Examples with XHTMLContentHandler org.apache.tika.sax.XHTMLContentHandler used on opensource projects

Example 11 with XHTMLContentHandler

use of org.apache.tika.sax.XHTMLContentHandler in project tika by apache.

the class MatParser method parse.

public void parse(InputStream stream, ContentHandler handler, Metadata metadata, ParseContext context) throws IOException, SAXException, TikaException {
    //Set MIME type as Matlab
    metadata.set(Metadata.CONTENT_TYPE, MATLAB_MIME_TYPE);
    TemporaryResources tmp = TikaInputStream.isTikaInputStream(stream) ? null : new TemporaryResources();
    try {
        // Use TIS so we can spool a temp file for parsing.
        TikaInputStream tis = TikaInputStream.get(stream, tmp);
        //Extract information from header file
        //input .mat file
        MatFileReader mfr = new MatFileReader(tis.getFile());
        //.mat header information
        MatFileHeader hdr = mfr.getMatFileHeader();
        // Example header: "MATLAB 5.0 MAT-file, Platform: MACI64, Created on: Sun Mar  2 23:41:57 2014"
        // Break header information into its parts
        String[] parts = hdr.getDescription().split(",");
        if (parts[2].contains("Created")) {
            int lastIndex1 = parts[2].lastIndexOf("Created on:");
            String dateCreated = parts[2].substring(lastIndex1 + "Created on:".length()).trim();
            metadata.set("createdOn", dateCreated);
        }
        if (parts[1].contains("Platform")) {
            int lastIndex2 = parts[1].lastIndexOf("Platform:");
            String platform = parts[1].substring(lastIndex2 + "Platform:".length()).trim();
            metadata.set("platform", platform);
        }
        if (parts[0].contains("MATLAB")) {
            metadata.set("fileType", parts[0]);
        }
        // Get endian indicator from header file
        // Retrieve endian bytes and convert to string
        String endianBytes = new String(hdr.getEndianIndicator(), UTF_8);
        // Convert bytes to characters to string
        String endianCode = String.valueOf(endianBytes.toCharArray());
        metadata.set("endian", endianCode);
        //Text output	
        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();
        xhtml.newline();
        //Loop through each variable
        for (Map.Entry<String, MLArray> entry : mfr.getContent().entrySet()) {
            String varName = entry.getKey();
            MLArray varData = entry.getValue();
            xhtml.element("p", varName + ":" + String.valueOf(varData));
            // If the variable is a structure, extract variable info from structure
            if (varData.isStruct()) {
                MLStructure mlStructure = (MLStructure) mfr.getMLArray(varName);
                xhtml.startElement("ul");
                xhtml.newline();
                for (MLArray element : mlStructure.getAllFields()) {
                    xhtml.startElement("li");
                    xhtml.characters(String.valueOf(element));
                    // If there is an embedded structure, extract variable info.
                    if (element.isStruct()) {
                        xhtml.startElement("ul");
                        // Should this actually be a recursive call?
                        xhtml.element("li", element.contentToString());
                        xhtml.endElement("ul");
                    }
                    xhtml.endElement("li");
                }
                xhtml.endElement("ul");
            }
        }
        xhtml.endDocument();
    } catch (IOException e) {
        throw new TikaException("Error parsing Matlab file with MatParser", e);
    } finally {
        if (tmp != null) {
            tmp.dispose();
        }
    }
}

Example 12 with XHTMLContentHandler

use of org.apache.tika.sax.XHTMLContentHandler in project tika by apache.

the class JackcessParser method parse.

@Override
public void parse(InputStream stream, ContentHandler handler, Metadata metadata, ParseContext context) throws IOException, SAXException, TikaException {
    TikaInputStream tis = TikaInputStream.get(stream);
    Database db = null;
    XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
    xhtml.startDocument();
    String password = null;
    PasswordProvider passwordProvider = context.get(PasswordProvider.class);
    if (passwordProvider != null) {
        password = passwordProvider.getPassword(metadata);
    }
    try {
        if (password == null) {
            //do this to ensure encryption/wrong password exception vs. more generic
            //"need right codec" error message.
            db = new DatabaseBuilder(tis.getFile()).setCodecProvider(new CryptCodecProvider()).setReadOnly(true).open();
        } else {
            db = new DatabaseBuilder(tis.getFile()).setCodecProvider(new CryptCodecProvider(password)).setReadOnly(true).open();
        }
        //just in case
        db.setLinkResolver(IGNORE_LINK_RESOLVER);
        JackcessExtractor ex = new JackcessExtractor(metadata, context, locale);
        ex.parse(db, xhtml);
    } catch (IllegalStateException e) {
        if (e.getMessage() != null && e.getMessage().contains("Incorrect password")) {
            throw new EncryptedDocumentException(e);
        }
        throw e;
    } finally {
        if (db != null) {
            try {
                db.close();
            } catch (IOException e) {
            //swallow = silent close
            }
        }
    }
    xhtml.endDocument();
}

Example 13 with XHTMLContentHandler

use of org.apache.tika.sax.XHTMLContentHandler in project tika by apache.

the class OldExcelParser method parse.

