0
|
1 /*--
|
|
2
|
|
3 $Id: XMLOutputter.java,v 1.117 2009/07/23 05:54:23 jhunter Exp $
|
|
4
|
|
5 Copyright (C) 2000-2007 Jason Hunter & Brett McLaughlin.
|
|
6 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 disclaimer that follows
|
|
17 these conditions in the documentation and/or other materials
|
|
18 provided with the distribution.
|
|
19
|
|
20 3. The name "JDOM" must not be used to endorse or promote products
|
|
21 derived from this software without prior written permission. For
|
|
22 written permission, please contact <request_AT_jdom_DOT_org>.
|
|
23
|
|
24 4. Products derived from this software may not be called "JDOM", nor
|
|
25 may "JDOM" appear in their name, without prior written permission
|
|
26 from the JDOM Project Management <request_AT_jdom_DOT_org>.
|
|
27
|
|
28 In addition, we request (but do not require) that you include in the
|
|
29 end-user documentation provided with the redistribution and/or in the
|
|
30 software itself an acknowledgement equivalent to the following:
|
|
31 "This product includes software developed by the
|
|
32 JDOM Project (http://www.jdom.org/)."
|
|
33 Alternatively, the acknowledgment may be graphical using the logos
|
|
34 available at http://www.jdom.org/images/logos.
|
|
35
|
|
36 THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
|
|
37 WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
|
|
38 OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
|
|
39 DISCLAIMED. IN NO EVENT SHALL THE JDOM AUTHORS OR THE PROJECT
|
|
40 CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
|
41 SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
|
42 LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
|
|
43 USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
|
|
44 ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
|
|
45 OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
|
|
46 OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
47 SUCH DAMAGE.
|
|
48
|
|
49 This software consists of voluntary contributions made by many
|
|
50 individuals on behalf of the JDOM Project and was originally
|
|
51 created by Jason Hunter <jhunter_AT_jdom_DOT_org> and
|
|
52 Brett McLaughlin <brett_AT_jdom_DOT_org>. For more information
|
|
53 on the JDOM Project, please see <http://www.jdom.org/>.
|
|
54
|
|
55 */
|
|
56
|
|
57 package org.jdom.output;
|
|
58
|
|
59 import java.io.*;
|
|
60 import java.util.*;
|
|
61
|
|
62 import javax.xml.transform.Result;
|
|
63
|
|
64 import org.jdom.*;
|
|
65
|
|
66 /**
|
|
67 * Outputs a JDOM document as a stream of bytes. The outputter can manage many
|
|
68 * styles of document formatting, from untouched to pretty printed. The default
|
|
69 * is to output the document content exactly as created, but this can be changed
|
|
70 * by setting a new Format object. For pretty-print output, use
|
|
71 * <code>{@link Format#getPrettyFormat()}</code>. For whitespace-normalized
|
|
72 * output, use <code>{@link Format#getCompactFormat()}</code>.
|
|
73 * <p>
|
|
74 * There are <code>{@link #output output(...)}</code> methods to print any of
|
|
75 * the standard JDOM classes, including Document and Element, to either a Writer
|
|
76 * or an OutputStream. <b>Warning</b>: When outputting to a Writer, make sure
|
|
77 * the writer's encoding matches the encoding setting in the Format object. This
|
|
78 * ensures the encoding in which the content is written (controlled by the
|
|
79 * Writer configuration) matches the encoding placed in the document's XML
|
|
80 * declaration (controlled by the XMLOutputter). Because a Writer cannot be
|
|
81 * queried for its encoding, the information must be passed to the Format
|
|
82 * manually in its constructor or via the
|
|
83 * <code>{@link Format#setEncoding}</code> method. The default encoding is
|
|
84 * UTF-8.
|
|
85 * <p>
|
|
86 * The methods <code>{@link #outputString outputString(...)}</code> are for
|
|
87 * convenience only; for top performance you should call one of the <code>{@link
|
|
88 * #output output(...)}</code> methods and pass in your own Writer or
|
|
89 * OutputStream if possible.
|
|
90 * <p>
|
|
91 * XML declarations are always printed on their own line followed by a line
|
|
92 * seperator (this doesn't change the semantics of the document). To omit
|
|
93 * printing of the declaration use
|
|
94 * <code>{@link Format#setOmitDeclaration}</code>. To omit printing of the
|
|
95 * encoding in the declaration use <code>{@link Format#setOmitEncoding}</code>.
|
|
96 * Unfortunatly there is currently no way to know the original encoding of the
|
|
97 * document.
|
|
98 * <p>
|
|
99 * Empty elements are by default printed as <empty/>, but this can be
|
|
100 * configured with <code>{@link Format#setExpandEmptyElements}</code> to cause
|
|
101 * them to be expanded to <empty></empty>.
|
|
102 *
|
|
103 * @version $Revision: 1.117 $, $Date: 2009/07/23 05:54:23 $
|
|
104 * @author Brett McLaughlin
|
|
105 * @author Jason Hunter
|
|
106 * @author Jason Reid
|
|
107 * @author Wolfgang Werner
|
|
108 * @author Elliotte Rusty Harold
|
|
109 * @author David & Will (from Post Tool Design)
|
|
110 * @author Dan Schaffer
|
|
111 * @author Alex Chaffee
|
|
112 * @author Bradley S. Huffman
|
|
113 */
|
|
114
|
|
115 public class XMLOutputter implements Cloneable {
|
|
116
|
|
117 private static final String CVS_ID =
|
|
118 "@(#) $RCSfile: XMLOutputter.java,v $ $Revision: 1.117 $ $Date: 2009/07/23 05:54:23 $ $Name: jdom_1_1_1 $";
|
|
119
|
|
120 // For normal output
|
|
121 private Format userFormat = Format.getRawFormat();
|
|
122
|
|
123 // For xml:space="preserve"
|
|
124 protected static final Format preserveFormat = Format.getRawFormat();
|
|
125
|
|
126 // What's currently in use
|
|
127 protected Format currentFormat = userFormat;
|
|
128
|
|
129 /** Whether output escaping is enabled for the being processed
|
|
130 * Element - default is <code>true</code> */
|
|
131 private boolean escapeOutput = true;
|
|
132
|
|
133 // * * * * * * * * * * Constructors * * * * * * * * * *
|
|
134 // * * * * * * * * * * Constructors * * * * * * * * * *
|
|
135
|
|
136 /**
|
|
137 * This will create an <code>XMLOutputter</code> with the default
|
|
138 * {@link Format} matching {@link Format#getRawFormat}.
|
|
139 */
|
|
140 public XMLOutputter() {
|
|
141 }
|
|
142
|
|
143 /**
|
|
144 * This will create an <code>XMLOutputter</code> with the specified
|
|
145 * format characteristics. Note the format object is cloned internally
|
|
146 * before use.
|
|
147 */
|
|
148 public XMLOutputter(Format format) {
|
|
149 userFormat = (Format) format.clone();
|
|
150 currentFormat = userFormat;
|
|
151 }
|
|
152
|
|
153 /**
|
|
154 * This will create an <code>XMLOutputter</code> with all the
|
|
155 * options as set in the given <code>XMLOutputter</code>. Note
|
|
156 * that <code>XMLOutputter two = (XMLOutputter)one.clone();</code>
|
|
157 * would work equally well.
|
|
158 *
|
|
159 * @param that the XMLOutputter to clone
|
|
160 */
|
|
161 public XMLOutputter(XMLOutputter that) {
|
|
162 this.userFormat = (Format) that.userFormat.clone();
|
|
163 currentFormat = userFormat;
|
|
164 }
|
|
165
|
|
166 // * * * * * * * * * * Set parameters methods * * * * * * * * * *
|
|
167 // * * * * * * * * * * Set parameters methods * * * * * * * * * *
|
|
168
|
|
169 /**
|
|
170 * Sets the new format logic for the outputter. Note the Format
|
|
171 * object is cloned internally before use.
|
|
172 *
|
|
173 * @param newFormat the format to use for output
|
|
174 */
|
|
175 public void setFormat(Format newFormat) {
|
|
176 this.userFormat = (Format) newFormat.clone();
|
|
177 this.currentFormat = userFormat;
|
|
178 }
|
|
179
|
|
180 /**
|
|
181 * Returns the current format in use by the outputter. Note the
|
|
182 * Format object returned is a clone of the one used internally.
|
|
183 */
|
|
184 public Format getFormat() {
|
|
185 return (Format) userFormat.clone();
|
|
186 }
|
|
187
|
|
188 // * * * * * * * * * * Output to a OutputStream * * * * * * * * * *
|
|
189 // * * * * * * * * * * Output to a OutputStream * * * * * * * * * *
|
|
190
|
|
191 /**
|
|
192 * This will print the <code>Document</code> to the given output stream.
