Examples with MediaType org.apache.tika.mime.MediaType used on opensource projects

Example 1 with MediaType

use of org.apache.tika.mime.MediaType in project che by eclipse.

the class MediaTypeFilter method accept.

@Override
public boolean accept(VirtualFile file) {
    try (InputStream content = file.getContent()) {
        TikaConfig tikaConfig = new TikaConfig();
        MediaType mimeType = tikaConfig.getDetector().detect(content, new Metadata());
        if (excludedMediaTypes.contains(mimeType) || excludedTypes.contains(mimeType.getType())) {
            return true;
        }
        return false;
    } catch (TikaException | ForbiddenException | ServerException | IOException e) {
        return true;
    }
}

Example 2 with MediaType

use of org.apache.tika.mime.MediaType in project camel by apache.

the class TikaProducer method doDetect.

private Object doDetect(Exchange exchange) throws IOException {
    InputStream inputStream = exchange.getIn().getBody(InputStream.class);
    Metadata metadata = new Metadata();
    MediaType result = this.detector.detect(inputStream, metadata);
    convertMetadataToHeaders(metadata, exchange);
    return result.toString();
}

Example 3 with MediaType

use of org.apache.tika.mime.MediaType in project jackrabbit-oak by apache.

the class MediaRange method parse.

public static MediaRange parse(String range, MediaTypeRegistry registry) {
    MediaType type = MediaType.parse(range);
    if (type == null) {
        return null;
    }
    type = registry.normalize(type);
    Map<String, String> parameters = new HashMap<String, String>(type.getParameters());
    String q = parameters.remove("q");
    if (q != null) {
        try {
            return new MediaRange(new MediaType(type.getBaseType(), parameters), Double.parseDouble(q));
        } catch (NumberFormatException e) {
            return null;
        }
    }
    return new MediaRange(type, 1.0);
}

Example 4 with MediaType

use of org.apache.tika.mime.MediaType in project lucene-solr by apache.

the class ExtractingDocumentLoader method load.

@Override
public void load(SolrQueryRequest req, SolrQueryResponse rsp, ContentStream stream, UpdateRequestProcessor processor) throws Exception {
    Parser parser = null;
    String streamType = req.getParams().get(ExtractingParams.STREAM_TYPE, null);
    if (streamType != null) {
        //Cache?  Parsers are lightweight to construct and thread-safe, so I'm told
        MediaType mt = MediaType.parse(streamType.trim().toLowerCase(Locale.ROOT));
        parser = new DefaultParser(config.getMediaTypeRegistry()).getParsers().get(mt);
    } else {
        parser = autoDetectParser;
    }
    if (parser != null) {
        Metadata metadata = new Metadata();
        // If you specify the resource name (the filename, roughly) with this parameter,
        // then Tika can make use of it in guessing the appropriate MIME type:
        String resourceName = req.getParams().get(ExtractingParams.RESOURCE_NAME, null);
        if (resourceName != null) {
            metadata.add(TikaMetadataKeys.RESOURCE_NAME_KEY, resourceName);
        }
        // Provide stream's content type as hint for auto detection
        if (stream.getContentType() != null) {
            metadata.add(HttpHeaders.CONTENT_TYPE, stream.getContentType());
        }
        InputStream inputStream = null;
        try {
            inputStream = stream.getStream();
            metadata.add(ExtractingMetadataConstants.STREAM_NAME, stream.getName());
            metadata.add(ExtractingMetadataConstants.STREAM_SOURCE_INFO, stream.getSourceInfo());
            metadata.add(ExtractingMetadataConstants.STREAM_SIZE, String.valueOf(stream.getSize()));
            metadata.add(ExtractingMetadataConstants.STREAM_CONTENT_TYPE, stream.getContentType());
            // HtmlParser and TXTParser regard Metadata.CONTENT_ENCODING in metadata
            String charset = ContentStreamBase.getCharsetFromContentType(stream.getContentType());
            if (charset != null) {
                metadata.add(HttpHeaders.CONTENT_ENCODING, charset);
            }
            String xpathExpr = params.get(ExtractingParams.XPATH_EXPRESSION);
            boolean extractOnly = params.getBool(ExtractingParams.EXTRACT_ONLY, false);
            SolrContentHandler handler = factory.createSolrContentHandler(metadata, params, req.getSchema());
            ContentHandler parsingHandler = handler;
            StringWriter writer = null;
            BaseMarkupSerializer serializer = null;
            if (extractOnly == true) {
                String extractFormat = params.get(ExtractingParams.EXTRACT_FORMAT, "xml");
                writer = new StringWriter();
                if (extractFormat.equals(TEXT_FORMAT)) {
                    serializer = new TextSerializer();
                    serializer.setOutputCharStream(writer);
                    serializer.setOutputFormat(new OutputFormat("Text", "UTF-8", true));
                } else {
                    serializer = new XMLSerializer(writer, new OutputFormat("XML", "UTF-8", true));
                }
                if (xpathExpr != null) {
                    Matcher matcher = PARSER.parse(xpathExpr);
                    //The MatchingContentHandler does not invoke startDocument.  See http://tika.markmail.org/message/kknu3hw7argwiqin
                    serializer.startDocument();
                    parsingHandler = new MatchingContentHandler(serializer, matcher);
                } else {
                    parsingHandler = serializer;
                }
            } else if (xpathExpr != null) {
                Matcher matcher = PARSER.parse(xpathExpr);
                parsingHandler = new MatchingContentHandler(handler, matcher);
            }
            try {
                //potentially use a wrapper handler for parsing, but we still need the SolrContentHandler for getting the document.
                ParseContext context = parseContextConfig.create();
                context.set(Parser.class, parser);
                context.set(HtmlMapper.class, MostlyPassthroughHtmlMapper.INSTANCE);
                // Password handling
                RegexRulesPasswordProvider epp = new RegexRulesPasswordProvider();
                String pwMapFile = params.get(ExtractingParams.PASSWORD_MAP_FILE);
                if (pwMapFile != null && pwMapFile.length() > 0) {
                    InputStream is = req.getCore().getResourceLoader().openResource(pwMapFile);
                    if (is != null) {
                        log.debug("Password file supplied: " + pwMapFile);
                        epp.parse(is);
                    }
                }
                context.set(PasswordProvider.class, epp);
                String resourcePassword = params.get(ExtractingParams.RESOURCE_PASSWORD);
                if (resourcePassword != null) {
                    epp.setExplicitPassword(resourcePassword);
                    log.debug("Literal password supplied for file " + resourceName);
                }
                parser.parse(inputStream, parsingHandler, metadata, context);
            } catch (TikaException e) {
                if (ignoreTikaException)
                    log.warn(new StringBuilder("skip extracting text due to ").append(e.getLocalizedMessage()).append(". metadata=").append(metadata.toString()).toString());
                else
                    throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, e);
            }
            if (extractOnly == false) {
                addDoc(handler);
            } else {
                //serializer is not null, so we need to call endDoc on it if using xpath
                if (xpathExpr != null) {
                    serializer.endDocument();
                }
                rsp.add(stream.getName(), writer.toString());
                writer.close();
                String[] names = metadata.names();
                NamedList metadataNL = new NamedList();
                for (int i = 0; i < names.length; i++) {
                    String[] vals = metadata.getValues(names[i]);
                    metadataNL.add(names[i], vals);
                }
                rsp.add(stream.getName() + "_metadata", metadataNL);
            }
        } catch (SAXException e) {
            throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, e);
        } finally {
            IOUtils.closeQuietly(inputStream);
        }
    } else {
        throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "Stream type of " + streamType + " didn't match any known parsers.  Please supply the " + ExtractingParams.STREAM_TYPE + " parameter.");
    }
}

