// Presentation Formats
        extensionMap.put("odp", "pdf");
        extensionMap.put("otp", "pdf");
        extensionMap.put("sxi", "pdf");
        extensionMap.put("ppt", "pdf");
        extensionMap.put("pptx", "pdf");
        // Drawing Formats
        extensionMap.put("odg", "svg");
        extensionMap.put("otg", "svg");

        extractorMap.put("pdf", new PdfExtractor());

        final AtomicBoolean streamClosed = new AtomicBoolean(false);

        try (final InputStream originalStream = ResourceUtil.getResourceAsStream("extractor/msoffice/test.ppt")) {
            final InputStream trackableStream = createTrackableInputStream(originalStream, streamClosed);
            final ExtractData result = extractor.getText(trackableStream, null);

            assertNotNull(result);

- Apache POI (for MS Office tests)
- Apache Commons Compress (for archive tests)

### Test Resources:
- `extractor/msoffice/test.doc`
- `extractor/msoffice/test.xls`
- `extractor/msoffice/test.ppt`
- `extractor/test.txt`
- `extractor/zip/test.zip`

## Key Improvements Tested

### 1. Resource Management
- ✅ Try-with-resources usage in MS Office extractors
- ✅ Proper closure of POI objects

# Select supported compute capabilities (supported graphics cards).
# This is the same as the official TensorFlow builds.
# See https://developer.nvidia.com/cuda-gpus#compute
# `compute_XY` enables PTX embedding in addition to SASS. PTX
# is forward compatible beyond the current compute capability major
# release while SASS is only forward compatible inside the current
# major release. Example: sm_80 kernels can run on sm_89 GPUs but

				"application/vnd.sealed.csf",
				"application/vnd.sealed.doc",
				"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",