HTML


1   /*
2    * Copyright  1999-2004 The Apache Software Foundation
3    *
4    *  Licensed under the Apache License, Version 2.0 (the "License");
5    *  you may not use this file except in compliance with the License.
6    *  You may obtain a copy of the License at
7    *
8    *      http://www.apache.org/licenses/LICENSE-2.0
9    *
10   *  Unless required by applicable law or agreed to in writing, software
11   *  distributed under the License is distributed on an "AS IS" BASIS,
12   *  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13   *  See the License for the specific language governing permissions and
14   *  limitations under the License.
15   *
16   */
17  
18  /* $Id: HTML.java 42598 2004-03-01 16:18:28Z gregor $  */
19  
20  package org.apache.lenya.util;
21  
22  import java.io.FileReader  ;
23  import java.io.IOException  ;
24  import java.io.InputStreamReader  ;
25  import java.io.Reader  ;
26  import java.net.URL  ;
27  import java.net.URLConnection  ;
28  import java.util.Iterator  ;
29  import java.util.List  ;
30  
31  import javax.swing.text.html.parser.ParserDelegator  ;
32  
33  
34  /**
35   * http://developer.java.sun/developer/TechTips/1999/tt0923.html
36   */
37  public class HTML {
38      HTMLHandler htmlHandler;
39  
40      /**
41       * Creates a new HTML object.
42       *
43       * @param uri DOCUMENT ME!
44       *
45       * @throws IOException DOCUMENT ME!
46       */
47      public HTML(String   uri) throws IOException   {
48          ParserDelegator   pd = new ParserDelegator  ();
49          htmlHandler = new HTMLHandler();
50          pd.parse(getReader(uri), htmlHandler, true);
51      }
52  
53      /**
54       * DOCUMENT ME!
55       *
56       * @param args DOCUMENT ME!
57       */
58      public static void main(String  [] args) {
59          if (args.length != 1) {
60              System.err.println("Usage: HTML uri (file or url)");
61  
62              return;
63          }
64  
65          try {
66              HTML html = new HTML(args[0]);
67  
68              List   img_src_list = html.getImageSrcs(false);
69              System.out.println("<im src");
70  
71              Iterator   img_src_iterator = img_src_list.iterator();
72  
73              while (img_src_iterator.hasNext()) {
74                  System.out.println((String  ) img_src_iterator.next());
75              }
76  
77              List   a_href_list = html.getAnchorHRefs(false);
78              System.out.println("<a href");
79  
80              Iterator   a_href_iterator = a_href_list.iterator();
81  
82              while (a_href_iterator.hasNext()) {
83                  System.out.println((String  ) a_href_iterator.next());
84              }
85  
86              List   link_href_list = html.getLinkHRefs(false);
87              System.out.println("<link href");
88  
89              Iterator   link_href_iterator = link_href_list.iterator();
90  
91              while (link_href_iterator.hasNext()) {
92                  System.out.println((String  ) link_href_iterator.next());
93              }
94          } catch (Exception   e) {
95              System.err.println(".main(): " + e);
96          }
97      }
98  
99      /**
100      * DOCUMENT ME!
101      *
102      * @param duplicate DOCUMENT ME!
103      *
104      * @return DOCUMENT ME!
105      */
106     public List   getAnchorHRefs(boolean duplicate) {
107         if (duplicate) {
108             return htmlHandler.getAllAHRefs();
109         } else {
110             return htmlHandler.getAHRefs();
111         }
112     }
113 
114     /**
115      * DOCUMENT ME!
116      *
117      * @param duplicate DOCUMENT ME!
118      *
119      * @return DOCUMENT ME!
120      */
121     public List   getLinkHRefs(boolean duplicate) {
122         if (duplicate) {
123             return htmlHandler.getAllLinkHRefs();
124         } else {
125             return htmlHandler.getLinkHRefs();
126         }
127     }
128 
129     /**
130      * DOCUMENT ME!
131      *
132      * @param duplicate DOCUMENT ME!
133      *
134      * @return DOCUMENT ME!
135      */
136     public List   getImageSrcs(boolean duplicate) {
137         if (duplicate) {
138             return htmlHandler.getAllImageSrcs();
139         } else {
140             return htmlHandler.getImageSrcs();
141         }
142     }
143 
144     private Reader   getReader(String   uri) throws IOException   {
145         if (uri.startsWith("http:")) {
146             // uri is url
147             URLConnection   connection = new URL  (uri).openConnection();
148 
149             return new InputStreamReader  (connection.getInputStream());
150         } else {
151             // uri is file
152             return new FileReader  (uri);
153         }
154     }
155 }
156
A to Z: JavaDoc & Examples Daily Java News & Articles Open Source Projects Open Source Codes Free Computer Books Remove Frame
Popular Tags