|
|
193 * The characters are printed using the encoding specified in the
|
|
194 * constructor, or a default of UTF-8.
|
|
195 *
|
|
196 * @param doc <code>Document</code> to format.
|
|
197 * @param out <code>OutputStream</code> to use.
|
|
198 * @throws IOException - if there's any problem writing.
|
|
199 */
|
|
200 public void output(Document doc, OutputStream out)
|
|
201 throws IOException {
|
|
202 Writer writer = makeWriter(out);
|
|
203 output(doc, writer); // output() flushes
|
|
204 }
|
|
205
|
|
206 /**
|
|
207 * Print out the <code>{@link DocType}</code>.
|
|
208 *
|
|
209 * @param doctype <code>DocType</code> to output.
|
|
210 * @param out <code>OutputStream</code> to use.
|
|
211 */
|
|
212 public void output(DocType doctype, OutputStream out) throws IOException {
|
|
213 Writer writer = makeWriter(out);
|
|
214 output(doctype, writer); // output() flushes
|
|
215 }
|
|
216
|
|
217 /**
|
|
218 * Print out an <code>{@link Element}</code>, including
|
|
219 * its <code>{@link Attribute}</code>s, and all
|
|
220 * contained (child) elements, etc.
|
|
221 *
|
|
222 * @param element <code>Element</code> to output.
|
|
223 * @param out <code>Writer</code> to use.
|
|
224 */
|
|
225 public void output(Element element, OutputStream out) throws IOException {
|
|
226 Writer writer = makeWriter(out);
|
|
227 output(element, writer); // output() flushes
|
|
228 }
|
|
229
|
|
230 /**
|
|
231 * This will handle printing out an <code>{@link
|
|
232 * Element}</code>'s content only, not including its tag, and
|
|
233 * attributes. This can be useful for printing the content of an
|
|
234 * element that contains HTML, like "<description>JDOM is
|
|
235 * <b>fun>!</description>".
|
|
236 *
|
|
237 * @param element <code>Element</code> to output.
|
|
238 * @param out <code>OutputStream</code> to use.
|
|
239 */
|
|
240 public void outputElementContent(Element element, OutputStream out)
|
|
241 throws IOException {
|
|
242 Writer writer = makeWriter(out);
|
|
243 outputElementContent(element, writer); // output() flushes
|
|
244 }
|
|
245
|
|
246 /**
|
|
247 * This will handle printing out a list of nodes.
|
|
248 * This can be useful for printing the content of an element that
|
|
249 * contains HTML, like "<description>JDOM is
|
|
250 * <b>fun>!</description>".
|
|
251 *
|
|
252 * @param list <code>List</code> of nodes.
|
|
253 * @param out <code>OutputStream</code> to use.
|
|
254 */
|
|
255 public void output(List list, OutputStream out)
|
|
256 throws IOException {
|
|
257 Writer writer = makeWriter(out);
|
|
258 output(list, writer); // output() flushes
|
|
259 }
|
|
260
|
|
261 /**
|
|
262 * Print out a <code>{@link CDATA}</code> node.
|
|
263 *
|
|
264 * @param cdata <code>CDATA</code> to output.
|
|
265 * @param out <code>OutputStream</code> to use.
|
|
266 */
|
|
267 public void output(CDATA cdata, OutputStream out) throws IOException {
|
|
268 Writer writer = makeWriter(out);
|
|
269 output(cdata, writer); // output() flushes
|
|
270 }
|
|
271
|
|
272 /**
|
|
273 * Print out a <code>{@link Text}</code> node. Perfoms
|
|
274 * the necessary entity escaping and whitespace stripping.
|
|
275 *
|
|
276 * @param text <code>Text</code> to output.
|
|
277 * @param out <code>OutputStream</code> to use.
|
|
278 */
|
|
279 public void output(Text text, OutputStream out) throws IOException {
|
|
280 Writer writer = makeWriter(out);
|
|
281 output(text, writer); // output() flushes
|
|
282 }
|
|
283
|
|
284 /**
|
|
285 * Print out a <code>{@link Comment}</code>.
|
|
286 *
|
|
287 * @param comment <code>Comment</code> to output.
|
|
288 * @param out <code>OutputStream</code> to use.
|
|
289 */
|
|
290 public void output(Comment comment, OutputStream out) throws IOException {
|
|
291 Writer writer = makeWriter(out);
|
|
292 output(comment, writer); // output() flushes
|
|
293 }
|
|
294
|
|
295 /**
|
|
296 * Print out a <code>{@link ProcessingInstruction}</code>.
|
|
297 *
|
|
298 * @param pi <code>ProcessingInstruction</code> to output.
|
|
299 * @param out <code>OutputStream</code> to use.
|
|
300 */
|
|
301 public void output(ProcessingInstruction pi, OutputStream out)
|
|
302 throws IOException {
|
|
303 Writer writer = makeWriter(out);
|
|
304 output(pi, writer); // output() flushes
|
|
305 }
|
|
306
|
|
307 /**
|
|
308 * Print out a <code>{@link EntityRef}</code>.
|
|
309 *
|
|
310 * @param entity <code>EntityRef</code> to output.
|
|
311 * @param out <code>OutputStream</code> to use.
|
|
312 */
|
|
313 public void output(EntityRef entity, OutputStream out) throws IOException {
|
|
314 Writer writer = makeWriter(out);
|
|
315 output(entity, writer); // output() flushes
|
|
316 }
|
|
317
|
|
318 /**
|
|
319 * Get an OutputStreamWriter, using prefered encoding
|
|
320 * (see {@link Format#setEncoding}).
|
|
321 */
|
|
322 private Writer makeWriter(OutputStream out)
|
|
323 throws java.io.UnsupportedEncodingException {
|
|
324 return makeWriter(out, userFormat.encoding);
|
|
325 }
|
|
326
|
|
327 /**
|
|
328 * Get an OutputStreamWriter, use specified encoding.
|
|
329 */
|
|
330 private static Writer makeWriter(OutputStream out, String enc)
|
|
331 throws java.io.UnsupportedEncodingException {
|
|
332 // "UTF-8" is not recognized before JDK 1.1.6, so we'll translate
|
|
333 // into "UTF8" which works with all JDKs.
|
|
334 if ("UTF-8".equals(enc)) {
|
|
335 enc = "UTF8";
|
|
336 }
|
|
337
|
|
338 Writer writer = new BufferedWriter(
|
|
339 (new OutputStreamWriter(
|
|
340 new BufferedOutputStream(out), enc)
|
|
341 ));
|
|
342 return writer;
|
|
343 }
|
|
344
|
|
345 // * * * * * * * * * * Output to a Writer * * * * * * * * * *
|
|
346 // * * * * * * * * * * Output to a Writer * * * * * * * * * *
|
|
347
|
|
348 /**
|
|
349 * This will print the <code>Document</code> to the given Writer.
|
|
350 *
|
|
351 * <p>
|
|
352 * Warning: using your own Writer may cause the outputter's
|
|
353 * preferred character encoding to be ignored. If you use
|
|
354 * encodings other than UTF-8, we recommend using the method that
|
|
355 * takes an OutputStream instead.
|
|
356 * </p>
|
|
357 *
|
|
358 * @param doc <code>Document</code> to format.
|
|
359 * @param out <code>Writer</code> to use.
|
|
360 * @throws IOException - if there's any problem writing.
|
|
361 */
|
|
362 public void output(Document doc, Writer out) throws IOException {
|
|
363
|
|
364 printDeclaration(out, doc, userFormat.encoding);
|
|
365
|
|
366 // Print out root element, as well as any root level
|
|
367 // comments and processing instructions,
|
|
368 // starting with no indentation
|
|
369 List content = doc.getContent();
|
|
370 int size = content.size();
|
|
371 for (int i = 0; i < size; i++) {
|
|
372 Object obj = content.get(i);
|
|
373
|
|
374 if (obj instanceof Element) {
|
|
375 printElement(out, doc.getRootElement(), 0,
|
|
376 createNamespaceStack());
|
|
377 }
|
|
378 else if (obj instanceof Comment) {
|
|
379 printComment(out, (Comment) obj);
|
|
380 }
|
|
381 else if (obj instanceof ProcessingInstruction) {
|
|
382 printProcessingInstruction(out, (ProcessingInstruction) obj);
|
|
383 }
|
|
384 else if (obj instanceof DocType) {
|
|
385 printDocType(out, doc.getDocType());
|
|
386 // Always print line separator after declaration, helps the
|
|
387 // output look better and is semantically inconsequential
|
|
388 out.write(currentFormat.lineSeparator);
|
|
389 }
|
|
390 else {
|
|
391 // XXX if we get here then we have a illegal content, for
|
|
392 // now we'll just ignore it
|
|
393 }
|
|
394
|
|
395 newline(out);
|
|
396 indent(out, 0);
|
|
397 }
|
|
398
|
|
399 // Output final line separator
|
|
400 // We output this no matter what the newline flags say
|
|
401 out.write(currentFormat.lineSeparator);
|
|
402
|
|
403 out.flush();
|
|
404 }
|
|
405
|
|
406 /**
|
|
407 * Print out the <code>{@link DocType}</code>.
