| Clover coverage report - dom4j - 1.6.1 |
|
 |
|
|
|||||||||||||||||||
| Source file | Conditionals | Statements | Methods | TOTAL | |||||||||||||||
| HTMLWriter.java | 29,5% | 37% | 37,5% | 35,5% |
|
||||||||||||||
| 1 | /* | |
| 2 | * Copyright 2001-2005 (C) MetaStuff, Ltd. All Rights Reserved. | |
| 3 | * | |
| 4 | * This software is open source. | |
| 5 | * See the bottom of this file for the licence. | |
| 6 | */ | |
| 7 | ||
| 8 | package org.dom4j.io; | |
| 9 | ||
| 10 | import java.io.IOException; | |
| 11 | import java.io.OutputStream; | |
| 12 | import java.io.StringWriter; | |
| 13 | import java.io.UnsupportedEncodingException; | |
| 14 | import java.io.Writer; | |
| 15 | import java.util.HashSet; | |
| 16 | import java.util.Iterator; | |
| 17 | import java.util.Set; | |
| 18 | import java.util.Stack; | |
| 19 | ||
| 20 | import org.dom4j.Document; | |
| 21 | import org.dom4j.DocumentHelper; | |
| 22 | import org.dom4j.Element; | |
| 23 | import org.dom4j.Entity; | |
| 24 | import org.dom4j.Node; | |
| 25 | ||
| 26 | import org.xml.sax.SAXException; | |
| 27 | ||
| 28 | /** | |
| 29 | * <p> | |
| 30 | * <code>HTMLWriter</code> takes a DOM4J tree and formats it to a stream as | |
| 31 | * HTML. This formatter is similar to XMLWriter but it outputs the text of CDATA | |
| 32 | * and Entity sections rather than the serialised format as in XML, it has an | |
| 33 | * XHTML mode, it retains whitespace in certain elements such as <PRE>, | |
| 34 | * and it supports certain elements which have no corresponding close tag such | |
| 35 | * as for <BR> and <P>. | |
| 36 | * </p> | |
| 37 | * | |
| 38 | * <p> | |
| 39 | * The OutputFormat passed in to the constructor is checked for isXHTML() and | |
| 40 | * isExpandEmptyElements(). See {@link OutputFormat OutputFormat}for details. | |
| 41 | * Here are the rules for <b>this class </b> based on an OutputFormat, "format", | |
| 42 | * passed in to the constructor: <br/><br/> | |
| 43 | * | |
| 44 | * <ul> | |
| 45 | * <li>If an element is in {@link #getOmitElementCloseSet() | |
| 46 | * getOmitElementCloseSet}, then it is treated specially: | |
| 47 | * | |
| 48 | * <ul> | |
| 49 | * <li>It never expands, since some browsers treat this as two separate | |
| 50 | * Horizontal Rules: <HR></HR></li> | |
| 51 | * <li>If {@link org.dom4j.io.OutputFormat#isXHTML() format.isXHTML()}, then | |
| 52 | * it has a space before the closing single-tag slash, since Netscape 4.x- | |
| 53 | * treats this: <HR /> as an element named "HR" with an attribute named | |
| 54 | * "/", but that's better than when it refuses to recognize this: <hr/> | |
| 55 | * which it thinks is an element named "HR/".</li> | |
| 56 | * </ul> | |
| 57 | * | |
| 58 | * </li> | |
| 59 | * <li>If {@link org.dom4j.io.OutputFormat#isXHTML() format.isXHTML()}, all | |
| 60 | * elements must have either a close element, or be a closed single tag.</li> | |
| 61 | * <li>If {@link org.dom4j.io.OutputFormat#isExpandEmptyElements() | |
| 62 | * format.isExpandEmptyElements()}() is true, all elements are expanded except | |
| 63 | * as above.</li> | |
| 64 | * </ul> | |
| 65 | * | |
| 66 | * <b>Examples </b> | |
| 67 | * </p> | |
| 68 | * | |
| 69 | * <p> | |
| 70 | * </p> | |
| 71 | * | |
| 72 | * <p> | |
| 73 | * If isXHTML == true, CDATA sections look like this: | |
| 74 | * | |
| 75 | * <PRE> | |
