1 /*
2 * The Apache Software License, Version 1.1
3 *
4 *
5 * Copyright (c) 1999, 2000 The Apache Software Foundation. All rights
6 * reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 *
12 * 1. Redistributions of source code must retain the above copyright
13 * notice, this list of conditions and the following disclaimer.
14 *
15 * 2. Redistributions in binary form must reproduce the above copyright
16 * notice, this list of conditions and the following disclaimer in
17 * the documentation and/or other materials provided with the
18 * distribution.
19 *
20 * 3. The end-user documentation included with the redistribution,
21 * if any, must include the following acknowledgment:
22 * "This product includes software developed by the
23 * Apache Software Foundation (http://www.apache.org/)."
24 * Alternately, this acknowledgment may appear in the software itself,
25 * if and wherever such third-party acknowledgments normally appear.
26 *
27 * 4. The names "Xerces" and "Apache Software Foundation" must
28 * not be used to endorse or promote products derived from this
29 * software without prior written permission. For written
30 * permission, please contact apache@apache.org.
31 *
32 * 5. Products derived from this software may not be called "Apache",
33 * nor may "Apache" appear in their name, without prior written
34 * permission of the Apache Software Foundation.
35 *
36 * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
37 * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
38 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
39 * DISCLAIMED. IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
40 * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
41 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
42 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
43 * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
44 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
45 * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
46 * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
47 * SUCH DAMAGE.
48 * ====================================================================
49 *
50 * This software consists of voluntary contributions made by many
51 * individuals on behalf of the Apache Software Foundation and was
52 * originally based on software copyright (c) 1999, International
53 * Business Machines, Inc., http://www.apache.org. For more
54 * information on the Apache Software Foundation, please see
55 * <http://www.apache.org/>.
56 */
57
58 package dom;
59
60 import util.Arguments;
61 import java.io.OutputStreamWriter;
62 import java.io.PrintWriter;
63 import java.io.UnsupportedEncodingException;
64
65 import org.apache.xerces.dom.TextImpl;
66
67 import org.w3c.dom.Attr;
68 import org.w3c.dom.Document;
69 import org.w3c.dom.NamedNodeMap;
70 import org.w3c.dom.Node;
71 import org.w3c.dom.NodeList;
72
73 /**
74 * A sample DOM counter. This sample program illustrates how to
75 * traverse a DOM tree in order to information about the document.
76 *
77 * @version $id$
78 */
79 public class DOMCount {
80
81 //
82 // Constants
83 //
84
85 /** Default parser name. */
86 private static final String
87 DEFAULT_PARSER_NAME = "dom.wrappers.DOMParser";
88
89 private static boolean setValidation = false; //defaults
90 private static boolean setNameSpaces = true;
91 private static boolean setSchemaSupport = true;
92 private static boolean setDeferredDOM = true;
93
94
95
96 //
97 // Data
98 //
99
100 /** Elements. */
101 private long elements;
102
103 /** Attributes. */
104 private long attributes;
105
106 /** Characters. */
107 private long characters;
108
109 /** Ignorable whitespace. */
110 private long ignorableWhitespace;
111
112
113 //
114 // Public static methods
115 //
116
117 /** Counts the resulting document tree. */
118 public static void count(String parserWrapperName, String uri) {
119
120 try {
121 DOMParserWrapper parser =
122 (DOMParserWrapper)Class.forName(parserWrapperName).newInstance();
123 DOMCount counter = new DOMCount();
124 long before = System.currentTimeMillis();
125 parser.setFeature( "http://apache.org/xml/features/dom/defer-node-expansion",
126
127 setDeferredDOM );
128 parser.setFeature( "http://xml.org/sax/features/validation",
129 setValidation );
130 parser.setFeature( "http://xml.org/sax/features/namespaces",
131 setNameSpaces );
132 parser.setFeature( "http://apache.org/xml/features/validation/schema",
133 setSchemaSupport );
134
135 Document document = parser.parse(uri);
136 counter.traverse(document);
137 long after = System.currentTimeMillis();
138 counter.printResults(uri, after - before);
139 } catch (org.xml.sax.SAXParseException spe) {
140 } catch (org.xml.sax.SAXNotRecognizedException ex ){
141 } catch (org.xml.sax.SAXNotSupportedException ex ){
142 } catch (org.xml.sax.SAXException se) {
143 if (se.getException() != null)
144 se.getException().printStackTrace(System.err);
145 else
146 se.printStackTrace(System.err);
147 } catch (Exception e) {
148 e.printStackTrace(System.err);
149 }
150
151 } // print(String,String,boolean)
152
153 //
154 // Public methods
155 //
156
157 /** Traverses the specified node, recursively. */
158 public void traverse(Node node) {
159
160 // is there anything to do?
