Source code: com/go/trove/io/SourceReader.java
1 /* ====================================================================
2 * Trove - Copyright (c) 1997-2000 Walt Disney Internet Group
3 * ====================================================================
4 * The Tea Software License, Version 1.1
5 *
6 * Copyright (c) 2000 Walt Disney Internet Group. All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * 1. Redistributions of source code must retain the above copyright
13 * notice, this list of conditions and the following disclaimer.
14 *
15 * 2. Redistributions in binary form must reproduce the above copyright
16 * notice, this list of conditions and the following disclaimer in
17 * the documentation and/or other materials provided with the
18 * distribution.
19 *
20 * 3. The end-user documentation included with the redistribution,
21 * if any, must include the following acknowledgment:
22 * "This product includes software developed by the
23 * Walt Disney Internet Group (http://opensource.go.com/)."
24 * Alternately, this acknowledgment may appear in the software itself,
25 * if and wherever such third-party acknowledgments normally appear.
26 *
27 * 4. The names "Tea", "TeaServlet", "Kettle", "Trove" and "BeanDoc" must
28 * not be used to endorse or promote products derived from this
29 * software without prior written permission. For written
30 * permission, please contact opensource@dig.com.
31 *
32 * 5. Products derived from this software may not be called "Tea",
33 * "TeaServlet", "Kettle" or "Trove", nor may "Tea", "TeaServlet",
34 * "Kettle", "Trove" or "BeanDoc" appear in their name, without prior
35 * written permission of the Walt Disney Internet Group.
36 *
37 * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
38 * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
39 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
40 * DISCLAIMED. IN NO EVENT SHALL THE WALT DISNEY INTERNET GROUP OR ITS
41 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
42 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
43 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
44 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
45 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
46 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
47 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
48 * ====================================================================
49 *
50 * For more information about Tea, please see http://opensource.go.com/.
51 */
52
53 package com.go.trove.io;
54
55 import java.io.*;
56
57 /******************************************************************************
58 * The SourceReader provides several services for reading source input.
59 * It calculates line numbers, position in the source file, supports two
60 * character pushback, extracts code from text that allows mixed code and
61 * plain text, and it processes unicode escape sequences that appear in
62 * source code.
63 *
64 * <p>Readers return -1 when the end of the stream, has been reached, and so
65 * does SourceReader. SourceReader will also return other special negative
66 * values to indicate a tag substitution. ENTER_CODE is returned to indicate
67 * that characters read are in source code, and ENTER_TEXT is returned to
68 * indicate that characters read are in plain text. The first character read
69 * from a SourceReader is either ENTER_CODE or ENTER_TEXT;
70 *
71 * @author Brian S O'Neill
72 * @version
73 * <!--$$Revision:--> 18 <!-- $-->, <!--$$JustDate:--> 4/26/01 <!-- $-->
74 */
75 public class SourceReader extends PushbackPositionReader {
76 public static final int ENTER_CODE = -2;
77 public static final int ENTER_TEXT = -3;
78
79 private static Reader createReader(Reader source,
80 String beginTag, String endTag) {
81 String[] tags = new String[4];
82 int[] codes = new int[4];
83
84 int i = 0;
85 // Convert different kinds of line breaks into the newline character.
86 tags[i] = "\r\n"; codes[i++] = '\n';
87 tags[i] = "\r"; codes[i++] = '\n';
88
89 if (beginTag != null && beginTag.length() > 0) {
90 tags[i] = beginTag; codes[i++] = ENTER_CODE;
91 }
92
93 if (endTag != null && endTag.length() > 0) {
94 tags[i] = endTag; codes[i++] = ENTER_TEXT;
95 }
96
97 if (i < 4) {
98 String[] newTags = new String[i];
99 System.arraycopy(tags, 0, newTags, 0, i);
100 tags = newTags;
101
102 int[] newCodes = new int[i];
103 System.arraycopy(codes, 0, newCodes, 0, i);
104 codes = newCodes;
105 }
106
107 return new UnicodeReader(new TagReader(source, tags, codes));
108 }
109
110 private UnicodeReader mUnicodeReader;
111 private TagReader mTagReader;
112 private boolean mClosed = false;
113
114 // The current line in the source. (1..)
115 private int mLine = 1;
116
117 private int mFirst;
118
119 private String mBeginTag;
120 private String mEndTag;
121
122 /**
123 * The begin and end tags for a SourceReader are optional. If the begin
124 * tag is null or has zero length, then the SourceReader starts reading
125 * characters as if they were source code.
126 *
127 * <p>If the end tag is null or has zero length, then a source code
128 * region continues to the end of the input Reader's characters.
129 *
130 * @param source the source reader
131 * @param beginTag tag that marks the beginning of a source code region
132 * @param endTag tag that marks the end of a source code region
133 */
134 public SourceReader(Reader source, String beginTag, String endTag) {
135 this(source, beginTag, endTag, false);
136 }
137
138 /**
139 * The begin and end tags for a SourceReader are optional. If the begin
140 * tag is null or has zero length, then the SourceReader starts reading
141 * characters as if they were source code.
