Source for gnu.javax.swing.text.html.parser.htmlValidator

   1: /* tagStack.java -- The HTML tag stack.
   2:    Copyright (C) 2005 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: 
  39: package gnu.javax.swing.text.html.parser;
  40: 
  41: import gnu.javax.swing.text.html.parser.models.node;
  42: import gnu.javax.swing.text.html.parser.models.transformer;
  43: 
  44: import java.util.BitSet;
  45: import java.util.Enumeration;
  46: import java.util.LinkedList;
  47: import java.util.ListIterator;
  48: 
  49: import javax.swing.text.SimpleAttributeSet;
  50: import javax.swing.text.html.HTML;
  51: import javax.swing.text.html.parser.*;
  52: 
  53: /**
  54:  * <p>The HTML content validator, is responsible for opening and
  55:  * closing elements with optional start/end tags, detecting
  56:  * the wrongly placed html tags and reporting errors. The working instance
  57:  * is the inner class inside the {@link javax.swing.text.html.parser.Parser }
  58:  * </p>
  59:  * <p>This class could potentially
  60:  * provide basis for automated closing and insertion of the html tags,
  61:  * correcting the found html errors.
  62:  * @author Audrius Meskauskas, Lithuania (AudriusA@Bioinformatics.org)
  63:  */
  64: public abstract class htmlValidator
  65: {
  66:   /**
  67:    * The tag reference, holding additional information that the tag
  68:    * has been forcibly closed.
  69:    */
  70:   protected class hTag
  71:   {
  72:     protected final Element element;
  73:     protected final HTML.Tag tag;
  74:     protected final TagElement tgElement;
  75:     protected boolean forcibly_closed;
  76:     protected node validationTrace;
  77: 
  78:     protected hTag(TagElement an_element)
  79:     {
  80:       element = an_element.getElement();
  81:       tag = an_element.getHTMLTag();
  82:       tgElement = an_element;
  83: 
  84:       if (element.content != null)
  85:         validationTrace = transformer.transform(element.content, dtd);
  86:     }
  87: 
  88:     /**
  89:      * This is called when the tag must be forcibly closed because
  90:      * it would make the newly appearing tag invalid.
  91:      * The parser is not notified about such event (just the error
  92:      * is reported). For such tags, the closing message does not
  93:      * appear when later reaching the end of stream. The exception is
  94:      * the &lt;head&gt; tag: the parser is notified about its silent closing
  95:      * when &lt;body&gt; or other html content appears.
  96:      */
  97:     protected void forciblyCloseDueContext()
  98:     {
  99:       forcibly_closed = true;
 100:     }
 101: 
 102:     /**
 103:      * This is called when the tag must be forcibly closed after
 104:      * reaching the end of stream. The parser is notified as if
 105:      * closing the tag explicitly.
 106:      */
 107:     protected void forciblyCloseDueEndOfStream()
 108:     {
 109:       forcibly_closed = true;
 110:       handleSupposedEndTag(element);
 111:     }
 112:   }
 113: 
 114:   /**
 115:    * The DTD, providing information about the valid document structure.
 116:    */
 117:   protected final DTD dtd;
 118: 
 119:   /**
 120:   * The stack, holding the current tag context.
 121:   */
 122:   protected final LinkedList stack = new LinkedList();
 123: 
 124:   /**
 125:    * Creates a new tag stack, using the given DTD.
 126:    * @param a_dtd A DTD, providing the information about the valid
 127:    * tag content.
 128:    */
 129:   public htmlValidator(DTD a_dtd)
 130:   {
 131:     dtd = a_dtd;
 132:   }
 133: 
 134:   /**
 135:    * Close all opened tags (called at the end of parsing).
 136:    */
 137:   public void closeAll()
 138:   {
 139:     hTag h;
 140:     while (!stack.isEmpty())
 141:       {
 142:         h = (hTag) stack.getLast();
 143:         if (!h.forcibly_closed && !h.element.omitEnd())
 144:           s_error("Unclosed <" + h.tag + ">, closing at the end of stream");
 145: 
 146:         handleSupposedEndTag(h.element);
 147: 
 148:         closeTag(h.tgElement);
 149:       }
 150:   }
 151: 
 152:   /**
 153:    * Remove the given tag from the stack or (if found) from the list
 154:    * of the forcibly closed tags.
