Examples with RecursiveParserWrapper org.apache.tika.parser.RecursiveParserWrapper used on opensource projects

Example 6 with RecursiveParserWrapper

use of org.apache.tika.parser.RecursiveParserWrapper in project tika by apache.

the class TikaCLI method handleRecursiveJson.

private void handleRecursiveJson(URL url, OutputStream output) throws IOException, SAXException, TikaException {
    Metadata metadata = new Metadata();
    RecursiveParserWrapper wrapper = new RecursiveParserWrapper(parser, getContentHandlerFactory(type));
    try (InputStream input = TikaInputStream.get(url, metadata)) {
        wrapper.parse(input, null, metadata, context);
    }
    JsonMetadataList.setPrettyPrinting(prettyPrint);
    Writer writer = getOutputWriter(output, encoding);
    try {
        JsonMetadataList.toJson(wrapper.getMetadata(), writer);
    } finally {
        writer.flush();
    }
}

Example 7 with RecursiveParserWrapper

use of org.apache.tika.parser.RecursiveParserWrapper in project tika by apache.

the class TikaGUI method handleStream.

private void handleStream(InputStream input, Metadata md) throws Exception {
    StringWriter htmlBuffer = new StringWriter();
    StringWriter textBuffer = new StringWriter();
    StringWriter textMainBuffer = new StringWriter();
    StringWriter xmlBuffer = new StringWriter();
    StringBuilder metadataBuffer = new StringBuilder();
    ContentHandler handler = new TeeContentHandler(getHtmlHandler(htmlBuffer), getTextContentHandler(textBuffer), getTextMainContentHandler(textMainBuffer), getXmlContentHandler(xmlBuffer));
    context.set(DocumentSelector.class, new ImageDocumentSelector());
    input = TikaInputStream.get(new ProgressMonitorInputStream(this, "Parsing stream", input));
    if (input.markSupported()) {
        int mark = -1;
        if (input instanceof TikaInputStream) {
            if (((TikaInputStream) input).hasFile()) {
                mark = (int) ((TikaInputStream) input).getLength();
            }
        }
        if (mark == -1) {
            mark = MAX_MARK;
        }
        input.mark(mark);
    }
    parser.parse(input, handler, md, context);
    String[] names = md.names();
    Arrays.sort(names);
    for (String name : names) {
        for (String val : md.getValues(name)) {
            metadataBuffer.append(name);
            metadataBuffer.append(": ");
            metadataBuffer.append(val);
            metadataBuffer.append("\n");
        }
    }
    String name = md.get(Metadata.RESOURCE_NAME_KEY);
    if (name != null && name.length() > 0) {
        setTitle("Apache Tika: " + name);
    } else {
        setTitle("Apache Tika: unnamed document");
    }
    setText(metadata, metadataBuffer.toString());
    setText(xml, xmlBuffer.toString());
    setText(text, textBuffer.toString());
    setText(textMain, textMainBuffer.toString());
    setText(html, htmlBuffer.toString());
    if (!input.markSupported()) {
        setText(json, "InputStream does not support mark/reset for Recursive Parsing");
        layout.show(cards, "metadata");
        return;
    }
    boolean isReset = false;
    try {
        input.reset();
        isReset = true;
    } catch (IOException e) {
        setText(json, "Error during stream reset.\n" + "There's a limit of " + MAX_MARK + " bytes for this type of processing in the GUI.\n" + "Try the app with command line argument of -J.");
    }
    if (isReset) {
        RecursiveParserWrapper wrapper = new RecursiveParserWrapper(parser, new BasicContentHandlerFactory(BasicContentHandlerFactory.HANDLER_TYPE.BODY, -1));
        wrapper.parse(input, null, new Metadata(), new ParseContext());
        StringWriter jsonBuffer = new StringWriter();
        JsonMetadataList.setPrettyPrinting(true);
        JsonMetadataList.toJson(wrapper.getMetadata(), jsonBuffer);
        setText(json, jsonBuffer.toString());
    }
    layout.show(cards, "metadata");
}

Example 8 with RecursiveParserWrapper

use of org.apache.tika.parser.RecursiveParserWrapper in project tika by apache.

the class RTFParserTest method testRegularImages.

