1
|
// XMLCatalogReader.java - Read XML 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.util.Vector;
|
9
|
import java.util.Enumeration;
|
10
|
import java.io.IOException;
|
11
|
import java.net.URL;
|
12
|
import java.net.MalformedURLException;
|
13
|
import com.arbortext.catalog.CatalogEntry;
|
14
|
import com.arbortext.catalog.CatalogReader;
|
15
|
import com.arbortext.catalog.InvalidCatalogEntryTypeException;
|
16
|
import com.arbortext.catalog.InvalidCatalogEntryException;
|
17
|
import com.arbortext.catalog.UnknownCatalogFormatException;
|
18
|
import com.arbortext.catalog.NotXMLCatalogException;
|
19
|
import com.arbortext.catalog.NoXMLParserException;
|
20
|
|
21
|
import org.xml.sax.*;
|
22
|
|
23
|
/**
|
24
|
* <p>Parses XML Catalog files.</p>
|
25
|
*
|
26
|
* <blockquote>
|
27
|
* <em>This module, both source code and documentation, is in the
|
28
|
* Public Domain, and comes with <strong>NO WARRANTY</strong>.</em>
|
29
|
* </blockquote>
|
30
|
*
|
31
|
* <p>This class reads XML Catalog files, returning a stream
|
32
|
* of tokens. At present, it recognizes John Cowan's
|
33
|
* <a href="http://www.ccil.org/~cowan/XML/XCatalog.html">XML Catalogs</a>
|
34
|
* (formerly XCatalogs). In the future, additional XML Catalog formats
|
35
|
* may be supported.</p>
|
36
|
*
|
37
|
* <p>This code interrogates the following non-standard system properties:</p>
|
38
|
*
|
39
|
* <dl>
|
40
|
* <dt><b>xml.catalog.debug</b></dt>
|
41
|
* <dd><p>Sets the debug level. A value of 0 is assumed if the
|
42
|
* property is not set or is not a number.</p></dd>
|
43
|
* </dl>
|
44
|
*
|
45
|
* @see Catalog
|
46
|
*
|
47
|
* @author Arbortext, Inc.
|
48
|
* @author Norman Walsh
|
49
|
* <a href="mailto:nwalsh@arbortext.com">nwalsh@arbortext.com</a>
|
50
|
* @version 1.0
|
51
|
*/
|
52
|
public class XMLCatalogReader implements DocumentHandler {
|
53
|
// These are class variables so that several methods can access them
|
54
|
/** The filename (URL) of the catalog being read */
|
55
|
private String catfilename = null;
|
56
|
|
57
|
/**
|
58
|
* <p>The debug level</p>
|
59
|
*
|
60
|
* <p>In general, higher numbers produce more information:</p>
|
61
|
* <ul>
|
62
|
* <li>0, no messages
|
63
|
* <li>1, minimal messages (high-level status)
|
64
|
* <li>2, detailed messages
|
65
|
* </ul>
|
66
|
*/
|
67
|
public int debug = 0;
|
68
|
|
69
|
/**
|
70
|
* Indicates that the catalog type is not XML
|
71
|
*/
|
72
|
private static final int NOTXMLCATALOG = -1;
|
73
|
|
74
|
/**
|
75
|
* Indicates that the catalog type is unknown.
|
76
|
*/
|
77
|
private static final int UNKNOWNCATALOG = 0;
|
78
|
|
79
|
/**
|
80
|
* Indicates that the catalog type is an XML Catalog (John Cowan's
|
81
|
* XCatalog)
|
82
|
*/
|
83
|
private static final int XCATALOG = 1;
|
84
|
|
85
|
/**
|
86
|
* Indicates the catalog type.
|
87
|
*/
|
88
|
private int catalogType = NOTXMLCATALOG;
|
89
|
|
90
|
/**
|
91
|
* <p>The list of entries scanned from the catalog.</p>
|
92
|
*
|
93
|
* <p>The SAX Parser is event-driven, but the Catalog class expects
|
94
|
* to iterate through the entries with
|
95
|
* <a href="#nextToken()">nextToken()</a>. So this class builds a
|
96
|
* vector of entries during the parse and returns them sequentially
|
97
|
* when <a href="#nextToken()">nextToken()</a> is called.</p>
|
98
|
*
|
99
|
* @see Catalog
|
100
|
*/
|
101
|
private Vector catalogEntries = new Vector();
|
102
|
|
103
|
/**
|
104
|
* An enumerator for walking through the list of catalogEntries.
