1 /*
2 * Copyright (c) 2005, Oracle and/or its affiliates. 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. Oracle designates this
8 * particular file as subject to the "Classpath" exception as provided
9 * by Oracle 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 Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
22 * or visit www.oracle.com if you need additional information or have any
23 * 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 @SuppressWarnings("deprecation")
58 public class RuleCharacterIterator {
59
60 // TODO: Ideas for later. (Do not implement if not needed, lest the
61 // code coverage numbers go down due to unused methods.)
62 // 1. Add a copy constructor, equals() method, clone() method.
63 // 2. Rather than return DONE, throw an exception if the end
64 // is reached -- this is an alternate usage model, probably not useful.
65 // 3. Return isEscaped from next(). If this happens,
66 // don't keep an isEscaped member variable.
67
68 /**
69 * Text being iterated.
70 */
71 private String text;
72
73 /**
74 * Position of iterator.
75 */
76 private ParsePosition pos;
77
78 /**
79 * Symbol table used to parse and dereference variables. May be null.
80 */
81 private SymbolTable sym;
82
83 /**
84 * Current variable expansion, or null if none.
85 */
86 private char[] buf;
87
88 /**
89 * Position within buf[]. Meaningless if buf == null.
90 */
91 private int bufPos;
92
93 /**
94 * Flag indicating whether the last character was parsed from an escape.
95 */
96 private boolean isEscaped;
97
98 /**
99 * Value returned when there are no more characters to iterate.
100 */
101 public static final int DONE = -1;
102
103 /**
104 * Bitmask option to enable parsing of variable names.
105 * If {@code (options & PARSE_VARIABLES) != 0},
106 * then an embedded variable will be expanded to
107 * its value. Variables are parsed using the SymbolTable API.
108 */
109 public static final int PARSE_VARIABLES = 1;
110
111 /**
112 * Bitmask option to enable parsing of escape sequences.
113 * If {@code (options & PARSE_ESCAPES) != 0},
114 * then an embedded escape sequence will be expanded
115 * to its value. Escapes are parsed using Utility.unescapeAt().
116 */
117 public static final int PARSE_ESCAPES = 2;
118
119 /**
120 * Bitmask option to enable skipping of whitespace.
121 * If {@code (options & SKIP_WHITESPACE) != 0},
122 * then whitespace characters will be silently
123 * skipped, as if they were not present in the input. Whitespace
124 * characters are defined by UCharacterProperty.isRuleWhiteSpace().
125 */
126 public static final int SKIP_WHITESPACE = 4;
127
128 /**
129 * Constructs an iterator over the given text, starting at the given
130 * position.
131 * @param text the text to be iterated
132 * @param sym the symbol table, or null if there is none. If sym is null,
133 * then variables will not be deferenced, even if the PARSE_VARIABLES
134 * option is set.
135 * @param pos upon input, the index of the next character to return. If a
136 * variable has been dereferenced, then pos will <em>not</em> increment as
137 * characters of the variable value are iterated.
138 */
139 public RuleCharacterIterator(String text, SymbolTable sym,
140 ParsePosition pos) {
141 if (text == null || pos.getIndex() > text.length()) {
142 throw new IllegalArgumentException();
143 }
144 this.text = text;
145 this.sym = sym;
146 this.pos = pos;
147 buf = null;
148 }
149
150 /**
151 * Returns true if this iterator has no more characters to return.
152 */
153 public boolean atEnd() {
154 return buf == null && pos.getIndex() == text.length();
155 }
156
157 /**
158 * Returns the next character using the given options, or DONE if there
159 * are no more characters, and advance the position to the next
160 * character.
161 * @param options one or more of the following options, bitwise-OR-ed
162 * together: PARSE_VARIABLES, PARSE_ESCAPES, SKIP_WHITESPACE.
163 * @return the current 32-bit code point, or DONE
164 */
165 public int next(int options) {
166 int c = DONE;
167 isEscaped = false;
168
169 for (;;) {
170 c = _current();
171 _advance(UTF16.getCharCount(c));
172
173 if (c == SymbolTable.SYMBOL_REF && buf == null &&
174 (options & PARSE_VARIABLES) != 0 && sym != null) {
175 String name = sym.parseReference(text, pos, text.length());
176 // If name == null there was an isolated SYMBOL_REF;
177 // return it. Caller must be prepared for this.
178 if (name == null) {
179 break;
180 }
181 bufPos = 0;
182 buf = sym.lookup(name);
183 if (buf == null) {
184 throw new IllegalArgumentException(
185 "Undefined variable: " + name);
186 }
187 // Handle empty variable value
188 if (buf.length == 0) {
189 buf = null;
190 }
191 continue;
192 }
193
194 if ((options & SKIP_WHITESPACE) != 0 &&
195 UCharacterProperty.isRuleWhiteSpace(c)) {
196 continue;
197 }
198
199 if (c == '\\' && (options & PARSE_ESCAPES) != 0) {
200 int offset[] = new int[] { 0 };
201 c = Utility.unescapeAt(lookahead(), offset);
202 jumpahead(offset[0]);
203 isEscaped = true;
204 if (c < 0) {
205 throw new IllegalArgumentException("Invalid escape");
206 }
207 }
208
209 break;
210 }
211
212 return c;
213 }
214
215 /**
216 * Returns true if the last character returned by next() was
217 * escaped. This will only be the case if the option passed in to
218 * next() included PARSE_ESCAPED and the next character was an
219 * escape sequence.