 155:    */
 156:   public void closeTag(TagElement tElement)
 157:   {
 158:     HTML.Tag tag = tElement.getHTMLTag();
 159:     hTag x;
 160:     hTag close;
 161: 
 162:     if (!stack.isEmpty())
 163:       {
 164:         ListIterator iter = stack.listIterator(stack.size());
 165: 
 166:         while (iter.hasPrevious())
 167:           {
 168:             x = (hTag) iter.previous();
 169:             if (tag.equals(x.tag))
 170:               {
 171:                 if (x.forcibly_closed && !x.element.omitEnd())
 172:                   s_error("The tag <" + x.tag +
 173:                           "> has already been forcibly closed"
 174:                          );
 175: 
 176: 
 177:                 // If the tag has a content model defined, forcibly close all
 178:                 // tags that were opened after the tag being currently closed.
 179:                 closing: 
 180:                 if (x.element.content != null)
 181:                   {
 182:                     iter = stack.listIterator(stack.size());
 183:                     while (iter.hasPrevious())
 184:                       {
 185:                         close = (hTag) iter.previous();
 186:                         if (close == x)
 187:                           break closing;
 188:                         handleSupposedEndTag(close.element);
 189:                         iter.remove();
 190:                       }
 191:                   }
 192: 
 193:                 stack.remove(x);
 194:                 return;
 195:               }
 196:           }
 197:       }
 198:     s_error("Closing unopened <" + tag + ">");
 199:   }
 200: 
 201:   /**
 202:    * Add the given HTML tag to the stack of the opened tags. Forcibly closes
 203:    * all tags in the stack that does not allow this tag in they content (error
 204:    * is reported).
 205:    * @param element
 206:    */
 207:   public void openTag(TagElement tElement, htmlAttributeSet parameters)
 208:   {
 209:     // If this is a fictional call, the message from the parser
 210:     // has recursively returned - ignore.
 211:     if (tElement.fictional())
 212:       return;
 213: 
 214:     validateParameters(tElement, parameters);
 215: 
 216:     // If the stack is empty, start from HTML
 217:     if (stack.isEmpty() && tElement.getHTMLTag() != HTML.Tag.HTML)
 218:       {
 219:         Element html = dtd.getElement(HTML.Tag.HTML.toString());
 220:         openFictionalTag(html);
 221:       }
 222: 
 223:     Object v = tagIsValidForContext(tElement);
 224:     if (v != Boolean.TRUE)
 225:       {
 226:         // The tag is not valid for context, the content
 227:         // model suggest to open another tag.
 228:         if (v instanceof Element)
 229:           {
 230:             int n = 0;
 231:             while (v instanceof Element && (n++ < 100))
 232:               {
 233:                 Element fe = (Element) v;
 234: 
 235:                 // notify the content model that we add the proposed tag
 236:                 getCurrentContentModel().show(fe);
 237:                 openFictionalTag(fe);
 238: 
 239:                 Object vv = tagIsValidForContext(tElement);
 240:                 if (vv instanceof Element) // One level of nesting is supported.
 241:                   {
 242:                     openFictionalTag((Element) vv);
 243: 
 244:                     Object vx = tagIsValidForContext(tElement);
 245:                     if (vx instanceof Element)
 246:                       openFictionalTag((Element) vx);
 247:                   }
 248:                 else if (vv == Boolean.FALSE)
 249:                   {
 250:                     // The tag is still not valid for the current
 251:                     // content after opening a fictional element.
 252:                     if (fe.omitEnd())
 253:                       {
 254:                         // close the previously opened fictional tag.
 255:                         closeLast();
 256:                         vv = tagIsValidForContext(tElement);
 257:                         if (vv instanceof Element)
 258: 
 259:                           // another tag was suggested by the content model
 260:                           openFictionalTag((Element) vv);
 261:                       }
 262:                   }
 263:                 v = tagIsValidForContext(tElement);
 264:               }
 265:           }
 266:         else // If the current element has the optional end tag, close it.
