1 |
71
|
higgins
|
// CatalogReader.java - Read OASIS Catalog files
|
2 |
|
|
// Written by Norman Walsh, nwalsh@arbortext.com
|
3 |
|
|
// NO WARRANTY! This class is in the public domain.
|
4 |
|
|
|
5 |
|
|
package com.arbortext.catalog;
|
6 |
|
|
|
7 |
|
|
import java.lang.Integer;
|
8 |
|
|
import java.io.DataInputStream;
|
9 |
|
|
import java.io.IOException;
|
10 |
|
|
import java.io.FileNotFoundException;
|
11 |
|
|
import java.net.URL;
|
12 |
|
|
import java.net.MalformedURLException;
|
13 |
|
|
import com.arbortext.catalog.CatalogEntry;
|
14 |
|
|
import com.arbortext.catalog.InvalidCatalogEntryTypeException;
|
15 |
|
|
import com.arbortext.catalog.InvalidCatalogEntryException;
|
16 |
|
|
|
17 |
|
|
/**
|
18 |
|
|
* <p>Parses OASIS Open Catalog files.</p>
|
19 |
|
|
*
|
20 |
|
|
* <blockquote>
|
21 |
|
|
* <em>This module, both source code and documentation, is in the
|
22 |
|
|
* Public Domain, and comes with <strong>NO WARRANTY</strong>.</em>
|
23 |
|
|
* </blockquote>
|
24 |
|
|
*
|
25 |
|
|
* <p>This class reads OASIS Open Catalog files, returning a stream
|
26 |
|
|
* of tokens.</p>
|
27 |
|
|
*
|
28 |
|
|
* <p>This code interrogates the following non-standard system properties:</p>
|
29 |
|
|
*
|
30 |
|
|
* <dl>
|
31 |
|
|
* <dt><b>xml.catalog.debug</b></dt>
|
32 |
|
|
* <dd><p>Sets the debug level. A value of 0 is assumed if the
|
33 |
|
|
* property is not set or is not a number.</p></dd>
|
34 |
|
|
* </dl>
|
35 |
|
|
*
|
36 |
|
|
* @see Catalog
|
37 |
|
|
*
|
38 |
|
|
* @author Arbortext, Inc.
|
39 |
|
|
* @author Norman Walsh
|
40 |
|
|
* <a href="mailto:nwalsh@arbortext.com">nwalsh@arbortext.com</a>
|
41 |
|
|
* @version 1.0
|
42 |
|
|
*/
|
43 |
|
|
public class CatalogReader {
|
44 |
|
|
// These are class variables so that several methods can access them
|
45 |
|
|
/** The filename (URL) of the catalog being read */
|
46 |
|
|
private String catfilename = null;
|
47 |
|
|
|
48 |
|
|
/** The input stream used to read the catalog */
|
49 |
|
|
private DataInputStream catfile = null;
|
50 |
|
|
|
51 |
|
|
/**
|
52 |
|
|
* Lookahead stack. Reading a catalog sometimes requires up to
|
53 |
|
|
* two characters of lookahead.
|
54 |
|
|
*/
|
55 |
|
|
private int[] stack = new int[3];
|
56 |
|
|
|
57 |
|
|
/** The current position on the lookahead stack */
|
58 |
|
|
private int top = -1;
|
59 |
|
|
|
60 |
|
|
/**
|
61 |
|
|
* <p>The debug level</p>
|
62 |
|
|
*
|
63 |
|
|
* <p>In general, higher numbers produce more information:</p>
|
64 |
|
|
* <ul>
|
65 |
|
|
* <li>0, no messages
|
66 |
|
|
* <li>1, minimal messages (high-level status)
|
67 |
|
|
* <li>2, detailed messages
|
68 |
|
|
* </ul>
|
69 |
|
|
*/
|
70 |
|
|
public int debug = 0;
|
71 |
|
|
|
72 |
|
|
/**
|
73 |
|
|
* <p>Construct a CatalogReader object.</p>
|
74 |
|
|
*/
|
75 |
|
|
public CatalogReader() {
|
76 |
|
|
String property = System.getProperty("xml.catalog.debug");
|
77 |
|
|
|
78 |
|
|
if (property != null) {
|
79 |
|
|
try {
|
80 |
|
|
debug = Integer.parseInt(property);
|
81 |
|
|
} catch (NumberFormatException e) {
|
82 |
|
|
debug = 0;
|
83 |
|
|
}
|
84 |
|
|
}
|
85 |
|
|
}
|
86 |
|
|
|
87 |
|
|
/**
|
88 |
|
|
* <p>Start parsing an OASIS Open Catalog file. The file is
|
89 |
|
|
* actually read and parsed
|
90 |
|
|
* as needed by <code>nextEntry</code>.</p>