|
|
408 *
|
|
409 * @param doctype <code>DocType</code> to output.
|
|
410 * @param out <code>Writer</code> to use.
|
|
411 */
|
|
412 public void output(DocType doctype, Writer out) throws IOException {
|
|
413 printDocType(out, doctype);
|
|
414 out.flush();
|
|
415 }
|
|
416
|
|
417 /**
|
|
418 * Print out an <code>{@link Element}</code>, including
|
|
419 * its <code>{@link Attribute}</code>s, and all
|
|
420 * contained (child) elements, etc.
|
|
421 *
|
|
422 * @param element <code>Element</code> to output.
|
|
423 * @param out <code>Writer</code> to use.
|
|
424 */
|
|
425 public void output(Element element, Writer out) throws IOException {
|
|
426 // If this is the root element we could pre-initialize the
|
|
427 // namespace stack with the namespaces
|
|
428 printElement(out, element, 0, createNamespaceStack());
|
|
429 out.flush();
|
|
430 }
|
|
431
|
|
432 /**
|
|
433 * This will handle printing out an <code>{@link
|
|
434 * Element}</code>'s content only, not including its tag, and
|
|
435 * attributes. This can be useful for printing the content of an
|
|
436 * element that contains HTML, like "<description>JDOM is
|
|
437 * <b>fun>!</description>".
|
|
438 *
|
|
439 * @param element <code>Element</code> to output.
|
|
440 * @param out <code>Writer</code> to use.
|
|
441 */
|
|
442 public void outputElementContent(Element element, Writer out)
|
|
443 throws IOException {
|
|
444 List content = element.getContent();
|
|
445 printContentRange(out, content, 0, content.size(),
|
|
446 0, createNamespaceStack());
|
|
447 out.flush();
|
|
448 }
|
|
449
|
|
450 /**
|
|
451 * This will handle printing out a list of nodes.
|
|
452 * This can be useful for printing the content of an element that
|
|
453 * contains HTML, like "<description>JDOM is
|
|
454 * <b>fun>!</description>".
|
|
455 *
|
|
456 * @param list <code>List</code> of nodes.
|
|
457 * @param out <code>Writer</code> to use.
|
|
458 */
|
|
459 public void output(List list, Writer out)
|
|
460 throws IOException {
|
|
461 printContentRange(out, list, 0, list.size(),
|
|
462 0, createNamespaceStack());
|
|
463 out.flush();
|
|
464 }
|
|
465
|
|
466 /**
|
|
467 * Print out a <code>{@link CDATA}</code> node.
|
|
468 *
|
|
469 * @param cdata <code>CDATA</code> to output.
|
|
470 * @param out <code>Writer</code> to use.
|
|
471 */
|
|
472 public void output(CDATA cdata, Writer out) throws IOException {
|
|
473 printCDATA(out, cdata);
|
|
474 out.flush();
|
|
475 }
|
|
476
|
|
477 /**
|
|
478 * Print out a <code>{@link Text}</code> node. Perfoms
|
|
479 * the necessary entity escaping and whitespace stripping.
|
|
480 *
|
|
481 * @param text <code>Text</code> to output.
|
|
482 * @param out <code>Writer</code> to use.
|
|
483 */
|
|
484 public void output(Text text, Writer out) throws IOException {
|
|
485 printText(out, text);
|
|
486 out.flush();
|
|
487 }
|
|
488
|
|
489 /**
|
|
490 * Print out a <code>{@link Comment}</code>.
|
|
491 *
|
|
492 * @param comment <code>Comment</code> to output.
|
|
493 * @param out <code>Writer</code> to use.
|
|
494 */
|
|
495 public void output(Comment comment, Writer out) throws IOException {
|
|
496 printComment(out, comment);
|
|
497 out.flush();
|
|
498 }
|
|
499
|
|
500 /**
|
|
501 * Print out a <code>{@link ProcessingInstruction}</code>.
|
|
502 *
|
|
503 * @param pi <code>ProcessingInstruction</code> to output.
|
|
504 * @param out <code>Writer</code> to use.
|
|
505 */
|
|
506 public void output(ProcessingInstruction pi, Writer out)
|
|
507 throws IOException {
|
|
508 boolean currentEscapingPolicy = currentFormat.ignoreTrAXEscapingPIs;
|
|
509
|
|
510 // Output PI verbatim, disregarding TrAX escaping PIs.
|
|
511 currentFormat.setIgnoreTrAXEscapingPIs(true);
|
|
512 printProcessingInstruction(out, pi);
|
|
513 currentFormat.setIgnoreTrAXEscapingPIs(currentEscapingPolicy);
|
|
514
|
|
515 out.flush();
|
|
516 }
|
|
517
|
|
518 /**
|
|
519 * Print out a <code>{@link EntityRef}</code>.
|
|
520 *
|
|
521 * @param entity <code>EntityRef</code> to output.
|
|
522 * @param out <code>Writer</code> to use.
|
|
523 */
|
|
524 public void output(EntityRef entity, Writer out) throws IOException {
|
|
525 printEntityRef(out, entity);
|
|
526 out.flush();
|
|
527 }
|
|
528
|
|
529 // * * * * * * * * * * Output to a String * * * * * * * * * *
|
|
530 // * * * * * * * * * * Output to a String * * * * * * * * * *
|
|
531
|
|
532 /**
|
|
533 * Return a string representing a document. Uses an internal
|
|
534 * StringWriter. Warning: a String is Unicode, which may not match
|
|
535 * the outputter's specified encoding.
|
|
536 *
|
|
537 * @param doc <code>Document</code> to format.
|
|
538 */
|
|
539 public String outputString(Document doc) {
|
|
540 StringWriter out = new StringWriter();
|
|
541 try {
|
|
542 output(doc, out); // output() flushes
|
|
543 } catch (IOException e) { }
|
|
544 return out.toString();
|
|
545 }
|
|
546
|
|
547 /**
|
|
548 * Return a string representing a DocType. Warning: a String is
|
|
549 * Unicode, which may not match the outputter's specified
|
|
550 * encoding.
|
|
551 *
|
|
552 * @param doctype <code>DocType</code> to format.
|
|
553 */
|
|
554 public String outputString(DocType doctype) {
|
|
555 StringWriter out = new StringWriter();
|
|
556 try {
|
|
557 output(doctype, out); // output() flushes
|
|
558 } catch (IOException e) { }
|
|
559 return out.toString();
|
|
560 }
|
|
561
|
|
562 /**
|
|
563 * Return a string representing an element. Warning: a String is
|
|
564 * Unicode, which may not match the outputter's specified
|
|
565 * encoding.
|
|
566 *
|
|
567 * @param element <code>Element</code> to format.
|
|
568 */
|
|
569 public String outputString(Element element) {
|
|
570 StringWriter out = new StringWriter();
|
|
571 try {
|
|
572 output(element, out); // output() flushes
|
|
573 } catch (IOException e) { }
|
|
574 return out.toString();
|
|
575 }
|
|
576
|
|
577 /**
|
|
578 * Return a string representing a list of nodes. The list is
|
|
579 * assumed to contain legal JDOM nodes.
|
|
580 *
|
|
581 * @param list <code>List</code> to format.
|
|
582 */
|
|
583 public String outputString(List list) {
|
|
584 StringWriter out = new StringWriter();
|
|
585 try {
|
|
586 output(list, out); // output() flushes
|
|
587 } catch (IOException e) { }
|
|
588 return out.toString();
|
|
589 }
|
|
590
|
|
591 /**
|
|
592 * Return a string representing a CDATA node. Warning: a String is
|
|
593 * Unicode, which may not match the outputter's specified
|
|
594 * encoding.
|
|
595 *
|
|
596 * @param cdata <code>CDATA</code> to format.
|
|
597 */
|
|
598 public String outputString(CDATA cdata) {
|
|
599 StringWriter out = new StringWriter();
|
|
600 try {
|
|
601 output(cdata, out); // output() flushes
|
|
602 } catch (IOException e) { }
|
|
603 return out.toString();
|
|
604 }
|
|
605
|
|
606 /**
|
|
607 * Return a string representing a Text node. Warning: a String is
|
|
608 * Unicode, which may not match the outputter's specified
|
|
609 * encoding.
