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

Example 1 with AutoDetectParser

use of org.apache.tika.parser.AutoDetectParser in project lucene-solr by apache.

the class TikaEntityProcessor method nextRow.

@Override
public Map<String, Object> nextRow() {
    if (done)
        return null;
    Map<String, Object> row = new HashMap<>();
    DataSource<InputStream> dataSource = context.getDataSource();
    InputStream is = dataSource.getData(context.getResolvedEntityAttribute(URL));
    ContentHandler contentHandler = null;
    Metadata metadata = new Metadata();
    StringWriter sw = new StringWriter();
    try {
        if ("html".equals(format)) {
            contentHandler = getHtmlHandler(sw);
        } else if ("xml".equals(format)) {
            contentHandler = getXmlContentHandler(sw);
        } else if ("text".equals(format)) {
            contentHandler = getTextContentHandler(sw);
        } else if ("none".equals(format)) {
            contentHandler = new DefaultHandler();
        }
    } catch (TransformerConfigurationException e) {
        wrapAndThrow(SEVERE, e, "Unable to create content handler");
    }
    Parser tikaParser = null;
    if (parser.equals(AUTO_PARSER)) {
        tikaParser = new AutoDetectParser(tikaConfig);
    } else {
        tikaParser = context.getSolrCore().getResourceLoader().newInstance(parser, Parser.class);
    }
    try {
        ParseContext context = new ParseContext();
        if ("identity".equals(htmlMapper)) {
            context.set(HtmlMapper.class, IdentityHtmlMapper.INSTANCE);
        }
        if (extractEmbedded) {
            context.set(Parser.class, tikaParser);
        }
        tikaParser.parse(is, contentHandler, metadata, context);
    } catch (Exception e) {
        if (SKIP.equals(onError)) {
            throw new DataImportHandlerException(DataImportHandlerException.SKIP_ROW, "Document skipped :" + e.getMessage());
        }
        wrapAndThrow(SEVERE, e, "Unable to read content");
    }
    IOUtils.closeQuietly(is);
    for (Map<String, String> field : context.getAllEntityFields()) {
        if (!"true".equals(field.get("meta")))
            continue;
        String col = field.get(COLUMN);
        String s = metadata.get(col);
        if (s != null)
            row.put(col, s);
    }
    if (!"none".equals(format))
        row.put("text", sw.toString());
    tryToAddLatLon(metadata, row);
    done = true;
    return row;
}

Example 2 with AutoDetectParser

use of org.apache.tika.parser.AutoDetectParser 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 3 with AutoDetectParser

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

the class TikaTest method getRecursiveMetadata.

protected List<Metadata> getRecursiveMetadata(String filePath, ParseContext context) throws Exception {
    Parser p = new AutoDetectParser();
    RecursiveParserWrapper wrapper = new RecursiveParserWrapper(p, new BasicContentHandlerFactory(BasicContentHandlerFactory.HANDLER_TYPE.XML, -1));
    try (InputStream is = getResourceAsStream("/test-documents/" + filePath)) {
        wrapper.parse(is, new DefaultHandler(), new Metadata(), context);
    }
    return wrapper.getMetadata();
}

Example 4 with AutoDetectParser

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

the class TikaEncodingDetectorTest method testConfigurabilityOfUserSpecified.

@Test
public void testConfigurabilityOfUserSpecified() throws Exception {
    TikaConfig tikaConfig = new TikaConfig(getResourceAsStream("/org/apache/tika/config/TIKA-2273-encoding-detector-outside-static-init.xml"));
    AutoDetectParser p = new AutoDetectParser(tikaConfig);
    //make sure that all static and non-static parsers are using the same encoding detector!
    List<Parser> parsers = new ArrayList<>();
    findEncodingDetectionParsers(p, parsers);
    assertEquals(3, parsers.size());
    for (Parser encodingDetectingParser : parsers) {
        EncodingDetector encodingDetector = ((AbstractEncodingDetectorParser) encodingDetectingParser).getEncodingDetector();
        assertTrue(encodingDetector instanceof CompositeEncodingDetector);
        assertEquals(2, ((CompositeEncodingDetector) encodingDetector).getDetectors().size());
        for (EncodingDetector child : ((CompositeEncodingDetector) encodingDetector).getDetectors()) {
            assertNotContained("cu4j", child.getClass().getCanonicalName());
        }
    }
    //also just make sure this is still true
    try {
        Metadata metadata = getXML("english.cp500.txt", p).metadata;
        fail("can't detect w/out ICU");
    } catch (TikaException e) {
        assertContains("Failed to detect", e.getMessage());
    }
}

Example 5 with AutoDetectParser

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

the class TikaEncodingDetectorTest method testNonDetectingDetectorParams.

@Test
public void testNonDetectingDetectorParams() throws Exception {
    TikaConfig tikaConfig = new TikaConfig(getResourceAsStream("/org/apache/tika/config/TIKA-2273-non-detecting-params.xml"));
    AutoDetectParser p = new AutoDetectParser(tikaConfig);
    List<Parser> parsers = new ArrayList<>();
    findEncodingDetectionParsers(p, parsers);
    assertEquals(3, parsers.size());
    EncodingDetector encodingDetector = ((AbstractEncodingDetectorParser) parsers.get(0)).getEncodingDetector();
    assertTrue(encodingDetector instanceof CompositeEncodingDetector);
    assertEquals(1, ((CompositeEncodingDetector) encodingDetector).getDetectors().size());
    EncodingDetector child = ((CompositeEncodingDetector) encodingDetector).getDetectors().get(0);
    assertTrue(child instanceof NonDetectingEncodingDetector);
    assertEquals(StandardCharsets.UTF_16LE, ((NonDetectingEncodingDetector) child).getCharset());
}