Package org.cyberneko.html.parsers

Examples of org.cyberneko.html.parsers.DOMParser

34
35
36
37
38
39
40
41
42
43
44
    // MAIN
    //

    /** Main. */
    public static void main(String[] argv) throws Exception {
        DOMParser parser = new DOMParser();
        for (int i = 0; i < argv.length; i++) {
            parser.parse(argv[i]);
            print(parser.getDocument(), "");
        }
    } // main(String[])
View Full Code Here
26
27
28
29
30
31
32
33
34
35
36
    // MAIN
    //

    /** Main. */
    public static void main(String[] argv) throws Exception {
        DOMParser parser = new DOMParser();
        for (int i = 0; i < argv.length; i++) {
            parser.parse(argv[i]);
            print(parser.getDocument(), "");
        }
    } // main(String[])
View Full Code Here
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
  }

  private Document parseFragment(String source) throws SAXException, IOException {
    InputSource input = new InputSource(new StringReader(source));
    if (attemptFullDocParseFirst(source)) {
      DOMParser parser = new DOMParser();
      // Force parser not to use HTMLDocumentImpl as document implementation otherwise
      // it forces all element names to uppercase.
      parser.setProperty("http://apache.org/xml/properties/dom/document-class-name",
          "org.apache.xerces.dom.DocumentImpl");
      // Dont convert element names to upper/lowercase
      parser.setProperty("http://cyberneko.org/html/properties/names/elems", "default");
      // Preserve case of attributes
      parser.setProperty("http://cyberneko.org/html/properties/names/attrs", "no-change");
      // Record entity references
      parser.setFeature("http://apache.org/xml/features/scanner/notify-char-refs", true);
      parser.setFeature("http://cyberneko.org/html/features/scanner/notify-builtin-refs", true);
      // No need to defer as full DOM is walked later
      parser.setFeature("http://apache.org/xml/features/dom/defer-node-expansion", false);
      parser.parse(input);
      return parser.getDocument();
    } else {
      Document htmlDoc = documentProvider.createDocument(null, null, null);
      // Workaround for error check failure adding text node to entity ref as a child
      htmlDoc.setStrictErrorChecking(false);
      DOMFragmentParser parser = new DOMFragmentParser();
      parser.setProperty("http://cyberneko.org/html/properties/names/elems", "default");
      parser.setFeature("http://cyberneko.org/html/features/document-fragment", true);
      parser.setProperty("http://cyberneko.org/html/properties/names/attrs", "no-change");
      parser.setFeature("http://apache.org/xml/features/scanner/notify-char-refs", true);
      parser.setFeature("http://cyberneko.org/html/features/scanner/notify-builtin-refs", true);
      DocumentFragment fragment = htmlDoc.createDocumentFragment();
      parser.parse(input, fragment);
      normalizeFragment(htmlDoc, fragment);
      return htmlDoc;
    }
  }
View Full Code Here
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
                    fPrinter.flush();
                }
               
            }};
           
                DOMParser parser = new DOMParser();
            parser.setFeature("http://cyberneko.org/html/features/balance-tags", false);
            parser.setFeature("http://cyberneko.org/html/features/scanner/notify-builtin-refs", true);
            parser.setProperty("http://cyberneko.org/html/properties/default-encoding", fileEncoding);
            parser.setProperty("http://cyberneko.org/html/properties/names/elems", "lower");
            parser.setProperty("http://cyberneko.org/html/properties/filters", filters);
                parser.parse(is);
               
                str = "<!DOCTYPE html>" + baos.toString(fileEncoding);
           