|
|
610 *
|
|
611 * @param text <code>Text</code> to format.
|
|
612 */
|
|
613 public String outputString(Text text) {
|
|
614 StringWriter out = new StringWriter();
|
|
615 try {
|
|
616 output(text, out); // output() flushes
|
|
617 } catch (IOException e) { }
|
|
618 return out.toString();
|
|
619 }
|
|
620
|
|
621
|
|
622 /**
|
|
623 * Return a string representing a comment. Warning: a String is
|
|
624 * Unicode, which may not match the outputter's specified
|
|
625 * encoding.
|
|
626 *
|
|
627 * @param comment <code>Comment</code> to format.
|
|
628 */
|
|
629 public String outputString(Comment comment) {
|
|
630 StringWriter out = new StringWriter();
|
|
631 try {
|
|
632 output(comment, out); // output() flushes
|
|
633 } catch (IOException e) { }
|
|
634 return out.toString();
|
|
635 }
|
|
636
|
|
637 /**
|
|
638 * Return a string representing a PI. Warning: a String is
|
|
639 * Unicode, which may not match the outputter's specified
|
|
640 * encoding.
|
|
641 *
|
|
642 * @param pi <code>ProcessingInstruction</code> to format.
|
|
643 */
|
|
644 public String outputString(ProcessingInstruction pi) {
|
|
645 StringWriter out = new StringWriter();
|
|
646 try {
|
|
647 output(pi, out); // output() flushes
|
|
648 } catch (IOException e) { }
|
|
649 return out.toString();
|
|
650 }
|
|
651
|
|
652 /**
|
|
653 * Return a string representing an entity. Warning: a String is
|
|
654 * Unicode, which may not match the outputter's specified
|
|
655 * encoding.
|
|
656 *
|
|
657 * @param entity <code>EntityRef</code> to format.
|
|
658 */
|
|
659 public String outputString(EntityRef entity) {
|
|
660 StringWriter out = new StringWriter();
|
|
661 try {
|
|
662 output(entity, out); // output() flushes
|
|
663 } catch (IOException e) { }
|
|
664 return out.toString();
|
|
665 }
|
|
666
|
|
667 // * * * * * * * * * * Internal printing methods * * * * * * * * * *
|
|
668 // * * * * * * * * * * Internal printing methods * * * * * * * * * *
|
|
669
|
|
670 /**
|
|
671 * This will handle printing of the declaration.
|
|
672 * Assumes XML version 1.0 since we don't directly know.
|
|
673 *
|
|
674 * @param doc <code>Document</code> whose declaration to write.
|
|
675 * @param out <code>Writer</code> to use.
|
|
676 * @param encoding The encoding to add to the declaration
|
|
677 */
|
|
678 protected void printDeclaration(Writer out, Document doc,
|
|
679 String encoding) throws IOException {
|
|
680
|
|
681 // Only print the declaration if it's not being omitted
|
|
682 if (!userFormat.omitDeclaration) {
|
|
683 // Assume 1.0 version
|
|
684 out.write("<?xml version=\"1.0\"");
|
|
685 if (!userFormat.omitEncoding) {
|
|
686 out.write(" encoding=\"" + encoding + "\"");
|
|
687 }
|
|
688 out.write("?>");
|
|
689
|
|
690 // Print new line after decl always, even if no other new lines
|
|
691 // Helps the output look better and is semantically
|
|
692 // inconsequential
|
|
693 out.write(currentFormat.lineSeparator);
|
|
694 }
|
|
695 }
|
|
696
|
|
697 /**
|
|
698 * This handle printing the DOCTYPE declaration if one exists.
|
|
699 *
|
|
700 * @param docType <code>Document</code> whose declaration to write.
|
|
701 * @param out <code>Writer</code> to use.
|
|
702 */
|
|
703 protected void printDocType(Writer out, DocType docType)
|
|
704 throws IOException {
|
|
705
|
|
706 String publicID = docType.getPublicID();
|
|
707 String systemID = docType.getSystemID();
|
|
708 String internalSubset = docType.getInternalSubset();
|
|
709 boolean hasPublic = false;
|
|
710
|
|
711 out.write("<!DOCTYPE ");
|
|
712 out.write(docType.getElementName());
|
|
713 if (publicID != null) {
|
|
714 out.write(" PUBLIC \"");
|
|
715 out.write(publicID);
|
|
716 out.write("\"");
|
|
717 hasPublic = true;
|
|
718 }
|
|
719 if (systemID != null) {
|
|
720 if (!hasPublic) {
|
|
721 out.write(" SYSTEM");
|
|
722 }
|
|
723 out.write(" \"");
|
|
724 out.write(systemID);
|
|
725 out.write("\"");
|
|
726 }
|
|
727 if ((internalSubset != null) && (!internalSubset.equals(""))) {
|
|
728 out.write(" [");
|
|
729 out.write(currentFormat.lineSeparator);
|
|
730 out.write(docType.getInternalSubset());
|
|
731 out.write("]");
|
|
732 }
|
|
733 out.write(">");
|
|
734 }
|
|
735
|
|
736 /**
|
|
737 * This will handle printing of comments.
|
|
738 *
|
|
739 * @param comment <code>Comment</code> to write.
|
|
740 * @param out <code>Writer</code> to use.
|
|
741 */
|
|
742 protected void printComment(Writer out, Comment comment)
|
|
743 throws IOException {
|
|
744 out.write("<!--");
|
|
745 out.write(comment.getText());
|
|
746 out.write("-->");
|
|
747 }
|
|
748
|
|
749 /**
|
|
750 * This will handle printing of processing instructions.
|
|
751 *
|
|
752 * @param pi <code>ProcessingInstruction</code> to write.
|
|
753 * @param out <code>Writer</code> to use.
|
|
754 */
|
|
755 protected void printProcessingInstruction(Writer out, ProcessingInstruction pi
|
|
756 ) throws IOException {
|
|
757 String target = pi.getTarget();
|
|
758 boolean piProcessed = false;
|
|
759
|
|
760 if (currentFormat.ignoreTrAXEscapingPIs == false) {
|
|
761 if (target.equals(Result.PI_DISABLE_OUTPUT_ESCAPING)) {
|
|
762 escapeOutput = false;
|
|
763 piProcessed = true;
|
|
764 }
|
|
765 else if (target.equals(Result.PI_ENABLE_OUTPUT_ESCAPING)) {
|
|
766 escapeOutput = true;
|
|
767 piProcessed = true;
|
|
768 }
|
|
769 }
|
|
770 if (piProcessed == false) {
|
|
771 String rawData = pi.getData();
|
|
772
|
|
773 // Write <?target data?> or if no data then just <?target?>
|
|
774 if (!"".equals(rawData)) {
|
|
775 out.write("<?");
|
|
776 out.write(target);
|
|
777 out.write(" ");
|
|
778 out.write(rawData);
|
|
779 out.write("?>");
|
|
780 }
|
|
781 else {
|
|
782 out.write("<?");
|
|
783 out.write(target);
|
|
784 out.write("?>");
|
|
785 }
|
|
786 }
|
|
787 }
|
|
788
|
|
789 /**
|
|
790 * This will handle printing a <code>{@link EntityRef}</code>.
|
|
791 * Only the entity reference such as <code>&entity;</code>
|
|
792 * will be printed. However, subclasses are free to override
|
|
793 * this method to print the contents of the entity instead.
|
|
794 *
|
|
795 * @param entity <code>EntityRef</code> to output.
|
|
796 * @param out <code>Writer</code> to use. */
|
|
797 protected void printEntityRef(Writer out, EntityRef entity)
|
|
798 throws IOException {
|
|
799 out.write("&");
|
|
800 out.write(entity.getName());
|
|
801 out.write(";");
|
|
802 }
|
|
803
|
|
804 /**
|
|
805 * This will handle printing of <code>{@link CDATA}</code> text.
|
|
806 *
|
|
807 * @param cdata <code>CDATA</code> to output.
|
|
808 * @param out <code>Writer</code> to use.
|
|
809 */
|
|
810 protected void printCDATA(Writer out, CDATA cdata) throws IOException {
|
|
811 String str = (currentFormat.mode == Format.TextMode.NORMALIZE)
|
|
812 ? cdata.getTextNormalize()
|
|
813 : ((currentFormat.mode == Format.TextMode.TRIM) ?
|
|
814 cdata.getText().trim() : cdata.getText());
|
|
815 out.write("<![CDATA[");
|
|
816 out.write(str);
|
|
817 out.write("]]>");
|
|
818 }
|
|
819
|
|
820 /**
|
|
821 * This will handle printing of <code>{@link Text}</code> strings.