|
91 |
|
|
*
|
92 |
|
|
* @param fileUrl The URL or filename of the catalog file to process
|
93 |
|
|
*
|
94 |
|
|
* @throws MalformedURLException Improper fileUrl
|
95 |
|
|
* @throws IOException Error reading catalog file
|
96 |
|
|
*/
|
97 |
|
|
public void parseCatalog(String fileUrl)
|
98 |
|
|
throws MalformedURLException, IOException {
|
99 |
|
|
catfilename = fileUrl;
|
100 |
|
|
URL catalog;
|
101 |
|
|
|
102 |
|
|
try {
|
103 |
|
|
catalog = new URL(fileUrl);
|
104 |
|
|
} catch (MalformedURLException e) {
|
105 |
|
|
catalog = new URL("file:///" + fileUrl);
|
106 |
|
|
}
|
107 |
|
|
|
108 |
|
|
try {
|
109 |
|
|
catfile = new DataInputStream(catalog.openStream());
|
110 |
|
|
} catch (FileNotFoundException e) {
|
111 |
|
|
debug(1, "Failed to load catalog, file not found", catalog.toString());
|
112 |
|
|
}
|
113 |
|
|
}
|
114 |
|
|
|
115 |
|
|
/**
|
116 |
|
|
* <p>The destructor.</p>
|
117 |
|
|
*
|
118 |
|
|
* <p>Makes sure the catalog file is closed.</p>
|
119 |
|
|
*/
|
120 |
|
|
protected void finalize() throws IOException {
|
121 |
|
|
if (catfile != null) {
|
122 |
|
|
catfile.close();
|
123 |
|
|
}
|
124 |
|
|
catfile = null;
|
125 |
|
|
}
|
126 |
|
|
|
127 |
|
|
/**
|
128 |
|
|
* <p>Get the next entry from the file</p>
|
129 |
|
|
*
|
130 |
|
|
* @throws IOException Error reading catalog file
|
131 |
|
|
* @return A CatalogEntry object for the next entry in the catalog
|
132 |
|
|
*/
|
133 |
|
|
public CatalogEntry nextEntry() throws IOException {
|
134 |
|
|
if (catfile == null) {
|
135 |
|
|
return null;
|
136 |
|
|
}
|
137 |
|
|
|
138 |
|
|
boolean confused = false;
|
139 |
|
|
|
140 |
|
|
while (true) {
|
141 |
|
|
String token = nextToken();
|
142 |
|
|
|
143 |
|
|
if (token == null) {
|
144 |
|
|
catfile.close();
|
145 |
|
|
catfile = null;
|
146 |
|
|
return null;
|
147 |
|
|
}
|
148 |
|
|
|
149 |
|
|
if (token.equalsIgnoreCase("BASE")
|
150 |
|
|
|| token.equalsIgnoreCase("CATALOG")
|
151 |
|
|
|| token.equalsIgnoreCase("DOCUMENT")
|
152 |
|
|
|| token.equalsIgnoreCase("OVERRIDE")
|
153 |
|
|
|| token.equalsIgnoreCase("SGMLDECL")) {
|
154 |
|
|
String spec = nextToken();
|
155 |
|
|
confused = false;
|
156 |
|
|
|
157 |
|
|
try {
|
158 |
|
|
if (token.equalsIgnoreCase("BASE")) {
|
159 |
|
|
return new CatalogEntry(CatalogEntry.BASE, spec);
|
160 |
|
|
}
|
161 |
|
|
if (token.equalsIgnoreCase("CATALOG")) {
|
162 |
|
|
return new CatalogEntry(CatalogEntry.CATALOG, spec);
|
163 |
|
|
}
|
164 |
|
|
if (token.equalsIgnoreCase("DOCUMENT")) {
|
165 |
|
|
return new CatalogEntry(CatalogEntry.DOCUMENT, spec);
|
166 |
|
|
}
|
167 |
|
|
if (token.equalsIgnoreCase("OVERRIDE")) {
|
168 |
|
|
return new CatalogEntry(CatalogEntry.OVERRIDE, spec);
|
169 |
|
|
}
|
170 |
|
|
if (token.equalsIgnoreCase("SGMLDECL")) {
|
171 |
|
|
return new CatalogEntry(CatalogEntry.SGMLDECL, spec);
|
172 |
|
|
}
|
173 |
|
|
} catch (InvalidCatalogEntryTypeException icete) {
|
174 |
|
|
debug(1, "Invalid catalog entry type", token);
|
175 |
|
|
confused = true;
|
176 |
|
|
} catch (InvalidCatalogEntryException icete) {
|
177 |
|
|
debug(1, "Invalid catalog entry", token, spec);
|
178 |
|
|
confused = true;
|
179 |
|
|
}
|
180 |
|
|
}
|
181 |
|
|
|
182 |
|
|
if (token.equalsIgnoreCase("DELEGATE")
|
183 |
|
|
