HTMLFormatter


1   /*
2    * Enhydra Java Application Server Project
3    * 
4    * The contents of this file are subject to the Enhydra Public License
5    * Version 1.1 (the "License"); you may not use this file except in
6    * compliance with the License. You may obtain a copy of the License on
7    * the Enhydra web site ( http://www.enhydra.org/ ).
8    * 
9    * Software distributed under the License is distributed on an "AS IS"
10   * basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See 
11   * the License for the specific terms governing rights and limitations
12   * under the License.
13   * 
14   * The Initial Developer of the Enhydra Application Server is Lutris
15   * Technologies, Inc. The Enhydra Application Server and portions created
16   * by Lutris Technologies, Inc. are Copyright Lutris Technologies, Inc.
17   * All Rights Reserved.
18   * 
19   * Contributor(s):
20   * 
21   * $Id: HTMLFormatter.java,v 1.7 2005/01/26 08:29:24 jkjome Exp $
22   */
23  
24  package org.enhydra.xml.io;
25  
26  import java.io.IOException  ;
27  
28  import org.enhydra.xml.dom.DOMAccess;
29  import org.w3c.dom.Attr  ;
30  import org.w3c.dom.CDATASection  ;
31  import org.w3c.dom.Document  ;
32  import org.w3c.dom.DocumentFragment  ;
33  import org.w3c.dom.DocumentType  ;
34  import org.w3c.dom.Element  ;
35  import org.w3c.dom.Node  ;
36  import org.w3c.dom.ProcessingInstruction  ;
37  import org.w3c.dom.Text  ;
38  
39  /*
40   * FIXME:
41   *  - Need to check for optional attributes.
42   *  - Give control over newline in OutputOptions or maybe use a PrintWriter???
43   *  - Need option to include HTML 4.0 entity references.
44   *  - script and style child special handling.
45   *  - Need to output HTML header.
46   */
47  
48  
49  /**
50   * Formatter for outputting a HTML DOM as a HTML text document.
51   */
52  final class HTMLFormatter extends BaseDOMFormatter implements Formatter {
53      /**
54       * Default XML encoding.
55       */
56      private static final String   DEFAULT_XML_ENCODING = "ISO-8859-1";
57  
58      /**
59       * Table use to optimized checking for characters that should be
60       * represented as entity references.
61       */
62      private static final boolean[] fEntityQuickCheck
63          = new boolean[MAX_ENTITY_QUICK_CHECK_CHAR+1];
64  
65      /**
66       * Should SPAN ID attributes be dropped?
67       */
68      private final boolean fDropSpanIds;
69  
70      /**
71       * Flag that indicates the ID attribute should be dropped for the current
72       * element.
73       */
74      private boolean fDropThisId;
75      
76      /**
77       * Flag that indicates wether to use all named entites for all HTML 4.0 character
78       * entities or not.
79       */
80      private boolean fUseHTML4Entities;
81  
82      /**
83       * Nesting count for elements that don't have their content formatted.
84       * This is done for  script and style elements.  The contents of these
85       * elements are outputted as-is.  Its not legal for them to nest, but a
86       * count is used to keep code from being confused by a broken DOM.
87       */
88      private int fNoFormatNestCount;
89  
90      /**
91       * Indicates then a text has just been handled
92       */
93      private boolean fHandleText = false;
94  
95      /**
96       * Indicates the next Sibling is a Text node
97       */
98      private boolean fNextSiblingText = false;
99  
100     /**
101      * Static constructor.
102      */
103     static {
104         for (char ch = 0; ch <= MAX_ENTITY_QUICK_CHECK_CHAR; ch++) {
105             fEntityQuickCheck[ch] = (HTMLEntities.charToEntity(ch) != null);
106         }
107     }
108 
109     /**
110      * Constructor.
111      */
112     public HTMLFormatter(Node   node,
113                          OutputOptions outputOptions,
114                          boolean forPreFormatting) {
115         super(node, outputOptions, forPreFormatting, DEFAULT_XML_ENCODING, fEntityQuickCheck);
116         fDropSpanIds = fOptions.getDropHtmlSpanIds();
117         fUseHTML4Entities = fOptions.getUseHTML4Entities();
118     }
119 
120     /**
121      * Get the default OutputOptions for a document formatter with this
122      * formatter.  The encoding will not be set, which signals to use the
123      * default encoding.