| 76 | * | |
| 77 | * <b><myelement><![CDATA[My data]]></myelement> </b> | |
| 78 | * | |
| 79 | * </PRE> | |
| 80 | * | |
| 81 | * Otherwise, they look like this: | |
| 82 | * | |
| 83 | * <PRE> | |
| 84 | * | |
| 85 | * <b><myelement>My data</myelement> </b> | |
| 86 | * | |
| 87 | * </PRE> | |
| 88 | * | |
| 89 | * </p> | |
| 90 | * | |
| 91 | * <p> | |
| 92 | * Basically, {@link OutputFormat.isXHTML() OutputFormat.isXHTML()} == | |
| 93 | * <code>true</code> will produce valid XML, while {@link | |
| 94 | * org.dom4j.io.OutputFormat#isExpandEmptyElements() | |
| 95 | * format.isExpandEmptyElements()} determines whether empty elements are | |
| 96 | * expanded if isXHTML is true, excepting the special HTML single tags. | |
| 97 | * </p> | |
| 98 | * | |
| 99 | * <p> | |
| 100 | * Also, HTMLWriter handles tags whose contents should be preformatted, that is, | |
| 101 | * whitespace-preserved. By default, this set includes the tags <PRE>, | |
| 102 | * <SCRIPT>, <STYLE>, and <TEXTAREA>, case insensitively. It | |
| 103 | * does not include <IFRAME>. Other tags, such as <CODE>, | |
| 104 | * <KBD>, <TT>, <VAR>, are usually rendered in a different | |
| 105 | * font in most browsers, but don't preserve whitespace, so they also don't | |
| 106 | * appear in the default list. HTML Comments are always whitespace-preserved. | |
| 107 | * However, the parser you use may store comments with linefeed-only text nodes | |
| 108 | * (\n) even if your platform uses another line.separator character, and | |
| 109 | * HTMLWriter outputs Comment nodes exactly as the DOM is set up by the parser. | |
| 110 | * See examples and discussion here: {@link#setPreformattedTags(java.util.Set) | |
| 111 | * setPreformattedTags} | |
| 112 | * </p> | |
| 113 | * | |
| 114 | * <p> | |
| 115 | * <b>Examples </b> | |
| 116 | * </p> | |
| 117 | * <blockquote> | |
| 118 | * <p> | |
| 119 | * <b>Pretty Printing </b> | |
| 120 | * </p> | |
| 121 | * | |
| 122 | * <p> | |
| 123 | * This example shows how to pretty print a string containing a valid HTML | |
| 124 | * document to a string. You can also just call the static methods of this | |
| 125 | * class: <br> | |
| 126 | * {@link #prettyPrintHTML(String) prettyPrintHTML(String)}or <br> | |
| 127 | * {@link #prettyPrintHTML(String,boolean,boolean,boolean,boolean) | |
| 128 | * prettyPrintHTML(String,boolean,boolean,boolean,boolean)} or, <br> | |
| 129 | * {@link #prettyPrintXHTML(String) prettyPrintXHTML(String)}for XHTML (note | |
| 130 | * the X) | |
| 131 | * </p> | |
| 132 | * | |
| 133 | * <pre> | |
| 134 | * String testPrettyPrint(String html) { | |
| 135 | * StringWriter sw = new StringWriter(); | |
| 136 | * OutputFormat format = OutputFormat.createPrettyPrint(); | |
| 137 | * // These are the default values for createPrettyPrint, | |
| 138 | * // so you needn't set them: | |
| 139 | * // format.setNewlines(true); | |
| 140 | * // format.setTrimText(true);</font> | |
| 141 | * format.setXHTML(true); | |
| 142 | * HTMLWriter writer = new HTMLWriter(sw, format); | |
| 143 | * Document document = DocumentHelper.parseText(html); | |
| 144 | * writer.write(document); | |
| 145 | * writer.flush(); | |
| 146 | * return sw.toString(); | |
| 147 | * } | |
| 148 | * </pre> | |
| 149 | * | |
| 150 | * <p> | |
| 151 | * This example shows how to create a "squeezed" document, but one that will | |