|
105
|
*/
|
106
|
private Enumeration catalogEnum = null;
|
107
|
|
108
|
/**
|
109
|
* <p>The name of the parser class to load when parsing XML Catalogs.</p>
|
110
|
*
|
111
|
* <p>If a parser class is provided,
|
112
|
* subsequent attempts to parse Catalog files will begin
|
113
|
* by attemptiing an XML parse of the catalog file using a parser
|
114
|
* of this class.
|
115
|
* If the XML parse fails, the "default" text parse will be done
|
116
|
* instead.</p>
|
117
|
*/
|
118
|
private String parserClass = null;
|
119
|
|
120
|
/**
|
121
|
* <p>Construct an XMLCatalogReader object.</p>
|
122
|
*/
|
123
|
public XMLCatalogReader() {
|
124
|
String property = System.getProperty("xml.catalog.debug");
|
125
|
|
126
|
if (property != null) {
|
127
|
try {
|
128
|
debug = Integer.parseInt(property);
|
129
|
} catch (NumberFormatException e) {
|
130
|
debug = 0;
|
131
|
}
|
132
|
}
|
133
|
}
|
134
|
|
135
|
/**
|
136
|
* <p>Sets the parser class, enabling XML Catalog parsing.</p>
|
137
|
*
|
138
|
* <p>Sets the parser class that will be used for loading XML Catalogs.
|
139
|
* If this method is not called, all attempts to use the
|
140
|
* <code>XMLCatalogParser</code> will fail, throwing a
|
141
|
* <code>NoXMLParserException</code>.</p>
|
142
|
*
|
143
|
* @param parser The name of a class implementing the SAX Parser
|
144
|
* interface to be used for subsequent XML Catalog parsing.
|
145
|
*
|
146
|
* @see com.arbortext.catalog.NoXMLParserException
|
147
|
*/
|
148
|
public void setParserClass(String parser) {
|
149
|
parserClass = parser;
|
150
|
}
|
151
|
|
152
|
/**
|
153
|
* <p>Attempt to parse an XML Catalog file.</p>
|
154
|
*
|
155
|
* @param fileUrl The URL or filename of the catalog file to process
|
156
|
* @param catParser A SAX-compliant parser to use for reading the files
|
157
|
*
|
158
|
* @throws SAXException Error parsing catalog file.
|
159
|
* @throws IOException Error reading catalog file.
|
160
|
* @throws NoXMLParserException No Parser class provided.
|
161
|
* @throws NotXMLCatalogException The Catalog appears not to be XML.
|
162
|
* @throws UnknownCatalogFormatException Unexpected XML catalog type.
|
163
|
* @throws ClassNotFoundException Parser class can't be found.
|
164
|
* @throws InstantiationException Parser class can't be instantiated.
|
165
|
* @throws IllegalAccessException Error instantiating parser class.
|
166
|
* @throws ClassCastException Parser class isn't a SAX Parser.
|
167
|
*/
|
168
|
public void parseCatalog(String fileUrl)
|
169
|
throws SAXException, IOException,
|
170
|
NotXMLCatalogException, NoXMLParserException,
|
171
|
UnknownCatalogFormatException, ClassNotFoundException,
|
172
|
InstantiationException, IllegalAccessException,
|
173
|
ClassCastException {
|
174
|
// Create an instance of the parser
|
175
|
if (parserClass == null) {
|
176
|
throw new NoXMLParserException();
|
177
|
}
|
178
|
|
179
|
Parser parser = (Parser) Class.forName(parserClass).newInstance();
|
180
|
|
181
|
catfilename = fileUrl;
|
182
|
parser.setDocumentHandler(this);
|
183
|
parser.parse(fileUrl);
|
184
|
|
185
|
if (catalogType == NOTXMLCATALOG) {
|
186
|
// Why doesn't the attempt to parse this file throw a
|
187
|
// SAX Exception???
