Examples with TemporaryResources org.apache.tika.io.TemporaryResources used on opensource projects

Example 21 with TemporaryResources

use of org.apache.tika.io.TemporaryResources in project tika by apache.

the class ParserDecorator method withFallbacks.

/**
     * Decorates the given parsers into a virtual parser, where they'll
     *  be tried in preference order until one works without error.
     * TODO Is this the right name?
     * TODO Is this the right place to put this? Should it be in CompositeParser? Elsewhere?
     * TODO Should we reset the Metadata if we try another parser?
     * TODO Should we reset the ContentHandler if we try another parser?
     * TODO Should we log/report failures anywhere?
     * @deprecated Do not use until the TODOs are resolved, see TIKA-1509
     */
public static final Parser withFallbacks(final Collection<? extends Parser> parsers, final Set<MediaType> types) {
    Parser parser = EmptyParser.INSTANCE;
    if (!parsers.isEmpty())
        parser = parsers.iterator().next();
    return new ParserDecorator(parser) {

        private static final long serialVersionUID = 1625187131782069683L;

        @Override
        public Set<MediaType> getSupportedTypes(ParseContext context) {
            return types;
        }

        @Override
        public void parse(InputStream stream, ContentHandler handler, Metadata metadata, ParseContext context) throws IOException, SAXException, TikaException {
            // Must have a TikaInputStream, so we can re-use it if parsing fails
            // Need to close internally created tstream to release resources
            TemporaryResources tmp = (TikaInputStream.isTikaInputStream(stream)) ? null : new TemporaryResources();
            try {
                TikaInputStream tstream = TikaInputStream.get(stream, tmp);
                tstream.getFile();
                // Try each parser in turn
                for (Parser p : parsers) {
                    tstream.mark(-1);
                    try {
                        p.parse(tstream, handler, metadata, context);
                        return;
                    } catch (Exception e) {
                    // TODO How to log / record this failure?
                    }
                    // Prepare for the next parser, if present
                    tstream.reset();
                }
            } finally {
                if (tmp != null) {
                    tmp.dispose();
                }
            }
        }

        @Override
        public String getDecorationName() {
            return "With Fallback";
        }
    };
}

Example 22 with TemporaryResources

use of org.apache.tika.io.TemporaryResources in project tika by apache.

the class CompositeParser method parse.

/**
     * Delegates the call to the matching component parser.
     * <p>
     * Potential {@link RuntimeException}s, {@link IOException}s and
     * {@link SAXException}s unrelated to the given input stream and content
     * handler are automatically wrapped into {@link TikaException}s to better
     * honor the {@link Parser} contract.
     */
public void parse(InputStream stream, ContentHandler handler, Metadata metadata, ParseContext context) throws IOException, SAXException, TikaException {
    Parser parser = getParser(metadata, context);
    TemporaryResources tmp = new TemporaryResources();
    try {
        TikaInputStream taggedStream = TikaInputStream.get(stream, tmp);
        TaggedContentHandler taggedHandler = handler != null ? new TaggedContentHandler(handler) : null;
        if (parser instanceof ParserDecorator) {
            metadata.add("X-Parsed-By", ((ParserDecorator) parser).getWrappedParser().getClass().getName());
        } else {
            metadata.add("X-Parsed-By", parser.getClass().getName());
        }
        try {
            parser.parse(taggedStream, taggedHandler, metadata, context);
        } catch (RuntimeException e) {
            throw new TikaException("Unexpected RuntimeException from " + parser, e);
        } catch (IOException e) {
            taggedStream.throwIfCauseOf(e);
            throw new TikaException("TIKA-198: Illegal IOException from " + parser, e);
        } catch (SAXException e) {
            if (taggedHandler != null)
                taggedHandler.throwIfCauseOf(e);
            throw new TikaException("TIKA-237: Illegal SAXException from " + parser, e);
        }
    } finally {
        tmp.dispose();
    }
}