| 152 | * work in browsers even if the browser line length is limited. No newlines are | |
| 153 | * included, no extra whitespace at all, except where it it required by | |
| 154 | * {@link #setPreformattedTags(java.util.Set) setPreformattedTags}. | |
| 155 | * </p> | |
| 156 | * | |
| 157 | * <pre> | |
| 158 | * String testCrunch(String html) { | |
| 159 | * StringWriter sw = new StringWriter(); | |
| 160 | * OutputFormat format = OutputFormat.createPrettyPrint(); | |
| 161 | * format.setNewlines(false); | |
| 162 | * format.setTrimText(true); | |
| 163 | * format.setIndent(""); | |
| 164 | * format.setXHTML(true); | |
| 165 | * format.setExpandEmptyElements(false); | |
| 166 | * format.setNewLineAfterNTags(20); | |
| 167 | * org.dom4j.io.HTMLWriter writer = new HTMLWriter(sw, format); | |
| 168 | * org.dom4j.Document document = DocumentHelper.parseText(html); | |
| 169 | * writer.write(document); | |
| 170 | * writer.flush(); | |
| 171 | * return sw.toString(); | |
| 172 | * } | |
| 173 | * </pre> | |
| 174 | * | |
| 175 | * </blockquote> | |
| 176 | * | |
| 177 | * @author <a href="mailto:james.strachan@metastuff.com">James Strachan </a> | |
| 178 | * @author Laramie Crocker | |
| 179 | * @version $Revision: 1.21 $ | |
| 180 | */ | |
| 181 | public class HTMLWriter extends XMLWriter { | |
| 182 | private static String lineSeparator = System.getProperty("line.separator"); | |
| 183 | ||
| 184 | protected static final HashSet DEFAULT_PREFORMATTED_TAGS; | |
| 185 | ||
| 186 | static { | |
| 187 | // If you change this list, update the javadoc examples, above in the | |
| 188 | // class javadoc, in writeElement, and in setPreformattedTags(). | |
| 189 | 1 | DEFAULT_PREFORMATTED_TAGS = new HashSet(); |
| 190 | 1 | DEFAULT_PREFORMATTED_TAGS.add("PRE"); |
| 191 | 1 | DEFAULT_PREFORMATTED_TAGS.add("SCRIPT"); |
| 192 | 1 | DEFAULT_PREFORMATTED_TAGS.add("STYLE"); |
| 193 | 1 | DEFAULT_PREFORMATTED_TAGS.add("TEXTAREA"); |
| 194 | } | |
| 195 | ||
| 196 | protected static final OutputFormat DEFAULT_HTML_FORMAT; | |
| 197 | ||
| 198 | static { | |
| 199 | 1 | DEFAULT_HTML_FORMAT = new OutputFormat(" ", true); |
| 200 | 1 | DEFAULT_HTML_FORMAT.setTrimText(true); |
| 201 | 1 | DEFAULT_HTML_FORMAT.setSuppressDeclaration(true); |
| 202 | } | |
| 203 | ||
| 204 | private Stack formatStack = new Stack(); | |
| 205 | ||
| 206 | private String lastText = ""; | |
| 207 | ||
| 208 | private int tagsOuput = 0; | |
| 209 | ||
| 210 | // legal values are 0+, but -1 signifies lazy initialization. | |
| 211 | private int newLineAfterNTags = -1; | |
| 212 | ||
| 213 | private HashSet preformattedTags = DEFAULT_PREFORMATTED_TAGS; | |
| 214 | ||
| 215 | /** | |
| 216 | * Used to store the qualified element names which should have no close | |
| 217 | * element tag | |
| 218 | */ | |
| 219 | private HashSet omitElementCloseSet; | |
| 220 | ||
| 221 | 1 | public HTMLWriter(Writer writer) { |
| 222 | 1 | super(writer, DEFAULT_HTML_FORMAT); |
| 223 | } | |
| 224 | ||
| 225 | 5 | public HTMLWriter(Writer writer, OutputFormat format) { |
| 226 | 5 | super(writer, format); |
| 227 | } | |
| 228 | ||
| 229 | 0 | public HTMLWriter() throws UnsupportedEncodingException { |
| 230 | 0 | super(DEFAULT_HTML_FORMAT); |
| 231 | } | |
| 232 | ||
| 233 | 0 | public HTMLWriter(OutputFormat format) throws UnsupportedEncodingException { |
