|
1 |
// Catalog.java - Represents OASIS Open Catalog files.
|
|
2 |
|
|
3 |
// Written by Norman Walsh, nwalsh@arbortext.com
|
|
4 |
// NO WARRANTY! This class is in the public domain.
|
|
5 |
|
|
6 |
package com.arbortext.catalog;
|
|
7 |
|
|
8 |
import java.lang.Integer;
|
|
9 |
import java.io.IOException;
|
|
10 |
import java.io.FileNotFoundException;
|
|
11 |
import java.util.Enumeration;
|
|
12 |
import java.util.Hashtable;
|
|
13 |
import java.util.Vector;
|
|
14 |
import java.net.URL;
|
|
15 |
import java.net.MalformedURLException;
|
|
16 |
import com.arbortext.catalog.CatalogReader;
|
|
17 |
import com.arbortext.catalog.XMLCatalogReader;
|
|
18 |
import com.arbortext.catalog.NotXMLCatalogException;
|
|
19 |
import com.arbortext.catalog.NoXMLParserException;
|
|
20 |
import org.xml.sax.SAXException;
|
|
21 |
|
|
22 |
/**
|
|
23 |
* <p>Represents OASIS Open Catalog files.</p>
|
|
24 |
*
|
|
25 |
* <blockquote>
|
|
26 |
* <em>This module, both source code and documentation, is in the
|
|
27 |
* Public Domain, and comes with <strong>NO WARRANTY</strong>.</em>
|
|
28 |
* </blockquote>
|
|
29 |
*
|
|
30 |
* <p>This class loads one or more OASIS Open Catalog files
|
|
31 |
* (defined by
|
|
32 |
* <a href="http://www.oasis-open.org/html/a401.htm">OASIS Technical
|
|
33 |
* Resolution 9401:1997 (Amendment 2 to TR 9401)</a>)
|
|
34 |
* and provides
|
|
35 |
* methods for implementing the Catalog semantics.</p>
|
|
36 |
*
|
|
37 |
* <p>The primary purpose of the Catalog is to associate resources in the
|
|
38 |
* document with local system identifiers. Some entities
|
|
39 |
* (document types, XML entities, and notations) have names and all of them
|
|
40 |
* can have either public or system identifiers or both. (In XML, only a
|
|
41 |
* notation can have a public identifier without a system identifier, but
|
|
42 |
* the methods implemented in this class obey the Catalog semantics
|
|
43 |
* from the SGML
|
|
44 |
* days when system identifiers were optional.)</p>
|
|
45 |
*
|
|
46 |
* <p>The system identifiers returned by the resolution methods in this
|
|
47 |
* class are valid, i.e. usable by, and in fact constructed by, the
|
|
48 |
* <tt>java.net.URL</tt> class. Unfortunately, this class seems to behave in
|
|
49 |
* somewhat non-standard ways and the system identifiers returned may
|
|
50 |
* not be directly usable in a browser or filesystem context.
|
|
51 |
*
|
|
52 |
* <p>This class processes the following Catalog entries:</p>
|
|
53 |
*
|
|
54 |
* <ul>
|
|
55 |
* <li><b>BASE</b>
|
|
56 |
* changes the base URI for resolving relative system identifiers. The
|
|
57 |
* initial base URI is the URI of the location of the catalog (which is,
|
|
58 |
* in turn, relative to the location of the current working directory
|
|
59 |
* at startup, as returned by the <tt>user.dir</tt> system property).</li>
|
|
60 |
* <li><b>CATALOG</b>
|
|
61 |
* processes other catalog files. An included catalog occurs logically
|
|
62 |
* at the end of the including catalog.</li>
|
|
63 |
* <li><b>DELEGATE</b>
|
|
64 |
* specifies alternate catalogs for some public identifiers. The delegated
|
|
65 |
* catalogs are not loaded until they are needed, but they are cached
|
|
66 |
* once loaded.</li>
|
|
67 |
* <li><b>DOCTYPE</b>
|
|
68 |
* associates the names of root elements with URIs. (In other words, an XML
|
|
69 |
* processor might infer the doctype of an XML document that does not include
|
|
70 |
* a doctype declaration by looking for the DOCTYPE entry in the
|
|
71 |
* catalog which matches the name of the root element of the document.)</li>
|
|
72 |
* <li><b>DOCUMENT</b>
|
|
73 |
* provides a default document.</li>
|
|
74 |
* <li><b>DTDDECL</b>
|
|
75 |
* recognized and silently ignored. Not relevant for XML.</li>
|
|
76 |
* <li><b>ENTITY</b>
|
|
77 |
* associates entity names with URIs.</li>
|
|
78 |
* <li><b>LINKTYPE</b>
|
|
79 |
* recognized and silently ignored. Not relevant for XML.</li>
|
|
80 |
* <li><b>NOTATION</b>
|
|
81 |
* associates notation names with URIs.</li>
|
|
82 |
* <li><b>OVERRIDE</b>
|
|
83 |
* changes the override behavior. Initial behavior is set by the
|
|
84 |
* system property <tt>xml.catalog.override</tt>. The default initial
|
|
85 |
* behavior is 'YES', that is, entries in the catalog override
|
|
86 |
* system identifiers specified in the document.</li>
|
|
87 |
* <li><b>PUBLIC</b>
|
|
88 |
* maps a public identifier to a system identifier.</li>
|
|
89 |
* <li><b>SGMLDECL</b>
|
|
90 |
* recognized and silently ignored. Not relevant for XML.</li>
|
|
91 |
* <li><b>SYSTEM</b>
|
|
92 |
* maps a system identifier to another system identifier.</li>
|
|
93 |
* </ul>
|
|
94 |
*
|
|
95 |
* <p>Note that subordinate catalogs (all catalogs except the first,
|
|
96 |
* including CATALOG and DELEGATE catalogs) are only loaded if and when
|
|
97 |
* they are required.</p>
|
|
98 |
*
|
|
99 |
* <p>If provided with an SAX Parser class, this object can also load
|
|
100 |
* XML Catalogs. For the details about which XML Catalog formats are
|
|
101 |
* recognized, see {@link XMLCatalogReader}.
