Shinsuke Sugaya <******@****.***> 1444529815 +0900

Shinsuke Sugaya <******@****.***> 1444529815 +0900

        assertContentType("application/msword", "extractor/msoffice/test.doc", "hoge.doc");
        assertContentType("application/vnd.ms-excel", "test/text1.txt", "hoge.xls");
        assertContentType("application/vnd.ms-excel", "extractor/msoffice/test.xls", "hoge.xls");
        assertContentType("application/vnd.ms-powerpoint", "test/text1.txt", "hoge.ppt");

			return sizeS, nil
		}
		return
	}
	xls := xlStorage{drivePath: base, diskInfoCache: cachevalue.New[DiskInfo]()}
	xls.diskInfoCache.InitOnce(time.Second, cachevalue.Opts{}, func(ctx context.Context) (DiskInfo, error) {
		return DiskInfo{Total: 1 << 40, Free: 1 << 40}, nil
	})
	weSleep := func() bool { return false }

            assertEquals("filetype:pdf", queryMap.get("PDF"));
        }

        // Check for XLS - could be either key format
        if (queryMap.containsKey("labels.facet_filetype_xls")) {
            assertEquals("filetype:xls", queryMap.get("labels.facet_filetype_xls"));
        } else {
            assertEquals("filetype:xls", queryMap.get("XLS"));
        }

        // DOC type won't be duplicated since it already exists

    // public ResponseData call() throws Exception {
    // String[] urls =
    // new String[] {
    // "http://.../",
    // "http://.../test.pdf",
    // "http://.../test.doc",
    // "http://.../test.xls",
    // "http://.../test.ppt",
    // "http://.../test.txt", };
    // for (String url : urls) {
    // ResponseData responseData = httpClient.doGet(url);
    // // assertEquals(200, responseData.getHttpStatusCode());

        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);