | * under the terms of the GNU General Public License version 2 only, as |
| 7 | * published by the Free Software Foundation. Sun designates this |
| 8 | * particular file as subject to the "Classpath" exception as provided |
| 9 | * by Sun in the LICENSE file that accompanied this code. |
| 10 | * |
| 11 | * This code is distributed in the hope that it will be useful, but WITHOUT |
| 12 | * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or |
| 13 | * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License |
| 14 | * version 2 for more details (a copy is included in the LICENSE file that |
| 15 | * accompanied this code). |
| 16 | * |
| 17 | * You should have received a copy of the GNU General Public License version |
| 18 | * 2 along with this work; if not, write to the Free Software Foundation, |
| 19 | * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. |
| 20 | * |
| 21 | * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara, |
| 22 | * CA 95054 USA or visit www.sun.com if you need additional information or |
| 23 | * have any questions. |
| 24 | */ |
| 25 | |
| 26 | /* |
| 27 | ******************************************************************************* |
| 28 | * (C) Copyright IBM Corp. 1996-2005 - All Rights Reserved * |
| 29 | * * |
| 30 | * The original version of this source code and documentation is copyrighted * |
| 31 | * and owned by IBM, These materials are provided under terms of a License * |
| 32 | * Agreement between IBM and Sun. This technology is protected by multiple * |
| 33 | * US and International patents. This notice and attribution to IBM may not * |
| 34 | * to removed. * |
| 35 | ******************************************************************************* |
| 36 | */ |
| 37 | |
| 38 | /* |
| 39 | ********************************************************************** |
| 40 | * Author: Alan Liu |
| 41 | * Created: September 23 2003 |
| 42 | * Since: ICU 2.8 |
| 43 | ********************************************************************** |
| 44 | */ |
| 45 | |
| 46 | package sun.text.normalizer; |
| 47 | |
| 48 | import java.text.ParsePosition; |
| 49 | |
| 50 | /** |
| 51 | * An iterator that returns 32-bit code points. This class is deliberately |
| 52 | * <em>not</em> related to any of the JDK or ICU4J character iterator classes |
| 53 | * in order to minimize complexity. |
| 54 | * @author Alan Liu |
| 55 | * @since ICU 2.8 |
| 56 | */ |
| 57 | public class RuleCharacterIterator { |
| 58 | |
| 59 | // TODO: Ideas for later. (Do not implement if not needed, lest the |
| 60 | // code coverage numbers go down due to unused methods.) |
| 61 | // 1. Add a copy constructor, equals() method, clone() method. |
| 62 | // 2. Rather than return DONE, throw an exception if the end |
| 63 | // is reached -- this is an alternate usage model, probably not useful. |
| 64 | // 3. Return isEscaped from next(). If this happens, |
| 65 | // don't keep an isEscaped member variable. |
| 66 | |
| 67 | /** |
| 68 | * Text being iterated. |
| 69 | */ |
| 70 | private String text; |
| 71 | |
| 72 | /** |
| 73 | * Position of iterator. |
| 74 | */ |
| 75 | private ParsePosition pos; |
| 76 | |
| 77 | /** |
| 78 | * Symbol table used to parse and dereference variables. May be null. |
| 79 | */ |
| 80 | private SymbolTable sym; |
| 81 | |
| 82 | /** |
| 83 | * Current variable expansion, or null if none. |
| 84 | */ |
| 85 | private char[] buf; |