Example 5 with MediaType

use of org.apache.tika.mime.MediaType in project uPortal by Jasig.

the class JaxbPortalDataHandlerService method importDataArchive.

protected void importDataArchive(Resource archive, InputStream resourceStream, BatchImportOptions options) {
    BufferedInputStream bufferedResourceStream = null;
    try {
        //Make sure the stream is buffered
        if (resourceStream instanceof BufferedInputStream) {
            bufferedResourceStream = (BufferedInputStream) resourceStream;
        } else {
            bufferedResourceStream = new BufferedInputStream(resourceStream);
        }
        //Buffer up to 100MB, bad things will happen if we bust this buffer.
        //TODO see if there is a buffered stream that will write to a file once the buffer fills up
        bufferedResourceStream.mark(100 * 1024 * 1024);
        final MediaType type = getMediaType(bufferedResourceStream, archive.getFilename());
        if (MT_JAVA_ARCHIVE.equals(type)) {
            final ArchiveInputStream archiveStream = new JarArchiveInputStream(bufferedResourceStream);
            importDataArchive(archive, archiveStream, options);
        } else if (MediaType.APPLICATION_ZIP.equals(type)) {
            final ArchiveInputStream archiveStream = new ZipArchiveInputStream(bufferedResourceStream);
            importDataArchive(archive, archiveStream, options);
        } else if (MT_CPIO.equals(type)) {
            final ArchiveInputStream archiveStream = new CpioArchiveInputStream(bufferedResourceStream);
            importDataArchive(archive, archiveStream, options);
        } else if (MT_AR.equals(type)) {
            final ArchiveInputStream archiveStream = new ArArchiveInputStream(bufferedResourceStream);
            importDataArchive(archive, archiveStream, options);
        } else if (MT_TAR.equals(type)) {
            final ArchiveInputStream archiveStream = new TarArchiveInputStream(bufferedResourceStream);
            importDataArchive(archive, archiveStream, options);
        } else if (MT_BZIP2.equals(type)) {
            final CompressorInputStream compressedStream = new BZip2CompressorInputStream(bufferedResourceStream);
            importDataArchive(archive, compressedStream, options);
        } else if (MT_GZIP.equals(type)) {
            final CompressorInputStream compressedStream = new GzipCompressorInputStream(bufferedResourceStream);
            importDataArchive(archive, compressedStream, options);
        } else if (MT_PACK200.equals(type)) {
            final CompressorInputStream compressedStream = new Pack200CompressorInputStream(bufferedResourceStream);
            importDataArchive(archive, compressedStream, options);
        } else if (MT_XZ.equals(type)) {
            final CompressorInputStream compressedStream = new XZCompressorInputStream(bufferedResourceStream);
            importDataArchive(archive, compressedStream, options);
        } else {
            throw new RuntimeException("Unrecognized archive media type: " + type);
        }
    } catch (IOException e) {
        throw new RuntimeException("Could not load InputStream for resource: " + archive, e);
    } finally {
        IOUtils.closeQuietly(bufferedResourceStream);
    }
}