| 1 |
769 |
jeremybenn |
/* XmlReader.java --
|
| 2 |
|
|
Copyright (C) 1999,2000,2001 Free Software Foundation, Inc.
|
| 3 |
|
|
|
| 4 |
|
|
This file is part of GNU Classpath.
|
| 5 |
|
|
|
| 6 |
|
|
GNU Classpath is free software; you can redistribute it and/or modify
|
| 7 |
|
|
it under the terms of the GNU General Public License as published by
|
| 8 |
|
|
the Free Software Foundation; either version 2, or (at your option)
|
| 9 |
|
|
any later version.
|
| 10 |
|
|
|
| 11 |
|
|
GNU Classpath is distributed in the hope that it will be useful, but
|
| 12 |
|
|
WITHOUT ANY WARRANTY; without even the implied warranty of
|
| 13 |
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
| 14 |
|
|
General Public License for more details.
|
| 15 |
|
|
|
| 16 |
|
|
You should have received a copy of the GNU General Public License
|
| 17 |
|
|
along with GNU Classpath; see the file COPYING. If not, write to the
|
| 18 |
|
|
Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
|
| 19 |
|
|
02110-1301 USA.
|
| 20 |
|
|
|
| 21 |
|
|
Linking this library statically or dynamically with other modules is
|
| 22 |
|
|
making a combined work based on this library. Thus, the terms and
|
| 23 |
|
|
conditions of the GNU General Public License cover the whole
|
| 24 |
|
|
combination.
|
| 25 |
|
|
|
| 26 |
|
|
As a special exception, the copyright holders of this library give you
|
| 27 |
|
|
permission to link this library with independent modules to produce an
|
| 28 |
|
|
executable, regardless of the license terms of these independent
|
| 29 |
|
|
modules, and to copy and distribute the resulting executable under
|
| 30 |
|
|
terms of your choice, provided that you also meet, for each linked
|
| 31 |
|
|
independent module, the terms and conditions of the license of that
|
| 32 |
|
|
module. An independent module is a module which is not derived from
|
| 33 |
|
|
or based on this library. If you modify this library, you may extend
|
| 34 |
|
|
this exception to your version of the library, but you are not
|
| 35 |
|
|
obligated to do so. If you do not wish to do so, delete this
|
| 36 |
|
|
exception statement from your version. */
|
| 37 |
|
|
|
| 38 |
|
|
package gnu.xml.aelfred2;
|
| 39 |
|
|
|
| 40 |
|
|
import java.io.IOException;
|
| 41 |
|
|
import java.util.Locale;
|
| 42 |
|
|
|
| 43 |
|
|
import org.xml.sax.*;
|
| 44 |
|
|
import org.xml.sax.ext.*;
|
| 45 |
|
|
|
| 46 |
|
|
import gnu.xml.pipeline.EventFilter;
|
| 47 |
|
|
import gnu.xml.pipeline.ValidationConsumer;
|
| 48 |
|
|
|
| 49 |
|
|
|
| 50 |
|
|
/**
|
| 51 |
|
|
* This SAX2 parser optionally layers a validator over the Ælfred2
|
| 52 |
|
|
* SAX2 parser. While this will not evaluate every XML validity constraint,
|
| 53 |
|
|
* it does support all the validity constraints that are of any real utility
|
| 54 |
|
|
* outside the strict SGML-compatible world. See the documentation for the
|
| 55 |
|
|
* SAXDriver class for information about the SAX2 features and properties
|
| 56 |
|
|
* that are supported, and documentation for the ValidationConsumer for
|
| 57 |
|
|
* information about what validity constraints may not be supported.
|
| 58 |
|
|
* (Ælfred2 tests some of those, even in non-validating mode, to
|
| 59 |
|
|
* achieve better conformance.)
|
| 60 |
|
|
*
|
| 61 |
|
|
* <p> Note that due to its internal construction, you can't change most
|
| 62 |
|
|
* handlers until parse() returns. This diverges slightly from SAX, which
|
| 63 |
|
|
* expects later binding to be supported. Early binding involves less
|
| 64 |
|
|
* runtime overhead, which is an issue for event pipelines as used inside
|
| 65 |
|
|
* this parser. Rather than relying on the parser to handle late binding
|
| 66 |
|
|
* to your own handlers, do it yourself.
