Package org.apache.tika.sax

Examples of org.apache.tika.sax.WriteOutContentHandler

76
77
78
79
80
81
82
83
84
85
86
    public void testEmptyText() throws Exception {
        Metadata metadata = new Metadata();
        StringWriter writer = new StringWriter();
        parser.parse(
                new ByteArrayInputStream(new byte[0]),
                new WriteOutContentHandler(writer),
                metadata);
        String content = writer.toString();
        assertEquals("text/plain", metadata.get(Metadata.CONTENT_TYPE));
        assertEquals("", content);
    }
View Full Code Here
31
32
33
34
35
36
37
38
39
40
41
        InputStream input = ExcelParserTest.class.getResourceAsStream(
                "/test-documents/testEXCEL.xls");
        try {
            Metadata metadata = new Metadata();
            StringWriter writer = new StringWriter();
            ContentHandler handler = new WriteOutContentHandler(writer);
            new ExcelParser().parse(input, handler, metadata);

            assertEquals(
                    "application/vnd.ms-excel",
                    metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here
43
44
45
46
47
48
49
50
51
52
53
        StringWriter writer = new StringWriter();
        Metadata metadata = new Metadata();

        parser.parse(getStream("test-documents/testHTML.html"),
                new WriteOutContentHandler(writer), metadata);
        String content = writer.toString();

        assertTrue("Did not contain expected text:"
                + "Title : Test Indexation Html", content
                .contains("Title : Test Indexation Html"));
View Full Code Here
64
65
66
67
68
69
70
71
72
73
74
        StringWriter writer = new StringWriter();
        Metadata metadata = new Metadata();

        parser.parse(getStream("test-documents/testHTML_utf8.html"),
                new WriteOutContentHandler(writer), metadata);
        String content = writer.toString();

        assertTrue("Did not contain expected text:"
                + "Title : Tilte with UTF-8 chars öäå", content
                .contains("Title : Tilte with UTF-8 chars öäå"));
View Full Code Here
84
85
86
87
88
89
90
91
92
    public void testParseEmpty() throws Exception {
        Metadata metadata = new Metadata();
        StringWriter writer = new StringWriter();
        parser.parse(new ByteArrayInputStream(new byte[0]),
                new WriteOutContentHandler(writer), metadata);
        String content = writer.toString();
        assertEquals("", content);
    }
View Full Code Here
31
32
33
34
35
36
37
38
39
40
41
        InputStream input = WordParserTest.class.getResourceAsStream(
                "/test-documents/testWORD.doc");
        try {
            Metadata metadata = new Metadata();
            StringWriter writer = new StringWriter();
            ContentHandler handler = new WriteOutContentHandler(writer);
            new WordParser().parse(input, handler, metadata);

            assertEquals(
                    "application/msword",
                    metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here
31
32
33
34
35
36
37
38
39
40
41
        InputStream input = PowerPointParserTest.class.getResourceAsStream(
                "/test-documents/testPPT.ppt");
        try {
            Metadata metadata = new Metadata();
            StringWriter writer = new StringWriter();
            ContentHandler handler = new WriteOutContentHandler(writer);
            new PowerPointParser().parse(input, handler, metadata);

            assertEquals(
                    "application/vnd.ms-powerpoint",
                    metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here
169
170
171
172
173
174
175
176
177
178
179
            throws TikaException, IOException {
        try {
            Parser parser = config.getParser(mimeType);
            StringWriter writer = new StringWriter();
            parser.parse(
                    stream, new WriteOutContentHandler(writer), new Metadata());
            return writer.toString();
        } catch (SAXException e) {
            throw new TikaException("Unexpected SAX error", e);
        }
    }
View Full Code Here
53
54
55
56
57
58
59
60
61
62
63
    public void parse(
            InputStream stream, ContentHandler handler, Metadata metadata)
            throws IOException, SAXException, TikaException {
        StringWriter writer = new StringWriter();
        handler = new TeeContentHandler(
                handler, new WriteOutContentHandler(writer));
        super.parse(stream, handler, metadata);

        String content = writer.toString();
        metadata.set("fulltext", content);
View Full Code Here
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
     * @throws IOException if the document can not be read
     * @throws TikaException if the document can not be parsed
     */
    public String parseToString(InputStream stream, Metadata metadata)
            throws IOException, TikaException {
        WriteOutContentHandler handler =
            new WriteOutContentHandler(maxStringLength);
        try {
            ParseContext context = new ParseContext();
            context.set(Parser.class, parser);
            parser.parse(
                    stream, new BodyContentHandler(handler), metadata, context);
        } catch (SAXException e) {
            if (!handler.isWriteLimitReached(e)) {
                // This should never happen with BodyContentHandler...
                throw new TikaException("Unexpected SAX processing failure", e);
            }
        } finally {
            stream.close();
        }
        return handler.toString();
    }
View Full Code Here
TOP

Related Classes of org.apache.tika.sax.WriteOutContentHandler

  • bixo.examples.crawl.SimpleBodyContentHandler
  • org.apache.jackrabbit.core.query.lucene.LazyTextExtractorField$ParsingTask
  • org.apache.jackrabbit.oak.plugins.index.lucene.LuceneIndexEditor
  • org.apache.jackrabbit.oak.plugins.index.lucene.LuceneIndexUpdate
  • org.apache.tika.parser.AutoDetectParserTest
  • org.apache.tika.parser.html.BoilerpipeContentHandler
  • org.apache.tika.parser.html.HtmlParser
  • org.apache.tika.parser.html.HtmlParserTest
  • org.apache.tika.parser.microsoft.ExcelParserTest
  • org.apache.tika.parser.microsoft.PowerPointParserTest
    • 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.