 267:           {
 268:             if (!stack.isEmpty())
 269:               {
 270:                 closing: 
 271:                 do
 272:                   {
 273:                     hTag last = (hTag) stack.getLast();
 274:                     if (last.element.omitEnd())
 275:                       {
 276:                         closeLast();
 277:                         v = tagIsValidForContext(tElement);
 278:                         if (v instanceof Element) // another tag was suggested by the content model
 279:                           {
 280:                             openFictionalTag((Element) v);
 281:                             break closing;
 282:                           }
 283:                       }
 284:                     else
 285:                       break closing;
 286:                   }
 287:                 while (v == Boolean.FALSE && !stack.isEmpty());
 288:               }
 289:           }
 290:       }
 291: 
 292:     stack.add(new hTag(tElement));
 293:   }
 294: 
 295:   /**
 296:    * Clear the stack.
 297:    */
 298:   public void restart()
 299:   {
 300:     stack.clear();
 301:   }
 302: 
 303:   /**
 304:    * Check if this tag is valid for the current context.
 305:    * Return Boolean.True if it is OK, Boolean.False
 306:    * if it is surely not OK or the Element that the
 307:    * content model recommends to insert making the situation
 308:    * ok. If Boolean.True is returned, the content model current
 309:    * position is moved forward. Otherwise this position remains
 310:    * the same.
 311:    * @param tElement
 312:    * @return
 313:    */
 314:   public Object tagIsValidForContext(TagElement tElement)
 315:   {
 316:     // Check the current content model, if one is available.
 317:     node cv = getCurrentContentModel();
 318: 
 319:     if (cv != null)
 320:       return cv.show(tElement.getElement());
 321: 
 322:     // Check exclusions and inclusions.
 323:     ListIterator iter = stack.listIterator(stack.size());
 324:     hTag t;
 325:     final int idx = tElement.getElement().index;
 326: 
 327:     // Check only known tags.
 328:     if (idx >= 0)
 329:       {
 330:         BitSet inclusions = new BitSet();
 331:         while (iter.hasPrevious())
 332:           {
 333:             t = (hTag) iter.previous();
 334:             if (!t.forcibly_closed)
 335:               {
 336:                 if (t.element.exclusions != null &&
 337:                     t.element.exclusions.get(idx)
 338:                    )
 339:                   return Boolean.FALSE;
 340: 
 341:                 if (t.element.inclusions != null)
 342:                   inclusions.or(t.element.inclusions);
 343:               }
 344:           }
 345:         if (!inclusions.get(idx))
 346:           return Boolean.FALSE;
 347:       }
 348:     return Boolean.TRUE;
 349:   }
 350: 
 351:   /**
 352:    * Validate tag without storing in into the tag stack. This is called
 353:    * for the empty tags and results the subsequent calls to the openTag
 354:    * and closeTag.
 355:    */
 356:   public void validateTag(TagElement tElement, htmlAttributeSet parameters)
 357:   {
 358:     openTag(tElement, parameters);
 359:     closeTag(tElement);
 360:   }
 361: 
 362:   /**
 363:    * Check for mandatory elements, subsequent to the last tag:
 364:    * @param tElement The element that will be inserted next.
 365:    */
 366:   protected void checkContentModel(TagElement tElement, boolean first)
 367:   {
 368:     if (stack.isEmpty())
 369:       return;
 370: 
 371:     hTag last = (hTag) stack.getLast();
 372:     if (last.validationTrace == null)
 373:       return;
 374: 
 375:     Object r = last.validationTrace.show(tElement.getElement());
 376:     if (r == Boolean.FALSE)
 377:       s_error("The <" + last.element + "> does not match the content model " +
 378:               last.validationTrace
 379:              );
 380:     else if (r instanceof Element) // The content model recommends insertion of this element
 381:       {
 382:         if (!first)
 383:           closeTag(last.tgElement);
 384:         handleSupposedStartTag((Element) r);
 385:         openTag(new TagElement((Element) r), null);
 386:       }
 387:   }
 388: 
 389:   /**
 390:    * The method is called when the tag must be closed because
 391:    * it does not allow the subsequent elements inside its context
 392:    * or the end of stream has been reached. The parser is only
 393:    * informed if the element being closed does not require the
 394:    * end tag (the "omitEnd" flag is set).