|
|
102 |
*
|
|
103 |
* <p>This code interrogates the following non-standard system properties:</p>
|
|
104 |
*
|
|
105 |
* <dl>
|
|
106 |
* <dt><b>xml.catalog.debug</b></dt>
|
|
107 |
* <dd><p>Sets the debug level. A value of 0 is assumed if the
|
|
108 |
* property is not set or is not a number.</p></dd>
|
|
109 |
*
|
|
110 |
* <dt><b>xml.catalog.override</b></dt>
|
|
111 |
* <dd><p>Specifies the default override behavior. If override is true ("true",
|
|
112 |
* "yes", "1"), system identifiers in the catalog file are used in preference
|
|
113 |
* to system identifiers in the document. In other words, a value of false
|
|
114 |
* essentially disables catalog processing since almost all external
|
|
115 |
* entities are required to have a system identifier in XML.
|
|
116 |
* A value of true is assumed if the property is not set.</p></dd>
|
|
117 |
*
|
|
118 |
* <dt><b>xml.catalog.files</b></dt>
|
|
119 |
* <dd><p>Identifies the list of catalog <i>files</i> to parse initially.
|
|
120 |
* (Additional catalog files may be parsed if the CATALOG entry
|
|
121 |
* is used.) Components of the list should be separated by the system
|
|
122 |
* property "<code>path.separator</code>" character
|
|
123 |
* (typically ";" on DOS/Windows systems, ":" on Unix systems).</p>
|
|
124 |
*
|
|
125 |
* <p>Additional catalogs may also be loaded with the
|
|
126 |
* {@link #parseCatalog} method.</p>
|
|
127 |
* </dd>
|
|
128 |
* </dl>
|
|
129 |
*
|
|
130 |
* <p><b>Change Log:</b></p>
|
|
131 |
* <dl>
|
|
132 |
* <dt>1.0.1</dt>
|
|
133 |
* <dd><p>Fixed a bug in the calculation of the list of subordinate catalogs.
|
|
134 |
* This bug caused an infinite loop where parsing would alternately process
|
|
135 |
* two catalogs indefinitely.</p>
|
|
136 |
* </dd>
|
|
137 |
* </dl>
|
|
138 |
*
|
|
139 |
* @see CatalogReader
|
|
140 |
* @see XMLCatalogReader
|
|
141 |
* @see CatalogEntry
|
|
142 |
*
|
|
143 |
* @author Abortext, Inc.
|
|
144 |
* @author Norman Walsh
|
|
145 |
* <a href="mailto:nwalsh@arbortext.com">nwalsh@arbortext.com</a>
|
|
146 |
* @version 1.0.1
|
|
147 |
*/
|
|
148 |
public class Catalog {
|
|
149 |
/**
|
|
150 |
* The base URI for relative system identifiers in the catalog.
|
|
151 |
* This may be changed by BASE entries in the catalog.
|
|
152 |
*/
|
|
153 |
private URL base;
|
|
154 |
|
|
155 |
/**
|
|
156 |
* The base URI of the Catalog file currently being parsed.
|
|
157 |
*/
|
|
158 |
private URL catalogCwd;
|
|
159 |
|
|
160 |
/** The catalog entries currently known to the system. */
|
|
161 |
private Vector catalogEntries = new Vector();
|
|
162 |
|
|
163 |
/** The default initial override setting. */
|
|
164 |
private boolean default_override = true;
|
|
165 |
|
|
166 |
/**
|
|
167 |
* <p>The debug level.</p>
|
|
168 |
*
|
|
169 |
* <p>In general, higher numbers produce more information:</p>
|
|
170 |
* <ul>
|
|
171 |
* <li>0, no messages
|
|
172 |
* <li>1, minimal messages (high-level status)
|
|
173 |
* <li>2, more messages
|
|
174 |
* <li>3, detailed messages
|
|
175 |
* </ul>
|
|
176 |
*/
|
|
177 |
public int debug = 0;
|
|
178 |
|
|
179 |
/**
|
|
180 |
* <p>A vector of catalog files to be loaded.</p>
|
|
181 |
*
|
|
182 |
* <p>This list is initially established by
|
|
183 |
* <code>loadSystemCatalogs</code> when
|
|
184 |
* it parses the system catalog list, but CATALOG entries may
|
|
185 |
* contribute to it during the course of parsing.</p>
|
|
186 |
*
|
|
187 |
* @see #loadSystemCatalogs
|
|
188 |
* @see localCatalogFiles
|
|
189 |
*/
|
|
190 |
private Vector catalogFiles = new Vector();
|
|
191 |
|
|
192 |
/**
|
|
193 |
* <p>A vector of catalog files constructed during processing of
|
|
194 |
* CATALOG entries in the current catalog.</p>
|
|
195 |
*
|
|
196 |
* <p>This two-level system is actually necessary to correctly implement
|
|
197 |
* the semantics of the CATALOG entry. If one catalog file includes
|
|
198 |
* another with a CATALOG entry, the included catalog logically
|
|
199 |
* occurs <i>at the end</i> of the including catalog, and after any
|
|
200 |
* preceding CATALOG entries. In other words, the CATALOG entry
|
|
201 |
* cannot insert anything into the middle of a catalog file.</p>
|
|
202 |
*
|
|
203 |
* <p>When processing reaches the end of each catalog files, any
|
|
204 |
* elements on this vector are added to the front of the
|
|
205 |
* <code>catalogFiles</code> vector.</p>
|
|
206 |
*
|
|
207 |
* @see catalogFiles
|
|
208 |
*/
|
|
209 |
private Vector localCatalogFiles = new Vector();
|
|
210 |
|
|
211 |
/**
|
|
212 |
* <p>A vector of Catalogs.</p>
|
|
213 |
*
|
|
214 |
* <p>The semantics of Catalog resolution are such that each
|
|
215 |
* catalog is effectively a list of Catalogs (in other words,
|
|
216 |
* a recursive list of Catalog instances).</p>
|
|
217 |
*
|
|
218 |
* <p>Catalogs that are processed as the result of CATALOG or
|
|
219 |
* DELEGATE entries are subordinate to the catalog that contained
|
|
220 |
* them, but they may in turn have subordinate catalogs.</p>
|
|
221 |
*
|
|
222 |
* <p>Catalogs are only loaded when they are needed, so this vector
|
|
223 |
* initially contains a list of Catalog filenames (URLs). If, during
|
|
224 |
* processing, one of these catalogs has to be loaded, the resulting
|
|
225 |
* Catalog object is placed in the vector, effectively caching it
|
|
226 |
* for the next query.</p>
|
|
227 |
*/
|
|
228 |
private Vector catalogs = new Vector();
|
|
229 |
|
|
230 |
/**
|
|
231 |
* <p>A vector of DELEGATE Catalog entries constructed during
|
|
232 |
* processing of the Catalog.</p>
|
|
233 |
*
|
|
234 |
* <p>This two-level system has two purposes; first, it allows
|
|
235 |
* us to sort the DELEGATE entries by the length of the partial
|
|
236 |
* public identifier so that a linear search encounters them in
|
|
237 |
* the correct order and second, it puts them all at the end of
|
|
238 |
* the Catalog.</p>
|
|
239 |
*
|
|
240 |
* <p>When processing reaches the end of each catalog file, any
|
|
241 |
* elements on this vector are added to the end of the
|
|
242 |
* <code>catalogEntries</code> vector. This assures that matching
|
|
243 |
* PUBLIC keywords are encountered before DELEGATE entries.</p>
|
|
244 |
*/
|
|
245 |
private Vector localDelegate = new Vector();
|
|
246 |
|
|
247 |
/**
|
|
248 |
* <p>The name of the parser class to load when parsing XML Catalogs.</p>
|
|
249 |
*
|
|
250 |
* <p>If a parser class is provided,
|
|
251 |
* subsequent attempts to parse Catalog files will begin
|
|
252 |
* by attemptiing an XML parse of the catalog file using a parser
|
|
253 |
* of this class.
