Package org.apache.tika.sax

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

134
135
136
137
138
139
140
141
142
143
144
            int n = reader.read(buffer);
            while (n != -1) {
                xhtml.characters(buffer, 0, n);
                n = reader.read(buffer);
            }
            xhtml.endElement("p");

            xhtml.endDocument();
        } catch (UnsupportedEncodingException e) {
            throw new TikaException(
                    "Unsupported text encoding: " + encoding, e);
View Full Code Here
274
275
276
277
278
279
280
281
282
283
284
                // call next handler if no filtering
                if (completelyFiltered == 0) {
                    // special handling of text:h, that are directly passed
                    // to xhtml handler
                    if (TEXT_NS.equals(namespaceURI) && "h".equals(localName)) {
                        xhtml.endElement(headingStack.pop());
                    } else {
                        super.endElement(namespaceURI,localName,qName);
                    }

                    // special handling of tabulators
View Full Code Here
57
58
59
60
61
62
63
64
65
66
67
        XHTMLContentHandler xhtml =
            new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();
        xhtml.startElement("p");
        extractText(filesystem, new AppendableAdaptor(xhtml));
        xhtml.endElement("p");
        xhtml.endDocument();
    }

    /**
     * The content type of the document being parsed.
View Full Code Here
45
46
47
48
49
50
51
52
53
54
        xhtml.startElement("p");
        char[] buffer = new char[4096];
        for (int n = reader.read(buffer); n != -1; n = reader.read(buffer)) {
            xhtml.characters(buffer, 0, n);
        }
        xhtml.endElement("p");
        xhtml.endDocument();
    }

}
View Full Code Here
102
103
104
105
106
107
108
109
110
111
112
        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();
        xhtml.startElement("p");
        xp.concatOccurrence(xmlDoc, "//*", " ", new AppendableAdaptor(xhtml));
        xhtml.endElement("p");
        xhtml.endDocument();
    }

    public List unzip(InputStream is) {
        List res = new ArrayList();
View Full Code Here
86
87
88
89
90
91
92
93
94
95
96
        getSAXParser(context).parse(
                new CloseShieldInputStream(stream),
                new OfflineContentHandler(
                        getContentHandler(handler, metadata)));

        xhtml.endElement("p");
        xhtml.endDocument();
    }

    /**
     * @deprecated This method will be removed in Apache Tika 1.0.
View Full Code Here
142
143
144
145
146
147
148
149
150
151
152
            int n = reader.read(buffer);
            while (n != -1) {
                xhtml.characters(buffer, 0, n);
                n = reader.read(buffer);
            }
            xhtml.endElement("p");

            xhtml.endDocument();
        } catch (UnsupportedEncodingException e) {
            throw new TikaException(
                    "Unsupported text encoding: " + encoding, e);
View Full Code Here
140
141
142
143
144
145
146
147
148
149
150
                    multiLine = curLine;
                } else {
                    boolean quoted = curLine.startsWith(">");
                    if (inQuote) {
                        if (!quoted) {
                            xhtml.endElement("q");
                            inQuote = false;
                        }
                    } else if (quoted) {
                        xhtml.startElement("q");
                        inQuote = true;
View Full Code Here
71
72
73
74
75
76
77
78
79
80
81
        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();
        xhtml.startElement("p");
        concatOccurrence(xmlDoc, "//*", " ", new AppendableAdaptor(xhtml));
        xhtml.endElement("p");
        xhtml.endDocument();
    }

    public void concatOccurrence(Object xmlDoc, String xpath, String concatSep, Appendable chaineConcat) throws IOException {

View Full Code Here
117
118
119
120
121
122
123
124
125
126
127
            int n = reader.read(buffer);
            while (n != -1) {
                xhtml.characters(buffer, 0, n);
                n = reader.read(buffer);
            }
            xhtml.endElement("p");

            xhtml.endDocument();
        } catch (UnsupportedEncodingException e) {
            throw new TikaException(
                    "Unsupported text encoding: " + encoding, e);
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.