124      */
125     static OutputOptions getDefaultOutputOptions() {
126         return new OutputOptions();  // Nothing special
127     }
128 
129     /**
130      * @see BaseDOMFormatter#getCharacterEntity
131      */
132     protected final String   getCharacterEntity(char textChar) {
133         if (fUseHTML4Entities) { 
134             return HTMLEntities.charToEntity4(textChar); 
135         } else { 
136             return HTMLEntities.charToEntity(textChar);
137         }
138     }
139 
140     /**
141      * Determine if an attribute's value should be printer. Those that don't
142      * normally have values only get them if one was explictly supplied.
143      */
144     private boolean printableAttrValue(Attr   attr) {
145         return (!(HTMLElements.isBooleanAttr(attr.getName())));
146     }
147 
148     /**
149      * Output the DOCTYPE declaration, if the information is available.
150      */
151     private void outputDocType(Document   document) throws IOException   {
152         //FIXME: Don't currently have a way of getting doctype from the parser
153         //to here, the only way is via the outputOptions override.
154         if ((fPublicId != null) || (fSystemId != null)) {
155             fOut.write("<!DOCTYPE html");
156             if (fPublicId != null) {
157                 fOut.write(" PUBLIC \"");
158                 fOut.write(fPublicId);
159                 fOut.write("\"");
160             }
161             if (fSystemId != null) {
162                 if (fPublicId == null) {
163                     fOut.write(" SYSTEM");
164                 }
165                 fOut.write(" \"");
166                 fOut.write(fSystemId);
167                 fOut.write("\"");
168             }
169             fOut.write('>');
170             writeln();
171         }
172     }
173 
174     /**
175      * Handler called for Document nodes.
176      * @see org.enhydra.xml.dom.DOMTraversal.Handler#handleDocument
177      */
178     public void handleDocument(Document   document) throws IOException   {
179         if (!fOptions.getOmitDocType()) {
180             outputDocType(document);
181         }
182         fTraverser.processChildren(document);
183     }
184 
185     /**
186      * Handler called for Document nodes; should never be called.
187      * @see org.enhydra.xml.dom.DOMTraversal.Handler#handleDocumentType
188      */
189     public void handleDocumentType(DocumentType   documentType) throws IOException   {
190         throw new XMLIOError("Unexpected call to handleDocumentType");
191     }
192 
193     /**
194      * Handler called for DocumentFragment nodes; just process children.
195      * @see org.enhydra.xml.dom.DOMTraversal.Handler#handleDocumentFragment
196      */
197     public void handleDocumentFragment(DocumentFragment   documentFragment) {
198         fTraverser.processChildren(documentFragment);
199     }
200 
201     /**
202      * Handler called for Attr nodes.
203      * @see org.enhydra.xml.dom.DOMTraversal.Handler#handleAttr
204      */
205     public void handleAttr(Attr   attr) throws IOException   {
206         String   name = attr.getName();
207         if (!(fDropThisId && name.equals("id"))) {
208             fOut.write(' ');
209             fOut.write(name);
210             if (printableAttrValue(attr)) {
211                 writeAttributeValue(attr);
212             }
213         }
214     }
215 
216     /**
217      * Write an element open tag.  The hasChildren option is ignored.