|
| 67 |
|
|
*
|
| 68 |
|
|
* @see SAXDriver
|
| 69 |
|
|
* @see gnu.xml.pipeline.ValidationConsumer
|
| 70 |
|
|
*
|
| 71 |
|
|
* @author David Brownell
|
| 72 |
|
|
*/
|
| 73 |
|
|
public final class XmlReader
|
| 74 |
|
|
implements XMLReader
|
| 75 |
|
|
{
|
| 76 |
|
|
|
| 77 |
|
|
static class FatalErrorHandler
|
| 78 |
|
|
extends DefaultHandler2
|
| 79 |
|
|
{
|
| 80 |
|
|
|
| 81 |
|
|
public void error(SAXParseException e)
|
| 82 |
|
|
throws SAXException
|
| 83 |
|
|
{
|
| 84 |
|
|
throw e;
|
| 85 |
|
|
}
|
| 86 |
|
|
|
| 87 |
|
|
}
|
| 88 |
|
|
|
| 89 |
|
|
private SAXDriver aelfred2 = new SAXDriver();
|
| 90 |
|
|
private EventFilter filter = new EventFilter();
|
| 91 |
|
|
private boolean isValidating;
|
| 92 |
|
|
private boolean active;
|
| 93 |
|
|
|
| 94 |
|
|
/**
|
| 95 |
|
|
* Constructs a SAX Parser.
|
| 96 |
|
|
*/
|
| 97 |
|
|
public XmlReader()
|
| 98 |
|
|
{
|
| 99 |
|
|
}
|
| 100 |
|
|
|
| 101 |
|
|
/**
|
| 102 |
|
|
* Constructs a SAX Parser, optionally treating validity errors
|
| 103 |
|
|
* as if they were fatal errors.
|
| 104 |
|
|
*/
|
| 105 |
|
|
public XmlReader(boolean invalidIsFatal)
|
| 106 |
|
|
{
|
| 107 |
|
|
if (invalidIsFatal)
|
| 108 |
|
|
{
|
| 109 |
|
|
setErrorHandler(new FatalErrorHandler());
|
| 110 |
|
|
}
|
| 111 |
|
|
}
|
| 112 |
|
|
|
| 113 |
|
|
/**
|
| 114 |
|
|
* <b>SAX2</b>: Returns the object used to report the logical
|
| 115 |
|
|
* content of an XML document.
|
| 116 |
|
|
*/
|
| 117 |
|
|
public ContentHandler getContentHandler()
|
| 118 |
|
|
{
|
| 119 |
|
|
return filter.getContentHandler();
|
| 120 |
|
|
}
|
| 121 |
|
|
|
| 122 |
|
|
/**
|
| 123 |
|
|
* <b>SAX2</b>: Assigns the object used to report the logical
|
| 124 |
|
|
* content of an XML document.
|
| 125 |
|
|
* @exception IllegalStateException if called mid-parse
|
| 126 |
|
|
*/
|
| 127 |
|
|
public void setContentHandler(ContentHandler handler)
|
| 128 |
|
|
{
|
| 129 |
|
|
if (active)
|
| 130 |
|
|
{
|
| 131 |
|
|
throw new IllegalStateException("already parsing");
|
| 132 |
|
|
}
|
| 133 |
|
|
filter.setContentHandler(handler);
|
| 134 |
|
|
}
|
| 135 |
|
|
|
| 136 |
|
|
/**
|
| 137 |
|
|
* <b>SAX2</b>: Returns the object used to process declarations related
|
| 138 |
|
|
* to notations and unparsed entities.