          } catch (Exception e) {
            logger.error(e.getMessage(), e);
View Full Code Here
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
   *             if an IO failure occurs.
   * @throws SAXException
   *             if an exception occurs while parsing the HTML string.
   */
  public static Document asDocument(String html) throws IOException {
    DOMParser domParser = new DOMParser();
    try {
      domParser.setProperty("http://cyberneko.org/html/properties/names/elems", "match");
      domParser.setFeature("http://xml.org/sax/features/namespaces", false);
      domParser.parse(new InputSource(new StringReader(html)));
    } catch (SAXException e) {
      throw new IOException("Error while reading HTML: " + html, e);
    }
    return domParser.getDocument();
  }
View Full Code Here
91
92
93
94
95
96
97
98
99
100
101
   *             if an exception occurs while parsing the HTML string.
   * @throws IOException
   *             if an IO failure occurs.
   */
  public static Document getDocumentNoBalance(String html) throws SAXException, IOException {
    DOMParser domParser = new DOMParser();
    domParser.setProperty("http://cyberneko.org/html/properties/names/elems", "match");
    domParser.setFeature("http://cyberneko.org/html/features/balance-tags", false);
    domParser.parse(new InputSource(new StringReader(html)));
    return domParser.getDocument();
  }
View Full Code Here
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
   *             if an IO failure occurs.
   * @throws SAXException
   *             if an exception occurs while parsing the HTML string.
   */
  public static Document asDocument(String html) throws IOException {
    DOMParser domParser = new DOMParser();
    try {
      domParser.setProperty("http://cyberneko.org/html/properties/names/elems", "match");
      domParser.setFeature("http://xml.org/sax/features/namespaces", false);
      domParser.parse(new InputSource(new StringReader(html)));
    } catch (SAXException e) {
      throw new IOException("Error while reading HTML: " + html, e);
    }
    return domParser.getDocument();
  }
View Full Code Here
91
92
93
94
95
96
97
98
99
100
101
   *             if an exception occurs while parsing the HTML string.
   * @throws IOException
   *             if an IO failure occurs.
   */
  public static Document getDocumentNoBalance(String html) throws SAXException, IOException {
    DOMParser domParser = new DOMParser();
    domParser.setProperty("http://cyberneko.org/html/properties/names/elems", "match");
    domParser.setFeature("http://cyberneko.org/html/features/balance-tags", false);
    domParser.parse(new InputSource(new StringReader(html)));
    return domParser.getDocument();
  }
View Full Code Here
162
163
164
165
166
167
168
169
170
171
172
   *             if an IO failure occurs.
   * @throws SAXException
   *             if an exception occurs while parsing the HTML string.
   */
  public static Document getDocument(String html) throws SAXException, IOException {
    DOMParser domParser = new DOMParser();
    domParser.setProperty("http://cyberneko.org/html/properties/names/elems", "match");
    domParser.setFeature("http://xml.org/sax/features/namespaces", false);
    domParser.parse(new InputSource(new StringReader(html)));
    return domParser.getDocument();
  }
View Full Code Here
179
180
181
182
183
184
185
186
187
188
189
   *             if an exception occurs while parsing the HTML string.
   * @throws IOException
   *             if an IO failure occurs.
   */
  public static Document getDocumentNoBalance(String html) throws SAXException, IOException {
    DOMParser domParser = new DOMParser();
    domParser.setProperty("http://cyberneko.org/html/properties/names/elems", "match");
    domParser.setFeature("http://cyberneko.org/html/features/balance-tags", false);
    domParser.parse(new InputSource(new StringReader(html)));
    return domParser.getDocument();
  }
View Full Code Here
TOP

Related Classes of org.cyberneko.html.parsers.DOMParser

  • biz.webgate.domino.mywebgate.util.URLFetcher
  • com.asual.summer.core.faces.FacesResourceProcessor
  • com.crawljax.util.DomUtils
  • com.crawljax.util.Helper
  • com.huangzhimin.contacts.email.OneEightNineImporter
  • com.huangzhimin.contacts.email.OneSixThreeImporter
  • com.huangzhimin.contacts.email.OneTwoSixImporter
  • com.huangzhimin.contacts.email.TomImporter
  • com.huangzhimin.contacts.email.YahooImporter
  • com.huangzhimin.contacts.email.YeahImporter
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.