220 */
221 public boolean isEscaped() {
222 return isEscaped;
223 }
224
225 /**
226 * Returns true if this iterator is currently within a variable expansion.
227 */
228 public boolean inVariable() {
229 return buf != null;
230 }
231
232 /**
233 * Returns an object which, when later passed to setPos(), will
234 * restore this iterator's position. Usage idiom:
235 *
236 * RuleCharacterIterator iterator = ...;
237 * Object pos = iterator.getPos(null); // allocate position object
238 * for (;;) {
239 * pos = iterator.getPos(pos); // reuse position object
240 * int c = iterator.next(...);
241 * ...
242 * }
243 * iterator.setPos(pos);
244 *
245 * @param p a position object previously returned by getPos(),
246 * or null. If not null, it will be updated and returned. If
247 * null, a new position object will be allocated and returned.
248 * @return a position object which may be passed to setPos(),
249 * either `p,' or if `p' == null, a newly-allocated object
250 */
251 public Object getPos(Object p) {
252 if (p == null) {
253 return new Object[] {buf, new int[] {pos.getIndex(), bufPos}};
254 }
255 Object[] a = (Object[]) p;
256 a[0] = buf;
257 int[] v = (int[]) a[1];
258 v[0] = pos.getIndex();
259 v[1] = bufPos;
260 return p;
261 }
262
263 /**
264 * Restores this iterator to the position it had when getPos()
265 * returned the given object.
266 * @param p a position object previously returned by getPos()
267 */
268 public void setPos(Object p) {
269 Object[] a = (Object[]) p;
270 buf = (char[]) a[0];
271 int[] v = (int[]) a[1];
272 pos.setIndex(v[0]);
273 bufPos = v[1];
274 }
275
276 /**
277 * Skips ahead past any ignored characters, as indicated by the given
278 * options. This is useful in conjunction with the lookahead() method.
279 *
280 * Currently, this only has an effect for SKIP_WHITESPACE.
281 * @param options one or more of the following options, bitwise-OR-ed
282 * together: PARSE_VARIABLES, PARSE_ESCAPES, SKIP_WHITESPACE.
283 */
284 public void skipIgnored(int options) {
285 if ((options & SKIP_WHITESPACE) != 0) {
286 for (;;) {
287 int a = _current();
288 if (!UCharacterProperty.isRuleWhiteSpace(a)) break;
289 _advance(UTF16.getCharCount(a));
290 }
291 }
292 }
293
294 /**
295 * Returns a string containing the remainder of the characters to be
296 * returned by this iterator, without any option processing. If the
297 * iterator is currently within a variable expansion, this will only
298 * extend to the end of the variable expansion. This method is provided
299 * so that iterators may interoperate with string-based APIs. The typical
300 * sequence of calls is to call skipIgnored(), then call lookahead(), then
301 * parse the string returned by lookahead(), then call jumpahead() to
302 * resynchronize the iterator.
303 * @return a string containing the characters to be returned by future
304 * calls to next()
305 */
306 public String lookahead() {
307 if (buf != null) {
308 return new String(buf, bufPos, buf.length - bufPos);
309 } else {
310 return text.substring(pos.getIndex());
311 }
312 }
313
314 /**
315 * Advances the position by the given number of 16-bit code units.
316 * This is useful in conjunction with the lookahead() method.
317 * @param count the number of 16-bit code units to jump over
318 */
319 public void jumpahead(int count) {
320 if (count < 0) {
321 throw new IllegalArgumentException();
322 }
323 if (buf != null) {
324 bufPos += count;
325 if (bufPos > buf.length) {
326 throw new IllegalArgumentException();
327 }
328 if (bufPos == buf.length) {
329 buf = null;
330 }
331 } else {
332 int i = pos.getIndex() + count;
333 pos.setIndex(i);
334 if (i > text.length()) {
335 throw new IllegalArgumentException();
336 }
337 }
338 }
339
340 /**
341 * Returns the current 32-bit code point without parsing escapes, parsing
342 * variables, or skipping whitespace.
343 * @return the current 32-bit code point
344 */
345 private int _current() {
346 if (buf != null) {
347 return UTF16.charAt(buf, 0, buf.length, bufPos);
348 } else {
349 int i = pos.getIndex();
350 return (i < text.length()) ? UTF16.charAt(text, i) : DONE;
351 }
352 }
353
354 /**
355 * Advances the position by the given amount.
356 * @param count the number of 16-bit code units to advance past
357 */
358 private void _advance(int count) {
359 if (buf != null) {
360 bufPos += count;
361 if (bufPos == buf.length) {
362 buf = null;
363 }
364 } else {
365 pos.setIndex(pos.getIndex() + count);
366 if (pos.getIndex() > text.length()) {
367 pos.setIndex(text.length());
368 }
369 }
370 }
371 }