Search in sources :

Example 1 with MatFileReader

use of com.jmatio.io.MatFileReader 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();
        }
    }
}
Also used : MatFileReader(com.jmatio.io.MatFileReader) MLArray(com.jmatio.types.MLArray) TikaException(org.apache.tika.exception.TikaException) MatFileHeader(com.jmatio.io.MatFileHeader) TemporaryResources(org.apache.tika.io.TemporaryResources) TikaInputStream(org.apache.tika.io.TikaInputStream) IOException(java.io.IOException) XHTMLContentHandler(org.apache.tika.sax.XHTMLContentHandler) MLStructure(com.jmatio.types.MLStructure) Map(java.util.Map)

Aggregations

MatFileHeader (com.jmatio.io.MatFileHeader)1 MatFileReader (com.jmatio.io.MatFileReader)1 MLArray (com.jmatio.types.MLArray)1 MLStructure (com.jmatio.types.MLStructure)1 IOException (java.io.IOException)1 Map (java.util.Map)1 TikaException (org.apache.tika.exception.TikaException)1 TemporaryResources (org.apache.tika.io.TemporaryResources)1 TikaInputStream (org.apache.tika.io.TikaInputStream)1 XHTMLContentHandler (org.apache.tika.sax.XHTMLContentHandler)1