代码之家 › 专栏 › 技术社区 › Alberto Rechy

有没有办法用另一个库解压缩DynaZip Max文件?F.E.DotNetZip公司

dotnetzip zlib gzip zip c#

6

Alberto Rechy · 技术社区 · 12 年前

我有一个数据库,我们在其中存储了使用DynaZip Max Secure压缩的pdf文件,使用以下代码:

MemoryStream msIN = new System.IO.MemoryStream();   //Input MemoryStream
MemoryStream msZip = new System.IO.MemoryStream();  //Compressed MemoryStream
BinaryReader objBinaryReader;
BinaryWriter objBinaryWriter;


public MemoryStream CompressStream(byte[] vbuf)
{
    System.IO.BinaryWriter bw = new System.IO.BinaryWriter(msIN);
    bw.Write(vbuf);
    CDZipSNET dz1 = new CDZipSNET();
    dz1.ZipMemToMemCallback += new CDZipSNET.OnZipMemToMemCallback(this.ZipMemToMemCallback_event);
    dz1.ActionDZ = CDZipSNET.DZACTION.ZIP_MEMTOMEM;
    return msZip;
}

这是ZipMemToMemCallback_event代码:

public void ZipMemToMemCallback_event(CDZipSNET.MEMTOMEMACTION lAction,ref byte[] lpMemBuf,ref uint pdwSize,uint dwTotalReadL,uint dwTotalReadH,uint dwTotalWrittenL,uint dwTotalWrittenH,ref CDZipSNET.MEMTOMEMRESPONSE plRet)
{
    int bytesToRead;

    switch(lAction)
    {
        case CDZipSNET.MEMTOMEMACTION.MEM_READ_DATA:
            if((dwTotalReadL == 0) && (dwTotalReadH == 0))
            {
                msIN.Seek(0, System.IO.SeekOrigin.Begin);
                objBinaryReader = new System.IO.BinaryReader(msIN);
            }
            try
            {
                bytesToRead = (int)(objBinaryReader.BaseStream.Length - dwTotalReadL);
                if(bytesToRead > pdwSize)
                {
                    bytesToRead = (int)pdwSize;
                    plRet = CDZipSNET.MEMTOMEMRESPONSE.MEM_CONTINUE;
                }
                else
                { 
                    plRet = CDZipSNET.MEMTOMEMRESPONSE.MEM_DONE; 
                }
                pdwSize = (uint)bytesToRead;
                if(bytesToRead > 0)
                { 
                    objBinaryReader.Read(lpMemBuf, 0, bytesToRead); 
                }
            }
            catch 
            { 
                plRet = CDZipSNET.MEMTOMEMRESPONSE.MEM_ERROR; 
            }
            break;

        case CDZipSNET.MEMTOMEMACTION.MEM_WRITE_DATA:
            if((dwTotalWrittenL == 0) && (dwTotalWrittenH == 0))
            { 
                objBinaryWriter = new System.IO.BinaryWriter(msZip); 
            }
            try
            {
                objBinaryWriter.Write(lpMemBuf, 0, (int)pdwSize);
                plRet = CDZipSNET.MEMTOMEMRESPONSE.MEM_CONTINUE;
            }
            catch (System.Exception)
            { 
                plRet = CDZipSNET.MEMTOMEMRESPONSE.MEM_ERROR; 
            }
            break;

        default: plRet = CDZipSNET.MEMTOMEMRESPONSE.MEM_ERROR;
            break;
    }
}

我会提供任何其他必要的东西来解开这个谜题,我尝试过常规的Zip解压缩、Zlib、Gzip,但都无济于事。如果有任何帮助,我将不胜感激。非常感谢。

编辑: 问题是DinaZip是一个专有的、已停产的库,在发布它的公司没有帮助或故障排除的情况下,我被委托解压缩以前使用这个库压缩的一堆文件(使用avobe代码),我不再有这个库可用于解压缩,我想知道是否有人知道如何解压缩这些文件,也许可以使用另一个库或方法。

1 回复 | 直到 12 年前

1

9

osexpert 12 年前

此代码来自 https://zlibnet.codeplex.com 并且它可以解压缩未加密的DynaZip流:

public static class DynazipCompressor
{
    const int DZ_DEFLATE_POS = 46;

    public static bool IsDynazip(byte[] source)
    {
        return source.Length >= 4 && BitConverter.ToInt32(source, 0) == 0x02014b50;
    }

    public static byte[] DeCompress(byte[] source)
    {
        if (!IsDynazip(source))
            throw new InvalidDataException("not dynazip header");
        using (MemoryStream srcStream = new MemoryStream(source, DZ_DEFLATE_POS, source.Length - DZ_DEFLATE_POS))
        using (MemoryStream dstStream = DeCompress(srcStream))
            return dstStream.ToArray();
    }

    private static MemoryStream DeCompress(Stream source)
    {
        MemoryStream dest = new MemoryStream();
        DeCompress(source, dest);
        dest.Position = 0;
        return dest;
    }

    private static void DeCompress(Stream source, Stream dest)
    {
        using (DeflateStream zsSource = new DeflateStream(source, CompressionMode.Decompress, true))
        {
            zsSource.CopyTo(dest);
        }
    }
}

DynaZip流只是一个带有PKZIP标头的DeflateStream,因此此代码只跳过标头。