| 234 | 0 | super(format); |
| 235 | } | |
| 236 | ||
| 237 | 0 | public HTMLWriter(OutputStream out) throws UnsupportedEncodingException { |
| 238 | 0 | super(out, DEFAULT_HTML_FORMAT); |
| 239 | } | |
| 240 | ||
| 241 | 0 | public HTMLWriter(OutputStream out, OutputFormat format) |
| 242 | throws UnsupportedEncodingException { | |
| 243 | 0 | super(out, format); |
| 244 | } | |
| 245 | ||
| 246 | 0 | public void startCDATA() throws SAXException { |
| 247 | } | |
| 248 | ||
| 249 | 0 | public void endCDATA() throws SAXException { |
| 250 | } | |
| 251 | ||
| 252 | // Overloaded methods | |
| 253 | // added isXHTML() stuff so you get the CDATA brackets if you desire. | |
| 254 | 1 | protected void writeCDATA(String text) throws IOException { |
| 255 | // XXX: Should we escape entities? | |
| 256 | // writer.write( escapeElementEntities( text ) ); | |
| 257 | 1 | if (getOutputFormat().isXHTML()) { |
| 258 | 0 | super.writeCDATA(text); |
| 259 | } else { | |
| 260 | 1 | writer.write(text); |
| 261 | } | |
| 262 | ||
| 263 | 1 | lastOutputNodeType = Node.CDATA_SECTION_NODE; |
| 264 | } | |
| 265 | ||
| 266 | 0 | protected void writeEntity(Entity entity) throws IOException { |
| 267 | 0 | writer.write(entity.getText()); |
| 268 | 0 | lastOutputNodeType = Node.ENTITY_REFERENCE_NODE; |
| 269 | } | |
| 270 | ||
| 271 | 3 | protected void writeDeclaration() throws IOException { |
| 272 | } | |
| 273 | ||
| 274 | 4 | protected void writeString(String text) throws IOException { |
| 275 | /* | |
| 276 | * DOM stores \n at the end of text nodes that are newlines. This is | |
| 277 | * significant if we are in a PRE section. However, we only want to | |
| 278 | * output the system line.separator, not \n. This is a little brittle, | |
| 279 | * but this function appears to be called with these lineseparators as a | |
| 280 | * separate TEXT_NODE. If we are in a preformatted section, output the | |
| 281 | * right line.separator, otherwise ditch. If the single \n character is | |
| 282 | * not the text, then do the super thing to output the text. | |
| 283 | * | |
| 284 | * Also, we store the last text that was not a \n since it may be used | |
| 285 | * by writeElement in this class to line up preformatted tags. | |
| 286 | */ | |
| 287 | 4 | if (text.equals("\n")) { |
| 288 | 0 | if (!formatStack.empty()) { |
| 289 | 0 | super.writeString(lineSeparator); |
| 290 | } | |
| 291 | ||
| 292 | 0 | return; |
| 293 | } | |
| 294 | ||
| 295 | 4 | lastText = text; |
| 296 | ||
| 297 | 4 | if (formatStack.empty()) { |
| 298 | 4 | super.writeString(text.trim()); |
| 299 | } else { | |
| 300 | 0 | super.writeString(text); |
| 301 | } | |
| 302 | } | |
| 303 | ||
| 304 | /** | |
| 305 | * Overriden method to not close certain element names to avoid wierd | |
| 306 | * behaviour from browsers for versions up to 5.x | |
| 307 | * | |
| 308 | * @param qualifiedName | |
| 309 | * DOCUMENT ME! | |
| 310 | * | |
| 311 | * @throws IOException | |
| 312 | * DOCUMENT ME! | |
| 313 | */ | |
| 314 | 0 | protected void writeClose(String qualifiedName) throws IOException { |
| 315 | 0 | if (!omitElementClose(qualifiedName)) { |
| 316 | 0 | super.writeClose(qualifiedName); |
| 317 | } | |
| 318 | } | |
| 319 | ||
| 320 | 2 | protected void writeEmptyElementClose(String qualifiedName) |
| 321 | throws IOException { | |