|
|
254 |
* If the XML parse fails, the "default" text parse will be done
|
|
255 |
* instead.</p>
|
|
256 |
*/
|
|
257 |
private String parserClass = null;
|
|
258 |
|
|
259 |
/**
|
|
260 |
* <p>Constructs an empty Catalog.</p>
|
|
261 |
*
|
|
262 |
* <p>The constructor interrogates the relevant system properties
|
|
263 |
* and initializes the catalog data structures.</p>
|
|
264 |
*/
|
|
265 |
public Catalog() {
|
|
266 |
String property = System.getProperty("xml.catalog.debug");
|
|
267 |
|
|
268 |
if (property != null) {
|
|
269 |
try {
|
|
270 |
debug = Integer.parseInt(property);
|
|
271 |
} catch (NumberFormatException e) {
|
|
272 |
debug = 0;
|
|
273 |
}
|
|
274 |
}
|
|
275 |
|
|
276 |
property = System.getProperty("xml.catalog.override");
|
|
277 |
|
|
278 |
if (property != null) {
|
|
279 |
default_override = (property.equalsIgnoreCase("true")
|
|
280 |
|| property.equalsIgnoreCase("yes")
|
|
281 |
|| property.equalsIgnoreCase("1"));
|
|
282 |
}
|
|
283 |
}
|
|
284 |
|
|
285 |
/**
|
|
286 |
* <p>Sets the parser class, enabling XML Catalog parsing.</p>
|
|
287 |
*
|
|
288 |
* <p>Sets the parser class that will be used for loading XML Catalogs.
|
|
289 |
* If this method is not called, all catalogs will be parsed as
|
|
290 |
* plain text (and assumed to conform to the
|
|
291 |
* <a href="http://www.oasis-open.org/html/a401.htm">OASIS Catalog
|
|
292 |
* format</a>).</p>
|
|
293 |
*
|
|
294 |
* @param parser The name of a class implementing the SAX Parser
|
|
295 |
* interface to be used for subsequent XML Catalog parsing.
|
|
296 |
*/
|
|
297 |
public void setParserClass(String parser) {
|
|
298 |
parserClass = parser;
|
|
299 |
}
|
|
300 |
|
|
301 |
/**
|
|
302 |
* <p>Load the system catalog files.</p>
|
|
303 |
*
|
|
304 |
* <p>The method adds all of the
|
|
305 |
* catalogs specified in the <tt>xml.catalog.files</tt> property
|
|
306 |
* to the Catalog list.</p>
|
|
307 |
*
|
|
308 |
* @throws MalformedURLException One of the system catalogs is
|
|
309 |
* identified with a filename that is not a valid URL.
|
|
310 |
* @throws IOException One of the system catalogs cannot be read.
|
|
311 |
*/
|
|
312 |
public void loadSystemCatalogs()
|
|
313 |
throws MalformedURLException, IOException {
|
|
314 |
String PCS = System.getProperty("path.separator");
|
|
315 |
String catalog_files = System.getProperty("xml.catalog.files");
|
|
316 |
|
|
317 |
while (catalog_files != null) {
|
|
318 |
int pos = catalog_files.indexOf(PCS);
|
|
319 |
String catfile = null;
|
|
320 |
|
|
321 |
if (pos > 0) {
|
|
322 |
catfile = catalog_files.substring(0, pos);
|
|
323 |
catalog_files = catalog_files.substring(pos+1);
|
|
324 |
} else {
|
|
325 |
catfile = catalog_files;
|
|
326 |
catalog_files = null;
|
|
327 |
}
|
|
328 |
|
|
329 |
catalogFiles.addElement(catfile);
|
|
330 |
}
|
|
331 |
|
|
332 |
if (catalogFiles.size() > 0) {
|
|
333 |
// This is a little odd. The parseCatalog() method expects
|
|
334 |
// a filename, but it adds that name to the end of the
|
|
335 |
// catalogFiles vector, and then processes that vector.
|
|
336 |
// This allows the system to handle CATALOG entries
|
|
337 |
// correctly.
|
|
338 |
//
|
|
339 |
// In this init case, we take the last element off the
|
|
340 |
// catalogFiles vector and pass it to parseCatalog. This
|
|
341 |
// will "do the right thing" in the init case, and allow
|
|
342 |
// parseCatalog() to do the right thing in the non-init
|
|
343 |
// case. Honest.
|
|
344 |
//
|
|
345 |
String catfile = (String) catalogFiles.lastElement();
|
|
346 |
catalogFiles.removeElement(catfile);
|
|
347 |
parseCatalog(catfile);
|
|
348 |
}
|
|
349 |
}
|
|
350 |
|
|
351 |
/**
|
|
352 |
* <p>Parse a catalog file, augmenting internal data structures</p>
|
|
353 |
*
|
|
354 |
* @param fileName The filename of the catalog file to process
|
|
355 |
*
|
|
356 |
* @throws MalformedURLException The fileName cannot be turned into
|
|
357 |
* a valid URL.