/**
     * Extracts properties and text from an MS Document input stream
     */
public void parse(InputStream stream, ContentHandler handler, Metadata metadata, ParseContext context) throws IOException, SAXException, TikaException {
    // Open the POI provided extractor
    OldExcelExtractor extractor = new OldExcelExtractor(stream);
    // We can't do anything about metadata, as these old formats
    //  didn't have any stored with them
    // Set the content type
    // TODO Get the version and type, to set as the Content Type
    // Have the text extracted and given to our Content Handler
    XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
    parse(extractor, xhtml);
}

Example 14 with XHTMLContentHandler

use of org.apache.tika.sax.XHTMLContentHandler in project tika by apache.

the class NetCDFParser method parse.

/*
     * (non-Javadoc)
     * 
     * @see org.apache.tika.parser.Parser#parse(java.io.InputStream,
     * org.xml.sax.ContentHandler, org.apache.tika.metadata.Metadata,
     * org.apache.tika.parser.ParseContext)
     */
public void parse(InputStream stream, ContentHandler handler, Metadata metadata, ParseContext context) throws IOException, SAXException, TikaException {
    TemporaryResources tmp = TikaInputStream.isTikaInputStream(stream) ? null : new TemporaryResources();
    TikaInputStream tis = TikaInputStream.get(stream, tmp);
    NetcdfFile ncFile = null;
    try {
        ncFile = NetcdfFile.open(tis.getFile().getAbsolutePath());
        metadata.set("File-Type-Description", ncFile.getFileTypeDescription());
        // 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.element("h1", "dimensions");
        xhtml.startElement("ul");
        xhtml.newline();
        for (Dimension dim : ncFile.getDimensions()) {
            xhtml.element("li", dim.getFullName() + " = " + dim.getLength());
        }
        xhtml.endElement("ul");
        xhtml.element("h1", "variables");
        xhtml.startElement("ul");
        xhtml.newline();
        for (Variable var : ncFile.getVariables()) {
            xhtml.startElement("li");
            xhtml.characters(var.getDataType() + " " + var.getNameAndDimensions());
            xhtml.newline();
            List<Attribute> attributes = var.getAttributes();
            if (!attributes.isEmpty()) {
                xhtml.startElement("ul");
                for (Attribute element : attributes) {
                    xhtml.element("li", element.toString());
                }
                xhtml.endElement("ul");
            }
            xhtml.endElement("li");
        }
        xhtml.endElement("ul");
        xhtml.endDocument();
    } catch (IOException e) {
        throw new TikaException("NetCDF parse error", e);
    } finally {
        if (ncFile != null) {
            ncFile.close();
        }
        if (tmp != null) {
            tmp.dispose();
        }
    }
}

Example 15 with XHTMLContentHandler

use of org.apache.tika.sax.XHTMLContentHandler in project tika by apache.

the class TesseractOCRParser method parse.

@Override
public void parse(InputStream stream, ContentHandler handler, Metadata metadata, ParseContext parseContext) throws IOException, SAXException, TikaException {
    TesseractOCRConfig config = parseContext.get(TesseractOCRConfig.class, DEFAULT_CONFIG);
    //  occur if someone directly calls this parser, not via DefaultParser or similar
    if (!hasTesseract(config))
        return;
    TemporaryResources tmp = new TemporaryResources();
    try {
        TikaInputStream tikaStream = TikaInputStream.get(stream, tmp);
        //trigger the spooling to a tmp file if the stream wasn't
        //already a TikaInputStream that contained a file
        tikaStream.getPath();
        //this is the text output file name specified on the tesseract
        //commandline.  The actual output file name will have a suffix added.
        File tmpOCROutputFile = tmp.createTemporaryFile();
        // Temporary workaround for TIKA-1445 - until we can specify
        //  composite parsers with strategies (eg Composite, Try In Turn),
        //  always send the image onwards to the regular parser to have
        //  the metadata for them extracted as well
        _TMP_IMAGE_METADATA_PARSER.parse(tikaStream, new DefaultHandler(), metadata, parseContext);
        XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);
        xhtml.startDocument();
        parse(tikaStream, tmpOCROutputFile, parseContext, xhtml, config);
        xhtml.endDocument();
    } finally {
        tmp.dispose();
    }
}