extensionMap.put("html", "txt");
        // Spreadsheet Formats
        extensionMap.put("ods", "tsv");
        extensionMap.put("ots", "tsv");
        extensionMap.put("sxc", "tsv");
        extensionMap.put("xls", "tsv");
        extensionMap.put("xlsx", "tsv");
        extensionMap.put("csv", "tsv");
        extensionMap.put("tsv", "tsv");
        // Presentation Formats
        extensionMap.put("odp", "pdf");

        assertEquals("かいしゃ", extractData.getValues("dc:publisher")[0]);
    }

    public void test_getTika_msexcel() {
        final InputStream in = ResourceUtil.getResourceAsStream("extractor/msoffice/test.xls");
        final ExtractData extractData = tikaExtractor.getText(in, null);
        final String content = extractData.getContent();
        CloseableUtil.closeQuietly(in);
        logger.info(content);

				"application/vnd.sealed.eml",
				"application/vnd.sealed.mht",
				"application/vnd.sealed.net",
				"application/vnd.sealed.ppt",
				"application/vnd.sealed.tiff",
				"application/vnd.sealed.xls",
				"application/vnd.sealedmedia.softseal.html",
				"application/vnd.sealedmedia.softseal.pdf",
				"application/vnd.seemail",
				"application/vnd.sema",
				"application/vnd.semd",
				"application/vnd.semf",

    <root-XML localName="Workbook"/>
    <sub-class-of type="application/xml"/>
    <_comment>Excel 2003 xml format, pre-ooxml</_comment>
    <_comment>glob pattern typically *.xls</_comment>
  </mime-type>
  <mime-type type="application/vnd.ms-wordml">
    <root-XML localName="wordDocument" namespaceURI="http://schemas.microsoft.com/office/word/2003/wordml"/>
    <root-XML localName="wordDocument"/>