|
|
358 |
* @throws IOException Error reading catalog file.
|
|
359 |
*/
|
|
360 |
public synchronized void parseCatalog(String fileName)
|
|
361 |
throws MalformedURLException, IOException {
|
|
362 |
|
|
363 |
// Put the file into the list of catalogs to process...
|
|
364 |
// In all cases except the case when initCatalog() is the
|
|
365 |
// caller, this will be the only catalog initially in the list...
|
|
366 |
catalogFiles.addElement(fileName);
|
|
367 |
|
|
368 |
// Now process all the files on the catalogFiles vector. This
|
|
369 |
// vector can grow during processing if CATALOG entries are
|
|
370 |
// encountered in the catalog
|
|
371 |
int curCat = 0;
|
|
372 |
while (curCat < catalogFiles.size()) {
|
|
373 |
String catfile = (String) catalogFiles.elementAt(curCat++);
|
|
374 |
|
|
375 |
if (catalogEntries.size() == 0 && catalogs.size() == 0) {
|
|
376 |
// We haven't parsed any catalogs yet, let this
|
|
377 |
// catalog be the first...
|
|
378 |
parseCatalogFile(catfile);
|
|
379 |
} else {
|
|
380 |
// This is a subordinate catalog. We save its name,
|
|
381 |
// but don't bother to load it unless it's necessary.
|
|
382 |
catalogs.addElement(catfile);
|
|
383 |
}
|
|
384 |
|
|
385 |
if (!localCatalogFiles.isEmpty()) {
|
|
386 |
// Move all the localCatalogFiles into the front of
|
|
387 |
// the catalogFiles queue
|
|
388 |
Vector newQueue = new Vector();
|
|
389 |
Enumeration q = localCatalogFiles.elements();
|
|
390 |
while (q.hasMoreElements()) {
|
|
391 |
newQueue.addElement(q.nextElement());
|
|
392 |
}
|
|
393 |
|
|
394 |
// Put the rest of the catalogs on the end of the new list
|
|
395 |
while (curCat < catalogFiles.size()) {
|
|
396 |
catfile = (String) catalogFiles.elementAt(curCat++);
|
|
397 |
newQueue.addElement(catfile);
|
|
398 |
}
|
|
399 |
|
|
400 |
localCatalogFiles = new Vector();
|
|
401 |
catalogFiles = newQueue;
|
|
402 |
curCat = 0;
|
|
403 |
}
|
|
404 |
|
|
405 |
if (!localDelegate.isEmpty()) {
|
|
406 |
Enumeration e = localDelegate.elements();
|
|
407 |
while (e.hasMoreElements()) {
|
|
408 |
catalogEntries.addElement(e.nextElement());
|
|
409 |
}
|
|
410 |
localDelegate = new Vector();
|
|
411 |
}
|
|
412 |
}
|
|
413 |
|
|
414 |
// We've parsed them all, reinit the vector...
|
|
415 |
catalogFiles = new Vector();
|
|
416 |
}
|
|
417 |
|
|
418 |
/**
|
|
419 |
* <p>Parse a single catalog file, augmenting internal data structures</p>
|
|
420 |
*
|
|
421 |
* @param fileName The filename of the catalog file to process
|
|
422 |
*
|
|
423 |
* @throws MalformedURLException The fileName cannot be turned into
|
|
424 |
* a valid URL.
|
|
425 |
* @throws IOException Error reading catalog file.
|
|
426 |
*/
|
|
427 |
private synchronized void parseCatalogFile(String fileName)
|
|
428 |
throws MalformedURLException, IOException {
|
|
429 |
|
|
430 |
CatalogEntry entry;
|
|
431 |
|
|
432 |
// The base-base is the cwd. If the catalog file is specified
|
|
433 |
// with a relative path, this assures that it gets resolved
|
|
434 |
// properly...
|
|
435 |
try {
|
|
436 |
// tack on a basename because URLs point to files not dirs
|
|
437 |
String userdir = fixSlashes(System.getProperty("user.dir"));
|
|
438 |
catalogCwd = new URL("file:///" + userdir + "/basename");
|
|
439 |
} catch (MalformedURLException e) {
|
|
440 |
String userdir = fixSlashes(System.getProperty("user.dir"));
|
|
441 |
debug(1, "Malformed URL on cwd", userdir);
|
|
442 |
catalogCwd = null;
|
|
443 |
}
|
|
444 |
|
|
445 |
// The initial base URI is the location of the catalog file
|
|
446 |
try {
|
|
447 |
base = new URL(catalogCwd, fixSlashes(fileName));
|
|
448 |
} catch (MalformedURLException e) {
|
|
449 |
try {
|
|
450 |
base = new URL("file:///" + fixSlashes(fileName));
|
|
451 |
} catch (MalformedURLException e2) {
|
|
452 |
debug(1, "Malformed URL on catalog filename",
|
|
453 |
fixSlashes(fileName));
|
|
454 |
base = null;
|
|
455 |
}
|
|
456 |
}
|
|
457 |
|
|
458 |
debug(1, "Loading catalog", fileName);
|
|
459 |
debug(3, "Default BASE", base.toString());
|
|
460 |
|
|
461 |
fileName = base.toString();
|
|
462 |
|
|
463 |
if (parserClass != null) {
|
|
464 |
try {
|
|
465 |
XMLCatalogReader catfile = new XMLCatalogReader();
|
|
466 |
catfile.setParserClass(parserClass);
|
|
467 |
catfile.parseCatalog(fileName);
|
|
468 |
|
|
469 |
CatalogEntry ce = null;
|
|
470 |
while ((ce = catfile.nextEntry()) != null) {
|
|
471 |
addEntry(ce);
|
|
472 |
}
|
|
473 |
return;
|
|
474 |
} catch (SAXException e1) {
|
|
475 |
// not an XML catalog, continue with text parse
|
|
476 |
} catch (NoXMLParserException e2) {
|
|
477 |
// not an XML catalog, continue with text parse
|
|
478 |
} catch (NotXMLCatalogException e2) {
|
|
479 |
// not an XML catalog, continue with text parse
|
|
480 |
} catch (InstantiationException e3) {
|
|
481 |
debug(1, "Cannot instantiate XML Parser class", parserClass);
|
|
482 |
} catch (IllegalAccessException e4) {
|
|
483 |
debug(1, "Cannot access XML Parser class", parserClass);
|
|
484 |
} catch (ClassNotFoundException e5) {
|
|
485 |
debug(1, "Cannot load XML Parser class", parserClass);
|
|
486 |
} catch (UnknownCatalogFormatException e6) {
|
|
487 |
debug(1, "Unrecognized XML Catalog format.");
|
|
488 |
return;
|
|
489 |
}
|
|
490 |
}
|
|
491 |
|
|
492 |
CatalogReader catfile = new CatalogReader();
|
|
493 |
catfile.parseCatalog(fileName);
|
|
494 |
|
|
495 |
// Process the contents of the catalog file as a whitespace
|
|
496 |
// delimited set of tokens
|
|
497 |
while ((entry = catfile.nextEntry()) != null) {
|
|
498 |
addEntry(entry);
|
|
499 |
}
|
|
500 |
}
|
|
501 |
|
|
502 |
/**
|
|
503 |
* <p>Cleanup and process a Catalog entry.</p>
|
|
504 |
*
|
|
505 |
* <p>This method processes each Catalog entry, changing mapped
|
|
506 |
* relative system identifiers into absolute ones (based on the current
|
|
507 |
* base URI), and maintaining other information about the current
|
|
508 |
* catalog.</p>
|
|
509 |
*
|
|
510 |
* @param entry The CatalogEntry to process.
