Package org.apache.tika.sax

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

144
145
146
147
148
149
150
151
152
153
154
    public void parse(
            InputStream stream, ContentHandler handler,
            Metadata metadata, ParseContext context)
            throws IOException, SAXException, TikaException {
        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();

        final DirectoryNode root;
        TikaInputStream tstream = TikaInputStream.cast(stream);
        if (tstream == null) {
            root = new NPOIFSFileSystem(new CloseShieldInputStream(stream)).getRoot();
View Full Code Here
61
62
63
64
65
66
67
68
69
70
71
        // metadata
        metadata.set(Metadata.CONTENT_TYPE, "application/vnd.ms-htmlhelp");

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

        Iterator<DirectoryListingEntry> it =
                chmExtractor.getChmDirList().getDirectoryListingEntryList().iterator();
        while (it.hasNext()) {
            DirectoryListingEntry entry = it.next();
View Full Code Here
92
93
94
95
96
97
98
99
100
101
        metadata.set(
                TikaCoreProperties.MODIFIED,
                font.getHeader().getModified().getTime());

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

}
View Full Code Here
49
50
51
52
53
54
55
56
57
58
59
            Archive archive = new Archive(file);

            metadata.set(Metadata.CONTENT_TYPE, TYPE.toString());
            XHTMLContentHandler xhtml =
                new XHTMLContentHandler(handler, metadata);
            xhtml.startDocument();
            for (FileHeader header : archive.getFileHeaders()) {
                Metadata entrydata = new Metadata();
                entrydata.set(
                        Metadata.RESOURCE_NAME_KEY,
                        header.getFileNameString());
View Full Code Here
63
64
65
66
67
68
69
70
71
72
73
     *      org.apache.tika.metadata.Metadata)
     */
    public void getXHTML(ContentHandler handler, Metadata metadata)
            throws SAXException, XmlException, IOException {
        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();
        buildXHTML(xhtml);
        xhtml.endDocument();
    }

    /**
 
View Full Code Here
40
41
42
43
44
45
46
47
48
49
50
    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
108
109
110
111
112
113
114
115
116
117
118
                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
67
68
69
70
71
72
73
74
75
76
        } catch (MetadataException e) {
            throw new TikaException("Can't read JPEG metadata", e);
        }

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

}
View Full Code Here
51
52
53
54
55
56
57
58
59
60
61
                reader.dispose();
            }
        }

        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
45
46
47
48
49
50
51
52
53
54
55
            Metadata metadata, ParseContext context)
            throws IOException, SAXException, TikaException {
        metadata.set(Metadata.CONTENT_TYPE, "application/x-gzip");

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

        // At the end we want to close the gzip stream to release any associated
        // resources, but the underlying document stream should not be closed
        InputStream gzip =
            new GZIPInputStream(new CloseShieldInputStream(stream));
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.