|
188
|
throw new NotXMLCatalogException();
|
189
|
}
|
190
|
|
191
|
if (catalogType == UNKNOWNCATALOG) {
|
192
|
throw new UnknownCatalogFormatException();
|
193
|
}
|
194
|
}
|
195
|
|
196
|
/**
|
197
|
* <p>Get the next entry from the file</p>
|
198
|
*
|
199
|
* @throws IOException Error reading catalog file
|
200
|
* @return A CatalogEntry object for the next entry in the catalog
|
201
|
*/
|
202
|
public CatalogEntry nextEntry() throws IOException {
|
203
|
if (catalogEnum == null) {
|
204
|
catalogEnum = catalogEntries.elements();
|
205
|
}
|
206
|
|
207
|
if (catalogEnum.hasMoreElements()) {
|
208
|
return (CatalogEntry) catalogEnum.nextElement();
|
209
|
} else {
|
210
|
return null;
|
211
|
}
|
212
|
}
|
213
|
|
214
|
// ----------------------------------------------------------------------
|
215
|
|
216
|
/*
|
217
|
* <p>Parse elements from John Cowan's XML Catalog doctype.</p>
|
218
|
*
|
219
|
* <p>Each recognized element is turned into an appropriate
|
220
|
* CatalogEntry and put onto the entries vector for later
|
221
|
* retrieval.</p>
|
222
|
*
|
223
|
* @param name The name of the element.
|
224
|
* @param atts The list of attributes on the element.
|
225
|
*
|
226
|
* @see CatalogEntry
|
227
|
*/
|
228
|
private void xCatalogEntry (String name, AttributeList atts) {
|
229
|
CatalogEntry ce = null;
|
230
|
|
231
|
try {
|
232
|
if (name.equals("Base")) {
|
233
|
ce = new CatalogEntry(CatalogEntry.BASE,
|
234
|
atts.getValue("HRef"));
|
235
|
debug(3, "Base", atts.getValue("HRef"));
|
236
|
}
|
237
|
|
238
|
if (name.equals("Delegate")) {
|
239
|
ce = new CatalogEntry(CatalogEntry.DELEGATE,
|
240
|
CatalogReader.normalize(atts.getValue("PublicId")),
|
241
|
atts.getValue("HRef"));
|
242
|
debug(3, "Delegate",
|
243
|
CatalogReader.normalize(atts.getValue("PublicId")),
|
244
|
atts.getValue("HRef"));
|
245
|
}
|
246
|
|
247
|
if (name.equals("Extend")) {
|
248
|
ce = new CatalogEntry(CatalogEntry.CATALOG,
|
249
|
atts.getValue("HRef"));
|
250
|
debug(3, "Extend", atts.getValue("HRef"));
|
251
|
}
|
252
|
|
253
|
if (name.equals("Map")) {
|
254
|
ce = new CatalogEntry(CatalogEntry.PUBLIC,
|
255
|
CatalogReader.normalize(atts.getValue("PublicId")),
|
256
|
atts.getValue("HRef"));
|
257
|
debug(3, "Map",
|
258
|
CatalogReader.normalize(atts.getValue("PublicId")),
|
259
|
atts.getValue("HRef"));
|
260
|
}
|
261
|
|
262
|
if (name.equals("Remap")) {
|
263
|
ce = new CatalogEntry(CatalogEntry.SYSTEM,
|
264
|
atts.getValue("SystemId"),
|
265
|
atts.getValue("HRef"));
|
266
|
debug(3, "Remap",
|
267
|
CatalogReader.normalize(atts.getValue("SystemId")),
|
268
|
atts.getValue("HRef"));
|
269
|
}
|
270
|
|
271
|
if (ce == null) {
|
272
|
// This is equivalent to an invalid catalog entry type
|
273
|
debug(1, "Invalid catalog entry type", name);
|
274
|
}
|
275
|
} catch (InvalidCatalogEntryTypeException icete) {
|
276
|
debug(1, "Invalid catalog entry type", name);
|
277
|
} catch (InvalidCatalogEntryException icete) {
|
278
|
debug(1, "Invalid catalog entry", name);
|
279
|
}
|
280
|
|
281
|
if (ce != null) {
|
282
|
catalogEntries.addElement(ce);
|
283
|
}
|
284
|
}
|
285
|
|
286
|
// ----------------------------------------------------------------------
|
287
|
// Implement the SAX DocumentHandler interface
|
288
|
|
289
|
/** <p>The SAX <code>setDocumentLocator</code> method. Does nothing.</p> */
|
290
|
public void setDocumentLocator (Locator locator) {
|
291
|
return;
|
292
|
}
|
293
|
|
294
|
/** <p>The SAX <code>startDocument</code> method. Does nothing.</p> */
|
295
|
public void startDocument ()
|
296
|
throws SAXException {
|
297
|
return;
|
298
|
}
|
299
|
|
300
|
/** <p>The SAX <code>endDocument</code> method. Does nothing.</p> */
|
301
|
public void endDocument ()
|
302
|
throws SAXException {
|
303
|
return;
|
304
|
}
|
305
|
|
306
|
/**
|
307
|
* <p>The SAX <code>startElement</code> method.</p>
|
308
|
*
|
309
|
* <p>This element attempts to identify the type of catalog
|
310
|
* by looking at the name of the first element encountered.