|
|
511 |
*/
|
|
512 |
private void addEntry(CatalogEntry entry) {
|
|
513 |
switch (entry.entryType()) {
|
|
514 |
case CatalogEntry.BASE: {
|
|
515 |
String value = entry.formalSystemIdentifier();
|
|
516 |
URL newbase = null;
|
|
517 |
|
|
518 |
debug(3, "BASE", value);
|
|
519 |
|
|
520 |
try {
|
|
521 |
value = fixSlashes(value);
|
|
522 |
newbase = new URL(catalogCwd, value);
|
|
523 |
} catch (MalformedURLException e) {
|
|
524 |
try {
|
|
525 |
newbase = new URL("file:///" + value);
|
|
526 |
} catch (MalformedURLException e2) {
|
|
527 |
debug(1, "Malformed URL on base", value);
|
|
528 |
newbase = null;
|
|
529 |
}
|
|
530 |
}
|
|
531 |
|
|
532 |
if (newbase != null) {
|
|
533 |
base = newbase;
|
|
534 |
}
|
|
535 |
|
|
536 |
break;
|
|
537 |
}
|
|
538 |
|
|
539 |
case CatalogEntry.CATALOG: {
|
|
540 |
String fsi = makeAbsolute(entry.formalSystemIdentifier());
|
|
541 |
|
|
542 |
debug(3, "CATALOG", fsi);
|
|
543 |
|
|
544 |
localCatalogFiles.addElement(fsi);
|
|
545 |
break;
|
|
546 |
}
|
|
547 |
|
|
548 |
case CatalogEntry.DOCUMENT: {
|
|
549 |
String fsi = makeAbsolute(entry.formalSystemIdentifier());
|
|
550 |
entry.updateFormalSystemIdentifier(fsi);
|
|
551 |
|
|
552 |
debug(3, "DOCUMENT", fsi);
|
|
553 |
|
|
554 |
catalogEntries.addElement(entry);
|
|
555 |
break;
|
|
556 |
}
|
|
557 |
case CatalogEntry.OVERRIDE: {
|
|
558 |
debug(3, "OVERRIDE", entry.yes_or_no());
|
|
559 |
|
|
560 |
catalogEntries.addElement(entry);
|
|
561 |
break;
|
|
562 |
}
|
|
563 |
case CatalogEntry.SGMLDECL: {
|
|
564 |
// meaningless in XML
|
|
565 |
break;
|
|
566 |
}
|
|
567 |
case CatalogEntry.DELEGATE: {
|
|
568 |
String fsi = makeAbsolute(entry.formalSystemIdentifier());
|
|
569 |
entry.updateFormalSystemIdentifier(fsi);
|
|
570 |
|
|
571 |
debug(3, "DELEGATE", entry.partialPublicId(), fsi);
|
|
572 |
|
|
573 |
addDelegate(entry);
|
|
574 |
break;
|
|
575 |
}
|
|
576 |
case CatalogEntry.DOCTYPE: {
|
|
577 |
String fsi = makeAbsolute(entry.formalSystemIdentifier());
|
|
578 |
entry.updateFormalSystemIdentifier(fsi);
|
|
579 |
|
|
580 |
debug(3, "DOCTYPE", entry.publicId(), fsi);
|
|
581 |
|
|
582 |
catalogEntries.addElement(entry);
|
|
583 |
break;
|
|
584 |
}
|
|
585 |
case CatalogEntry.DTDDECL: {
|
|
586 |
// meaningless in XML
|
|
587 |
break;
|
|
588 |
}
|
|
589 |
case CatalogEntry.ENTITY: {
|
|
590 |
String fsi = makeAbsolute(entry.formalSystemIdentifier());
|
|
591 |
entry.updateFormalSystemIdentifier(fsi);
|
|
592 |
|
|
593 |
debug(3, "ENTITY", entry.entityName(), fsi);
|
|
594 |
|
|
595 |
catalogEntries.addElement(entry);
|
|
596 |
break;
|
|
597 |
}
|
|
598 |
case CatalogEntry.LINKTYPE: {
|
|
599 |
// meaningless in XML
|
|
600 |
break;
|
|
601 |
}
|
|
602 |
case CatalogEntry.NOTATION: {
|
|
603 |
String fsi = makeAbsolute(entry.formalSystemIdentifier());
|
|
604 |
entry.updateFormalSystemIdentifier(fsi);
|
|
605 |
|
|
606 |
debug(3, "NOTATION", entry.entityName(), fsi);
|
|
607 |
|
|
608 |
catalogEntries.addElement(entry);
|
|
609 |
break;
|
|
610 |
}
|
|
611 |
case CatalogEntry.PUBLIC: {
|
|
612 |
// This entry has to go in the vector because it would
|
|
613 |
// be relevant in subsequent searches for notations.