| 86 | |
| 87 | /** |
| 88 | * Position within buf[]. Meaningless if buf == null. |
| 89 | */ |
| 90 | private int bufPos; |
| 91 | |
| 92 | /** |
| 93 | * Flag indicating whether the last character was parsed from an escape. |
| 94 | */ |
| 95 | private boolean isEscaped; |
| 96 | |
| 97 | /** |
| 98 | * Value returned when there are no more characters to iterate. |
| 99 | */ |
| 100 | public static final int DONE = -1; |
| 101 | |
| 102 | /** |
| 103 | * Bitmask option to enable parsing of variable names. If (options & |
| 104 | * PARSE_VARIABLES) != 0, then an embedded variable will be expanded to |
| 105 | * its value. Variables are parsed using the SymbolTable API. |
| 106 | */ |
| 107 | public static final int PARSE_VARIABLES = 1; |
| 108 | |
| 109 | /** |
| 110 | * Bitmask option to enable parsing of escape sequences. If (options & |
| 111 | * PARSE_ESCAPES) != 0, then an embedded escape sequence will be expanded |
| 112 | * to its value. Escapes are parsed using Utility.unescapeAt(). |
| 113 | */ |
| 114 | public static final int PARSE_ESCAPES = 2; |
| 115 | |
| 116 | /** |
| 117 | * Bitmask option to enable skipping of whitespace. If (options & |
| 118 | * SKIP_WHITESPACE) != 0, then whitespace characters will be silently |
| 119 | * skipped, as if they were not present in the input. Whitespace |
| 120 | * characters are defined by UCharacterProperty.isRuleWhiteSpace(). |
| 121 | */ |
| 122 | public static final int SKIP_WHITESPACE = 4; |
| 123 | |
| 124 | /** |
| 125 | * Constructs an iterator over the given text, starting at the given |
| 126 | * position. |
| 127 | * @param text the text to be iterated |
| 128 | * @param sym the symbol table, or null if there is none. If sym is null, |
| 129 | * then variables will not be deferenced, even if the PARSE_VARIABLES |
| 130 | * option is set. |
| 131 | * @param pos upon input, the index of the next character to return. If a |
| 132 | * variable has been dereferenced, then pos will <em>not</em> increment as |
| 133 | * characters of the variable value are iterated. |
| 134 | */ |
| 135 | public RuleCharacterIterator(String text, SymbolTable sym, |
| 136 | ParsePosition pos) { |
| 137 | if (text == null || pos.getIndex() > text.length()) { |
| 138 | throw new IllegalArgumentException(); |
| 139 | } |
| 140 | this.text = text; |
| 141 | this.sym = sym; |
| 142 | this.pos = pos; |
| 143 | buf = null; |
| 144 | } |
| 145 | |
| 146 | /** |
| 147 | * Returns true if this iterator has no more characters to return. |
| 148 | */ |
| 149 | public boolean atEnd() { |
| 150 | return buf == null && pos.getIndex() == text.length(); |
| 151 | } |
| 152 | |
| 153 | /** |
| 154 | * Returns the next character using the given options, or DONE if there |
| 155 | * are no more characters, and advance the position to the next |
| 156 | * character. |
| 157 | * @param options one or more of the following options, bitwise-OR-ed |
| 158 | * together: PARSE_VARIABLES, PARSE_ESCAPES, SKIP_WHITESPACE. |
| 159 | * @return the current 32-bit code point, or DONE |
| 160 | */ |
| 161 | public int next(int options) { |
| 162 | int c = DONE; |
| 163 | isEscaped = false; |
| 164 | |
| 165 | for (;;) { |
| 166 | c = _current(); |
| 167 | _advance(UTF16.getCharCount(c)); |
| 168 | |
| 169 | if (c == SymbolTable.SYMBOL_REF && buf == null && |
| 170 | (options & PARSE_VARIABLES) != 0 && sym != null) { |
