解压缩从c＃发送到PHP的字符串

Question

This is the function which zip the string to be sent to php code

public string Zip(string value)
    {
        //Transform string into byte[]  
        byte[] byteArray = Encoding.UTF8.GetBytes(value);
        int indexBA = 0;
        foreach (char item in value.ToCharArray())
        {
            byteArray[indexBA++] = (byte)item;
        }
        //Prepare for compress
        System.IO.MemoryStream ms = new System.IO.MemoryStream();
        System.IO.Compression.GZipStream sw = new System.IO.Compression.GZipStream(ms,
        System.IO.Compression.CompressionMode.Compress);
        //Compress
        sw.Write(byteArray, 0, byteArray.Length);
        //Close, DO NOT FLUSH cause bytes will go missing...
        sw.Close();
        //Transform byte[] zip data to string
        byteArray = ms.ToArray();
        System.Text.StringBuilder sB = new System.Text.StringBuilder(byteArray.Length);
        foreach (byte item in byteArray)
        {
            sB.Append((char)item);
        }
        ms.Close();
        sw.Dispose();
        ms.Dispose();
        return sB.ToString();
    }

and am sending the request using

 string data = req.Zip(xml);
        string resp = req.post(url,"&Data="+data);

i tried to use gzuncompress , gzdecode but all produce a data error any body knows why ?

Answer 1

This code is bizarre to start with:

byte[] byteArray = Encoding.UTF8.GetBytes(value);
int indexBA = 0;
foreach (char item in value.ToCharArray())
{
    byteArray[indexBA++] = (byte)item;
}

You're converting it to a byte array using UTF-8 encoding... and then you're overwriting the contents of that array (or at least some of the contents of that array) by casting each character to a byte - which is effectively applying ISO-Latin-1 encoding.

Then later on, you're converting arbitrary binary data to a string like this:

byteArray = ms.ToArray();
System.Text.StringBuilder sB = new System.Text.StringBuilder(byteArray.Length);
foreach (byte item in byteArray)
{
    sB.Append((char)item);
}

Don't do that. It's opaque binary data - the chances of the "string" you're creating (again, effectively created via ISO-8859-1) being properly transferable are very slim.

When encoding arbitrary binary data as a string, you should almost always use Base64:

string base64 = Convert.ToBase64String(byteArray);

You're then also using the data as URL-encoded form data - despite the fact that the string could quite easily include characters such as & and % which have special meaning within URL encoded text. Don't do that either.

Basically, you should:

• Choose which encoding you want to use for the initial text to binary conversion. UTF-8 is a good choice here, as it can represent all of Unicode.
• Perform the compression (and no, flushing should not cause problems here, although you should also close anyway - ideally via a using statement)
• Transform the binary data back into text (assuming you really have to) using base64. If you're going to use this as a URL parameter, you should use a web-safe variant of base64, as described on the Wikipedia base64 page .

To decompress, you'll obviously need to perform the base64 to binary conversion before decompression.

If possible, it would be significantly more efficient (in terms of the transferred data) to post the compressed data as binary data rather than as a URL-encoded form parameter.