org.apache.poi.ooxml.util.SAXHelper Java Examples

/**
 * Read this shared strings table from an XML file.
 *
 * @param is The input stream containing the XML document.
 * @throws IOException  if an error occurs while reading.
 * @throws SAXException if parsing the XML data fails.
 */
public void readFrom(InputStream is) throws IOException, SAXException {
    // test if the file is empty, otherwise parse it
    PushbackInputStream pis = new PushbackInputStream(is, 1);
    int emptyTest = pis.read();
    if (emptyTest > -1) {
        pis.unread(emptyTest);
        InputSource sheetSource = new InputSource(pis);
        try {
            XMLReader sheetParser = SAXHelper.newXMLReader();
            sheetParser.setContentHandler(this);
            sheetParser.parse(sheetSource);
            stringsCache.finished();
        } catch (ParserConfigurationException e) {
            throw new RuntimeException("SAX parser appears to be broken - " + e.getMessage());
        }
    }
}
 

/**
 * Parses and shows the content of one sheet
 * using the specified styles and shared-strings tables.
 *
 * @param strings          The table of strings that may be referenced by cells in the sheet
 * @param sheetInputStream The stream to read the sheet-data from.
 * @throws java.io.IOException An IO exception from the parser,
 *                             possibly from a byte stream or character stream
 *                             supplied by the application.
 * @throws SAXException        if parsing the XML data fails.
 */
private void processSheet(
        SharedStrings strings,
        XSSFSheetXMLHandler.SheetContentsHandler sheetHandler,
        InputStream sheetInputStream) throws IOException, SAXException {
    DataFormatter formatter = new DataFormatter();
    InputSource sheetSource = new InputSource(sheetInputStream);
    try {
        XMLReader sheetParser = SAXHelper.newXMLReader();
        ContentHandler handler = new XSSFSheetXMLHandler(
                null, null, strings, sheetHandler, formatter, false);
        sheetParser.setContentHandler(handler);
        sheetParser.parse(sheetSource);
    } catch (ParserConfigurationException e) {
        throw new RuntimeException("SAX parser appears to be broken - " + e.getMessage());
    }
}
 

/**
 * Handles an individual Excel sheet from the entire Excel document. Each sheet will result in an individual flowfile.
 *
 * @param session
 *  The NiFi ProcessSession instance for the current invocation.
 */
private void handleExcelSheet(ProcessSession session, FlowFile originalParentFF, final InputStream sheetInputStream, ExcelSheetReadConfig readConfig,
                              CSVFormat csvFormat) throws IOException {

    FlowFile ff = session.create(originalParentFF);
    try {
        final DataFormatter formatter = new DataFormatter();
        final InputSource sheetSource = new InputSource(sheetInputStream);

        final SheetToCSV sheetHandler = new SheetToCSV(readConfig, csvFormat);

        final XMLReader parser = SAXHelper.newXMLReader();

        //If Value Formatting is set to false then don't pass in the styles table.
        // This will cause the XSSF Handler to return the raw value instead of the formatted one.
        final StylesTable sst = readConfig.getFormatValues()?readConfig.getStyles():null;

        final XSSFSheetXMLHandler handler = new XSSFSheetXMLHandler(
                sst, null, readConfig.getSharedStringsTable(), sheetHandler, formatter, false);

        parser.setContentHandler(handler);

        ff = session.write(ff, new OutputStreamCallback() {
            @Override
            public void process(OutputStream out) throws IOException {
                PrintStream outPrint = new PrintStream(out);
                sheetHandler.setOutput(outPrint);

                try {
                    parser.parse(sheetSource);

                    sheetInputStream.close();

                    sheetHandler.close();
                    outPrint.close();
                } catch (SAXException se) {
                    getLogger().error("Error occurred while processing Excel sheet {}", new Object[]{readConfig.getSheetName()}, se);
                }
            }
        });

        ff = session.putAttribute(ff, SHEET_NAME, readConfig.getSheetName());
        ff = session.putAttribute(ff, ROW_NUM, new Long(sheetHandler.getRowCount()).toString());

        if (StringUtils.isNotEmpty(originalParentFF.getAttribute(CoreAttributes.FILENAME.key()))) {
            ff = session.putAttribute(ff, SOURCE_FILE_NAME, originalParentFF.getAttribute(CoreAttributes.FILENAME.key()));
        } else {
            ff = session.putAttribute(ff, SOURCE_FILE_NAME, UNKNOWN_SHEET_NAME);
        }

        //Update the CoreAttributes.FILENAME to have the .csv extension now. Also update MIME.TYPE
        ff = session.putAttribute(ff, CoreAttributes.FILENAME.key(), updateFilenameToCSVExtension(ff.getAttribute(CoreAttributes.UUID.key()),
                ff.getAttribute(CoreAttributes.FILENAME.key()), readConfig.getSheetName()));
        ff = session.putAttribute(ff, CoreAttributes.MIME_TYPE.key(), CSV_MIME_TYPE);

        session.transfer(ff, SUCCESS);

    } catch (SAXException | ParserConfigurationException saxE) {
        getLogger().error("Failed to create instance of Parser.", saxE);
        ff = session.putAttribute(ff,
                ConvertExcelToCSVProcessor.class.getName() + ".error", saxE.getMessage());
        session.transfer(ff, FAILURE);
    } finally {
        sheetInputStream.close();
    }
}