 395:    * The closing message must be passed to the parser mechanism
 396:    * before passing message about the opening the next tag.
 397:    *
 398:    * @param element The tag being fictionally (forcibly) closed.
 399:    */
 400:   protected abstract void handleSupposedEndTag(Element element);
 401: 
 402:   /**
 403:    * The method is called when the validator decides to open the
 404:    * tag on its own initiative. This may happen if the content model
 405:    * includes the element with the optional (supposed) start tag.
 406:    *
 407:    * @param element The tag being opened.
 408:    */
 409:   protected abstract void handleSupposedStartTag(Element element);
 410: 
 411:   /**
 412:    * Handles the error message. This method must be overridden to pass
 413:    * the message where required.
 414:    * @param msg The message text.
 415:    */
 416:   protected abstract void s_error(String msg);
 417: 
 418:   /**
 419:    * Validate the parameters, report the error if the given parameter is
 420:    * not in the parameter set, valid for the given attribute. The information
 421:    * about the valid parameter set is taken from the Element, enclosed
 422:    * inside the tag. The method does not validate the default parameters.
 423:    * @param tag The tag
 424:    * @param parameters The parameters of this tag.
 425:    */
 426:   protected void validateParameters(TagElement tag, htmlAttributeSet parameters)
 427:   {
 428:     if (parameters == null ||
 429:         parameters == htmlAttributeSet.EMPTY_HTML_ATTRIBUTE_SET ||
 430:         parameters == SimpleAttributeSet.EMPTY
 431:        )
 432:       return;
 433: 
 434:     Enumeration enumeration = parameters.getAttributeNames();
 435: 
 436:     while (enumeration.hasMoreElements())
 437:       {
 438:         validateAttribute(tag, parameters, enumeration);
 439:       }
 440: 
 441:     // Check for missing required values.
 442:     AttributeList a = tag.getElement().getAttributes();
 443: 
 444:     while (a != null)
 445:       {
 446:         if (a.getModifier() == DTDConstants.REQUIRED)
 447:           if (parameters.getAttribute(a.getName()) == null)
 448:             {
 449:               s_error("Missing required attribute '" + a.getName() + "' for <" +
 450:                       tag.getHTMLTag() + ">"
 451:                      );
 452:             }
 453:         a = a.next;
 454:       }
 455:   }
 456: 
 457:   private node getCurrentContentModel()
 458:   {
 459:     if (!stack.isEmpty())
 460:       {
 461:         hTag last = (hTag) stack.getLast();
 462:         return last.validationTrace;
 463:       }
 464:     else
 465:       return null;
 466:   }
 467: 
 468:   private void closeLast()
 469:   {
 470:     handleSupposedEndTag(((hTag) stack.getLast()).element);
 471:     stack.removeLast();
 472:   }
 473: 
 474:   private void openFictionalTag(Element e)
 475:   {
 476:     handleSupposedStartTag(e);
 477:     stack.add(new hTag(new TagElement(e, true)));
 478:     if (!e.omitStart())
 479:       s_error("<" + e + "> is expected (supposing it)");
 480:   }
 481: 
 482:   private void validateAttribute(TagElement tag, htmlAttributeSet parameters,
 483:                                  Enumeration enumeration
 484:                                 )
 485:   {
 486:     Object foundAttribute;
 487:     AttributeList dtdAttribute;
 488:     foundAttribute = enumeration.nextElement();
 489:     dtdAttribute = tag.getElement().getAttribute(foundAttribute.toString());
 490:     if (dtdAttribute == null)
 491:       {
 492:         StringBuffer valid =
 493:           new StringBuffer("The tag <" + tag.getHTMLTag() +
 494:                            "> cannot contain the attribute '" + foundAttribute +
 495:                            "'. The valid attributes for this tag are: "
 496:                           );
 497: 
 498:         AttributeList a = tag.getElement().getAttributes();
 499: 
 500:         while (a != null)
 501:           {
 502:             valid.append(a.name.toUpperCase());
 503:             valid.append(' ');
 504:             a = a.next;
