Package org.w3c.tidy

Examples of org.w3c.tidy.Tidy

336
337
338
339
340
341
342
343
344
345
346
347
   */
  private void parseToDOM() {
    ByteArrayInputStream is = new ByteArrayInputStream(content);
   
    // set tidy parameters
    Tidy tidy = new Tidy();
    tidy.setUpperCaseTags(false);
    tidy.setUpperCaseAttrs(false);
    tidy.setErrout(new PrintWriter(System.err));
   
    domDoc = tidy.parseDOM(is,null);
  }
View Full Code Here
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
      return input;
    }

    // okay, parse the HTML code
    ByteArrayInputStream bis = new ByteArrayInputStream(input.getContent());
    Tidy tidy = new Tidy();
    tidy.setUpperCaseTags(false);
    tidy.setUpperCaseAttrs(false);
    tidy.setErrout(new PrintWriter(new NullWriter()));

    Document doc = tidy.parseDOM(bis,null);

    rewriteDOM(doc,input.getURL());

    ByteArrayOutputStream bos = new ByteArrayOutputStream();
    tidy.pprint(doc,bos);

    input.setContent(bos.toByteArray());
   
    return input;
  }
View Full Code Here
98
99
100
101
102
103
104
105
106
107
108
109
  public ByteArrayOutputStream transformHTML(String sHTML)
    throws IOException,FOPException,TransformerException,TransformerConfigurationException {
    ByteArrayOutputStream oXml = new ByteArrayOutputStream();
    StringBufferInputStream oHtm = new StringBufferInputStream(sHTML);
    Tidy oTdy = new Tidy();
    oTdy.setXmlOut(true);
    oTdy.setTidyMark(false);
    oTdy.setNumEntities(true);
  oTdy.parseDOM(oHtm, oXml);
  return transformXHTML(oXml.toString(Charset.defaultCharset().name()));
  }
View Full Code Here
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
      throw new NullPointerException("Input HTML code string is NULL");
    }
    StringReader stringReader = new StringReader(htmlCode);
    StringWriter tidyHtmlWriter = new StringWriter();
    try{
      Tidy tidy = new Tidy();
      tidy.setPrintBodyOnly(bodyOnly);
      tidy.setJoinClasses(true);
      tidy.setJoinStyles(true);
      tidy.setQuoteAmpersand(true);
      tidy.setHideComments(true);
      tidy.setInputEncoding(encoding);
      tidy.setOutputEncoding(encoding);   
      tidy.parse(stringReader,tidyHtmlWriter);
      return tidyHtmlWriter.getBuffer().toString();
    }catch(Exception e){
      logger.error("FormatUtil: tidyHTML error: " + e);
    }finally{   
      if(stringReader != null){
View Full Code Here
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
        }
        return false;
    }
   
    public InputStream tidy(InputStream is) {
        Tidy tidy = new Tidy();
        tidy.setXHTML(false);
        tidy.setPrintBodyOnly(true);
        tidy.setDocType("loose");
        tidy.setXHTML(true);
        tidy.setForceOutput(true);
        java.io.ByteArrayOutputStream out = new java.io.ByteArrayOutputStream(1024);
        tidy.parse(is,out);
        return new ByteArrayInputStream(out.toByteArray());
    }
View Full Code Here
21
22
23
24
25
26
27
28
29
30
31
    }
    public void testPageRender() throws IOException {
        HtmlCanvas html = new HtmlCanvas();
        html.render(new PersonalPage());
       
        Tidy tidy = new Tidy();
        tidy.setMessageListener(new TidyMessageCheck());
        tidy.setXHTML(true);
        tidy.setDocType("loose");
        tidy.parse(new ByteArrayInputStream(html.toHtml().getBytes()), System.out);       
    }
View Full Code Here
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
    public void testPageRenderWithError() throws IOException {
        HtmlCanvas html = new HtmlCanvas();
        html.tag("bogus");
        html.render(new PersonalPage());
       
        Tidy tidy = new Tidy();
        tidy.setMessageListener(new TidyMessageCheck());
        tidy.setXHTML(true);
        tidy.setDocType("loose");
        try {
            tidy.parse(new ByteArrayInputStream(html.toHtml().getBytes()), System.out);       
        } catch (AssertionFailedError err) {
            System.out.println(err);
            // good thing
        }
    }
View Full Code Here
127
128
129
130
131
132
133
134
135
136
137
138
      buf.append("<body>");
      buf.append(htmlContent);
      buf.append("</body>");
      buf.append("</html>");
     
      Tidy tidy = new Tidy();
    tidy.setXHTML(true);
    tidy.setQuiet(true);
    tidy.setShowWarnings(false);
     
      return tidy.parseDOM(new ByteArrayInputStream(buf.toString().getBytes()), null);
  }
View Full Code Here
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
        // there is also a javax.swing.text.Document class.
        org.w3c.dom.Document document = null;

        StringWriter sw = new StringWriter();
        Tidy tidy = XPathUtil.makeTidyParser(true, true, true, sw);
        document = tidy.parseDOM(baIS, null);
        document.normalize();
        if (tidy.getParseErrors() > 0) {
            showErrorMessageDialog(sw.toString(),
                    "Tidy: " + tidy.getParseErrors() + " errors, " + tidy.getParseWarnings() + " warnings",
                    JOptionPane.WARNING_MESSAGE);
        }

        JPanel domTreePanel = new DOMTreePanel(document);
        resultsScrollPane.setViewportView(domTreePanel);
View Full Code Here
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
     *
     * @return a <code>tidy</code> HTML parser
     */
    private static Tidy getTidyParser() {
        log.debug("Start : getParser");
        Tidy tidy = new Tidy();
        tidy.setInputEncoding("UTF8");
        tidy.setOutputEncoding("UTF8");
        tidy.setQuiet(true);
        tidy.setShowWarnings(false);
        if (log.isDebugEnabled()) {
            log.debug("getParser : tidy parser created - " + tidy);
        }
        log.debug("End   : getParser");
        return tidy;
View Full Code Here
TOP

Related Classes of org.w3c.tidy.Tidy

  • ag.kcmedia.JavaDocViewer
  • ariba.ui.aribaweb.util.AWUtil
  • br.com.caelum.tubaina.util.XHTMLValidator
  • ch.entwine.weblounge.cache.impl.filter.PrettyPrinter
  • ch.entwine.weblounge.cache.impl.filter.TidyReport$TidyThread
  • ch.entwine.weblounge.common.impl.util.TestUtils
  • com.alibaba.antx.config.resource.util.TextBasedPageParser
  • com.betfair.testing.utils.cougar.helpers.HttpHelpers
  • com.betfair.testing.utils.cougar.misc.DocumentHelpers
  • com.cardence.lawshelf.handler.UscParserHandler
    • 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.