|| token.equalsIgnoreCase("DOCTYPE")
|
184 |
|
|
|| token.equalsIgnoreCase("DTDDECL")
|
185 |
|
|
|| token.equalsIgnoreCase("ENTITY")
|
186 |
|
|
|| token.equalsIgnoreCase("LINKTYPE")
|
187 |
|
|
|| token.equalsIgnoreCase("NOTATION")
|
188 |
|
|
|| token.equalsIgnoreCase("PUBLIC")
|
189 |
|
|
|| token.equalsIgnoreCase("SYSTEM")) {
|
190 |
|
|
String spec1 = nextToken();
|
191 |
|
|
String spec2 = nextToken();
|
192 |
|
|
confused = false;
|
193 |
|
|
try {
|
194 |
|
|
if (token.equalsIgnoreCase("DELEGATE")) {
|
195 |
|
|
return new CatalogEntry(CatalogEntry.DELEGATE,
|
196 |
|
|
normalize(spec1), spec2);
|
197 |
|
|
}
|
198 |
|
|
if (token.equalsIgnoreCase("DOCTYPE")) {
|
199 |
|
|
return new CatalogEntry(CatalogEntry.DOCTYPE,
|
200 |
|
|
spec1, spec2);
|
201 |
|
|
}
|
202 |
|
|
if (token.equalsIgnoreCase("DTDDECL")) {
|
203 |
|
|
return new CatalogEntry(CatalogEntry.DTDDECL,
|
204 |
|
|
normalize(spec1), spec2);
|
205 |
|
|
}
|
206 |
|
|
if (token.equalsIgnoreCase("ENTITY")) {
|
207 |
|
|
return new CatalogEntry(CatalogEntry.ENTITY,
|
208 |
|
|
spec1, spec2);
|
209 |
|
|
}
|
210 |
|
|
if (token.equalsIgnoreCase("LINKTYPE")) {
|
211 |
|
|
return new CatalogEntry(CatalogEntry.LINKTYPE,
|
212 |
|
|
spec1, spec2);
|
213 |
|
|
}
|
214 |
|
|
if (token.equalsIgnoreCase("NOTATION")) {
|
215 |
|
|
return new CatalogEntry(CatalogEntry.NOTATION,
|
216 |
|
|
spec1, spec2);
|
217 |
|
|
}
|
218 |
|
|
if (token.equalsIgnoreCase("PUBLIC")) {
|
219 |
|
|
return new CatalogEntry(CatalogEntry.PUBLIC,
|
220 |
|
|
normalize(spec1), spec2);
|
221 |
|
|
}
|
222 |
|
|
if (token.equalsIgnoreCase("SYSTEM")) {
|
223 |
|
|
return new CatalogEntry(CatalogEntry.SYSTEM,
|
224 |
|
|
spec1, spec2);
|
225 |
|
|
}
|
226 |
|
|
} catch (InvalidCatalogEntryTypeException icete) {
|
227 |
|
|
debug(1, "Invalid catalog entry type", token);
|
228 |
|
|
confused = true;
|
229 |
|
|
} catch (InvalidCatalogEntryException icete) {
|
230 |
|
|
debug(1, "Invalid catalog entry", token, spec1, spec2);
|
231 |
|
|
confused = true;
|
232 |
|
|
}
|
233 |
|
|
}
|
234 |
|
|
|
235 |
|
|
if (!confused) {
|
236 |
|
|
if (debug > 1) {
|
237 |
|
|
System.out.println("Unrecognized token parsing catalog: '"
|
238 |
|
|
+ catfilename
|
239 |
|
|
+ "': "
|
240 |
|
|
+ token);
|
241 |
|
|
System.out.println("\tSkipping to next recognized token.");
|
242 |
|
|
}
|
243 |
|
|
confused = true;
|
244 |
|
|
}
|
245 |
|
|
}
|
246 |
|
|
}
|
247 |
|
|
|
248 |
|
|
// -----------------------------------------------------------------
|
249 |
|
|
|
250 |
|
|
/**
|
251 |
|
|
* <p>Normalize a public identifier.</p>
|
252 |
|
|
*
|
253 |
|
|
* <p>Public identifiers must be normalized according to the following
|
254 |
|
|
* rules before comparisons between them can be made:</p>
|
255 |
|
|
*
|
256 |
|
|
* <ul>
|
257 |
|
|
* <li>Whitespace characters are normalized to spaces (e.g., line feeds,
|
258 |
|
|
* tabs, etc. become spaces).</li>
|
259 |
|
|
* <li>Leading and trailing whitespace is removed.</li>
|
260 |
|
|
* <li>Multiple internal whitespaces are normalized to a single
|
261 |
|
|
* space.</li>
|
262 |
|
|
* </ul>
|
263 |
|
|
*
|
264 |
|
|
* <p>This method is declared static so that other classes
|
265 |
|
|
* can use it directly.</p>
|
266 |
|
|
*
|
267 |
|
|
* @param publicId The unnormalized public identifier.