| 171 | String name = sym.parseReference(text, pos, text.length()); |
| 172 | // If name == null there was an isolated SYMBOL_REF; |
| 173 | // return it. Caller must be prepared for this. |
| 174 | if (name == null) { |
| 175 | break; |
| 176 | } |
| 177 | bufPos = 0; |
| 178 | buf = sym.lookup(name); |
| 179 | if (buf == null) { |
| 180 | throw new IllegalArgumentException( |
| 181 | "Undefined variable: " + name); |
| 182 | } |
| 183 | // Handle empty variable value |
| 184 | if (buf.length == 0) { |
| 185 | buf = null; |
| 186 | } |
| 187 | continue; |
| 188 | } |
| 189 | |
| 190 | if ((options & SKIP_WHITESPACE) != 0 && |
| 191 | UCharacterProperty.isRuleWhiteSpace(c)) { |
| 192 | continue; |
| 193 | } |
| 194 | |
| 195 | if (c == '\\' && (options & PARSE_ESCAPES) != 0) { |
| 196 | int offset[] = new int[] { 0 }; |
| 197 | c = Utility.unescapeAt(lookahead(), offset); |
| 198 | jumpahead(offset[0]); |
| 199 | isEscaped = true; |
| 200 | if (c < 0) { |
| 201 | throw new IllegalArgumentException("Invalid escape"); |
| 202 | } |
| 203 | } |
| 204 | |
| 205 | break; |
| 206 | } |
| 207 | |
| 208 | return c; |
| 209 | } |
| 210 | |
| 211 | /** |
| 212 | * Returns true if the last character returned by next() was |
| 213 | * escaped. This will only be the case if the option passed in to |
| 214 | * next() included PARSE_ESCAPED and the next character was an |
| 215 | * escape sequence. |
| 216 | */ |
| 217 | public boolean isEscaped() { |
| 218 | return isEscaped; |
| 219 | } |
| 220 | |
| 221 | /** |
| 222 | * Returns true if this iterator is currently within a variable expansion. |
| 223 | */ |
| 224 | public boolean inVariable() { |
| 225 | return buf != null; |
| 226 | } |
| 227 | |
| 228 | /** |
| 229 | * Returns an object which, when later passed to setPos(), will |
| 230 | * restore this iterator's position. Usage idiom: |
| 231 | * |
| 232 | * RuleCharacterIterator iterator = ...; |
| 233 | * Object pos = iterator.getPos(null); // allocate position object |
| 234 | * for (;;) { |
| 235 | * pos = iterator.getPos(pos); // reuse position object |
| 236 | * int c = iterator.next(...); |
| 237 | * ... |
| 238 | * } |
| 239 | * iterator.setPos(pos); |
| 240 | * |
| 241 | * @param p a position object previously returned by getPos(), |
| 242 | * or null. If not null, it will be updated and returned. If |
| 243 | * null, a new position object will be allocated and returned. |
| 244 | * @return a position object which may be passed to setPos(), |
| 245 | * either `p,' or if `p' == null, a newly-allocated object |
| 246 | */ |
| 247 | public Object getPos(Object p) { |
| 248 | if (p == null) { |
| 249 | return new Object[] {buf, new int[] {pos.getIndex(), bufPos}}; |
| 250 | } |
| 251 | Object[] a = (Object[]) p; |
| 252 | a[0] = buf; |
| 253 | int[] v = (int[]) a[1]; |
| 254 | v[0] = pos.getIndex(); |
| 255 | v[1] = bufPos; |
| 256 | return p; |
| 257 | } |
| 258 | |
| 259 | /** |
| 260 | * Restores this iterator to the position it had when getPos() |
| 261 | * returned the given object. |
| 262 | * @param p a position object previously returned by getPos() |
| 263 | */ |
| 264 | public void setPos(Object p) { |
| 265 | Object[] a = (Object[]) p; |
| 266 | buf = (char[]) a[0]; |
| 267 | int[] v = (int[]) a[1]; |
| 268 | pos.setIndex(v[0]); |
| 269 | bufPos = v[1]; |
| 270 | } |
| 271 | |
| 272 | /** |
| 273 | * Skips ahead past any ignored characters, as indicated by the given |
