Examples of AutoDetectParser

Examples of org.apache.tika.parser.AutoDetectParser

34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
    /**
     * Test that with only ID3v1 tags, we get some information out  
     */
    public void testMp3ParsingID3v1() throws Exception {
        Parser parser = new AutoDetectParser(); // Should auto-detect!
        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();

        InputStream stream = Mp3ParserTest.class.getResourceAsStream(
                "/test-documents/testMP3id3v1.mp3");
        try {
            parser.parse(stream, handler, metadata);
        } finally {
            stream.close();
        }

        assertEquals("audio/mpeg", metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here

Examples of org.apache.tika.parser.AutoDetectParser

68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
    /**
     * Test that with only ID3v2 tags, we get the full
     *  set of information out.
     */
    public void testMp3ParsingID3v2() throws Exception {
        Parser parser = new AutoDetectParser(); // Should auto-detect!
        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();

        InputStream stream = Mp3ParserTest.class.getResourceAsStream(
                "/test-documents/testMP3id3v2.mp3");
        try {
            parser.parse(stream, handler, metadata);
        } finally {
            stream.close();
        }

        assertEquals("audio/mpeg", metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here

Examples of org.apache.tika.parser.AutoDetectParser

102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
    /**
     * Test that with both id3v2 and id3v1, we prefer the
     *  details from id3v2
     */
    public void testMp3ParsingID3v1v2() throws Exception {
        Parser parser = new AutoDetectParser(); // Should auto-detect!
        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();

        InputStream stream = Mp3ParserTest.class.getResourceAsStream(
                "/test-documents/testMP3id3v1_v2.mp3");
        try {
            parser.parse(stream, handler, metadata);
        } finally {
            stream.close();
        }

        assertEquals("audio/mpeg", metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here

Examples of org.apache.tika.parser.AutoDetectParser

71
72
73
74
75
76
77
78
79
     */
    public static void main(String[] args) throws Exception {
        UIManager.setLookAndFeel(UIManager.getSystemLookAndFeelClassName());
        SwingUtilities.invokeLater(new Runnable() {
            public void run() {
                new TikaGUI(new AutoDetectParser()).setVisible(true);
            }
        });
    }
View Full Code Here

Examples of org.apache.tika.parser.AutoDetectParser

69
70
71
72
73
74
75
     *
     * @param config Tika configuration
     */
    public Tika(TikaConfig config) {
        this.detector = config.getMimeRepository();
        this.parser = new AutoDetectParser(config);
    }
View Full Code Here

Examples of org.apache.tika.parser.AutoDetectParser

32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
* Test case for parsing Outlook files.
*/
public class OutlookParserTest extends TestCase {

    public void testOutlookParsing() throws Exception {
        Parser parser = new AutoDetectParser(); // Should auto-detect!
        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();

        InputStream stream = OutlookParserTest.class.getResourceAsStream(
                "/test-documents/test-outlook.msg");
        try {
            parser.parse(stream, handler, metadata);
        } finally {
            stream.close();
        }

        assertEquals(
View Full Code Here

Examples of org.apache.tika.parser.AutoDetectParser

68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
     * Test case for TIKA-197
     *
     * @see <a href="https://issues.apache.org/jira/browse/TIKA-197">TIKA-197</a>
     */
    public void testMultipleCopies() throws Exception {
        Parser parser = new AutoDetectParser();
        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();

        InputStream stream = OutlookParserTest.class.getResourceAsStream(
                "/test-documents/testMSG.msg");
        try {
            parser.parse(stream, handler, metadata);
        } finally {
            stream.close();
        }

        assertEquals(
View Full Code Here

Examples of org.apache.tika.parser.AutoDetectParser

97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
     * Test case for TIKA-395, to ensure parser works for new Outlook formats.
     *
     * @see <a href="https://issues.apache.org/jira/browse/TIKA-395">TIKA-395</a>
     */
    public void testOutlookNew() throws Exception {
        Parser parser = new AutoDetectParser();
        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();

        InputStream stream = OutlookParserTest.class.getResourceAsStream(
                "/test-documents/test-outlook2003.msg");
        try {
            parser.parse(stream, handler, metadata);
        } finally {
            stream.close();
        }

        assertEquals(
View Full Code Here

Examples of org.apache.tika.parser.AutoDetectParser

45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
    public void testExcel() throws Exception {
        InputStream input = OOXMLParserTest.class
                .getResourceAsStream("/test-documents/testEXCEL.xlsx");

        Parser parser = new AutoDetectParser();
        Metadata metadata = new Metadata();
        // TODO: should auto-detect without the resource name
        metadata.set(Metadata.RESOURCE_NAME_KEY, "testEXCEL.xlsx");
        ContentHandler handler = new BodyContentHandler();

        try {
            parser.parse(input, handler, metadata);

            assertEquals(
                    "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
                    metadata.get(Metadata.CONTENT_TYPE));
            assertEquals("Simple Excel document", metadata.get(Metadata.TITLE));
View Full Code Here

Examples of org.apache.tika.parser.AutoDetectParser

76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
    public void testExcelFormats() throws Exception {
        InputStream input = OOXMLParserTest.class
                .getResourceAsStream("/test-documents/testEXCEL-formats.xlsx");

        Parser parser = new AutoDetectParser();
        Metadata metadata = new Metadata();
        // TODO: should auto-detect without the resource name
        metadata.set(Metadata.RESOURCE_NAME_KEY, "testEXCEL-formats.xlsx");
        ContentHandler handler = new BodyContentHandler();

        try {
            parser.parse(input, handler, metadata);

            assertEquals(
                    "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
                    metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. 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.