| 322 | 2 | if (getOutputFormat().isXHTML()) { |
| 323 | // xhtml, always check with format object whether to expand or not. | |
| 324 | 0 | if (omitElementClose(qualifiedName)) { |
| 325 | // it was a special omit tag, do it the XHTML way: "<br/>", | |
| 326 | // ignoring the expansion option, since <br></br> is OK XML, | |
| 327 | // but produces twice the linefeeds desired in the browser. | |
| 328 | // for netscape 4.7, though all are fine with it, write a space | |
| 329 | // before the close slash. | |
| 330 | 0 | writer.write(" />"); |
| 331 | } else { | |
| 332 | 0 | super.writeEmptyElementClose(qualifiedName); |
| 333 | } | |
| 334 | } else { | |
| 335 | // html, not xhtml | |
| 336 | 2 | if (omitElementClose(qualifiedName)) { |
| 337 | // it was a special omit tag, do it the old html way: "<br>". | |
| 338 | 1 | writer.write(">"); |
| 339 | } else { | |
| 340 | // it was NOT a special omit tag, check with format object | |
| 341 | // whether to expand or not. | |
| 342 | 1 | super.writeEmptyElementClose(qualifiedName); |
| 343 | } | |
| 344 | } | |
| 345 | } | |
| 346 | ||
| 347 | 2 | protected boolean omitElementClose(String qualifiedName) { |
| 348 | 2 | return internalGetOmitElementCloseSet().contains( |
| 349 | qualifiedName.toUpperCase()); | |
| 350 | } | |
| 351 | ||
| 352 | 2 | private HashSet internalGetOmitElementCloseSet() { |
| 353 | 2 | if (omitElementCloseSet == null) { |
| 354 | 2 | omitElementCloseSet = new HashSet(); |
| 355 | 2 | loadOmitElementCloseSet(omitElementCloseSet); |
| 356 | } | |
| 357 | ||
| 358 | 2 | return omitElementCloseSet; |
| 359 | } | |
| 360 | ||
| 361 | // If you change this, change the javadoc for getOmitElementCloseSet. | |
| 362 | 2 | protected void loadOmitElementCloseSet(Set set) { |
| 363 | 2 | set.add("AREA"); |
| 364 | 2 | set.add("BASE"); |
| 365 | 2 | set.add("BR"); |
| 366 | 2 | set.add("COL"); |
| 367 | 2 | set.add("HR"); |
| 368 | 2 | set.add("IMG"); |
| 369 | 2 | set.add("INPUT"); |
| 370 | 2 | set.add("LINK"); |
| 371 | 2 | set.add("META"); |
| 372 | 2 | set.add("P"); |
| 373 | 2 | set.add("PARAM"); |
| 374 | } | |
| 375 | ||
| 376 | // let the people see the set, but not modify it. | |
| 377 | ||
| 378 | /** | |
| 379 | * A clone of the Set of elements that can have their close-tags omitted. By | |
| 380 | * default it should be "AREA", "BASE", "BR", "COL", "HR", "IMG", "INPUT", | |
| 381 | * "LINK", "META", "P", "PARAM" | |
| 382 | * | |
| 383 | * @return A clone of the Set. | |
| 384 | */ | |
| 385 | 0 | public Set getOmitElementCloseSet() { |
| 386 | 0 | return (Set) (internalGetOmitElementCloseSet().clone()); |
| 387 | } | |
| 388 | ||
| 389 | /** | |
| 390 | * To use the empty set, pass an empty Set, or null: | |
| 391 | * | |
| 392 | * <pre> | |
| 393 | * | |
| 394 | * | |
| 395 | * setOmitElementCloseSet(new HashSet()); | |
| 396 | * or | |
| 397 | * setOmitElementCloseSet(null); | |
| 398 | * | |
| 399 | * | |
| 400 | * </pre> | |
| 401 | * | |
| 402 | * @param newSet | |
| 403 | * DOCUMENT ME! | |
| 404 | */ | |
| 405 | 0 | public void setOmitElementCloseSet(Set newSet) { |
| 406 | // resets, and safely empties it out if newSet is null. | |
| 407 | 0 | omitElementCloseSet = new HashSet(); |
| 408 | ||
| 409 | 0 | if (newSet != null) { |
| 410 | 0 | omitElementCloseSet = new HashSet(); |
| 411 | ||
| 412 | 0 | Object aTag; |