218      */
219     protected final void writeOpenTag(Element   element, 
220                                       String   tagName,
221                                       boolean hasChildren) throws IOException   {
222         String   formattedTag = null;
223         if (fPrettyPrinting) {
224             if (fNextSiblingText) {
225                 fOut.write('\n');
226             }
227             fNextSiblingText = (element.getNextSibling() instanceof Text  );
228         } // end of if ()
229         
230         if (fUsePreFormattedElements && (element instanceof PreFormattedText)) {
231             formattedTag = ((PreFormattedText)element).getPreFormattedText();
232         }
233         if (formattedTag != null) {
234             fOut.write(formattedTag);
235             fPreFormattedElementCount++;
236         } else {
237             if (fPrettyPrinting && !(element.getPreviousSibling() instanceof Text  )) {
238                 printIndent();                
239             } // end of if ()
240             
241             fDropThisId = fDropSpanIds && tagName.equals("SPAN");
242             fOut.write('<');
243             fOut.write(tagName);
244             fTraverser.processAttributes(element);
245             fOut.write('>');
246             fDynamicFormattedElementCount++;
247             if (fPrettyPrinting && !(element.getFirstChild() instanceof Text  )) {
248                 fOut.write('\n');
249             } // end of if ()
250         }
251     }
252 
253     /**
254      * Write an element close tag.
255      */
256     private void writeCloseTag(String   tagName) throws IOException   {
257         // Output end tag when legal.
258         if (fHandleText) {
259             fHandleText = false;
260         } else {
261             printIndent();             
262         } // end of else
263         
264         if (HTMLElements.hasCloseTag(tagName)) {
265             fOut.write("</");
266             fOut.write(tagName);
267             fOut.write('>');
268         }
269         if (fPrettyPrinting && !fNextSiblingText) {
270             fOut.write('\n');
271         } // end of if ()
272     }
273 
274     /**
275      * Handler called for Element nodes.
276      * <p>
277      * This optionally corrects problem cases for browsers:
278      * <UL>
279      * <LI> ID attributes are dropped from SPAN tags.  This cause Internet
280      *      Explorer 4.0 to get confused on keep-alive connections.
281      * </UL>
282      * @see org.enhydra.xml.dom.DOMTraversal.Handler#handleElement
283      */
284     public void handleElement(Element   element) throws IOException   {
285         String   tagName = element.getTagName();
286 
287         // Start Barracuda Kludge ======
288         // check the element to see if it contains a
289         // attribute "visdom".
290         // (org.enhydra.barracuda.core.comp.BComponent.VISIBILITY_MARKER)
291         // This controls DOM visibility. If this value exists and does not
292         // match to true, don't print this particular node.
293         // Note: This should be made generic, but for now...
294         Attr   attr = DOMAccess.accessAttribute(fDocument, element, null, "visdom");
295         if (attr != null && !(Boolean.valueOf(attr.getValue()).booleanValue())) return;        
296         // End Barracuda Kludge ======
297 
298         // HTML version doesn't care if it has childrne.
299         writeOpenTag(element, tagName, false);
300 
301         // Output childern and close
302         boolean isScriptStyle = HTMLElements.isScriptStyle(element);
303         if (isScriptStyle) {
304             fNoFormatNestCount++;
305         }
306         fTraverser.processChildren(element);
307         if (isScriptStyle) {
308             fNoFormatNestCount--;
309         }
310         writeCloseTag(tagName);
311     }
312 
313     /**
314      * Handler called for ProcessingInstruction nodes.
315      * @see org.enhydra.xml.dom.DOMTraversal.Handler#handleProcessingInstruction
316      */
317     public void handleProcessingInstruction(ProcessingInstruction   pi) throws IOException   {
318         throw new XMLIOError("Unexpected call to handleProcessingInstruction");
319     }
320 
321     /**
322      * Handler called for CDATASection nodes.
323      * Non-standard extension: outputs data as-is.
324      * @see org.enhydra.xml.dom.DOMTraversal.Handler#handleCDATASection
325      */
326     public void handleCDATASection(CDATASection   cdata) throws IOException   {
327         fOut.write(cdata.getData());
328     }
329 
330     /**
331      * Handler called for Text nodes.
332      * @see org.enhydra.xml.dom.DOMTraversal.Handler#handleText
333      */
334     public final void handleText(Text   text) throws IOException   {
335         fHandleText = true;
336         if (fNoFormatNestCount > 0) {
337             fOut.write(text.getData());
338         } else {
339             super.handleText(text);
340         }
341     }
342 }
343
A to Z: JavaDoc & Examples Daily Java News & Articles Open Source Projects Open Source Codes Free Computer Books Remove Frame
Popular Tags