142 *
143 * <p>If the end tag is null or has zero length, then a source code
144 * region continues to the end of the input Reader's characters.
145 *
146 * @param source the source reader
147 * @param beginTag tag that marks the beginning of a source code region
148 * @param endTag tag that marks the end of a source code region
149 * @param inCode flag that indicates if the stream is starting in code
150 */
151 public SourceReader(Reader source, String beginTag, String endTag,
152 boolean inCode) {
153 super(createReader(source, beginTag, endTag), 2);
154 mUnicodeReader = (UnicodeReader)in;
155 mTagReader = (TagReader)mUnicodeReader.getOriginalSource();
156
157 boolean codeMode = ((beginTag == null || beginTag.length() == 0) ||
158 inCode);
159 mFirst = (codeMode) ? ENTER_CODE : ENTER_TEXT;
160
161 mBeginTag = beginTag;
162 mEndTag = endTag;
163 }
164
165 public String getBeginTag() {
166 return mBeginTag;
167 }
168
169 public String getEndTag() {
170 return mEndTag;
171 }
172
173 /**
174 * All newline character patterns are are converted to \n.
175 */
176 public int read() throws IOException {
177 int c;
178
179 if (mFirst != 0) {
180 c = mFirst;
181 mFirst = 0;
182 }
183 else {
184 c = super.read();
185 }
186
187 if (c == '\n') {
188 mLine++;
189 }
190 else if (c == ENTER_CODE) {
191 mUnicodeReader.setEscapesEnabled(true);
192 }
193 else if (c == ENTER_TEXT) {
194 mUnicodeReader.setEscapesEnabled(false);
195 }
196
197 return c;
198 }
199
200 public int getLineNumber() {
201 return mLine;
202 }
203
204 /**
205 * The position in the reader where the last read character ended. The
206 * position of the first character read from a Reader is zero.
207 *
208 * <p>The end position is usually the same as the start position, but
209 * sometimes a SourceReader may combine multiple characters into a
210 * single one.
211 *
212 * @return the end position where the last character was read
213 */
214 public int getEndPosition() {
215 int e = getNextPosition() - 1;
216 return (e < getStartPosition()) ? getStartPosition() : e;
217 }
218
219 public void ignoreTags(boolean ignore) {
220 mTagReader.setEscapesEnabled(!ignore);
221 }
222
223 public boolean isClosed() {
224 return mClosed;
225 }
226
227 public void close() throws IOException {
228 mClosed = true;
229 super.close();
230 }
231
232 protected void unreadHook(int c) {
233 if (c == '\n') {
234 mLine--;
235 }
236 else if (c == ENTER_CODE) {
237 mUnicodeReader.setEscapesEnabled(false);
238 }
239 else if (c == ENTER_TEXT) {
240 mUnicodeReader.setEscapesEnabled(true);
241 }
242 }
243
244 /**
245 * Simple test program
246 */
247 public static void main(String[] arg) throws Exception {
248 Tester.test(arg);
249 }
250
251 private static class Tester {
252 public static void test(String[] arg) throws Exception {
253 String str =
254 "This is \\" + "u0061 test.\n" +
255 "This is \\" + "u00612 test.\n" +
256 "This is \\" + "u0061" + "\\" + "u0061" + " test.\n" +
257 "This is \\" + "u061 test.\n" +
258 "This is \\\\" + "u0061 test.\n" +
259 "This is \\" + "a test.\n" +
260 "Plain text <%put code here%> plain text.\n" +
261 "Plain text <\\" + "u0025put code here%> plain text.\n" +
262 "Plain text <%put code here\\" + "u0025> plain text.\n";
263
264 Reader reader;
265
266 if (arg.length > 0) {
267 reader = new java.io.FileReader(arg[0]);
268 }
269 else {
270 System.out.println("\nOriginal:\n");
271
272 reader = new StringReader(str);
273
274 int c;
275 while ( (c = reader.read()) != -1 ) {
276 System.out.print((char)c);
277 }
278 }
279
280 System.out.println("\nTest 1:\n");
281
282 if (arg.length > 0) {
283 reader = new java.io.FileReader(arg[0]);
284 }
285 else {
286 reader = new StringReader(str);
287 }
288
289 SourceReader sr = new SourceReader(reader, "<%", "%>");
290
291 int c;
292 while ( (c = sr.read()) != -1 ) {
293 System.out.print((char)c);
294 System.out.print("\t" + c);
295 System.out.print("\t" + sr.getLineNumber());
296 System.out.print("\t" + sr.getStartPosition());
297 System.out.println("\t" + sr.getEndPosition());
298 }
299
300 System.out.println("\nTest 2:\n");
301 if (arg.length > 0) {
302 reader = new java.io.FileReader(arg[0]);
303 }
304 else {
305 reader = new StringReader(str);
306 }
307
308 sr = new SourceReader(reader, null, null);
309
310 while ( (c = sr.read()) != -1 ) {
311 System.out.print((char)c);
312 System.out.print("\t" + c);
313 System.out.print("\t" + sr.getLineNumber());
314 System.out.print("\t" + sr.getStartPosition());
315 System.out.println("\t" + sr.getEndPosition());
316 }
317 }
318 }
319 }