|
| 139 |
|
|
*/
|
| 140 |
|
|
public DTDHandler getDTDHandler()
|
| 141 |
|
|
{
|
| 142 |
|
|
return filter.getDTDHandler();
|
| 143 |
|
|
}
|
| 144 |
|
|
|
| 145 |
|
|
/**
|
| 146 |
|
|
* <b>SAX1</b> Assigns DTD handler
|
| 147 |
|
|
* @exception IllegalStateException if called mid-parse
|
| 148 |
|
|
*/
|
| 149 |
|
|
public void setDTDHandler(DTDHandler handler)
|
| 150 |
|
|
{
|
| 151 |
|
|
if (active)
|
| 152 |
|
|
{
|
| 153 |
|
|
throw new IllegalStateException("already parsing");
|
| 154 |
|
|
}
|
| 155 |
|
|
filter.setDTDHandler(handler);
|
| 156 |
|
|
}
|
| 157 |
|
|
|
| 158 |
|
|
/**
|
| 159 |
|
|
* <b>SAX2</b>: Returns the object used when resolving external
|
| 160 |
|
|
* entities during parsing (both general and parameter entities).
|
| 161 |
|
|
*/
|
| 162 |
|
|
public EntityResolver getEntityResolver()
|
| 163 |
|
|
{
|
| 164 |
|
|
return aelfred2.getEntityResolver();
|
| 165 |
|
|
}
|
| 166 |
|
|
|
| 167 |
|
|
/**
|
| 168 |
|
|
* <b>SAX1</b> Assigns parser's entity resolver
|
| 169 |
|
|
*/
|
| 170 |
|
|
public void setEntityResolver(EntityResolver handler)
|
| 171 |
|
|
{
|
| 172 |
|
|
aelfred2.setEntityResolver(handler);
|
| 173 |
|
|
}
|
| 174 |
|
|
|
| 175 |
|
|
/**
|
| 176 |
|
|
* <b>SAX2</b>: Returns the object used to receive callbacks for XML
|
| 177 |
|
|
* errors of all levels (fatal, nonfatal, warning); this is never null;
|
| 178 |
|
|
*/
|
| 179 |
|
|
public ErrorHandler getErrorHandler()
|
| 180 |
|
|
{
|
| 181 |
|
|
return aelfred2.getErrorHandler();
|
| 182 |
|
|
}
|
| 183 |
|
|
|
| 184 |
|
|
/**
|
| 185 |
|
|
* <b>SAX1</b> Assigns error handler
|
| 186 |
|
|
* @exception IllegalStateException if called mid-parse
|
| 187 |
|
|
*/
|
| 188 |
|
|
public void setErrorHandler(ErrorHandler handler)
|
| 189 |
|
|
{
|
| 190 |
|
|
if (active)
|
| 191 |
|
|
{
|
| 192 |
|
|
throw new IllegalStateException("already parsing");
|
| 193 |
|
|
}
|
| 194 |
|
|
aelfred2.setErrorHandler(handler);
|
| 195 |
|
|
}
|
| 196 |
|
|
|
| 197 |
|
|
/**
|
| 198 |
|
|
* <b>SAX2</b>: Assigns the specified property.
|
| 199 |
|
|
* @exception IllegalStateException if called mid-parse
|
| 200 |
|
|
*/
|
| 201 |
|
|
public void setProperty(String propertyId, Object value)
|
| 202 |
|
|
throws SAXNotRecognizedException, SAXNotSupportedException
|
| 203 |
|
|
{
|
| 204 |
|
|
if (active)
|
| 205 |
|
|
{
|
| 206 |
|
|
throw new IllegalStateException("already parsing");
|
| 207 |
|
|
}
|
| 208 |
|
|
if (getProperty(propertyId) != value)
|
| 209 |
|
|
{
|
| 210 |
|
|
filter.setProperty(propertyId, value);
|
| 211 |
|
|
}
|
| 212 |
|
|
}
|
| 213 |
|
|
|
| 214 |
|
|
/**
|
| 215 |
|
|
* <b>SAX2</b>: Returns the specified property.