| 274 | * options. This is useful in conjunction with the lookahead() method. |
| 275 | * |
| 276 | * Currently, this only has an effect for SKIP_WHITESPACE. |
| 277 | * @param options one or more of the following options, bitwise-OR-ed |
| 278 | * together: PARSE_VARIABLES, PARSE_ESCAPES, SKIP_WHITESPACE. |
| 279 | */ |
| 280 | public void skipIgnored(int options) { |
| 281 | if ((options & SKIP_WHITESPACE) != 0) { |
| 282 | for (;;) { |
| 283 | int a = _current(); |
| 284 | if (!UCharacterProperty.isRuleWhiteSpace(a)) break; |
| 285 | _advance(UTF16.getCharCount(a)); |
| 286 | } |
| 287 | } |
| 288 | } |
| 289 | |
| 290 | /** |
| 291 | * Returns a string containing the remainder of the characters to be |
| 292 | * returned by this iterator, without any option processing. If the |
| 293 | * iterator is currently within a variable expansion, this will only |
| 294 | * extend to the end of the variable expansion. This method is provided |
| 295 | * so that iterators may interoperate with string-based APIs. The typical |
| 296 | * sequence of calls is to call skipIgnored(), then call lookahead(), then |
| 297 | * parse the string returned by lookahead(), then call jumpahead() to |
| 298 | * resynchronize the iterator. |
| 299 | * @return a string containing the characters to be returned by future |
| 300 | * calls to next() |
| 301 | */ |
| 302 | public String lookahead() { |
| 303 | if (buf != null) { |
| 304 | return new String(buf, bufPos, buf.length - bufPos); |
| 305 | } else { |
| 306 | return text.substring(pos.getIndex()); |
| 307 | } |
| 308 | } |
| 309 | |
| 310 | /** |
| 311 | * Advances the position by the given number of 16-bit code units. |
| 312 | * This is useful in conjunction with the lookahead() method. |
| 313 | * @param count the number of 16-bit code units to jump over |
| 314 | */ |
| 315 | public void jumpahead(int count) { |
| 316 | if (count < 0) { |
| 317 | throw new IllegalArgumentException(); |
| 318 | } |
| 319 | if (buf != null) { |
| 320 | bufPos += count; |
| 321 | if (bufPos > buf.length) { |
| 322 | throw new IllegalArgumentException(); |
| 323 | } |
| 324 | if (bufPos == buf.length) { |
| 325 | buf = null; |
| 326 | } |
| 327 | } else { |
| 328 | int i = pos.getIndex() + count; |
| 329 | pos.setIndex(i); |
| 330 | if (i > text.length()) { |
| 331 | throw new IllegalArgumentException(); |
| 332 | } |
| 333 | } |
| 334 | } |
| 335 | |
| 336 | /** |
| 337 | * Returns the current 32-bit code point without parsing escapes, parsing |
| 338 | * variables, or skipping whitespace. |
| 339 | * @return the current 32-bit code point |
| 340 | */ |
| 341 | private int _current() { |
| 342 | if (buf != null) { |
| 343 | return UTF16.charAt(buf, 0, buf.length, bufPos); |
| 344 | } else { |
| 345 | int i = pos.getIndex(); |
| 346 | return (i < text.length()) ? UTF16.charAt(text, i) : DONE; |
| 347 | } |
| 348 | } |
| 349 | |
| 350 | /** |
| 351 | * Advances the position by the given amount. |
| 352 | * @param count the number of 16-bit code units to advance past |
| 353 | */ |
| 354 | private void _advance(int count) { |
| 355 | if (buf != null) { |
| 356 | bufPos += count; |
| 357 | if (bufPos == buf.length) { |
| 358 | buf = null; |
| 359 | } |
| 360 | } else { |
| 361 | pos.setIndex(pos.getIndex() + count); |
| 362 | if (pos.getIndex() > text.length()) { |
| 363 | pos.setIndex(text.length()); |
| 364 | } |
| 365 | } |
| 366 | } |
| 367 | } |