Package org.xml.sax

Examples of org.xml.sax.ContentHandler

78
79
80
81
82
83
84
85
86
87
88
89
90
    String html = "<html><head><title>The Big Brown Shoe</title></head><body><p>The best pizza place " +
            "in the US is <a href=\"http://antoniospizzas.com/\">Antonio's Pizza</a>.</p>" +
            "<p>It is located in Amherst, MA.</p></body></html>";
    //<start id="tika-html"/>
    InputStream input = new ByteArrayInputStream(html.getBytes(Charset.forName("UTF-8")));
    ContentHandler text = new BodyContentHandler();//<co id="html.text.co"/>
    LinkContentHandler links = new LinkContentHandler();//<co id="html.link.co"/>
    ContentHandler handler = new TeeContentHandler(links, text);//<co id="html.merge"/>
    Metadata metadata = new Metadata();//<co id="html.store"/>
    Parser parser = new HtmlParser();//<co id="html.parser"/>
    ParseContext context = new ParseContext();
    parser.parse(input, handler, metadata, context);//<co id="html.parse"/>
    System.out.println("Title: " + metadata.get(Metadata.TITLE));
View Full Code Here
687
688
689
690
691
692
693
694
695
696
697
            final SAXSource saxSource = (SAXSource) source;
            final SAXResult saxResult = (SAXResult) result;
           
            LexicalHandler lh = null;
            if (result != null) {
                ContentHandler ch = saxResult.getHandler();
                lh = saxResult.getLexicalHandler();
                /** If the lexical handler is not set try casting the ContentHandler. **/
                if (lh == null && ch instanceof LexicalHandler) {
                    lh = (LexicalHandler) ch;
                }
View Full Code Here
71
72
73
74
75
76
77
78
79
  public XMLReader fetchSheetParser(SharedStringsTable sst) throws SAXException {
    XMLReader parser =
      XMLReaderFactory.createXMLReader(
          "org.apache.xerces.parsers.SAXParser"
      );
    ContentHandler handler = new SheetHandler(sst);
    parser.setContentHandler(handler);
    return parser;
  }
View Full Code Here
476
477
478
479
480
481
482
483
484
485
486
    @Override
    public void importXML(String parentAbsPath, InputStream in, int uuidBehavior)
            throws IOException, RepositoryException {
        try {
            ContentHandler handler = getImportContentHandler(parentAbsPath, uuidBehavior);
            new ParsingContentHandler(handler).parse(in);
        } catch (SAXException e) {
            Throwable exception = e.getException();
            if (exception instanceof RepositoryException) {
                throw (RepositoryException) exception;
View Full Code Here
526
527
528
529
530
531
532
533
534
535
536
    @Override
    public void exportSystemView(String absPath, OutputStream out, boolean skipBinary, boolean noRecurse)
            throws IOException, RepositoryException {
        try {
            ContentHandler handler = new ToXmlContentHandler(out);
            export(absPath, new SystemViewExporter(this, handler, !noRecurse, !skipBinary));
        } catch (SAXException e) {
            Exception exception = e.getException();
            if (exception instanceof RepositoryException) {
                throw (RepositoryException) exception;
View Full Code Here
550
551
552
553
554
555
556
557
558
559
560
    @Override
    public void exportDocumentView(String absPath, OutputStream out, boolean skipBinary, boolean noRecurse)
            throws IOException, RepositoryException {
        try {
            ContentHandler handler = new ToXmlContentHandler(out);
            export(absPath, new DocumentViewExporter(this, handler, !noRecurse, !skipBinary));
        } catch (SAXException e) {
            Exception exception = e.getException();
            if (exception instanceof RepositoryException) {
                throw (RepositoryException) exception;
View Full Code Here
260
261
262
263
264
265
266
267
268
269
270
    @Override
    public void importXML(String parentAbsPath, InputStream in, int uuidBehavior) throws IOException, RepositoryException {
        ensureIsAlive();

        try {
            ContentHandler handler = getImportContentHandler(parentAbsPath, uuidBehavior);
            new ParsingContentHandler(handler).parse(in);
        } catch (SAXException e) {
            Throwable exception = e.getException();
            if (exception instanceof RepositoryException) {
                throw (RepositoryException) exception;
View Full Code Here
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
    public void testExcelParser() throws Exception {
        InputStream input = ExcelParserTest.class.getResourceAsStream(
                "/test-documents/testEXCEL.xls");
        try {
            Metadata metadata = new Metadata();
            ContentHandler handler = new BodyContentHandler();
            new OfficeParser().parse(input, handler, metadata);

            assertEquals(
                    "application/vnd.ms-excel",
                    metadata.get(Metadata.CONTENT_TYPE));
            assertEquals("Simple Excel document", metadata.get(Metadata.TITLE));
            assertEquals("Keith Bennett", metadata.get(Metadata.AUTHOR));
            String content = handler.toString();
            assertTrue(content.contains("Sample Excel Worksheet"));
            assertTrue(content.contains("Numbers and their Squares"));
            assertTrue(content.contains("\t\tNumber\tSquare"));
            assertTrue(content.contains("9"));
            assertFalse(content.contains("9.0"));
View Full Code Here
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
    public void testExcelParserFormatting() throws Exception {
        InputStream input = ExcelParserTest.class.getResourceAsStream(
                "/test-documents/testEXCEL-formats.xls");
        try {
            Metadata metadata = new Metadata();
            ContentHandler handler = new BodyContentHandler();
            new OfficeParser().parse(input, handler, metadata);

            assertEquals(
                    "application/vnd.ms-excel",
                    metadata.get(Metadata.CONTENT_TYPE));

            String content = handler.toString();

            // Number #,##0.00
            assertTrue(content.contains("1,599.99"));
            assertTrue(content.contains("-1,599.99"));
View Full Code Here
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
    }

    public void testUTF8Text() throws Exception {
        String text = "I\u00F1t\u00EBrn\u00E2ti\u00F4n\u00E0liz\u00E6ti\u00F8n";

        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();
        parser.parse(
                new ByteArrayInputStream(text.getBytes("UTF-8")),
                handler, metadata, new ParseContext());
        assertEquals("text/plain", metadata.get(Metadata.CONTENT_TYPE));
        assertEquals("UTF-8", metadata.get(Metadata.CONTENT_ENCODING));

        assertTrue(handler.toString().contains(text));
    }
View Full Code Here
TOP

Related Classes of org.xml.sax.ContentHandler

  • com.alibaba.citrus.asm.xml.Processor
  • com.nidhinova.tika.server.TikaService
  • com.sun.tools.internal.ws.wsdl.parser.DOMForest
  • com.sun.tools.internal.xjc.reader.internalizer.DOMForest
  • com.tamingtext.tika.TikaTest
  • cx.fbn.nevernote.threads.IndexRunner
  • net.bnubot.vercheck.XMLElementDecorator
  • net.javacrumbs.json2xml.JsonSaxAdapterTest
  • nu.validator.messages.MessageEmitterAdapter
  • nu.validator.servlet.VerifierServletTransaction
    • 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.