|
|
822 *
|
|
823 * @param text <code>Text</code> to write.
|
|
824 * @param out <code>Writer</code> to use.
|
|
825 */
|
|
826 protected void printText(Writer out, Text text) throws IOException {
|
|
827 String str = (currentFormat.mode == Format.TextMode.NORMALIZE)
|
|
828 ? text.getTextNormalize()
|
|
829 : ((currentFormat.mode == Format.TextMode.TRIM) ?
|
|
830 text.getText().trim() : text.getText());
|
|
831 out.write(escapeElementEntities(str));
|
|
832 }
|
|
833
|
|
834 /**
|
|
835 * This will handle printing a string. Escapes the element entities,
|
|
836 * trims interior whitespace, etc. if necessary.
|
|
837 */
|
|
838 private void printString(Writer out, String str) throws IOException {
|
|
839 if (currentFormat.mode == Format.TextMode.NORMALIZE) {
|
|
840 str = Text.normalizeString(str);
|
|
841 }
|
|
842 else if (currentFormat.mode == Format.TextMode.TRIM) {
|
|
843 str = str.trim();
|
|
844 }
|
|
845 out.write(escapeElementEntities(str));
|
|
846 }
|
|
847
|
|
848 /**
|
|
849 * This will handle printing of a <code>{@link Element}</code>,
|
|
850 * its <code>{@link Attribute}</code>s, and all contained (child)
|
|
851 * elements, etc.
|
|
852 *
|
|
853 * @param element <code>Element</code> to output.
|
|
854 * @param out <code>Writer</code> to use.
|
|
855 * @param level <code>int</code> level of indention.
|
|
856 * @param namespaces <code>List</code> stack of Namespaces in scope.
|
|
857 */
|
|
858 protected void printElement(Writer out, Element element,
|
|
859 int level, NamespaceStack namespaces)
|
|
860 throws IOException {
|
|
861
|
|
862 List attributes = element.getAttributes();
|
|
863 List content = element.getContent();
|
|
864
|
|
865 // Check for xml:space and adjust format settings
|
|
866 String space = null;
|
|
867 if (attributes != null) {
|
|
868 space = element.getAttributeValue("space",
|
|
869 Namespace.XML_NAMESPACE);
|
|
870 }
|
|
871
|
|
872 Format previousFormat = currentFormat;
|
|
873
|
|
874 if ("default".equals(space)) {
|
|
875 currentFormat = userFormat;
|
|
876 }
|
|
877 else if ("preserve".equals(space)) {
|
|
878 currentFormat = preserveFormat;
|
|
879 }
|
|
880
|
|
881 // Print the beginning of the tag plus attributes and any
|
|
882 // necessary namespace declarations
|
|
883 out.write("<");
|
|
884 printQualifiedName(out, element);
|
|
885
|
|
886 // Mark our namespace starting point
|
|
887 int previouslyDeclaredNamespaces = namespaces.size();
|
|
888
|
|
889 // Print the element's namespace, if appropriate
|
|
890 printElementNamespace(out, element, namespaces);
|
|
891
|
|
892 // Print out additional namespace declarations
|
|
893 printAdditionalNamespaces(out, element, namespaces);
|
|
894
|
|
895 // Print out attributes
|
|
896 if (attributes != null)
|
|
897 printAttributes(out, attributes, element, namespaces);
|
|
898
|
|
899 // Depending on the settings (newlines, textNormalize, etc), we may
|
|
900 // or may not want to print all of the content, so determine the
|
|
901 // index of the start of the content we're interested
|
|
902 // in based on the current settings.
|
|
903
|
|
904 int start = skipLeadingWhite(content, 0);
|
|
905 int size = content.size();
|
|
906 if (start >= size) {
|
|
907 // Case content is empty or all insignificant whitespace
|
|
908 if (currentFormat.expandEmptyElements) {
|
|
909 out.write("></");
|
|
910 printQualifiedName(out, element);
|
|
911 out.write(">");
|
|
912 }
|
|
913 else {
|
|
914 out.write(" />");
|
|
915 }
|
|
916 }
|
|
917 else {
|
|
918 out.write(">");
|
|
919
|
|
920 // For a special case where the content is only CDATA
|
|
921 // or Text we don't want to indent after the start or
|
|
922 // before the end tag.
|
|
923
|
|
924 if (nextNonText(content, start) < size) {
|
|
925 // Case Mixed Content - normal indentation
|
|
926 newline(out);
|
|
927 printContentRange(out, content, start, size,
|
|
928 level + 1, namespaces);
|
|
929 newline(out);
|
|
930 indent(out, level);
|
|
931 }
|
|
932 else {
|
|
933 // Case all CDATA or Text - no indentation
|
|
934 printTextRange(out, content, start, size);
|
|
935 }
|
|
936 out.write("</");
|
|
937 printQualifiedName(out, element);
|
|
938 out.write(">");
|
|
939 }
|
|
940
|
|
941 // remove declared namespaces from stack
|
|
942 while (namespaces.size() > previouslyDeclaredNamespaces) {
|
|
943 namespaces.pop();
|
|
944 }
|
|
945
|
|
946 // Restore our format settings
|
|
947 currentFormat = previousFormat;
|
|
948 }
|
|
949
|
|
950 /**
|
|
951 * This will handle printing of content within a given range.
|
|
952 * The range to print is specified in typical Java fashion; the
|
|
953 * starting index is inclusive, while the ending index is
|
|
954 * exclusive.
|
|
955 *
|
|
956 * @param content <code>List</code> of content to output
|
|
957 * @param start index of first content node (inclusive.
|
|
958 * @param end index of last content node (exclusive).
|
|
959 * @param out <code>Writer</code> to use.
|
|
960 * @param level <code>int</code> level of indentation.
|
|
961 * @param namespaces <code>List</code> stack of Namespaces in scope.
|
|
962 */
|
|
963 private void printContentRange(Writer out, List content,
|
|
964 int start, int end, int level,
|
|
965 NamespaceStack namespaces)
|
|
966 throws IOException {
|
|
967 boolean firstNode; // Flag for 1st node in content
|
|
968 Object next; // Node we're about to print
|
|
969 int first, index; // Indexes into the list of content
|
|
970
|
|
971 index = start;
|
|
972 while (index < end) {
|
|
973 firstNode = (index == start) ? true : false;
|
|
974 next = content.get(index);
|
|
975
|
|
976 //
|
|
977 // Handle consecutive CDATA, Text, and EntityRef nodes all at once
|
|
978 //
|
|
979 if ((next instanceof Text) || (next instanceof EntityRef)) {
|
|
980 first = skipLeadingWhite(content, index);
|
|
981 // Set index to next node for loop
|
|
982 index = nextNonText(content, first);
|
|
983
|
|
984 // If it's not all whitespace - print it!
|
|
985 if (first < index) {
|
|
986 if (!firstNode)
|
|
987 newline(out);
|
|
988 indent(out, level);
|
|
989 printTextRange(out, content, first, index);
|
|
990 }
|
|
991 continue;
|
|
992 }
|
|
993
|
|
994 //
|
|
995 // Handle other nodes
|
|
996 //
|
|
997 if (!firstNode) {
|
|
998 newline(out);
|
|
999 }
|
|
1000
|
|
1001 indent(out, level);
|
|
1002
|
|
1003 if (next instanceof Comment) {
|
|
1004 printComment(out, (Comment)next);
|
|
1005 }
|
|
1006 else if (next instanceof Element) {
|
|
1007 printElement(out, (Element)next, level, namespaces);
|
|
1008 }
|
|
1009 else if (next instanceof ProcessingInstruction) {
|
|
1010 printProcessingInstruction(out, (ProcessingInstruction)next);
|
|
1011 }
|
|
1012 else {
|
|
1013 // XXX if we get here then we have a illegal content, for
|
|
1014 // now we'll just ignore it (probably should throw
|
|
1015 // a exception)
|
|
1016 }
|
|
1017
|
|
1018 index++;
|
|
1019 } /* while */
|
|
1020 }
|
|
1021
|
|
1022 /**
|
|
1023 * This will handle printing of a sequence of <code>{@link CDATA}</code>
|
|
1024 * or <code>{@link Text}</code> nodes. It is an error to have any other
|
|
1025 * pass this method any other type of node.
|
|
1026 *
|
|
1027 * @param content <code>List</code> of content to output
|
|
1028 * @param start index of first content node (inclusive).
|
|
1029 * @param end index of last content node (exclusive).
|
|
1030 * @param out <code>Writer</code> to use.