|
| 216 |
|
|
*/
|
| 217 |
|
|
public Object getProperty(String propertyId)
|
| 218 |
|
|
throws SAXNotRecognizedException
|
| 219 |
|
|
{
|
| 220 |
|
|
if ((SAXDriver.PROPERTY + "declaration-handler").equals(propertyId)
|
| 221 |
|
|
|| (SAXDriver.PROPERTY + "lexical-handler").equals(propertyId))
|
| 222 |
|
|
{
|
| 223 |
|
|
return filter.getProperty(propertyId);
|
| 224 |
|
|
}
|
| 225 |
|
|
throw new SAXNotRecognizedException(propertyId);
|
| 226 |
|
|
}
|
| 227 |
|
|
|
| 228 |
|
|
private void forceValidating()
|
| 229 |
|
|
throws SAXNotRecognizedException, SAXNotSupportedException
|
| 230 |
|
|
{
|
| 231 |
|
|
aelfred2.setFeature(SAXDriver.FEATURE + "namespace-prefixes",
|
| 232 |
|
|
true);
|
| 233 |
|
|
aelfred2.setFeature(SAXDriver.FEATURE + "external-general-entities",
|
| 234 |
|
|
true);
|
| 235 |
|
|
aelfred2.setFeature(SAXDriver.FEATURE + "external-parameter-entities",
|
| 236 |
|
|
true);
|
| 237 |
|
|
}
|
| 238 |
|
|
|
| 239 |
|
|
/**
|
| 240 |
|
|
* <b>SAX2</b>: Sets the state of features supported in this parser.
|
| 241 |
|
|
* Note that this parser requires reporting of namespace prefixes when
|
| 242 |
|
|
* validating.
|
| 243 |
|
|
*/
|
| 244 |
|
|
public void setFeature(String featureId, boolean state)
|
| 245 |
|
|
throws SAXNotRecognizedException, SAXNotSupportedException
|
| 246 |
|
|
{
|
| 247 |
|
|
boolean value = getFeature(featureId);
|
| 248 |
|
|
|
| 249 |
|
|
if (state == value)
|
| 250 |
|
|
{
|
| 251 |
|
|
return;
|
| 252 |
|
|
}
|
| 253 |
|
|
|
| 254 |
|
|
if ((SAXDriver.FEATURE + "validation").equals(featureId))
|
| 255 |
|
|
{
|
| 256 |
|
|
if (active)
|
| 257 |
|
|
{
|
| 258 |
|
|
throw new SAXNotSupportedException("already parsing");
|
| 259 |
|
|
}
|
| 260 |
|
|
if (state)
|
| 261 |
|
|
{
|
| 262 |
|
|
forceValidating();
|
| 263 |
|
|
}
|
| 264 |
|
|
isValidating = state;
|
| 265 |
|
|
}
|
| 266 |
|
|
else
|
| 267 |
|
|
{
|
| 268 |
|
|
aelfred2.setFeature(featureId, state);
|
| 269 |
|
|
}
|
| 270 |
|
|
}
|
| 271 |
|
|
|
| 272 |
|
|
/**
|
| 273 |
|
|
* <b>SAX2</b>: Tells whether this parser supports the specified feature.
|
| 274 |
|
|
* At this time, this directly parallels the underlying SAXDriver,
|
| 275 |
|
|
* except that validation is optionally supported.
|
| 276 |
|
|
*
|
| 277 |
|
|
* @see SAXDriver
|
| 278 |
|
|
*/
|
| 279 |
|
|
public boolean getFeature(String featureId)
|
| 280 |
|
|
throws SAXNotRecognizedException, SAXNotSupportedException
|
| 281 |
|
|
{
|
| 282 |
|
|
if ((SAXDriver.FEATURE + "validation").equals(featureId))
|
| 283 |
|
|
{
|
| 284 |
|
|
return isValidating;
|
| 285 |
|
|
}
|
| 286 |
|
|
|
| 287 |
|
|
return aelfred2.getFeature(featureId);
|
| 288 |
|
|
}
|
| 289 |
|
|
|
| 290 |
|
|
/**
|
| 291 |
|
|
* <b>SAX1</b>: Sets the locale used for diagnostics; currently,
|
| 292 |
|
|
* only locales using the English language are supported.
