Package org.apache.tika.sax

Examples of org.apache.tika.sax.XHTMLContentHandler.startDocument()

76
77
78
79
80
81
82
83
84
85
86
                throw new TikaException(type + " parse error", e);
            }
        }

        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();
        xhtml.endDocument();
    }

    /**
     * @deprecated This method will be removed in Apache Tika 1.0.
View Full Code Here
133
134
135
136
137
138
139
140
141
142
143
                reader.reset();
            }

            XHTMLContentHandler xhtml =
                new XHTMLContentHandler(handler, metadata);
            xhtml.startDocument();

            xhtml.startElement("p");
            char[] buffer = new char[4096];
            int n = reader.read(buffer);
            while (n != -1) {
View Full Code Here
78
79
80
81
82
83
84
85
86
87
88
            metadata.set(Metadata.CONTENT_TYPE, "application/xml");
        }

        final XHTMLContentHandler xhtml =
            new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();
        xhtml.startElement("p");

        getSAXParser(context).parse(
                new CloseShieldInputStream(stream),
                new OfflineContentHandler(
View Full Code Here
63
64
65
66
67
68
69
70
71
72
73
    }

    public void parse(InputStream stream)
            throws IOException, SAXException, TikaException {
        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();

        // At the end we want to close the package/compression stream to
        // release any associated resources, but the underlying document
        // stream should not be closed
        stream = new CloseShieldInputStream(stream);
View Full Code Here
67
68
69
70
71
72
73
74
75
76
77
    public void parse(
            InputStream stream, ContentHandler handler,
            Metadata metadata, ParseContext context)
            throws IOException, SAXException, TikaException {
        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();

        POIFSFileSystem filesystem = new POIFSFileSystem(stream);

        // Parse summary entries first, to make metadata available early
        new SummaryExtractor(metadata).parseSummaries(filesystem);
View Full Code Here
56
57
58
59
60
61
62
63
64
65
66
            Metadata metadata, ParseContext context)
            throws IOException, SAXException, TikaException {
        metadata.set(Metadata.CONTENT_TYPE, "audio/mpeg");

        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();

        // Create handlers for the various kinds of ID3 tags
        ID3TagsAndAudio audioAndTags = getAllTagHandlers(stream, handler);

        if (audioAndTags.tags.length > 0) {
View Full Code Here
80
81
82
83
84
85
86
87
88
89
90
        metadata.set(Metadata.CONTENT_TYPE, MBOX_MIME_TYPE);
        metadata.set(Metadata.CONTENT_ENCODING, "us-ascii");

        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();

        ParseStates parseState = ParseStates.START;
        String multiLine = null;
        boolean inQuote = false;
        int numEmails = 0;
View Full Code Here
47
48
49
50
51
52
53
54
55
56
57
    public void parse(
            InputStream stream, ContentHandler handler,
            Metadata metadata, ParseContext context)
            throws SAXException {
        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();
        xhtml.endDocument();
    }

    /**
     * @deprecated This method will be removed in Apache Tika 1.0.
View Full Code Here
68
69
70
71
72
73
74
75
76
77
78
        extractContent(xmlDoc, Metadata.IDENTIFIER, "//dc:identifier", metadata);
        extractContent(xmlDoc, Metadata.LANGUAGE, "//dc:language", metadata);
        extractContent(xmlDoc, Metadata.RIGHTS, "//dc:rights", metadata);

        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();
        xhtml.startElement("p");
        concatOccurrence(xmlDoc, "//*", " ", new AppendableAdaptor(xhtml));
        xhtml.endElement("p");
        xhtml.endDocument();
    }
View Full Code Here
195
196
197
198
199
200
201
202
203
204
205
        metadata.set(Metadata.CONTENT_TYPE, "video/x-flv");
        metadata.set("hasVideo", Boolean.toString((typeFlags & MASK_VIDEO) != 0));
        metadata.set("hasAudio", Boolean.toString((typeFlags & MASK_AUDIO) != 0));

        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();

        // flv tag stream follows...
        while (true) {
            int type = datainput.read();
            if (type == -1) {
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.