 505:           }
 506:         s_error(valid.toString());
 507:       }
 508: 
 509:     else
 510:       {
 511:         String value = parameters.getAttribute(foundAttribute).toString();
 512: 
 513:         if (dtdAttribute.type == DTDConstants.NUMBER)
 514:           validateNumberAttribute(tag, foundAttribute, value);
 515: 
 516:         if (dtdAttribute.type == DTDConstants.NAME ||
 517:             dtdAttribute.type == DTDConstants.ID
 518:            )
 519:           validateNameOrIdAttribute(tag, foundAttribute, value);
 520: 
 521:         if (dtdAttribute.values != null)
 522:           validateAttributeWithValueList(tag, foundAttribute, dtdAttribute,
 523:                                          value
 524:                                         );
 525:       }
 526:   }
 527: 
 528:   private void validateAttributeWithValueList(TagElement tag,
 529:                                               Object foundAttribute,
 530:                                               AttributeList dtdAttribute,
 531:                                               String value
 532:                                              )
 533:   {
 534:     if (!dtdAttribute.values.contains(value.toLowerCase()) &&
 535:         !dtdAttribute.values.contains(value.toUpperCase())
 536:        )
 537:       {
 538:         StringBuffer valid;
 539:         if (dtdAttribute.values.size() == 1)
 540:           valid =
 541:             new StringBuffer("The attribute '" + foundAttribute +
 542:                              "' of the tag <" + tag.getHTMLTag() +
 543:                              "> cannot have the value '" + value +
 544:                              "'. The only valid value is "
 545:                             );
 546:         else
 547:           valid =
 548:             new StringBuffer("The attribute '" + foundAttribute +
 549:                              "' of the tag <" + tag.getHTMLTag() +
 550:                              "> cannot have the value '" + value + "'. The " +
 551:                              dtdAttribute.values.size() +
 552:                              " valid values are: "
 553:                             );
 554: 
 555:         Enumeration vv = dtdAttribute.values.elements();
 556:         while (vv.hasMoreElements())
 557:           {
 558:             valid.append('"');
 559:             valid.append(vv.nextElement());
 560:             valid.append("\"  ");
 561:           }
 562:         s_error(valid.toString());
 563:       }
 564:   }
 565: 
 566:   private void validateNameOrIdAttribute(TagElement tag, Object foundAttribute,
 567:                                          String value
 568:                                         )
 569:   {
 570:     boolean ok = true;
 571: 
 572:     if (!Character.isLetter(value.charAt(0)))
 573:       ok = false;
 574: 
 575:     char c;
 576:     for (int i = 0; i < value.length(); i++)
 577:       {
 578:         c = value.charAt(i);
 579:         if (!(
 580:               Character.isLetter(c) || Character.isDigit(c) ||
 581:               "".indexOf(c) >= 0
 582:             )
 583:            )
 584:           ok = false;
 585:       }
 586:     if (!ok)
 587:       s_error("The '" + foundAttribute + "' attribute of the tag <" +
 588:               tag.getHTMLTag() + "> must start from letter and consist of " +
 589:               "letters, digits, hypens, colons, underscores and periods. " +
 590:               "It cannot be '" + value + "'"
 591:              );
 592:   }
 593: 
 594:   private void validateNumberAttribute(TagElement tag, Object foundAttribute,
 595:                                        String value
 596:                                       )
 597:   {
 598:     try
 599:       {
 600:         Integer.parseInt(value);
 601:       }
 602:     catch (NumberFormatException ex)
 603:       {
 604:         s_error("The '" + foundAttribute + "' attribute of the tag <" +
 605:                 tag.getHTMLTag() + "> must be a valid number and not '" +
 606:                 value + "'"
 607:                );
 608:       }
 609:   }
 610: }