|
| 293 |
|
|
* @param locale The locale for which diagnostics will be generated
|
| 294 |
|
|
*/
|
| 295 |
|
|
public void setLocale(Locale locale)
|
| 296 |
|
|
throws SAXException
|
| 297 |
|
|
{
|
| 298 |
|
|
aelfred2.setLocale(locale);
|
| 299 |
|
|
}
|
| 300 |
|
|
|
| 301 |
|
|
/**
|
| 302 |
|
|
* <b>SAX1</b>: Preferred API to parse an XML document, using a
|
| 303 |
|
|
* system identifier (URI).
|
| 304 |
|
|
*/
|
| 305 |
|
|
public void parse(String systemId)
|
| 306 |
|
|
throws SAXException, IOException
|
| 307 |
|
|
{
|
| 308 |
|
|
parse(new InputSource(systemId));
|
| 309 |
|
|
}
|
| 310 |
|
|
|
| 311 |
|
|
/**
|
| 312 |
|
|
* <b>SAX1</b>: Underlying API to parse an XML document, used
|
| 313 |
|
|
* directly when no URI is available. When this is invoked,
|
| 314 |
|
|
* and the parser is set to validate, some features will be
|
| 315 |
|
|
* automatically reset to appropriate values: for reporting
|
| 316 |
|
|
* namespace prefixes, and incorporating external entities.
|
| 317 |
|
|
*
|
| 318 |
|
|
* @param source The XML input source.
|
| 319 |
|
|
*
|
| 320 |
|
|
* @exception IllegalStateException if called mid-parse
|
| 321 |
|
|
* @exception SAXException The handlers may throw any SAXException,
|
| 322 |
|
|
* and the parser normally throws SAXParseException objects.
|
| 323 |
|
|
* @exception IOException IOExceptions are normally through through
|
| 324 |
|
|
* the parser if there are problems reading the source document.
|
| 325 |
|
|
*/
|
| 326 |
|
|
public void parse(InputSource source)
|
| 327 |
|
|
throws SAXException, IOException
|
| 328 |
|
|
{
|
| 329 |
|
|
EventFilter next;
|
| 330 |
|
|
boolean nsdecls;
|
| 331 |
|
|
|
| 332 |
|
|
synchronized (aelfred2)
|
| 333 |
|
|
{
|
| 334 |
|
|
if (active)
|
| 335 |
|
|
{
|
| 336 |
|
|
throw new IllegalStateException("already parsing");
|
| 337 |
|
|
}
|
| 338 |
|
|
active = true;
|
| 339 |
|
|
}
|
| 340 |
|
|
|
| 341 |
|
|
// set up the output pipeline
|
| 342 |
|
|
if (isValidating)
|
| 343 |
|
|
{
|
| 344 |
|
|
forceValidating();
|
| 345 |
|
|
next = new ValidationConsumer(filter);
|
| 346 |
|
|
}
|
| 347 |
|
|
else
|
| 348 |
|
|
{
|
| 349 |
|
|
next = filter;
|
| 350 |
|
|
}
|
| 351 |
|
|
|
| 352 |
|
|
// connect pipeline and error handler
|
| 353 |
|
|
// don't let _this_ call to bind() affect xmlns* attributes
|
| 354 |
|
|
nsdecls = aelfred2.getFeature(SAXDriver.FEATURE + "namespace-prefixes");
|
| 355 |
|
|
EventFilter.bind(aelfred2, next);
|
| 356 |
|
|
if (!nsdecls)
|
| 357 |
|
|
{
|
| 358 |
|
|
aelfred2.setFeature(SAXDriver.FEATURE + "namespace-prefixes",
|
| 359 |
|
|
false);
|
| 360 |
|
|
}
|
| 361 |
|
|
|
| 362 |
|
|
// parse, clean up
|
| 363 |
|
|
try
|
| 364 |
|
|
{
|
| 365 |
|
|
aelfred2.parse(source);
|
| 366 |
|
|
}
|
| 367 |
|
|
finally
|
| 368 |
|
|
{
|
| 369 |
|
|
active = false;
|
| 370 |
|
|
}
|
| 371 |
|
|
}
|
| 372 |
|
|
|
| 373 |
|
|
}
|