|
311
|
* If it recognizes the element, it sets the <code>catalogType</code>
|
312
|
* appropriately.</p>
|
313
|
*
|
314
|
* <p>After the catalog type has been identified, the appropriate
|
315
|
* entry parser is called for each subsequent element in the
|
316
|
* catalog.</p>
|
317
|
*
|
318
|
* @param name The name of the element.
|
319
|
* @param atts The list of attributes on the element.
|
320
|
*
|
321
|
*/
|
322
|
public void startElement (String name, AttributeList atts)
|
323
|
throws SAXException {
|
324
|
|
325
|
if (catalogType == UNKNOWNCATALOG || catalogType == NOTXMLCATALOG) {
|
326
|
if (name.equals("XMLCatalog")) {
|
327
|
catalogType = XCATALOG;
|
328
|
return;
|
329
|
}
|
330
|
}
|
331
|
|
332
|
if (catalogType == XCATALOG) {
|
333
|
xCatalogEntry(name, atts);
|
334
|
}
|
335
|
}
|
336
|
|
337
|
/** <p>The SAX <code>endElement</code> method. Does nothing.</p> */
|
338
|
public void endElement (String name)
|
339
|
throws SAXException {
|
340
|
return;
|
341
|
}
|
342
|
|
343
|
/** <p>The SAX <code>characters</code> method. Does nothing.</p> */
|
344
|
public void characters (char ch[], int start, int length)
|
345
|
throws SAXException {
|
346
|
return;
|
347
|
}
|
348
|
|
349
|
/** <p>The SAX <code>ignorableWhitespace</code> method. Does nothing.</p> */
|
350
|
public void ignorableWhitespace (char ch[], int start, int length)
|
351
|
throws SAXException {
|
352
|
return;
|
353
|
}
|
354
|
|
355
|
/** <p>The SAX <code>processingInstruction</code> method. Does nothing.</p> */
|
356
|
public void processingInstruction (String target, String data)
|
357
|
throws SAXException {
|
358
|
return;
|
359
|
}
|
360
|
|
361
|
// -----------------------------------------------------------------
|
362
|
|
363
|
/**
|
364
|
* <p>Print debug message (if the debug level is high enough).</p>
|
365
|
*
|
366
|
* @param level The debug level of this message. This message
|
367
|
* will only be
|
368
|
* displayed if the current debug level is at least equal to this
|
369
|
* value.
|
370
|
* @param message The text of the message.
|
371
|
* @param token The catalog file token being processed.
|
372
|
*/
|
373
|
private void debug(int level, String message, String token) {
|
374
|
if (debug >= level) {
|
375
|
System.out.println(message + ": " + token);
|
376
|
}
|
377
|
}
|
378
|
|
379
|
/**
|
380
|
* <p>Print debug message (if the debug level is high enough).</p>
|
381
|
*
|
382
|
* @param level The debug level of this message. This message
|
383
|
* will only be
|
384
|
* displayed if the current debug level is at least equal to this
|
385
|
* value.
|
386
|
* @param message The text of the message.
|
387
|
* @param token The catalog file token being processed.
|
388
|
* @param spec The argument to the token.
|
389
|
*/
|
390
|
private void debug(int level, String message, String token, String spec) {
|
391
|
if (debug >= level) {
|
392
|
System.out.println(message + ": " + token + " " + spec);
|
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 spec1 The first argument to the token.
|
406
|
* @param spec2 The second argument to the token.
|
407
|
*/
|
408
|
private void debug(int level, String message,
|
409
|
String token, String spec1, String spec2) {
|
410
|
if (debug >= level) {
|
411
|
System.out.println(message + ": " + token + " " + spec1);
|
412
|
System.out.println("\t" + spec2);
|
413
|
}
|
414
|
}
|
415
|
}
|