|
|
614 |
String publicid = entry.publicId();
|
|
615 |
String systemid = makeAbsolute(entry.formalSystemIdentifier());
|
|
616 |
|
|
617 |
debug(3, "PUBLIC", publicid, systemid);
|
|
618 |
|
|
619 |
entry.updateFormalSystemIdentifier(systemid);
|
|
620 |
catalogEntries.addElement(entry);
|
|
621 |
break;
|
|
622 |
}
|
|
623 |
case CatalogEntry.SYSTEM: {
|
|
624 |
String systemid = entry.systemId();
|
|
625 |
String fsi = makeAbsolute(entry.formalSystemIdentifier());
|
|
626 |
|
|
627 |
debug(3, "SYSTEM", systemid, fsi);
|
|
628 |
|
|
629 |
entry.updateFormalSystemIdentifier(fsi);
|
|
630 |
catalogEntries.addElement(entry);
|
|
631 |
break;
|
|
632 |
}
|
|
633 |
}
|
|
634 |
}
|
|
635 |
|
|
636 |
/**
|
|
637 |
* <p>Parse all subordinate catalogs.</p>
|
|
638 |
*
|
|
639 |
* <p>This method recursively parses all of the subordinate catalogs.
|
|
640 |
* If this method does not throw an exception, you can be confident that
|
|
641 |
* no subsequent call to any resolve*() method will either, with two
|
|
642 |
* possible exceptions:</p>
|
|
643 |
*
|
|
644 |
* <ol>
|
|
645 |
* <li><p>Delegated catalogs are re-parsed each time they are needed
|
|
646 |
* (because a variable list of them may be needed in each case,
|
|
647 |
* depending on the length of the matching partial public identifier).</p>
|
|
648 |
* <p>But they are parsed by this method, so as long as they don't
|
|
649 |
* change or disappear while the program is running, they shouldn't
|
|
650 |
* generate errors later if they don't generate errors now.</p>
|
|
651 |
* <li><p>If you add new catalogs with <code>parseCatalog</code>, they
|
|
652 |
* won't be loaded until they are needed or until you call
|
|
653 |
* <code>parseAllCatalogs</code> again.</p>
|
|
654 |
* </ol>
|
|
655 |
*
|
|
656 |
* <p>On the other hand, if you don't call this method, you may
|
|
657 |
* successfully parse documents without having to load all possible
|
|
658 |
* catalogs.</p>
|
|
659 |
*
|
|
660 |
* @throws MalformedURLException The filename (URL) for a
|
|
661 |
* subordinate or delegated catalog is not a valid URL.
|
|
662 |
* @throws IOException Error reading some subordinate or delegated
|
|
663 |
* catalog file.
|
|
664 |
*/
|
|
665 |
public void parseAllCatalogs()
|
|
666 |
throws MalformedURLException, IOException {
|
|
667 |
|
|
668 |
// Parse all the subordinate catalogs
|
|
669 |
for (int catPos = 0; catPos < catalogs.size(); catPos++) {
|
|
670 |
Catalog c = null;
|
|
671 |
|
|
672 |
try {
|
|
673 |
c = (Catalog) catalogs.elementAt(catPos);
|
|
674 |
} catch (ClassCastException e) {
|
|
675 |
String catfile = (String) catalogs.elementAt(catPos);
|
|
676 |
c = new Catalog();
|
|
677 |
c.setParserClass(parserClass);
|
|
678 |
c.debug = debug;
|
|
679 |
|
|
680 |
c.parseCatalog(catfile);
|
|
681 |
catalogs.setElementAt(c, catPos);
|
|
682 |
c.parseAllCatalogs();
|
|
683 |
}
|
|
684 |
}
|
|
685 |
|
|
686 |
// Parse all the DELEGATE catalogs
|
|
687 |
Enumeration enum = catalogEntries.elements();
|
|
688 |
while (enum.hasMoreElements()) {
|
|
689 |
CatalogEntry e = (CatalogEntry) enum.nextElement();
|
|
690 |
if (e.entryType() == CatalogEntry.DELEGATE) {
|
|
691 |
Catalog dcat = new Catalog();
|
|
692 |
dcat.setParserClass(parserClass);
|
|
693 |
dcat.debug = debug;
|
|
694 |
dcat.parseCatalog(e.formalSystemIdentifier());
|
|
695 |
}
|
|
696 |
}
|
|
697 |
}
|
|
698 |
|
|
699 |
|
|
700 |
/**
|
|
701 |
* <p>Return the applicable DOCTYPE system identifier.</p>
|
|
702 |
*
|
|
703 |
* @param entityName The name of the entity (element) for which
|
|
704 |
* a doctype is required.
|
|
705 |
* @param publicId The nominal public identifier for the doctype
|
|
706 |
* (as provided in the source document).
|
|
707 |
* @param systemId The nominal system identifier for the doctype
|
|
708 |
* (as provided in the source document).
|
|
709 |
*
|
|
710 |
* @return The system identifier to use for the doctype.
|
|
711 |
*
|
|
712 |
* @throws MalformedURLException The formal system identifier of a
|
|
713 |
* subordinate catalog cannot be turned into a valid URL.
|
|
714 |
* @throws IOException Error reading subordinate catalog file.
|
|
715 |
*/
|
|
716 |
public String resolveDoctype(String entityName,
|
|
717 |
String publicId,
|
|
718 |
String systemId)
|
|
719 |
throws MalformedURLException, IOException {
|
|
720 |
String resolved = null;
|
|
721 |
|
|
722 |
if (systemId != null) {
|
|
723 |
// If there's a SYSTEM entry in this catalog, use it
|
|
724 |
resolved = resolveLocalSystem(systemId);
|
|
725 |
if (resolved != null) {
|
|
726 |
return resolved;
|
|
727 |
}
|
|
728 |
}
|
|
729 |
|
|
730 |
if (publicId != null) {
|
|
731 |
// If there's a PUBLIC entry in this catalog, use it
|
|
732 |
resolved = resolveLocalPublic(CatalogEntry.DOCTYPE,
|
|
733 |
entityName,
|
|
734 |
publicId,
|
|
735 |
systemId);
|
|
736 |
if (resolved != null) {
|
|
737 |
return resolved;
|
|
738 |
}
|
|
739 |
}
|
|
740 |
|
|
741 |
// If there's a DOCTYPE entry in this catalog, use it
|
|
742 |
boolean over = default_override;
|
|
743 |
Enumeration enum = catalogEntries.elements();
|
|
744 |
while (enum.hasMoreElements()) {
|
|
745 |
CatalogEntry e = (CatalogEntry) enum.nextElement();
|
|
746 |
if (e.entryType() == CatalogEntry.OVERRIDE) {
|
|
747 |
over = e.yes_or_no().equalsIgnoreCase("YES");
|
|
748 |
continue;
|
|
749 |
}
|
|
750 |
|
|
751 |
if (e.entryType() == CatalogEntry.DOCTYPE
|
|
752 |
&& e.entityName().equals(entityName)) {
|
|
753 |
if (over || systemId == null) {
|
|
754 |
return e.formalSystemIdentifier();
|
|
755 |
}
|
|
756 |
}
|
|
757 |
}
|
|
758 |
|
|
759 |
// Otherwise, look in the subordinate catalogs
|
|
760 |
return resolveSubordinateCatalogs(CatalogEntry.DOCTYPE,
|
|
761 |
entityName,
|
|
762 |
publicId,
|
|
763 |
systemId);
|
|
764 |
}
|
|
765 |
|
|
766 |
/**
|
|
767 |
* <p>Return the applicable DOCUMENT entry.</p>
|
|
768 |
*
|
|
769 |
* @return The system identifier to use for the doctype.