//TIKA-1010 test regular (not "embedded") images/picts
@Test
public void testRegularImages() throws Exception {
    Parser base = new AutoDetectParser();
    ParseContext ctx = new ParseContext();
    RecursiveParserWrapper parser = new RecursiveParserWrapper(base, new BasicContentHandlerFactory(BasicContentHandlerFactory.HANDLER_TYPE.IGNORE, -1));
    ContentHandler handler = new BodyContentHandler();
    Metadata rootMetadata = new Metadata();
    rootMetadata.add(Metadata.RESOURCE_NAME_KEY, "testRTFRegularImages.rtf");
    try (TikaInputStream tis = TikaInputStream.get(getResourceAsStream("/test-documents/testRTFRegularImages.rtf"))) {
        parser.parse(tis, handler, rootMetadata, ctx);
    }
    List<Metadata> metadatas = parser.getMetadata();
    //("testJPEG_EXIF_普林斯顿.jpg");
    Metadata meta_jpg_exif = metadatas.get(1);
    //("testJPEG_普林斯顿.jpg");
    Metadata meta_jpg = metadatas.get(3);
    assertTrue(meta_jpg_exif != null);
    assertTrue(meta_jpg != null);
    assertTrue(Arrays.asList(meta_jpg_exif.getValues("dc:subject")).contains("serbor"));
    assertTrue(meta_jpg.get("Comments").contains("Licensed to the Apache"));
    //make sure old metadata doesn't linger between objects
    assertFalse(Arrays.asList(meta_jpg.getValues("dc:subject")).contains("serbor"));
    assertEquals("false", meta_jpg.get(RTFMetadata.THUMBNAIL));
    assertEquals("false", meta_jpg_exif.get(RTFMetadata.THUMBNAIL));
    assertEquals(49, meta_jpg.names().length);
    assertEquals(113, meta_jpg_exif.names().length);
}

Example 9 with RecursiveParserWrapper

use of org.apache.tika.parser.RecursiveParserWrapper in project tika by apache.

the class RecursiveMetadataResource method parseMetadata.

private MetadataList parseMetadata(InputStream is, MultivaluedMap<String, String> httpHeaders, UriInfo info, String handlerTypeName) throws Exception {
    final Metadata metadata = new Metadata();
    final ParseContext context = new ParseContext();
    Parser parser = TikaResource.createParser();
    // TODO: parameterize choice of max chars/max embedded attachments
    BasicContentHandlerFactory.HANDLER_TYPE type = BasicContentHandlerFactory.parseHandlerType(handlerTypeName, DEFAULT_HANDLER_TYPE);
    RecursiveParserWrapper wrapper = new RecursiveParserWrapper(parser, new BasicContentHandlerFactory(type, -1));
    TikaResource.fillMetadata(parser, metadata, context, httpHeaders);
    // no need to add parser to parse recursively
    TikaResource.fillParseContext(context, httpHeaders, null);
    TikaResource.logRequest(LOG, info, metadata);
    TikaResource.parse(wrapper, LOG, info.getPath(), is, new LanguageHandler() {

        public void endDocument() {
            metadata.set("language", getLanguage().getLanguage());
        }
    }, metadata, context);
    return new MetadataList(wrapper.getMetadata());
}

Example 10 with RecursiveParserWrapper

use of org.apache.tika.parser.RecursiveParserWrapper in project tika by apache.

the class ParsingExample method recursiveParserWrapperExample.

/**
     * For documents that may contain embedded documents, it might be helpful
     * to create list of metadata objects, one for the container document and
     * one for each embedded document.  This allows easy access to both the
     * extracted content and the metadata of each embedded document.
     * Note that many document formats can contain embedded documents,
     * including traditional container formats -- zip, tar and others -- but also
     * common office document formats including: MSWord, MSExcel,
     * MSPowerPoint, RTF, PDF, MSG and several others.
     * <p>
     * The "content" format is determined by the ContentHandlerFactory, and
     * the content is stored in {@link org.apache.tika.parser.RecursiveParserWrapper#TIKA_CONTENT}
     * <p>
     * The drawback to the RecursiveParserWrapper is that it caches metadata and contents
     * in memory.  This should not be used on files whose contents are too big to be handled
     * in memory.
     *
     * @return a list of metadata object, one each for the container file and each embedded file
     * @throws IOException
     * @throws SAXException
     * @throws TikaException
     */
public List<Metadata> recursiveParserWrapperExample() throws IOException, SAXException, TikaException {
    Parser p = new AutoDetectParser();
    ContentHandlerFactory factory = new BasicContentHandlerFactory(BasicContentHandlerFactory.HANDLER_TYPE.HTML, -1);
    RecursiveParserWrapper wrapper = new RecursiveParserWrapper(p, factory);
    Metadata metadata = new Metadata();
    metadata.set(Metadata.RESOURCE_NAME_KEY, "test_recursive_embedded.docx");
    ParseContext context = new ParseContext();
    try (InputStream stream = ParsingExample.class.getResourceAsStream("test_recursive_embedded.docx")) {
        wrapper.parse(stream, new DefaultHandler(), metadata, context);
    }
    return wrapper.getMetadata();
}