|
|
1031 */
|
|
1032 private void printTextRange(Writer out, List content, int start, int end
|
|
1033 ) throws IOException {
|
|
1034 String previous; // Previous text printed
|
|
1035 Object node; // Next node to print
|
|
1036 String next; // Next text to print
|
|
1037
|
|
1038 previous = null;
|
|
1039
|
|
1040 // Remove leading whitespace-only nodes
|
|
1041 start = skipLeadingWhite(content, start);
|
|
1042
|
|
1043 int size = content.size();
|
|
1044 if (start < size) {
|
|
1045 // And remove trialing whitespace-only nodes
|
|
1046 end = skipTrailingWhite(content, end);
|
|
1047
|
|
1048 for (int i = start; i < end; i++) {
|
|
1049 node = content.get(i);
|
|
1050
|
|
1051 // Get the unmangled version of the text
|
|
1052 // we are about to print
|
|
1053 if (node instanceof Text) {
|
|
1054 next = ((Text) node).getText();
|
|
1055 }
|
|
1056 else if (node instanceof EntityRef) {
|
|
1057 next = "&" + ((EntityRef) node).getValue() + ";";
|
|
1058 }
|
|
1059 else {
|
|
1060 throw new IllegalStateException("Should see only " +
|
|
1061 "CDATA, Text, or EntityRef");
|
|
1062 }
|
|
1063
|
|
1064 // This may save a little time
|
|
1065 if (next == null || "".equals(next)) {
|
|
1066 continue;
|
|
1067 }
|
|
1068
|
|
1069 // Determine if we need to pad the output (padding is
|
|
1070 // only need in trim or normalizing mode)
|
|
1071 if (previous != null) { // Not 1st node
|
|
1072 if (currentFormat.mode == Format.TextMode.NORMALIZE ||
|
|
1073 currentFormat.mode == Format.TextMode.TRIM) {
|
|
1074 if ((endsWithWhite(previous)) ||
|
|
1075 (startsWithWhite(next))) {
|
|
1076 out.write(" ");
|
|
1077 }
|
|
1078 }
|
|
1079 }
|
|
1080
|
|
1081 // Print the node
|
|
1082 if (node instanceof CDATA) {
|
|
1083 printCDATA(out, (CDATA) node);
|
|
1084 }
|
|
1085 else if (node instanceof EntityRef) {
|
|
1086 printEntityRef(out, (EntityRef) node);
|
|
1087 }
|
|
1088 else {
|
|
1089 printString(out, next);
|
|
1090 }
|
|
1091
|
|
1092 previous = next;
|
|
1093 }
|
|
1094 }
|
|
1095 }
|
|
1096
|
|
1097 /**
|
|
1098 * This will handle printing of any needed <code>{@link Namespace}</code>
|
|
1099 * declarations.
|
|
1100 *
|
|
1101 * @param ns <code>Namespace</code> to print definition of
|
|
1102 * @param out <code>Writer</code> to use.
|
|
1103 */
|
|
1104 private void printNamespace(Writer out, Namespace ns,
|
|
1105 NamespaceStack namespaces)
|
|
1106 throws IOException {
|
|
1107 String prefix = ns.getPrefix();
|
|
1108 String uri = ns.getURI();
|
|
1109
|
|
1110 // Already printed namespace decl?
|
|
1111 if (uri.equals(namespaces.getURI(prefix))) {
|
|
1112 return;
|
|
1113 }
|
|
1114
|
|
1115 out.write(" xmlns");
|
|
1116 if (!prefix.equals("")) {
|
|
1117 out.write(":");
|
|
1118 out.write(prefix);
|
|
1119 }
|
|
1120 out.write("=\"");
|
|
1121 out.write(escapeAttributeEntities(uri));
|
|
1122 out.write("\"");
|
|
1123 namespaces.push(ns);
|
|
1124 }
|
|
1125
|
|
1126 /**
|
|
1127 * This will handle printing of a <code>{@link Attribute}</code> list.
|
|
1128 *
|
|
1129 * @param attributes <code>List</code> of Attribute objcts
|
|
1130 * @param out <code>Writer</code> to use
|
|
1131 */
|
|
1132 protected void printAttributes(Writer out, List attributes, Element parent,
|
|
1133 NamespaceStack namespaces)
|
|
1134 throws IOException {
|
|
1135
|
|
1136 // I do not yet handle the case where the same prefix maps to
|
|
1137 // two different URIs. For attributes on the same element
|
|
1138 // this is illegal; but as yet we don't throw an exception
|
|
1139 // if someone tries to do this
|
|
1140 // Set prefixes = new HashSet();
|
|
1141 for (int i = 0; i < attributes.size(); i++) {
|
|
1142 Attribute attribute = (Attribute) attributes.get(i);
|
|
1143 Namespace ns = attribute.getNamespace();
|
|
1144 if ((ns != Namespace.NO_NAMESPACE) &&
|
|
1145 (ns != Namespace.XML_NAMESPACE)) {
|
|
1146 printNamespace(out, ns, namespaces);
|
|
1147 }
|
|
1148
|
|
1149 out.write(" ");
|
|
1150 printQualifiedName(out, attribute);
|
|
1151 out.write("=");
|
|
1152
|
|
1153 out.write("\"");
|
|
1154 out.write(escapeAttributeEntities(attribute.getValue()));
|
|
1155 out.write("\"");
|
|
1156 }
|
|
1157 }
|
|
1158
|
|
1159 private void printElementNamespace(Writer out, Element element,
|
|
1160 NamespaceStack namespaces)
|
|
1161 throws IOException {
|
|
1162 // Add namespace decl only if it's not the XML namespace and it's
|
|
1163 // not the NO_NAMESPACE with the prefix "" not yet mapped
|
|
1164 // (we do output xmlns="" if the "" prefix was already used and we
|
|
1165 // need to reclaim it for the NO_NAMESPACE)
|
|
1166 Namespace ns = element.getNamespace();
|
|
1167 if (ns == Namespace.XML_NAMESPACE) {
|
|
1168 return;
|
|
1169 }
|
|
1170 if ( !((ns == Namespace.NO_NAMESPACE) &&
|
|
1171 (namespaces.getURI("") == null))) {
|
|
1172 printNamespace(out, ns, namespaces);
|
|
1173 }
|
|
1174 }
|
|
1175
|
|
1176 private void printAdditionalNamespaces(Writer out, Element element,
|
|
1177 NamespaceStack namespaces)
|
|
1178 throws IOException {
|
|
1179 List list = element.getAdditionalNamespaces();
|
|
1180 if (list != null) {
|
|
1181 for (int i = 0; i < list.size(); i++) {
|
|
1182 Namespace additional = (Namespace)list.get(i);
|
|
1183 printNamespace(out, additional, namespaces);
|
|
1184 }
|
|
1185 }
|
|
1186 }
|
|
1187
|
|
1188 // * * * * * * * * * * Support methods * * * * * * * * * *
|
|
1189 // * * * * * * * * * * Support methods * * * * * * * * * *
|
|
1190
|
|
1191 /**
|
|
1192 * This will print a newline only if indent is not null.
|
|
1193 *
|
|
1194 * @param out <code>Writer</code> to use
|
|
1195 */
|
|
1196 private void newline(Writer out) throws IOException {
|
|
1197 if (currentFormat.indent != null) {
|
|
1198 out.write(currentFormat.lineSeparator);
|
|
1199 }
|
|
1200 }
|
|
1201
|
|
1202 /**
|
|
1203 * This will print indents only if indent is not null or the empty string.
|
|
1204 *
|
|
1205 * @param out <code>Writer</code> to use
|
|
1206 * @param level current indent level
|
|
1207 */
|
|
1208 private void indent(Writer out, int level) throws IOException {
|
|
1209 if (currentFormat.indent == null ||
|
|
1210 currentFormat.indent.equals("")) {
|
|
1211 return;
|
|
1212 }
|
|
1213
|
|
1214 for (int i = 0; i < level; i++) {
|
|
1215 out.write(currentFormat.indent);
|
|
1216 }
|
|
1217 }
|
|
1218
|
|
1219 // Returns the index of the first non-all-whitespace CDATA or Text,
|
|
1220 // index = content.size() is returned if content contains
|
|
1221 // all whitespace.
|
|
1222 // @param start index to begin search (inclusive)
|
|
1223 private int skipLeadingWhite(List content, int start) {
|
|
1224 if (start < 0) {
|
|
1225 start = 0;
|
|
1226 }
|
|
1227
|
|
1228 int index = start;
|
|
1229 int size = content.size();
|
|
1230 if (currentFormat.mode == Format.TextMode.TRIM_FULL_WHITE
|
|
1231 || currentFormat.mode == Format.TextMode.NORMALIZE
|
|
1232 || currentFormat.mode == Format.TextMode.TRIM) {
|
|
1233 while (index < size) {
|
|
1234 if (!isAllWhitespace(content.get(index))) {
|
|
1235 return index;
|
|
1236 }
|
|
1237 index++;
|
|
1238 }
|
|
1239 }
|
|
1240 return index;
|
|
1241 }
|
|
1242
|
|
1243 // Return the index + 1 of the last non-all-whitespace CDATA or
|
|
1244 // Text node, index < 0 is returned
|
|
1245 // if content contains all whitespace.