|
|
770 |
*
|
|
771 |
* @throws MalformedURLException The formal system identifier of a
|
|
772 |
* subordinate catalog cannot be turned into a valid URL.
|
|
773 |
* @throws IOException Error reading subordinate catalog file.
|
|
774 |
*/
|
|
775 |
public String resolveDocument()
|
|
776 |
throws MalformedURLException, IOException {
|
|
777 |
// If there's a DOCUMENT entry, return it
|
|
778 |
Enumeration enum = catalogEntries.elements();
|
|
779 |
while (enum.hasMoreElements()) {
|
|
780 |
CatalogEntry e = (CatalogEntry) enum.nextElement();
|
|
781 |
if (e.entryType() == CatalogEntry.DOCUMENT) {
|
|
782 |
return e.formalSystemIdentifier();
|
|
783 |
}
|
|
784 |
}
|
|
785 |
|
|
786 |
return resolveSubordinateCatalogs(CatalogEntry.DOCUMENT,
|
|
787 |
null, null, null);
|
|
788 |
}
|
|
789 |
|
|
790 |
/**
|
|
791 |
* <p>Return the applicable ENTITY system identifier.</p>
|
|
792 |
*
|
|
793 |
* @param entityName The name of the entity for which
|
|
794 |
* a system identifier is required.
|
|
795 |
* @param publicId The nominal public identifier for the entity
|
|
796 |
* (as provided in the source document).
|
|
797 |
* @param systemId The nominal system identifier for the entity
|
|
798 |
* (as provided in the source document).
|
|
799 |
*
|
|
800 |
* @return The system identifier to use for the entity.
|
|
801 |
*
|
|
802 |
* @throws MalformedURLException The formal system identifier of a
|
|
803 |
* subordinate catalog cannot be turned into a valid URL.
|
|
804 |
* @throws IOException Error reading subordinate catalog file.
|
|
805 |
*/
|
|
806 |
public String resolveEntity(String entityName,
|
|
807 |
String publicId,
|
|
808 |
String systemId)
|
|
809 |
throws MalformedURLException, IOException {
|
|
810 |
String resolved = null;
|
|
811 |
|
|
812 |
if (systemId != null) {
|
|
813 |
// If there's a SYSTEM entry in this catalog, use it
|
|
814 |
resolved = resolveLocalSystem(systemId);
|
|
815 |
if (resolved != null) {
|
|
816 |
return resolved;
|
|
817 |
}
|
|
818 |
}
|
|
819 |
|
|
820 |
if (publicId != null) {
|
|
821 |
// If there's a PUBLIC entry in this catalog, use it
|
|
822 |
resolved = resolveLocalPublic(CatalogEntry.ENTITY,
|
|
823 |
entityName,
|
|
824 |
publicId,
|
|
825 |
systemId);
|
|
826 |
if (resolved != null) {
|
|
827 |
return resolved;
|
|
828 |
}
|
|
829 |
}
|
|
830 |
|
|
831 |
// If there's a ENTITY entry in this catalog, use it
|
|
832 |
boolean over = default_override;
|
|
833 |
Enumeration enum = catalogEntries.elements();
|
|
834 |
while (enum.hasMoreElements()) {
|
|
835 |
CatalogEntry e = (CatalogEntry) enum.nextElement();
|
|
836 |
if (e.entryType() == CatalogEntry.OVERRIDE) {
|
|
837 |
over = e.yes_or_no().equalsIgnoreCase("YES");
|
|
838 |
continue;
|
|
839 |
}
|
|
840 |
|
|
841 |
if (e.entryType() == CatalogEntry.ENTITY
|
|
842 |
&& e.entityName().equals(entityName)) {
|
|
843 |
if (over || systemId == null) {
|
|
844 |
return e.formalSystemIdentifier();
|
|
845 |
}
|
|
846 |
}
|
|
847 |
}
|
|
848 |
|
|
849 |
// Otherwise, look in the subordinate catalogs
|
|
850 |
return resolveSubordinateCatalogs(CatalogEntry.ENTITY,
|
|
851 |
entityName,
|
|
852 |
publicId,
|
|
853 |
systemId);
|
|
854 |
}
|
|
855 |
|
|
856 |
/**
|
|
857 |
* <p>Return the applicable NOTATION system identifier.</p>
|
|
858 |
*
|
|
859 |
* @param notationName The name of the notation for which
|
|
860 |
* a doctype is required.
|
|
861 |
* @param publicId The nominal public identifier for the notation
|
|
862 |
* (as provided in the source document).
|
|
863 |
* @param systemId The nominal system identifier for the notation
|
|
864 |
* (as provided in the source document).
|
|
865 |
*
|
|
866 |
* @return The system identifier to use for the notation.
|
|
867 |
*
|
|
868 |
* @throws MalformedURLException The formal system identifier of a
|
|
869 |
* subordinate catalog cannot be turned into a valid URL.
|
|
870 |
* @throws IOException Error reading subordinate catalog file.
