001: /*
002: * Portions Copyright 2005 Sun Microsystems, Inc. All Rights Reserved.
003: * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
004: *
005: * This code is free software; you can redistribute it and/or modify it
006: * under the terms of the GNU General Public License version 2 only, as
007: * published by the Free Software Foundation. Sun designates this
008: * particular file as subject to the "Classpath" exception as provided
009: * by Sun in the LICENSE file that accompanied this code.
010: *
011: * This code is distributed in the hope that it will be useful, but WITHOUT
012: * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
013: * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
014: * version 2 for more details (a copy is included in the LICENSE file that
015: * accompanied this code).
016: *
017: * You should have received a copy of the GNU General Public License version
018: * 2 along with this work; if not, write to the Free Software Foundation,
019: * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
020: *
021: * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
022: * CA 95054 USA or visit www.sun.com if you need additional information or
023: * have any questions.
024: */
025:
026: /*
027: *******************************************************************************
028: * (C) Copyright IBM Corp. 1996-2005 - All Rights Reserved *
029: * *
030: * The original version of this source code and documentation is copyrighted *
031: * and owned by IBM, These materials are provided under terms of a License *
032: * Agreement between IBM and Sun. This technology is protected by multiple *
033: * US and International patents. This notice and attribution to IBM may not *
034: * to removed. *
035: *******************************************************************************
036: */
037:
038: /*
039: **********************************************************************
040: * Author: Alan Liu
041: * Created: September 23 2003
042: * Since: ICU 2.8
043: **********************************************************************
044: */
045:
046: package sun.text.normalizer;
047:
048: import java.text.ParsePosition;
049:
050: /**
051: * An iterator that returns 32-bit code points. This class is deliberately
052: * <em>not</em> related to any of the JDK or ICU4J character iterator classes
053: * in order to minimize complexity.
054: * @author Alan Liu
055: * @since ICU 2.8
056: */
057: public class RuleCharacterIterator {
058:
059: // TODO: Ideas for later. (Do not implement if not needed, lest the
060: // code coverage numbers go down due to unused methods.)
061: // 1. Add a copy constructor, equals() method, clone() method.
062: // 2. Rather than return DONE, throw an exception if the end
063: // is reached -- this is an alternate usage model, probably not useful.
064: // 3. Return isEscaped from next(). If this happens,
065: // don't keep an isEscaped member variable.
066:
067: /**
068: * Text being iterated.
069: */
070: private String text;
071:
072: /**
073: * Position of iterator.
074: */
075: private ParsePosition pos;
076:
077: /**
078: * Symbol table used to parse and dereference variables. May be null.
079: */
080: private SymbolTable sym;
081:
082: /**
083: * Current variable expansion, or null if none.
084: */
085: private char[] buf;
086:
087: /**
088: * Position within buf[]. Meaningless if buf == null.
089: */
090: private int bufPos;
091:
092: /**
093: * Flag indicating whether the last character was parsed from an escape.
094: */
095: private boolean isEscaped;
096:
097: /**
098: * Value returned when there are no more characters to iterate.
099: */
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
172: .length());
173: // If name == null there was an isolated SYMBOL_REF;
174: // return it. Caller must be prepared for this.
175: if (name == null) {
176: break;
177: }
178: bufPos = 0;
179: buf = sym.lookup(name);
180: if (buf == null) {
181: throw new IllegalArgumentException(
182: "Undefined variable: " + name);
183: }
184: // Handle empty variable value
185: if (buf.length == 0) {
186: buf = null;
187: }
188: continue;
189: }
190:
191: if ((options & SKIP_WHITESPACE) != 0
192: && UCharacterProperty.isRuleWhiteSpace(c)) {
193: continue;
194: }
195:
196: if (c == '\\' && (options & PARSE_ESCAPES) != 0) {
197: int offset[] = new int[] { 0 };
198: c = Utility.unescapeAt(lookahead(), offset);
199: jumpahead(offset[0]);
200: isEscaped = true;
201: if (c < 0) {
202: throw new IllegalArgumentException("Invalid escape");
203: }
204: }
205:
206: break;
207: }
208:
209: return c;
210: }
211:
212: /**
213: * Returns true if the last character returned by next() was
214: * escaped. This will only be the case if the option passed in to
215: * next() included PARSE_ESCAPED and the next character was an
216: * escape sequence.
217: */
218: public boolean isEscaped() {
219: return isEscaped;
220: }
221:
222: /**
223: * Returns true if this iterator is currently within a variable expansion.