|
|
1246 // @param start index to begin search (exclusive)
|
|
1247 private int skipTrailingWhite(List content, int start) {
|
|
1248 int size = content.size();
|
|
1249 if (start > size) {
|
|
1250 start = size;
|
|
1251 }
|
|
1252
|
|
1253 int index = start;
|
|
1254 if (currentFormat.mode == Format.TextMode.TRIM_FULL_WHITE
|
|
1255 || currentFormat.mode == Format.TextMode.NORMALIZE
|
|
1256 || currentFormat.mode == Format.TextMode.TRIM) {
|
|
1257 while (index >= 0) {
|
|
1258 if (!isAllWhitespace(content.get(index - 1)))
|
|
1259 break;
|
|
1260 --index;
|
|
1261 }
|
|
1262 }
|
|
1263 return index;
|
|
1264 }
|
|
1265
|
|
1266 // Return the next non-CDATA, non-Text, or non-EntityRef node,
|
|
1267 // index = content.size() is returned if there is no more non-CDATA,
|
|
1268 // non-Text, or non-EntiryRef nodes
|
|
1269 // @param start index to begin search (inclusive)
|
|
1270 private static int nextNonText(List content, int start) {
|
|
1271 if (start < 0) {
|
|
1272 start = 0;
|
|
1273 }
|
|
1274
|
|
1275 int index = start;
|
|
1276 int size = content.size();
|
|
1277 while (index < size) {
|
|
1278 Object node = content.get(index);
|
|
1279 if (!((node instanceof Text) || (node instanceof EntityRef))) {
|
|
1280 return index;
|
|
1281 }
|
|
1282 index++;
|
|
1283 }
|
|
1284 return size;
|
|
1285 }
|
|
1286
|
|
1287 // Determine if a Object is all whitespace
|
|
1288 private boolean isAllWhitespace(Object obj) {
|
|
1289 String str = null;
|
|
1290
|
|
1291 if (obj instanceof String) {
|
|
1292 str = (String) obj;
|
|
1293 }
|
|
1294 else if (obj instanceof Text) {
|
|
1295 str = ((Text) obj).getText();
|
|
1296 }
|
|
1297 else if (obj instanceof EntityRef) {
|
|
1298 return false;
|
|
1299 }
|
|
1300 else {
|
|
1301 return false;
|
|
1302 }
|
|
1303
|
|
1304 for (int i = 0; i < str.length(); i++) {
|
|
1305 if (!Verifier.isXMLWhitespace(str.charAt(i)))
|
|
1306 return false;
|
|
1307 }
|
|
1308 return true;
|
|
1309 }
|
|
1310
|
|
1311 // Determine if a string starts with a XML whitespace.
|
|
1312 private boolean startsWithWhite(String str) {
|
|
1313 if ((str != null) &&
|
|
1314 (str.length() > 0) &&
|
|
1315 Verifier.isXMLWhitespace(str.charAt(0))) {
|
|
1316 return true;
|
|
1317 }
|
|
1318 return false;
|
|
1319 }
|
|
1320
|
|
1321 // Determine if a string ends with a XML whitespace.
|
|
1322 private boolean endsWithWhite(String str) {
|
|
1323 if ((str != null) &&
|
|
1324 (str.length() > 0) &&
|
|
1325 Verifier.isXMLWhitespace(str.charAt(str.length() - 1))) {
|
|
1326 return true;
|
|
1327 }
|
|
1328 return false;
|
|
1329 }
|
|
1330
|
|
1331 /**
|
|
1332 * This will take the pre-defined entities in XML 1.0 and
|
|
1333 * convert their character representation to the appropriate
|
|
1334 * entity reference, suitable for XML attributes. It does not convert
|
|
1335 * the single quote (') because it's not necessary as the outputter
|
|
1336 * writes attributes surrounded by double-quotes.
|
|
1337 *
|
|
1338 * @param str <code>String</code> input to escape.
|
|
1339 * @return <code>String</code> with escaped content.
|
|
1340 * @throws IllegalArgumentException if an entity can not be escaped
|
|
1341 */
|
|
1342 public String escapeAttributeEntities(String str) {
|
|
1343 StringBuffer buffer;
|
|
1344 int ch, pos;
|
|
1345 String entity;
|
|
1346 EscapeStrategy strategy = currentFormat.escapeStrategy;
|
|
1347
|
|
1348 buffer = null;
|
|
1349 for (int i = 0; i < str.length(); i++) {
|
|
1350 ch = str.charAt(i);
|
|
1351 pos = i;
|
|
1352 switch(ch) {
|
|
1353 case '<' :
|
|
1354 entity = "<";
|
|
1355 break;
|
|
1356 case '>' :
|
|
1357 entity = ">";
|
|
1358 break;
|
|
1359 /*
|
|
1360 case '\'' :
|
|
1361 entity = "'";
|
|
1362 break;
|
|
1363 */
|
|
1364 case '\"' :
|
|
1365 entity = """;
|
|
1366 break;
|
|
1367 case '&' :
|
|
1368 entity = "&";
|
|
1369 break;
|
|
1370 case '\r' :
|
|
1371 entity = "
";
|
|
1372 break;
|
|
1373 case '\t' :
|
|
1374 entity = "	";
|
|
1375 break;
|
|
1376 case '\n' :
|
|
1377 entity = "
";
|
|
1378 break;
|
|
1379 default :
|
|
1380
|
|
1381 if (strategy.shouldEscape((char) ch)) {
|
|
1382 // Make sure what we are escaping is not the
|
|
1383 // Beginning of a multi-byte character.
|
|
1384 if (Verifier.isHighSurrogate((char) ch)) {
|
|
1385 // This is a the high of a surrogate pair
|
|
1386 i++;
|
|
1387 if (i < str.length()) {
|
|
1388 char low = str.charAt(i);
|
|
1389 if(!Verifier.isLowSurrogate(low)) {
|
|
1390 throw new IllegalDataException("Could not decode surrogate pair 0x" +
|
|
1391 Integer.toHexString(ch) + " / 0x" + Integer.toHexString(low));
|
|
1392 }
|
|
1393 ch = Verifier.decodeSurrogatePair((char) ch, low);
|
|
1394 } else {
|
|
1395 throw new IllegalDataException("Surrogate pair 0x" +
|
|
1396 Integer.toHexString(ch) + " truncated");
|
|
1397 }
|
|
1398 }
|
|
1399 entity = "&#x" + Integer.toHexString(ch) + ";";
|
|
1400 }
|
|
1401 else {
|
|
1402 entity = null;
|
|
1403 }
|
|
1404 break;
|
|
1405 }
|
|
1406 if (buffer == null) {
|
|
1407 if (entity != null) {
|
|
1408 // An entity occurred, so we'll have to use StringBuffer
|
|
1409 // (allocate room for it plus a few more entities).
|
|
1410 buffer = new StringBuffer(str.length() + 20);
|
|
1411 // Copy previous skipped characters and fall through
|
|
1412 // to pickup current character
|
|
1413 buffer.append(str.substring(0, pos));
|
|
1414 buffer.append(entity);
|
|
1415 }
|
|
1416 }
|
|
1417 else {
|
|
1418 if (entity == null) {
|
|
1419 buffer.append((char) ch);
|
|
1420 }
|
|
1421 else {
|
|
1422 buffer.append(entity);
|
|
1423 }
|
|
1424 }
|
|
1425 }
|
|
1426
|
|
1427 // If there were any entities, return the escaped characters
|
|
1428 // that we put in the StringBuffer. Otherwise, just return
|
|
1429 // the unmodified input string.
|
|
1430 return (buffer == null) ? str : buffer.toString();
|
|
1431 }
|
|
1432
|
|
1433
|
|
1434 /**
|
|
1435 * This will take the three pre-defined entities in XML 1.0
|
|
1436 * (used specifically in XML elements) and convert their character
|
|
1437 * representation to the appropriate entity reference, suitable for
|
|
1438 * XML element content.
|
|
1439 *
|
|
1440 * @param str <code>String</code> input to escape.
|
|
1441 * @return <code>String</code> with escaped content.