|
|
871 |
*/
|
|
872 |
public String resolveNotation(String notationName,
|
|
873 |
String publicId,
|
|
874 |
String systemId)
|
|
875 |
throws MalformedURLException, IOException {
|
|
876 |
String resolved = null;
|
|
877 |
|
|
878 |
if (systemId != null) {
|
|
879 |
// If there's a SYSTEM entry in this catalog, use it
|
|
880 |
resolved = resolveLocalSystem(systemId);
|
|
881 |
if (resolved != null) {
|
|
882 |
return resolved;
|
|
883 |
}
|
|
884 |
}
|
|
885 |
|
|
886 |
if (publicId != null) {
|
|
887 |
// If there's a PUBLIC entry in this catalog, use it
|
|
888 |
resolved = resolveLocalPublic(CatalogEntry.NOTATION,
|
|
889 |
notationName,
|
|
890 |
publicId,
|
|
891 |
systemId);
|
|
892 |
if (resolved != null) {
|
|
893 |
return resolved;
|
|
894 |
}
|
|
895 |
}
|
|
896 |
|
|
897 |
// If there's a NOTATION entry in this catalog, use it
|
|
898 |
boolean over = default_override;
|
|
899 |
Enumeration enum = catalogEntries.elements();
|
|
900 |
while (enum.hasMoreElements()) {
|
|
901 |
CatalogEntry e = (CatalogEntry) enum.nextElement();
|
|
902 |
if (e.entryType() == CatalogEntry.OVERRIDE) {
|
|
903 |
over = e.yes_or_no().equalsIgnoreCase("YES");
|
|
904 |
continue;
|
|
905 |
}
|
|
906 |
|
|
907 |
if (e.entryType() == CatalogEntry.NOTATION
|
|
908 |
&& e.entityName().equals(notationName)) {
|
|
909 |
if (over || systemId == null) {
|
|
910 |
return e.formalSystemIdentifier();
|
|
911 |
}
|
|
912 |
}
|
|
913 |
}
|
|
914 |
|
|
915 |
// Otherwise, look in the subordinate catalogs
|
|
916 |
return resolveSubordinateCatalogs(CatalogEntry.NOTATION,
|
|
917 |
notationName,
|
|
918 |
publicId,
|
|
919 |
systemId);
|
|
920 |
}
|
|
921 |
|
|
922 |
/**
|
|
923 |
* <p>Return the applicable PUBLIC or SYSTEM identifier.</p>
|
|
924 |
*
|
|
925 |
* <p>This method searches the Catalog and returns the system
|
|
926 |
* identifier specified for the given system or
|
|
927 |
* public identifiers. If
|
|
928 |
* no appropriate PUBLIC or SYSTEM entry is found in the Catalog,
|
|
929 |
* null is returned.</p>
|
|
930 |
*
|
|
931 |
* @param publicId The public identifier to locate in the catalog.
|
|
932 |
* Public identifiers are normalized before comparison.
|
|
933 |
* @param systemId The nominal system identifier for the entity
|
|
934 |
* in question (as provided in the source document).
|
|
935 |
*
|
|
936 |
* @throws MalformedURLException The formal system identifier of a
|
|
937 |
* subordinate catalog cannot be turned into a valid URL.
|
|
938 |
* @throws IOException Error reading subordinate catalog file.
|
|
939 |
*
|
|
940 |
* @return The system identifier to use.
|
|
941 |
* Note that the nominal system identifier is not returned if a
|
|
942 |
* match is not found in the catalog, instead null is returned
|
|
943 |
* to indicate that no match was found.
|
|
944 |
*/
|
|
945 |
public String resolvePublic(String publicId, String systemId)
|
|
946 |
throws MalformedURLException, IOException {
|
|
947 |
|
|
948 |
// If there's a SYSTEM entry in this catalog, use it
|
|
949 |
if (systemId != null) {
|
|
950 |
String resolved = resolveLocalSystem(systemId);
|
|
951 |
if (resolved != null) {
|
|
952 |
return resolved;
|
|
953 |
}
|
|
954 |
}
|
|
955 |
|
|
956 |
// If there's a PUBLIC entry in this catalog, use it
|
|
957 |
String resolved = resolveLocalPublic(CatalogEntry.PUBLIC,
|
|
958 |
null,
|
|
959 |
publicId,
|
|
960 |
systemId);
|
|
961 |
if (resolved != null) {
|
|
962 |
return resolved;
|
|
963 |
}
|
|
964 |
|
|
965 |
// Otherwise, look in the subordinate catalogs
|
|
966 |
return resolveSubordinateCatalogs(CatalogEntry.PUBLIC,
|
|
967 |
null,
|
|
968 |
publicId,
|
|
969 |
systemId);
|
|
970 |
}
|
|
971 |
|
|
972 |
/**
|
|
973 |
* <p>Return the applicable PUBLIC or SYSTEM identifier</p>
|
|
974 |
*
|
|
975 |
* <p>This method searches the Catalog and returns the system
|
|
976 |
* identifier specified for the given system or public identifiers.
|
|
977 |
* If no appropriate PUBLIC or SYSTEM entry is found in the Catalog,
|
|
978 |
* delegated Catalogs are interrogated.</p>
|
|
979 |
*
|
|
980 |
* <p>There are four possible cases:</p>
|
|
981 |
*
|
|
982 |
* <ul>
|
|
983 |
* <li>If the system identifier provided matches a SYSTEM entry
|
|
984 |
* in the current catalog, the SYSTEM entry is returned.
|
|
985 |
* <li>If the system identifier is not null, the PUBLIC entries
|
|
986 |
* that were encountered when OVERRIDE YES was in effect are
|
|
987 |
* interrogated and the first matching entry is returned.</li>
|
|
988 |
* <li>If the system identifier is null, then all of the PUBLIC
|
|
989 |
* entries are interrogated and the first matching entry
|
|
990 |
* is returned. This may not be the same as the preceding case, if
|
|
991 |
* some PUBLIC entries are encountered when OVERRIDE NO is in effect. In
|
|
992 |
* XML, the only place where a public identifier may occur without
|
|
993 |
* a system identifier is in a notation declaration.</li>
|
|
994 |
* <li>Finally, if the public identifier matches one of the partial
|
|
995 |
* public identifiers specified in a DELEGATE entry in
|
|
996 |
* the Catalog, the delegated catalog is interrogated. The first
|
|
997 |
* time that the delegated catalog is required, it will be
|
|
998 |
* retrieved and parsed. It is subsequently cached.
|
Catalog files from arbortext added