|
268 |
|
|
*
|
269 |
|
|
* @return The normalized identifier.
|
270 |
|
|
*/
|
271 |
|
|
public static String normalize(String publicId) {
|
272 |
|
|
String normal = publicId.replace('\t', ' ');
|
273 |
|
|
normal = normal.replace('\r', ' ');
|
274 |
|
|
normal = normal.replace('\n', ' ');
|
275 |
|
|
normal = normal.trim();
|
276 |
|
|
|
277 |
|
|
int pos;
|
278 |
|
|
|
279 |
|
|
while ((pos = normal.indexOf(" ")) >= 0) {
|
280 |
|
|
normal = normal.substring(0, pos) + normal.substring(pos+1);
|
281 |
|
|
}
|
282 |
|
|
|
283 |
|
|
return normal;
|
284 |
|
|
}
|
285 |
|
|
|
286 |
|
|
// -----------------------------------------------------------------
|
287 |
|
|
|
288 |
|
|
/**
|
289 |
|
|
* <p>Return the next token in the catalog file.</p>
|
290 |
|
|
*
|
291 |
|
|
* @return The Catalog file token from the input stream.
|
292 |
|
|
* @throws IOException If an error occurs reading from the stream.
|
293 |
|
|
*/
|
294 |
|
|
private String nextToken() throws IOException {
|
295 |
|
|
String token = "";
|
296 |
|
|
int ch, nextch;
|
297 |
|
|
|
298 |
|
|
// Skip over leading whitespace and comments
|
299 |
|
|
while (true) {
|
300 |
|
|
// skip leading whitespace
|
301 |
|
|
ch = catfile.read();
|
302 |
|
|
while (ch <= ' ') { // all ctrls are whitespace
|
303 |
|
|
ch = catfile.read();
|
304 |
|
|
if (ch < 0) {
|
305 |
|
|
return null;
|
306 |
|
|
}
|
307 |
|
|
}
|
308 |
|
|
|
309 |
|
|
// now 'ch' is the current char from the file
|
310 |
|
|
nextch = catfile.read();
|
311 |
|
|
if (nextch < 0) {
|
312 |
|
|
return null;
|
313 |
|
|
}
|
314 |
|
|
|
315 |
|
|
if (ch == '-' && nextch == '-') {
|
316 |
|
|
// we've found a comment, skip it...
|
317 |
|
|
ch = ' ';
|
318 |
|
|
nextch = nextChar();
|
319 |
|
|
while (ch != '-' || nextch != '-') {
|
320 |
|
|
ch = nextch;
|
321 |
|
|
nextch = nextChar();
|
322 |
|
|
}
|
323 |
|
|
|
324 |
|
|
// Ok, we've found the end of the comment,
|
325 |
|
|
// loop back to the top and start again...
