|
1
17
18
19
20 package org.apache.lenya.lucene.index;
21
22 import java.io.File  ;
23 import java.io.FileWriter ;
24 import java.io.IOException ;
25 import java.io.Reader ;
26 import java.io.StringReader ;
27 import java.io.StringWriter ;
28 import java.io.Writer ;
29 import java.lang.reflect.Method ;
30
31 import javax.xml.parsers.DocumentBuilder ;
32 import javax.xml.parsers.DocumentBuilderFactory ;
33 import javax.xml.transform.OutputKeys ;
34 import javax.xml.transform.Transformer ;
35 import javax.xml.transform.TransformerFactory ;
36 import javax.xml.transform.dom.DOMSource ;
37 import javax.xml.transform.stream.StreamResult ;
38 import javax.xml.transform.stream.StreamSource ;
39
40 import org.apache.lenya.lucene.parser.HTMLParser;
41 import org.apache.lenya.lucene.parser.HTMLParserFactory;
42 import org.apache.lenya.lucene.parser.StringCleaner;
43 import org.apache.lenya.xml.DocumentHelper;
44 import org.apache.lenya.xml.NamespaceHelper;
45 import org.apache.log4j.Category;
46 import org.apache.lucene.document.Document;
47 import org.apache.lucene.document.Field;
48 import org.w3c.dom.Element ;
49 import org.w3c.dom.Node ;
50 import org.w3c.dom.NodeList ;
51 import org.xml.sax.InputSource ;
52
53
56 public class ConfigurableDocumentCreator extends AbstractDocumentCreator {
57 Category log = Category.getInstance(ConfigurableDocumentCreator.class);
58
59 public static final String LUCENE_NAMESPACE = "http://apache.org/cocoon/lenya/lucene/1.0";
60 public static final String XHTML_NAMESPACE = "http://www.w3.org/1999/xhtml";
61
62
67 public ConfigurableDocumentCreator(String stylesheet) {
68 this.stylesheet = stylesheet;
69 }
70
71 private String stylesheet;
72
73
78 public String getStylesheet() {
79 return stylesheet;
80 }
81
82
92 public Document getDocument(File file, File htdocsDumpDir) throws Exception {
93 log.debug(".getDocument() : indexing " + file.getAbsolutePath());
94 try {
95
96 org.w3c.dom.Document sourceDocument = null;
97 DocumentBuilderFactory parserFactory = DocumentBuilderFactory.newInstance();
98 parserFactory.setValidating(false);
99 parserFactory.setNamespaceAware(true);
100 parserFactory.setIgnoringElementContentWhitespace(true);
101 DocumentBuilder mybuilder = parserFactory.newDocumentBuilder();
102 sourceDocument = mybuilder.parse(file.getAbsolutePath());
103
104
105
116
117
118
119
120 DOMSource documentSource = new DOMSource (sourceDocument);
121 Writer documentWriter = new StringWriter ();
122
123 TransformerFactory tFactory = TransformerFactory.newInstance();
124 Transformer documentTransformer = tFactory.newTransformer(new StreamSource (new StringReader (getStylesheet())));
125 documentTransformer.setOutputProperty(OutputKeys.INDENT, "yes");
126 documentTransformer.setOutputProperty(OutputKeys.ENCODING, "ISO-8859-1");
127
128 String fileName = file.getName();
129
130 if (fileName.endsWith(".pdf.txt")) {
131 fileName = fileName.substring(0, fileName.lastIndexOf(".txt"));
132 }
133
134 documentTransformer.setParameter("filename", fileName);
135 documentTransformer.transform(documentSource, new StreamResult (documentWriter));
136
137
140 DocumentBuilder builder = DocumentHelper.createBuilder();
141 org.w3c.dom.Document luceneDocument = builder.parse(new InputSource (new StringReader (documentWriter.toString())));
142
143 NamespaceHelper helper = new NamespaceHelper(LUCENE_NAMESPACE, "luc", luceneDocument);
144 Element root = luceneDocument.getDocumentElement();
145 Element [] fieldElements = helper.getChildren(root, "field");
146
147 Document document = super.getDocument(file, htdocsDumpDir);
148
149 Class [] parameterTypes = { String .class, String .class };
150
151 for (int i = 0; i < fieldElements.length; i++) {
152 String name = fieldElements[i].getAttribute("name");
153 String type = fieldElements[i].getAttribute("type");
154 String text = getText(fieldElements[i]);
155
156 Method method = Field.class.getMethod(type, parameterTypes);
157
158 String [] args = { name, text };
159
160 Field field = (Field) method.invoke(null, args);
161 document.add(field);
162
163 }
164
165 return document;
166 } catch (Exception e) {
167 throw e;
168 }
169 }
170
171
174 protected void dumpLuceneDocument(File file, Writer writer) throws IOException {
175 log.debug(".dumpLuceneDocument(): Dump document: " + file.getAbsolutePath());
176
177 File luceneDocumentFile = new File (file.getAbsolutePath() + ".xluc");
178 luceneDocumentFile.createNewFile();
179
180 FileWriter fileWriter = new FileWriter (luceneDocumentFile);
181 fileWriter.write(writer.toString());
182 fileWriter.close();
183 }
184
185
192 public static String getText(Node node) {
193 StringBuffer result = new StringBuffer ();
194
195 if (!node.hasChildNodes()) {
196 return "";
197 }
198
199 NodeList list = node.getChildNodes();
200
201 for (int i = 0; i < list.getLength(); i++) {
202 Node subnode = list.item(i);
203
204 if (subnode.getNodeType() == Node.TEXT_NODE) {
205 result.append(subnode.getNodeValue());
206 } else if (subnode.getNodeType() == Node.CDATA_SECTION_NODE) {
207 result.append(subnode.getNodeValue());
208 } else if (subnode.getNodeType() == Node.ENTITY_REFERENCE_NODE) {
209 result.append(getText(subnode));
212 }
213 }
214
215 return result.toString();
216 }
217
218
227 public static String getBodyText(File file) throws Exception {
228 HTMLParser parser = HTMLParserFactory.newInstance(file);
229 parser.parse(file);
230
231 Reader reader = parser.getReader();
232 Writer writer = new StringWriter ();
233
234 int c;
235
236 while ((c = reader.read()) != -1)
237 writer.write(c);
238
239 String content = writer.toString();
240 reader.close();
241 writer.close();
242
243 content = StringCleaner.clean(content);
244
245 return content;
246 }
247 }
248 |
Popular Tags |
Java API By Example, From Geeks To Geeks.