Package org.apache.tika.parser

Examples of org.apache.tika.parser.ParseContext

149
150
151
152
153
154
155
     * @deprecated This method will be removed in Apache Tika 1.0.
     */
    public void parse(
            InputStream stream, ContentHandler handler, Metadata metadata)
    throws IOException, SAXException, TikaException {
        parse(stream, handler, metadata, new ParseContext());
    }
View Full Code Here
181
182
183
184
185
186
187
     * @deprecated This method will be removed in Apache Tika 1.0.
     */
    public void parse(
            InputStream stream, ContentHandler handler, Metadata metadata)
    throws IOException, SAXException, TikaException {
        parse(stream, handler, metadata, new ParseContext());
    }
View Full Code Here
74
75
76
77
78
79
80
     * @deprecated This method will be removed in Apache Tika 1.0.
     */
    public void parse(
            InputStream stream, ContentHandler handler, Metadata metadata)
            throws IOException, SAXException, TikaException {
        parse(stream, handler, metadata, new ParseContext());
    }
View Full Code Here
62
63
64
65
66
67
68
     * @deprecated This method will be removed in Apache Tika 1.0.
     */
    public void parse(
            InputStream stream, ContentHandler handler, Metadata metadata)
            throws IOException, SAXException, TikaException {
        parse(stream, handler, metadata, new ParseContext());
    }
View Full Code Here
60
61
62
63
64
65
66
67
68
69
70
                    }
                }
            };
            new HtmlParser().parse(
                    stream, new TeeContentHandler(body, link),
                    metadata, new ParseContext());
        } finally {
            stream.close();
        }

        assertEquals(
View Full Code Here
120
121
122
123
124
125
126
127
    public void testParseEmpty() throws Exception {
        ContentHandler handler = new BodyContentHandler();
        new HtmlParser().parse(
                new ByteArrayInputStream(new byte[0]),
                handler,  new Metadata(), new ParseContext());
        assertEquals("", handler.toString());
    }
View Full Code Here
198
199
200
201
202
203
204
205
206
                            links.add(atts.getValue("", "href"));
                        }
                    }
                },
                new Metadata(),
                new ParseContext());
        assertEquals(1, links.size());
        assertEquals(url, links.get(0));
    }
View Full Code Here
230
231
232
233
234
235
236
237
            + "<title>the name is \u00e1ndre</title>"
            + "</head><body></body></html>";
        Metadata metadata = new Metadata();
        new HtmlParser().parse (
                new ByteArrayInputStream(test.getBytes("UTF-8")),
                new BodyContentHandler(),  metadata, new ParseContext());
        assertEquals("ISO-8859-1", metadata.get(Metadata.CONTENT_ENCODING));
    }
View Full Code Here
244
245
246
247
248
249
250
251
        String test =
            "<html><head><title>\u017d</title></head><body></body></html>";
        Metadata metadata = new Metadata();
        new HtmlParser().parse (
                new ByteArrayInputStream(test.getBytes("UTF-8")),
                new BodyContentHandler(),  metadata, new ParseContext());
        assertEquals("\u017d", metadata.get(Metadata.TITLE));
    }
View Full Code Here
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
            + "<body></body></html>";

        Metadata metadata = new Metadata();
        new HtmlParser().parse (
                new ByteArrayInputStream(test.getBytes("UTF-8")),
                new BodyContentHandler(),  metadata, new ParseContext());
        assertEquals("UTF-8", metadata.get(Metadata.CONTENT_ENCODING));

        metadata = new Metadata();
        metadata.set(Metadata.CONTENT_TYPE, "text/html; charset=ISO-8859-1");
        new HtmlParser().parse (
                new ByteArrayInputStream(test.getBytes("UTF-8")),
                new BodyContentHandler(),  metadata, new ParseContext());
        assertEquals("ISO-8859-1", metadata.get(Metadata.CONTENT_ENCODING));
    }
View Full Code Here
TOP

Related Classes of org.apache.tika.parser.ParseContext

  • com.cloudera.cdk.morphline.solrcell.SolrCellBuilder$SolrCell
  • com.nidhinova.tika.server.TikaService
  • com.tamingtext.tika.TikaTest
  • cx.fbn.nevernote.threads.IndexRunner
  • gov.nara.nwts.ftappImg.jpeg.JpegExtractor
  • org.apache.chemistry.opencmis.client.parser.MetadataParserTika
  • org.apache.jackrabbit.core.query.lucene.JackrabbitParser
  • org.apache.jackrabbit.core.query.lucene.LazyTextExtractorField$ParsingTask
  • org.apache.jackrabbit.oak.plugins.index.lucene.LuceneIndexEditor
  • org.apache.nutch.parse.tika.DOMContentUtilsTest
    • 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.