| 413 | 0 | Iterator iter = newSet.iterator(); |
| 414 | ||
| 415 | 0 | while (iter.hasNext()) { |
| 416 | 0 | aTag = iter.next(); |
| 417 | ||
| 418 | 0 | if (aTag != null) { |
| 419 | 0 | omitElementCloseSet.add(aTag.toString().toUpperCase()); |
| 420 | } | |
| 421 | } | |
| 422 | } | |
| 423 | } | |
| 424 | ||
| 425 | /** | |
| 426 | * @see #setPreformattedTags(java.util.Set) setPreformattedTags | |
| 427 | */ | |
| 428 | 0 | public Set getPreformattedTags() { |
| 429 | 0 | return (Set) (preformattedTags.clone()); |
| 430 | } | |
| 431 | ||
| 432 | /** | |
| 433 | * <p> | |
| 434 | * Override the default set, which includes PRE, SCRIPT, STYLE, and | |
| 435 | * TEXTAREA, case insensitively. | |
| 436 | * </p> | |
| 437 | * | |
| 438 | * <p> | |
| 439 | * <b>Setting Preformatted Tags </b> | |
| 440 | * </p> | |
| 441 | * | |
| 442 | * <p> | |
| 443 | * Pass in a Set of Strings, one for each tag name that should be treated | |
| 444 | * like a PRE tag. You may pass in null or an empty Set to assign the empty | |
| 445 | * set, in which case no tags will be treated as preformatted, except that | |
| 446 | * HTML Comments will continue to be preformatted. If a tag is included in | |
| 447 | * the set of preformatted tags, all whitespace within the tag will be | |
| 448 | * preserved, including whitespace on the same line preceding the close tag. | |
| 449 | * This will generally make the close tag not line up with the start tag, | |
| 450 | * but it preserves the intention of the whitespace within the tag. | |
| 451 | * </p> | |
| 452 | * | |
| 453 | * <p> | |
| 454 | * The browser considers leading whitespace before the close tag to be | |
| 455 | * significant, but leading whitespace before the open tag to be | |
| 456 | * insignificant. For example, if the HTML author doesn't put the close | |
| 457 | * TEXTAREA tag flush to the left margin, then the TEXTAREA control in the | |
| 458 | * browser will have spaces on the last line inside the control. This may be | |
| 459 | * the HTML author's intent. Similarly, in a PRE, the browser treats a | |
| 460 | * flushed left close PRE tag as different from a close tag with leading | |
| 461 | * whitespace. Again, this must be left up to the HTML author. | |
| 462 | * </p> | |
| 463 | * | |
| 464 | * <p> | |
| 465 | * <b>Examples </b> | |
| 466 | * </p> | |
| 467 | * <blockquote> | |
| 468 | * <p> | |
| 469 | * Here is an example of how you can set the PreformattedTags list using | |
| 470 | * setPreformattedTags to include IFRAME, as well as the default set, if you | |
| 471 | * have an instance of this class named myHTMLWriter: | |
| 472 | * | |
| 473 | * <pre> | |
| 474 | * Set current = myHTMLWriter.getPreformattedTags(); | |
| 475 | * current.add("IFRAME"); | |
| 476 | * myHTMLWriter.setPreformattedTags(current); | |
| 477 | * | |
| 478 | * //The set is now <b>PRE, SCRIPT, STYLE, TEXTAREA, IFRAME</b> | |
| 479 | * | |
| 480 | * | |
| 481 | * </pre> | |
| 482 | * | |
| 483 | * Similarly, you can simply replace it with your own: | |
| 484 | * | |
| 485 | * <pre> | |
| 486 | * | |
| 487 | * | |
| 488 | * HashSet newset = new HashSet(); | |
| 489 | * newset.add("PRE"); | |
| 490 | * newset.add("TEXTAREA"); | |
| 491 | * myHTMLWriter.setPreformattedTags(newset); | |
| 492 | * | |
| 493 | * //The set is now <b>{PRE, TEXTAREA}</b> | |
| 494 | * | |
| 495 | * | |
| 496 | * & |