|
326 |
|
|
} else {
|
327 |
|
|
stack[++top] = nextch;
|
328 |
|
|
stack[++top] = ch;
|
329 |
|
|
break;
|
330 |
|
|
}
|
331 |
|
|
}
|
332 |
|
|
|
333 |
|
|
ch = nextChar();
|
334 |
|
|
if (ch == '"' || ch == '\'') {
|
335 |
|
|
int quote = ch;
|
336 |
|
|
while ((ch = nextChar()) != quote) {
|
337 |
|
|
char[] chararr = new char[1];
|
338 |
|
|
chararr[0] = (char) ch;
|
339 |
|
|
String s = new String(chararr);
|
340 |
|
|
token = token.concat(s);
|
341 |
|
|
}
|
342 |
|
|
return token;
|
343 |
|
|
} else {
|
344 |
|
|
// return the next whitespace or comment delimited
|
345 |
|
|
// string
|
346 |
|
|
while (ch > ' ') {
|
347 |
|
|
nextch = nextChar();
|
348 |
|
|
if (ch == '-' && nextch == '-') {
|
349 |
|
|
stack[++top] = ch;
|
350 |
|
|
stack[++top] = nextch;
|
351 |
|
|
return token;
|
352 |
|
|
} else {
|
353 |
|
|
char[] chararr = new char[1];
|
354 |
|
|
chararr[0] = (char) ch;
|
355 |
|
|
String s = new String(chararr);
|
356 |
|
|
token = token.concat(s);
|
357 |
|
|
ch = nextch;
|
358 |
|
|
}
|
359 |
|
|
}
|
360 |
|
|
return token;
|
361 |
|
|
}
|
362 |
|
|
}
|
363 |
|
|
|
364 |
|
|
/**
|
365 |
|
|
* <p>Return the next logical character from the input stream.</p>
|
366 |
|
|
*
|
367 |
|
|
* @return The next (logical) character from the input stream. The
|
368 |
|
|
* character may be buffered from a previous lookahead.
|
369 |
|
|
*
|
370 |
|
|
* @throws IOException If an error occurs reading from the stream.
|
371 |
|
|
*/
|
372 |
|
|
private int nextChar() throws IOException {
|
373 |
|
|
if (top < 0) {
|
374 |
|
|
return catfile.read();
|
375 |
|
|
} else {
|
376 |
|
|
return stack[top--];
|
377 |
|
|
}
|
378 |
|
|
}
|
379 |
|
|
|
380 |
|
|
/**
|
381 |
|
|
* <p>Print debug message (if the debug level is high enough).</p>
|
382 |
|
|
*
|
383 |
|
|
* @param level The debug level of this message. This message
|
384 |
|
|
* will only be
|
385 |
|
|
* displayed if the current debug level is at least equal to this
|
386 |
|
|
* value.
|
387 |
|
|
* @param message The text of the message.
|
388 |
|
|
* @param token The catalog file token being processed.
|
389 |
|
|
*/
|
390 |
|
|
private void debug(int level, String message, String token) {
|
391 |
|
|
if (debug >= level) {
|
392 |
|
|
System.out.println(message + ": " + token);
|
393 |
|
|
}
|
394 |
|
|
}
|
395 |
|
|
|
396 |
|
|
/**
|
397 |
|
|
* <p>Print debug message (if the debug level is high enough).</p>
|
398 |
|
|
*
|
399 |
|
|
* @param level The debug level of this message. This message
|
400 |
|
|
* will only be
|
401 |
|
|
* displayed if the current debug level is at least equal to this
|
402 |
|
|
* value.
|
403 |
|
|
* @param message The text of the message.
|
404 |
|
|
* @param token The catalog file token being processed.
|
405 |
|
|
* @param spec The argument to the token.
|
406 |
|
|
*/
|
407 |
|
|
private void debug(int level, String message, String token, String spec) {
|
408 |
|
|
if (debug >= level) {
|
409 |
|
|
System.out.println(message + ": " + token + " " + spec);
|
410 |
|
|
}
|
411 |
|
|
}
|
412 |
|
|
|
413 |
|
|
/**
|
414 |
|
|
* <p>Print debug message (if the debug level is high enough).</p>
|
415 |
|
|
*
|
416 |
|
|
* @param level The debug level of this message. This message
|
417 |
|
|
* will only be
|
418 |
|
|
* displayed if the current debug level is at least equal to this
|
419 |
|
|
* value.
|
420 |
|
|
* @param message The text of the message.
|
421 |
|
|
* @param token The catalog file token being processed.
|
422 |
|
|
* @param spec1 The first argument to the token.
|
423 |
|
|
* @param spec2 The second argument to the token.
|
424 |
|
|
*/
|
425 |
|
|
private void debug(int level, String message,
|
426 |
|
|
String token, String spec1, String spec2) {
|
427 |
|
|
if (debug >= level) {
|
428 |
|
|
System.out.println(message + ": " + token + " " + spec1);
|
429 |
|
|
System.out.println("\t" + spec2);
|
430 |
|
|
}
|
431 |
|
|
}
|
432 |
|
|
}
|