Package net.htmlparser.jericho

Examples of net.htmlparser.jericho.Renderer$AlternateTextElementHandler

60
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
   * @param html
   * @return
   */
  public static String html2text(String html) {

    Renderer renderer = new Renderer(new Source(html)) {

      {
        setIncludeHyperlinkURLs(true);
      }

      @Override
      public String renderHyperlinkURL(StartTag startTag) {
        final String href = startTag.getAttributeValue("href");
        if (href == null || href.equals("#") || href.startsWith("javascript:")) {
          return null;
        }
        // TODO customize?
        return '<' + href + '>';
      }

    };

    String text = renderer.toString();

    String newLine = renderer.getNewLine();

    // strip leading new lines
    while (text != (text = StringUtils.stripLeading(text, newLine))) {
    }

View Full Code Here
156
157
158
159
160
161
162
163
164
165
166
167
168
    }
    return output.toString();
  }

  private String cleanHTML(String input) {
    Renderer renderer=new Source(input).getRenderer();

    // We don't want it to be wrapped
    renderer.setMaxLineLength(Integer.MAX_VALUE);
    renderer.setIncludeHyperlinkURLs(false);
   
    return renderer.toString();
  }
View Full Code Here
45
46
47
48
49
50
51
52
     * @return
     */
    public static String htmlToText(String html) {
        Source htmlSource = new Source(html);
        Segment htmlSeg = new Segment(htmlSource, 0, htmlSource.length());
        Renderer htmlRend = new Renderer(htmlSeg);
        return htmlRend.toString();
    }
View Full Code Here
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
    {
        final Source source = new Source ( html );
        final Tag[] tags = source.fullSequentialParse ();
        if ( tags.length > 0 )
        {
            final Renderer renderer = source.getRenderer ();
            renderer.setIncludeHyperlinkURLs ( false );
            renderer.setIncludeAlternateText ( false );
            renderer.setDecorateFontStyles ( false );
            renderer.setMaxLineLength ( Integer.MAX_VALUE );
            renderer.setBlockIndentSize ( 4 );
            renderer.setConvertNonBreakingSpaces ( false );
            renderer.setNewLine ( lineSeparator );
            return renderer.toString ();
        }
        else
        {
            return html;
        }
View Full Code Here
143
144
145
146
147
148
149
150
151
152
153
154
    }
   
    // Extract text from the source, nicely formatted with whitespace and
    // newlines where appropriate.
    private String renderHTMLAsPlainText(Source source) {
        Renderer renderer = source.getRenderer();
        renderer.setNewLine("\n");
        renderer.setIncludeHyperlinkURLs(false);
        renderer.setDecorateFontStyles(false);
        renderer.setIncludeAlternateText(false);
        return renderer.toString();
    }
View Full Code Here
TOP

Related Classes of net.htmlparser.jericho.Renderer$AlternateTextElementHandler

  • at.molindo.notify.util.NotifyUtils
  • com.alee.utils.HtmlUtils
  • org.openntf.news.monster.RSSReader
  • org.sleuthkit.autopsy.keywordsearch.JerichoParserWrapper
  • org.zanata.email.EmailUtil
    • 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.