|
1
16 package org.outerj.daisy.htmlcleaner;
17
18 import org.cyberneko.dtd.parsers.DOMParser;
19 import org.xml.sax.InputSource  ;
20 import org.w3c.dom.Document ;
21 import org.w3c.dom.Element ;
22 import org.w3c.dom.NodeList ;
23 import org.w3c.dom.Node ;
24 import org.w3c.dom.traversal.DocumentTraversal;
25 import org.w3c.dom.traversal.NodeFilter;
26 import org.w3c.dom.traversal.TreeWalker;
27
28 import java.net.URL ;
29 import java.util.ArrayList ;
30 import java.util.HashMap ;
31 import java.util.Map ;
32
33 class XhtmlDescriptorBuilder {
34
37 public Map build() throws Exception {
38 DOMParser parser = new DOMParser();
39
40 URL dtdURL = getClass().getClassLoader().getResource("org/outerj/daisy/htmlcleaner/xhtml1-strict.dtd");
41
42 InputSource inputSource = new InputSource ();
43 inputSource.setByteStream(dtdURL.openStream());
44 inputSource.setSystemId(dtdURL.toExternalForm());
45
46 parser.parse(inputSource);
47 Document document = parser.getDocument();
48
49 Element dtdElement = document.getDocumentElement();
50 Element externalSubset = findChildElement(dtdElement, "externalSubset");
51
52 Element [] elementDecls = findChildElements(externalSubset, "elementDecl");
54 HashMap elementDescriptors = new HashMap ();
55 for (int i = 0; i < elementDecls.length; i++) {
56 Element elementDecl = elementDecls[i];
57 String name = elementDecl.getAttribute("ename");
58 elementDescriptors.put(name, new ElementDescriptor(name));
59 }
60
61 Element [] attlists = findChildElements(externalSubset, "attlist");
63 for (int i = 0; i < attlists.length; i++) {
64 Element attlist = attlists[i];
65 String ename = attlist.getAttribute("ename");
66 Element [] attributeDecls = findChildElements(attlist, "attributeDecl");
67 ElementDescriptor descriptor = (ElementDescriptor)elementDescriptors.get(ename);
68 for (int j = 0; j < attributeDecls.length; j++) {
69 String aname = attributeDecls[j].getAttribute("aname");
70 if (!aname.equals("xmlns") && !aname.startsWith("xml:"))
71 descriptor.addAttribute(aname);
72 }
73 }
74
75 Element [] contentModels = findChildElements(externalSubset, "contentModel");
77 for (int i = 0; i < contentModels.length; i++) {
78 Element contentModel = contentModels[i];
79
80 String ename = contentModel.getAttribute("ename");
81 Element [] elements = findDescendants(contentModel, "element");
82 ElementDescriptor descriptor = (ElementDescriptor)elementDescriptors.get(ename);
83 for (int j = 0; j < elements.length; j++) {
84 String name = elements[j].getAttribute("name");
85 descriptor.addChild(name);
86 }
87 }
88
89 return elementDescriptors;
90 }
91
92 private Element findChildElement(Element element, String name) {
93 NodeList children = element.getChildNodes();
94 for (int i = 0; i < children.getLength(); i++) {
95 Node node = children.item(i);
96 if (node instanceof Element && node.getNodeName().equals(name))
97 return (Element )node;
98 }
99 throw new RuntimeException ("Did not find expected element: " + name);
100 }
101
102 private Element [] findChildElements(Element element, String name) {
103 ArrayList foundElements = new ArrayList ();
104 NodeList children = element.getChildNodes();
105 for (int i = 0; i < children.getLength(); i++) {
106 Node node = children.item(i);
107 if (node instanceof Element && node.getNodeName().equals(name))
108 foundElements.add(node);
109 }
110 return (Element [])foundElements.toArray(new Element [foundElements.size()]);
111 }
112
113 private Element [] findDescendants(Element element, String name) {
114 ArrayList foundElements = new ArrayList ();
115 TreeWalker walker = ((DocumentTraversal)element.getOwnerDocument()).createTreeWalker(element, NodeFilter.SHOW_ELEMENT, null, false);
116 while (walker.nextNode() != null) {
117 Element currentEl = (Element )walker.getCurrentNode();
118 if (currentEl.getNodeName().equals(name))
119 foundElements.add(currentEl);
120 }
121 return (Element [])foundElements.toArray(new Element [foundElements.size()]);
122 }
123 }
124 |
Popular Tags |
Java API By Example, From Geeks To Geeks.