161 if (node == null) {
162 return;
163 }
164
165 int type = node.getNodeType();
166 switch (type) {
167 // print document
168 case Node.DOCUMENT_NODE: {
169 elements = 0;
170 attributes = 0;
171 characters = 0;
172 ignorableWhitespace = 0;
173 traverse(((Document)node).getDocumentElement());
174 break;
175 }
176
177 // print element with attributes
178 case Node.ELEMENT_NODE: {
179 elements++;
180 NamedNodeMap attrs = node.getAttributes();
181 if (attrs != null) {
182 attributes += attrs.getLength();
183 }
184 NodeList children = node.getChildNodes();
185 if (children != null) {
186 int len = children.getLength();
187 for (int i = 0; i < len; i++) {
188 traverse(children.item(i));
189 }
190 }
191 break;
192 }
193
194 // handle entity reference nodes
195 case Node.ENTITY_REFERENCE_NODE: {
196 NodeList children = node.getChildNodes();
197 if (children != null) {
198 int len = children.getLength();
199 for (int i = 0; i < len; i++) {
200 traverse(children.item(i));
201 }
202 }
203 break;
204 }
205
206 // print text
207 case Node.CDATA_SECTION_NODE: {
208 characters += node.getNodeValue().length();
209 break;
210 }
211 case Node.TEXT_NODE: {
212 if (node instanceof TextImpl) {
213 if (((TextImpl)node).isIgnorableWhitespace())
214 ignorableWhitespace += node.getNodeValue().length();
215 else
216 characters += node.getNodeValue().length();
217 } else
218 characters += node.getNodeValue().length();
219 break;
220 }
221 }
222
223 } // traverse(Node)
224
225 /** Prints the results. */
226 public void printResults(String uri, long time) {
227
228 // filename.xml: 631 ms (4 elems, 0 attrs, 78 spaces, 0 chars)
229 System.out.print(uri);
230 System.out.print(": ");
231 System.out.print(time);
232 System.out.print(" ms (");
233 System.out.print(elements);
234 System.out.print(" elems, ");
235 System.out.print(attributes);
236 System.out.print(" attrs, ");
237 System.out.print(ignorableWhitespace);
238 System.out.print(" spaces, ");
239 System.out.print(characters);
240 System.out.print(" chars)");
241 System.out.println();
242
243 } // printResults(String,long)
244
245 //
246 // Main
247 //
248
249 /** Main program entry point. */
250 public static void main(String argv[]) {
251
252 Arguments argopt = new Arguments();
253 argopt.setUsage( new String[] {
254 "usage: java dom.DOMCount (options) uri ...",
255 "",
256 "options:",
257 " -p name Specify DOM parser wrapper by name.",
258 " -n | -N Turn on/off namespace [default=on]",
259 " -v | -V Turn on/off validation [default=on]",
260 " -s | -S Turn on/off Schema support [default=on]",
261 " -d | -D Turn on/off deferred DOM [default=on]",
262 " -h This help screen."} );
263
264
265 // is there anything to do?
266 if (argv.length == 0) {
267 argopt.printUsage();
268 System.exit(1);
269 }
270
271 // vars
272 String parserName = DEFAULT_PARSER_NAME;
273
274 argopt.parseArgumentTokens(argv , new char[] { 'p'} );
275
276 int c;
277 String arg = null;
278 while ( ( arg = argopt.getlistFiles() ) != null ) {
279 outer:
280 while ( (c = argopt.getArguments()) != -1 ){
281 switch (c) {
282 case 'v':
283 setValidation = true;
284 //System.out.println( "v" );
285 break;
286 case 'V':
287 setValidation = false;
288 //System.out.println( "V" );
289 break;
290 case 'N':
291 setNameSpaces = false;
292 break;
293 case 'n':
294 setNameSpaces = true;
295 break;
296 case 'p':
297 //System.out.println('p');
298 parserName = argopt.getStringParameter();
299 //System.out.println( "parserName = " + parserName );
300 break;
301 case 'd':
302 setDeferredDOM = true;
303 break;
304 case 'D':
305 setDeferredDOM = false;
306 break;
307 case 's':
308 //System.out.println("s" );
309 setSchemaSupport = true;
310 break;
311 case 'S':
312 //System.out.println("S" );
313 setSchemaSupport = false;
314 break;
315 case '?':
316 case 'h':
317 case '-':
318 argopt.printUsage();
319 System.exit(1);
320 break;
321 case -1:
322 //System.out.println( "-1" );
323 break outer;
324 default:
325
326 break;
327 }
328 }
329
330 count(parserName, arg ); //count uri
331 }
332
333 } // main(String[])
334
335 } // class DOMCount