} catch (final Exception e) {
            throw new CrawlingAccessException("Failed to parse " + url, e);
        }
    }

    /**
     * Decodes a similar document hash from its compressed and encoded form.
     * Reverses the encoding process applied by encodeSimilarDocHash.
     *
     * @param hash the encoded hash string to decode

	VADDPD 15(CX)(AX*2), X29, X28 // 6261950058a4410f000000
	VADDPD 15(CX)(AX*2), X29, X10 // 627195005894410f000000
	VADDPD 15(CX)(AX*2), X10, X28 // 6261ad0858a4410f000000
	// EVEX: compressed displacement (Disp8).
	VADDPD 2032(DX), X29, X0        // 62f1950058427f
	VADDPD 2032(DX), X1, X29        // 6261f508586a7f
	VADDPD 2032(DX), X29, X28       // 6261950058627f

		t.Errorf("unexpected method and flags: %v", b[6:10])
	}

	if !bytes.Equal(b[14:26], make([]byte, 12)) { // FileHeader.{CRC32,CompressSize,UncompressedSize} all zero.
		t.Errorf("unexpected crc, compress and uncompressed size to be 0 was: %v", b[14:26])
	}

	binary.LittleEndian.PutUint32(sig[:], uint32(dataDescriptorSignature))
	if bytes.Contains(b, sig[:]) {
		t.Error("there should be no data descriptor")
	}
}

    /**
     * Files is marked to be archived
     */
    int ATTR_ARCHIVE = 0x20;

    // extended file attribute encoding(others same as above)
    /**
     * File is compressed.
     */
    int ATTR_COMPRESSED = 0x800;
    /**
     * File is a normal file.
     */
    int ATTR_NORMAL = 0x080;
    /**
     * File is temporary.
     */
    int ATTR_TEMPORARY = 0x100;

        }
        return baos.toByteArray();
    }

    /**
     * Mock HttpURLConnection that returns GZIP-compressed content for successful responses.
     */
    class GzipSuccessMockHttpURLConnection extends HttpURLConnection {
        private final byte[] gzippedData;

        GzipSuccessMockHttpURLConnection(URL u, byte[] gzippedData) {
            super(u);

	"github.com/minio/sio"
)

// ETag is a single S3 ETag.
//
// An S3 ETag sometimes corresponds to the MD5 of
// the S3 object content. However, when an object
// is encrypted, compressed or uploaded using
// the S3 multipart API then its ETag is not
// necessarily the MD5 of the object content.
//
// For a more detailed description of S3 ETags
// take a look at the package documentation.

        val firstByte = readByte()
        val huffmanDecode = firstByte and 0x80 == 0x80 // 1NNNNNNN
        val length = readInt(firstByte, PREFIX_7_BITS).toLong()

        // If the compressed or decompressed length exceeds the limit, don't even bother.
        if (headerListByteCount + length > HEADER_LIMIT) {
          throw IOException("header byte count limit of $HEADER_LIMIT exceeded")
        }

- RTF, WordPerfect

#### PDFs and Images
- PDF documents (text and metadata extraction)
- Images (JPEG, PNG, GIF, TIFF, BMP)
- Image metadata (EXIF, IPTC, XMP)

#### Archives and Compressed Files
- ZIP, TAR, GZ archives
- LHA compression format
- Nested archive extraction

#### Web and Markup
- HTML, XHTML with XPath support
- XML documents
- JSON and structured data

// will return during reading. It returns -1 for unlimited
// data.
func (r *Reader) Size() int64 { return r.size }

// ActualSize returns the pre-modified size of the object.
// DecompressedSize - For compressed objects.
func (r *Reader) ActualSize() int64 { return r.actualSize }

// ETag returns the ETag computed by an underlying etag.Tagger.
// If the underlying io.Reader does not implement etag.Tagger
// it returns nil.

}

// MultipartInfo captures metadata information about the uploadId
// this data structure is used primarily for some internal purposes
// for verifying upload type such as was the upload
// - encrypted
// - compressed
type MultipartInfo struct {
	// Name of the bucket.
	Bucket string

	// Name of the object.
	Object string

	// Upload ID identifying the multipart upload whose parts are being listed.