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

Example 76 with AutoDetectParser

use of org.apache.tika.parser.AutoDetectParser in project jackrabbit by apache.

the class SearchIndex method createParser.

private Parser createParser() {
    URL url = null;
    if (tikaConfigPath != null) {
        File file = new File(tikaConfigPath);
        if (file.exists()) {
            try {
                url = file.toURI().toURL();
            } catch (MalformedURLException e) {
                log.warn("Invalid Tika configuration path: " + file, e);
            }
        } else {
            ClassLoader loader = SearchIndex.class.getClassLoader();
            url = loader.getResource(tikaConfigPath);
        }
    }
    if (url == null) {
        url = SearchIndex.class.getResource("tika-config.xml");
    }
    TikaConfig config = null;
    if (url != null) {
        try {
            config = new TikaConfig(url);
        } catch (Exception e) {
            log.warn("Tika configuration not available: " + url, e);
        }
    }
    if (config == null) {
        config = TikaConfig.getDefaultConfig();
    }
    if (forkJavaCommand != null) {
        ForkParser forkParser = new ForkParser(SearchIndex.class.getClassLoader(), new AutoDetectParser(config));
        forkParser.setJavaCommand(forkJavaCommand);
        forkParser.setPoolSize(extractorPoolSize);
        return forkParser;
    } else {
        return new AutoDetectParser(config);
    }
}

Example 77 with AutoDetectParser

use of org.apache.tika.parser.AutoDetectParser in project jmeter by apache.

the class Document method getTextFromDocument.

/**
     * Convert to text plain a lot of kind of document (like odt, ods, odp,
     * doc(x), xls(x), ppt(x), pdf, mp3, mp4, etc.) with Apache Tika
     *
     * @param document
     *            binary representation of the document
     * @return text from document without format
     */
public static String getTextFromDocument(byte[] document) {
    // $NON-NLS-1$
    String errMissingTika = JMeterUtils.getResString("view_results_response_missing_tika");
    String response = errMissingTika;
    Parser parser = new AutoDetectParser();
    // -1 to disable the write limit
    ContentHandler handler = new BodyContentHandler(MAX_DOCUMENT_SIZE > 0 ? MAX_DOCUMENT_SIZE : -1);
    Metadata metadata = new Metadata();
    ParseContext context = new ParseContext();
    // open the stream
    InputStream stream = new ByteArrayInputStream(document);
    try {
        parser.parse(stream, handler, metadata, context);
        response = handler.toString();
    } catch (Exception e) {
        response = e.toString();
        log.warn("Error document parsing.", e);
    } catch (NoClassDefFoundError e) {
        // put a warning if tika-app.jar missing (or some dependencies in only tika-core|parsers packages are using)
        if (!System.getProperty("java.class.path").contains("tika-app")) {
            // $NON-NLS-1$ $NON-NLS-2$ 
            log.warn(errMissingTika);
        } else {
            log.warn(errMissingTika, e);
        }
    } finally {
        try {
            // close the stream
            stream.close();
        } catch (IOException ioe) {
            // $NON-NLS-1$
            log.warn("Error closing document stream", ioe);
        }
    }
    if (response.length() == 0 && document.length > 0) {
        // $NON-NLS-1$
        log.warn("Probably: {}", errMissingTika);
        response = errMissingTika;
    }
    return response;
}

Example 78 with AutoDetectParser

use of org.apache.tika.parser.AutoDetectParser in project jackrabbit-oak by apache.

the class BinaryTextExtractor method createDefaultParser.

private static AutoDetectParser createDefaultParser() {
    ClassLoader current = Thread.currentThread().getContextClassLoader();
    URL configUrl = LuceneIndexEditorContext.class.getResource("tika-config.xml");
    InputStream is = null;
    if (configUrl != null) {
        try {
            Thread.currentThread().setContextClassLoader(LuceneIndexEditorContext.class.getClassLoader());
            is = configUrl.openStream();
            TikaConfig config = new TikaConfig(is);
            log.info("Loaded default Tika Config from classpath {}", configUrl);
            return new AutoDetectParser(config);
        } catch (Exception e) {
            log.warn("Tika configuration not available : " + configUrl, e);
        } finally {
            IOUtils.closeQuietly(is);
            Thread.currentThread().setContextClassLoader(current);
        }
    } else {
        log.warn("Default Tika configuration not found");
    }
    return new AutoDetectParser();
}

Example 79 with AutoDetectParser

use of org.apache.tika.parser.AutoDetectParser in project jackrabbit-oak by apache.

the class BinaryTextExtractor method initializeTikaParser.

private static Parser initializeTikaParser(IndexDefinition definition) {
    ClassLoader current = Thread.currentThread().getContextClassLoader();
    try {
        if (definition.hasCustomTikaConfig()) {
            log.debug("[{}] Using custom tika config", definition.getIndexName());
            Thread.currentThread().setContextClassLoader(LuceneIndexEditorContext.class.getClassLoader());
            InputStream is = definition.getTikaConfig();
            try {
                return new AutoDetectParser(getTikaConfig(is, definition));
            } finally {
                IOUtils.closeQuietly(is);
            }
        }
    } finally {
        Thread.currentThread().setContextClassLoader(current);
    }
    return defaultParser;
}

Example 80 with AutoDetectParser

use of org.apache.tika.parser.AutoDetectParser in project ddf by codice.

the class PdfInputTransformer method transformWithExtractors.

private Metacard transformWithExtractors(InputStream input, String id) throws IOException, CatalogTransformerException {
    try (TemporaryFileBackedOutputStream fbos = new TemporaryFileBackedOutputStream()) {
        try {
            IOUtils.copy(input, fbos);
        } catch (IOException e) {
            throw new CatalogTransformerException("Could not copy bytes of content message.", e);
        }
        String plainText = null;
        try (InputStream isCopy = fbos.asByteSource().openStream()) {
            Parser parser = new AutoDetectParser();
            ContentHandler contentHandler = new ToTextContentHandler();
            TikaMetadataExtractor tikaMetadataExtractor = new TikaMetadataExtractor(parser, contentHandler);
            tikaMetadataExtractor.parseMetadata(isCopy, new ParseContext());
            plainText = contentHandler.toString();
        } catch (CatalogTransformerException e) {
            LOGGER.warn("Cannot extract metadata from pdf", e);
        }
        try (InputStream isCopy = fbos.asByteSource().openStream();
            PDDocument pdfDocument = pdDocumentGenerator.apply(isCopy)) {
            return transformPdf(id, pdfDocument, plainText);
        } catch (InvalidPasswordException e) {
            LOGGER.debug("Cannot transform encrypted pdf", e);
            return initializeMetacard(id);
        }
    }
}