Example 23 with TemporaryResources

use of org.apache.tika.io.TemporaryResources in project tika by apache.

the class NetworkParser method parse.

public void parse(InputStream stream, ContentHandler handler, Metadata metadata, ParseContext context) throws IOException, SAXException, TikaException {
    TemporaryResources tmp = new TemporaryResources();
    try {
        TikaInputStream tis = TikaInputStream.get(stream, tmp);
        parse(tis, handler, metadata, context);
    } finally {
        tmp.dispose();
    }
}

Example 24 with TemporaryResources

use of org.apache.tika.io.TemporaryResources in project tika by apache.

the class GeographicInformationParser method parse.

@Override
public void parse(InputStream inputStream, ContentHandler contentHandler, Metadata metadata, ParseContext parseContext) throws IOException, SAXException, TikaException {
    metadata.set(Metadata.CONTENT_TYPE, geoInfoType);
    DataStore dataStore = null;
    DefaultMetadata defaultMetadata = null;
    XHTMLContentHandler xhtmlContentHandler = new XHTMLContentHandler(contentHandler, metadata);
    TemporaryResources tmp = TikaInputStream.isTikaInputStream(inputStream) ? null : new TemporaryResources();
    try {
        TikaInputStream tikaInputStream = TikaInputStream.get(inputStream, tmp);
        File file = tikaInputStream.getFile();
        dataStore = DataStores.open(file);
        defaultMetadata = new DefaultMetadata(dataStore.getMetadata());
        if (defaultMetadata != null)
            extract(xhtmlContentHandler, metadata, defaultMetadata);
    } catch (UnsupportedStorageException e) {
        throw new TikaException("UnsupportedStorageException", e);
    } catch (DataStoreException e) {
        throw new TikaException("DataStoreException", e);
    } finally {
        if (tmp != null) {
            tmp.dispose();
        }
    }
}

Example 25 with TemporaryResources

use of org.apache.tika.io.TemporaryResources in project tika by apache.

the class GribParser method parse.

public void parse(InputStream stream, ContentHandler handler, Metadata metadata, ParseContext context) throws IOException, SAXException, TikaException {
    //Set MIME type as grib2
    metadata.set(Metadata.CONTENT_TYPE, GRIB_MIME_TYPE);
    TikaInputStream tis = TikaInputStream.get(stream, new TemporaryResources());
    File gribFile = tis.getFile();
    try {
        NetcdfFile ncFile = NetcdfDataset.openFile(gribFile.getAbsolutePath(), null);
        // first parse out the set of global attributes
        for (Attribute attr : ncFile.getGlobalAttributes()) {
            Property property = resolveMetadataKey(attr.getFullName());
            if (attr.getDataType().isString()) {
                metadata.add(property, attr.getStringValue());
            } else if (attr.getDataType().isNumeric()) {
                int value = attr.getNumericValue().intValue();
                metadata.add(property, String.valueOf(value));
            }
        }
        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();
        xhtml.newline();
        xhtml.startElement("ul");
        xhtml.characters("dimensions:");
        xhtml.newline();
        for (Dimension dim : ncFile.getDimensions()) {
            xhtml.element("li", dim.getFullName() + "=" + String.valueOf(dim.getLength()) + ";");
            xhtml.newline();
        }
        xhtml.startElement("ul");
        xhtml.characters("variables:");
        xhtml.newline();
        for (Variable var : ncFile.getVariables()) {
            xhtml.element("p", String.valueOf(var.getDataType()) + var.getNameAndDimensions() + ";");
            for (Attribute element : var.getAttributes()) {
                xhtml.element("li", " :" + element + ";");
                xhtml.newline();
            }
        }
        xhtml.endElement("ul");
        xhtml.endElement("ul");
        xhtml.endDocument();
    } catch (IOException e) {
        throw new TikaException("NetCDF parse error", e);
    }
}