224: */
225: public boolean inVariable() {
226: return buf != null;
227: }
228:
229: /**
230: * Returns an object which, when later passed to setPos(), will
231: * restore this iterator's position. Usage idiom:
232: *
233: * RuleCharacterIterator iterator = ...;
234: * Object pos = iterator.getPos(null); // allocate position object
235: * for (;;) {
236: * pos = iterator.getPos(pos); // reuse position object
237: * int c = iterator.next(...);
238: * ...
239: * }
240: * iterator.setPos(pos);
241: *
242: * @param p a position object previously returned by getPos(),
243: * or null. If not null, it will be updated and returned. If
244: * null, a new position object will be allocated and returned.
245: * @return a position object which may be passed to setPos(),
246: * either `p,' or if `p' == null, a newly-allocated object
247: */
248: public Object getPos(Object p) {
249: if (p == null) {
250: return new Object[] { buf,
251: new int[] { pos.getIndex(), bufPos } };
252: }
253: Object[] a = (Object[]) p;
254: a[0] = buf;
255: int[] v = (int[]) a[1];
256: v[0] = pos.getIndex();
257: v[1] = bufPos;
258: return p;
259: }
260:
261: /**
262: * Restores this iterator to the position it had when getPos()
263: * returned the given object.
264: * @param p a position object previously returned by getPos()
265: */
266: public void setPos(Object p) {
267: Object[] a = (Object[]) p;
268: buf = (char[]) a[0];
269: int[] v = (int[]) a[1];
270: pos.setIndex(v[0]);
271: bufPos = v[1];
272: }
273:
274: /**
275: * Skips ahead past any ignored characters, as indicated by the given
276: * options. This is useful in conjunction with the lookahead() method.
277: *
278: * Currently, this only has an effect for SKIP_WHITESPACE.
279: * @param options one or more of the following options, bitwise-OR-ed
280: * together: PARSE_VARIABLES, PARSE_ESCAPES, SKIP_WHITESPACE.
281: */
282: public void skipIgnored(int options) {
283: if ((options & SKIP_WHITESPACE) != 0) {
284: for (;;) {
285: int a = _current();
286: if (!UCharacterProperty.isRuleWhiteSpace(a))
287: break;
288: _advance(UTF16.getCharCount(a));
289: }
290: }
291: }
292:
293: /**
294: * Returns a string containing the remainder of the characters to be
295: * returned by this iterator, without any option processing. If the
296: * iterator is currently within a variable expansion, this will only
297: * extend to the end of the variable expansion. This method is provided
298: * so that iterators may interoperate with string-based APIs. The typical
299: * sequence of calls is to call skipIgnored(), then call lookahead(), then
300: * parse the string returned by lookahead(), then call jumpahead() to
301: * resynchronize the iterator.
302: * @return a string containing the characters to be returned by future
303: * calls to next()
304: */
305: public String lookahead() {
306: if (buf != null) {
307: return new String(buf, bufPos, buf.length - bufPos);
308: } else {
309: return text.substring(pos.getIndex());
310: }
311: }
312:
313: /**
314: * Advances the position by the given number of 16-bit code units.
315: * This is useful in conjunction with the lookahead() method.
316: * @param count the number of 16-bit code units to jump over
317: */
318: public void jumpahead(int count) {
319: if (count < 0) {
320: throw new IllegalArgumentException();
321: }
322: if (buf != null) {
323: bufPos += count;
324: if (bufPos > buf.length) {
325: throw new IllegalArgumentException();
326: }
327: if (bufPos == buf.length) {
328: buf = null;
329: }
330: } else {
331: int i = pos.getIndex() + count;
332: pos.setIndex(i);
333: if (i > text.length()) {
334: throw new IllegalArgumentException();
335: }
336: }
337: }
338:
339: /**
340: * Returns the current 32-bit code point without parsing escapes, parsing
341: * variables, or skipping whitespace.
342: * @return the current 32-bit code point
343: */
344: private int _current() {
345: if (buf != null) {
346: return UTF16.charAt(buf, 0, buf.length, bufPos);
347: } else {
348: int i = pos.getIndex();
349: return (i < text.length()) ? UTF16.charAt(text, i) : DONE;
350: }
351: }
352:
353: /**
354: * Advances the position by the given amount.
355: * @param count the number of 16-bit code units to advance past
356: */
357: private void _advance(int count) {
358: if (buf != null) {
359: bufPos += count;
360: if (bufPos == buf.length) {
361: buf = null;
362: }
363: } else {
364: pos.setIndex(pos.getIndex() + count);
365: if (pos.getIndex() > text.length()) {
366: pos.setIndex(text.length());
367: }
368: }
369: }
370: }
|