|
|
1442 * @throws IllegalArgumentException if an entity can not be escaped
|
|
1443 */
|
|
1444 public String escapeElementEntities(String str) {
|
|
1445 if (escapeOutput == false) return str;
|
|
1446
|
|
1447 StringBuffer buffer;
|
|
1448 int ch, pos;
|
|
1449 String entity;
|
|
1450 EscapeStrategy strategy = currentFormat.escapeStrategy;
|
|
1451
|
|
1452 buffer = null;
|
|
1453 for (int i = 0; i < str.length(); i++) {
|
|
1454 ch = str.charAt(i);
|
|
1455 pos = i;
|
|
1456 switch(ch) {
|
|
1457 case '<' :
|
|
1458 entity = "<";
|
|
1459 break;
|
|
1460 case '>' :
|
|
1461 entity = ">";
|
|
1462 break;
|
|
1463 case '&' :
|
|
1464 entity = "&";
|
|
1465 break;
|
|
1466 case '\r' :
|
|
1467 entity = "
";
|
|
1468 break;
|
|
1469 case '\n' :
|
|
1470 entity = currentFormat.lineSeparator;
|
|
1471 break;
|
|
1472 default :
|
|
1473
|
|
1474 if (strategy.shouldEscape((char) ch)) {
|
|
1475
|
|
1476 //make sure what we are escaping is not the
|
|
1477 //beginning of a multi-byte character.
|
|
1478 if(Verifier.isHighSurrogate((char) ch)) {
|
|
1479 //this is a the high of a surrogate pair
|
|
1480 i++;
|
|
1481 if (i < str.length()) {
|
|
1482 char low = str.charAt(i);
|
|
1483 if(!Verifier.isLowSurrogate(low)) {
|
|
1484 throw new IllegalDataException("Could not decode surrogate pair 0x" +
|
|
1485 Integer.toHexString(ch) + " / 0x" + Integer.toHexString(low));
|
|
1486 }
|
|
1487 ch = Verifier.decodeSurrogatePair((char) ch, low);
|
|
1488 } else {
|
|
1489 throw new IllegalDataException("Surrogate pair 0x" +
|
|
1490 Integer.toHexString(ch) + " truncated");
|
|
1491 }
|
|
1492 }
|
|
1493 entity = "&#x" + Integer.toHexString(ch) + ";";
|
|
1494 }
|
|
1495 else {
|
|
1496 entity = null;
|
|
1497 }
|
|
1498 break;
|
|
1499 }
|
|
1500 if (buffer == null) {
|
|
1501 if (entity != null) {
|
|
1502 // An entity occurred, so we'll have to use StringBuffer
|
|
1503 // (allocate room for it plus a few more entities).
|
|
1504 buffer = new StringBuffer(str.length() + 20);
|
|
1505 // Copy previous skipped characters and fall through
|
|
1506 // to pickup current character
|
|
1507 buffer.append(str.substring(0, pos));
|
|
1508 buffer.append(entity);
|
|
1509 }
|
|
1510 }
|
|
1511 else {
|
|
1512 if (entity == null) {
|
|
1513 buffer.append((char) ch);
|
|
1514 }
|
|
1515 else {
|
|
1516 buffer.append(entity);
|
|
1517 }
|
|
1518 }
|
|
1519 }
|
|
1520
|
|
1521 // If there were any entities, return the escaped characters
|
|
1522 // that we put in the StringBuffer. Otherwise, just return
|
|
1523 // the unmodified input string.
|
|
1524 return (buffer == null) ? str : buffer.toString();
|
|
1525 }
|
|
1526
|
|
1527 /**
|
|
1528 * Returns a copy of this XMLOutputter.
|
|
1529 */
|
|
1530 public Object clone() {
|
|
1531 // Implementation notes: Since all state of an XMLOutputter is
|
|
1532 // embodied in simple private instance variables, Object.clone
|
|
1533 // can be used. Note that since Object.clone is totally
|
|
1534 // broken, we must catch an exception that will never be
|
|
1535 // thrown.
|
|
1536 try {
|
|
1537 return super.clone();
|
|
1538 }
|
|
1539 catch (java.lang.CloneNotSupportedException e) {
|
|
1540 // even though this should never ever happen, it's still
|
|
1541 // possible to fool Java into throwing a
|
|
1542 // CloneNotSupportedException. If that happens, we
|
|
1543 // shouldn't swallow it.
|
|
1544 throw new RuntimeException(e.toString());
|
|
1545 }
|
|
1546 }
|
|
1547
|
|
1548 /**
|
|
1549 * Return a string listing of the settings for this
|
|
1550 * XMLOutputter instance.
|
|
1551 *
|
|
1552 * @return a string listing the settings for this XMLOutputter instance
|
|
1553 */
|
|
1554 public String toString() {
|
|
1555 StringBuffer buffer = new StringBuffer();
|
|
1556 for (int i = 0; i < userFormat.lineSeparator.length(); i++) {
|
|
1557 char ch = userFormat.lineSeparator.charAt(i);
|
|
1558 switch (ch) {
|
|
1559 case '\r': buffer.append("\\r");
|
|
1560 break;
|
|
1561 case '\n': buffer.append("\\n");
|
|
1562 break;
|
|
1563 case '\t': buffer.append("\\t");
|
|
1564 break;
|
|
1565 default: buffer.append("[" + ((int)ch) + "]");
|
|
1566 break;
|
|
1567 }
|
|
1568 }
|
|
1569
|
|
1570 return (
|
|
1571 "XMLOutputter[omitDeclaration = " + userFormat.omitDeclaration + ", " +
|
|
1572 "encoding = " + userFormat.encoding + ", " +
|
|
1573 "omitEncoding = " + userFormat.omitEncoding + ", " +
|
|
1574 "indent = '" + userFormat.indent + "'" + ", " +
|
|
1575 "expandEmptyElements = " + userFormat.expandEmptyElements + ", " +
|
|
1576 "lineSeparator = '" + buffer.toString() + "', " +
|
|
1577 "textMode = " + userFormat.mode + "]"
|
|
1578 );
|
|
1579 }
|
|
1580
|
|
1581 /**
|
|
1582 * Factory for making new NamespaceStack objects. The NamespaceStack
|
|
1583 * created is actually an inner class extending the package protected
|
|
1584 * NamespaceStack, as a way to make NamespaceStack "friendly" toward
|
|
1585 * subclassers.
|
|
1586 */
|
|
1587 private NamespaceStack createNamespaceStack() {
|
|
1588 // actually returns a XMLOutputter.NamespaceStack (see below)
|
|
1589 return new NamespaceStack();
|
|
1590 }
|
|
1591
|
|
1592 /**
|
|
1593 * Our own null subclass of NamespaceStack. This plays a little
|
|
1594 * trick with Java access protection. We want subclasses of
|
|
1595 * XMLOutputter to be able to override protected methods that
|
|
1596 * declare a NamespaceStack parameter, but we don't want to
|
|
1597 * declare the parent NamespaceStack class as public.
|
|
1598 */
|
|
1599 protected class NamespaceStack
|
|
1600 extends org.jdom.output.NamespaceStack
|
|
1601 {
|
|
1602 }
|
|
1603
|
|
1604 // Support method to print a name without using elt.getQualifiedName()
|
|
1605 // and thus avoiding a StringBuffer creation and memory churn
|
|
1606 private void printQualifiedName(Writer out, Element e) throws IOException {
|
|
1607 if (e.getNamespace().getPrefix().length() == 0) {
|
|
1608 out.write(e.getName());
|
|
1609 }
|
|
1610 else {
|
|
1611 out.write(e.getNamespace().getPrefix());
|
|
1612 out.write(':');
|
|
1613 out.write(e.getName());
|
|
1614 }
|
|
1615 }
|
|
1616
|
|
1617 // Support method to print a name without using att.getQualifiedName()
|
|
1618 // and thus avoiding a StringBuffer creation and memory churn
|
|
1619 private void printQualifiedName(Writer out, Attribute a) throws IOException {
|
|
1620 String prefix = a.getNamespace().getPrefix();
|
|
1621 if ((prefix != null) && (!prefix.equals(""))) {
|
|
1622 out.write(prefix);
|
|
1623 out.write(':');
|
|
1624 out.write(a.getName());
|
|
1625 }
|
|
1626 else {
|
|
1627 out.write(a.getName());
|
|
1628 }
|
|
1629 }
|
|
1630
|
|
1631 // * * * * * * * * * * Deprecated methods * * * * * * * * * *
|
|
1632
|
|
1633 /* The methods below here are deprecations of protected methods. We
|
|
1634 * don't usually deprecate protected methods, so they're commented out.
|
|
1635 * They're left here in case this mass deprecation causes people trouble.
|
|
1636 * Since we're getting close to 1.0 it's actually better for people to
|
|
1637 * raise issues early though.
|
|
1638 */
|
|
1639
|
|
1640 }
|