(MFC) Crypt2 Compression Example

The Chilkat Crypt2 class include a few methods for compression and decompression (inflate). These are legacy method that existed long before the Chilkat.Compression class was first introduced. These methods should be avoided in favor of using the Chilkat.Compression methods. The reason is that they are limited to bzip2, and the compressed output includes an 8-byte header that is composed of a magic number (0xB394A7E1) and the size of the original uncompressed data (4 bytes and thus limited to 4GB).

This example demonstrates compression and decompression, and also how to compress using Crypt2 and decompress using Chilkat.Compression. The Chilkat.Compression class does not emit any header, and supports other compression methods.

Chilkat C/C++ Library Downloads MS Visual C/C++ Libs

See Also: Using MFC CString in Chilkat

#include <CkCrypt2.h>
#include <CkStringBuilder.h>
#include <CkCompression.h>
#include <CkBinData.h>

void ChilkatSample(void)
    {
    CkString strOut;

    // This example assumes the Chilkat API to have been previously unlocked.
    // See Global Unlock Sample for sample code.

    CkCrypt2 crypt;

    CkStringBuilder sb;
    int i = 0;
    while (i < 10) {
        sb.AppendInt(i);
        sb.Append(" the quick brown fox jumps over the crazy dog\r\n");
        i = i + 1;
    }

    strOut.append("Before compression:");
    strOut.append("\r\n");
    strOut.append(sb.getAsString());
    strOut.append("\r\n");

    // Before compression:
    // 0 the quick brown fox jumps over the crazy dog
    // 1 the quick brown fox jumps over the crazy dog
    // 2 the quick brown fox jumps over the crazy dog
    // 3 the quick brown fox jumps over the crazy dog
    // 4 the quick brown fox jumps over the crazy dog
    // 5 the quick brown fox jumps over the crazy dog
    // 6 the quick brown fox jumps over the crazy dog
    // 7 the quick brown fox jumps over the crazy dog
    // 8 the quick brown fox jumps over the crazy dog
    // 9 the quick brown fox jumps over the crazy dog

    crypt.put_EncodingMode("base64");
    // Compress the utf-8 representation of the string.
    // Note: Choosing "Unicode" or "utf-16" is a poor choice for text that is mostly us-ascii.
    // The utf-8 representation will result in a smaller size.
    crypt.put_Charset("utf-8");
    const char *compressedStr = crypt.compressStringENC(sb.getAsString());

    strOut.append("After compression:");
    strOut.append("\r\n");
    strOut.append(compressedStr);
    strOut.append("\r\n");

    // After compression:
    // 4aeUs+ABAABCWmgzMUFZJlNZT/+H9wAAN1mAABJAAH/gP/v/8CAAkDGExNBgjEMjCYFKqGoAaDygNDamyTYmKe6Ypqn0mqZpgmaZd/hv8fLz9McsU7kyTgmwnAmibU4pyJom5PxNqapxTBO1M05JuT4TrT+TRMCapqm9VzTmn+LuSKcKEgn/8P7g

    // Note: The compressed data includes an 8-byte header composed of a magic number (0xB394A7E1) and the original data byte count.
    // Therefore, the Crypt2 compress/inflate methods are considered legacy and should only be used for data already
    // compressed with these methods.  The Chilkat.Compression class is generalized, emits no headers, and can interoperate with
    // with other software systems without trouble.  

    // If we re-encode from base64 to hex, then we can see the header:
    strOut.append(crypt.reEncode(compressedStr,"base64","hex"));
    strOut.append("\r\n");

    // The hex compressed string is:
    // E1A794B3E0010000425A68333141592653594...
    // We can see the 8-byte header in the hex:  magic number = 0xE1A794B3E (little-endian), and byte count = 0xE0010000 (little-endian)

    const char *originalStr = crypt.inflateStringENC(compressedStr);
    strOut.append("After inflate:");
    strOut.append("\r\n");
    strOut.append(originalStr);
    strOut.append("\r\n");

    // ---------------------------------------------------------------------------
    // To inflate with Chilkat.Compression:

    CkCompression compress;

    compress.put_Algorithm("bzip2");
    compress.put_Charset("utf-8");
    compress.put_EncodingMode("base64");

    CkBinData bd;
    bd.AppendEncoded(compressedStr,"base64");
    // Remove the 1st 8 bytes.
    bd.RemoveChunk(0,8);

    compress.DecompressBd(bd);

    strOut.append("After decompression:");
    strOut.append("\r\n");
    strOut.append(bd.getString("utf-8"));
    strOut.append("\r\n");


    SetDlgItemText(IDC_EDIT1,strOut.getUnicode());

    }