使用7z SDK解压缩多个文件

Question

I am trying to implement the 7z SDK to compress and decompress multiple files at once.

I tried with one file at a time and both methods worked, now I implemented some changes on the compress method to support multiple files being compressed.

public static void CompressFile(List<string> inFiles, string outFile)
{
        SevenZip.Compression.LZMA.Encoder coder = new SevenZip.Compression.LZMA.Encoder();

        FileStream output = new FileStream(outFile, FileMode.Create);

        // Write the encoder properties
        coder.WriteCoderProperties(output);

        int listSize = inFiles.Count;

        // get the size of my list to loop through all the items
        // Writing each file on the compressed file at a time

        for(int i = 0; i < listSize; i++)
        {
            FileStream input = new FileStream(inFiles[i], FileMode.Open);

            // Write the decompressed file size.
            output.Write(BitConverter.GetBytes(input.Length), 0, 8);

            // Encode the file.
            coder.Code(input, output, input.Length, -1, null);
        }
        output.Flush();
        output.Close();
    }

I got a compressed file as expected but I need to implement the decompress method to test if everything went well.

I got stuck on figuring out how to implement the changes I need to decompress multiple files:

public static void DecompressFile(string inFile, string outFile)
{
        //Original method used to decompress one file worked.

        SevenZip.Compression.LZMA.Decoder coder = new SevenZip.Compression.LZMA.Decoder();


            FileStream input = new FileStream(inFile, FileMode.Open);
            FileStream output = new FileStream(outFile, FileMode.Create);

            // Read the decoder properties
            byte[] properties = new byte[5];
            input.Read(properties, 0, 5);

            // Read in the decompress file size.
            byte[] fileLengthBytes = new byte[8];
            input.Read(fileLengthBytes, 0, 8);
            long fileLength = BitConverter.ToInt64(fileLengthBytes, 0);

            coder.SetDecoderProperties(properties);
            coder.Code(input, output, input.Length, fileLength, null);
            output.Flush();
            output.Close();

    }

I had the idea (dont know if good or not) to compress all the files using a loop structure, but to decompress it I cant. Which approach would be the ideal to decompress a file that is supposed to contain several files?

Answer 1

I did something like this with gzipstream. but its all about how you want pack the files.

In this case something like this would be ideal.

string payload = Path.GetTempFileName();
using (FileStream temp_fs = new FileStream(payload, FileMode.OpenOrCreate))
{
    using (BinaryWriter output_s = new BinaryWriter(new FileStream("target.out", FileMode.OpenOrCreate)))
    {
        //Write a blank. must be a long!
        output_s.Write((long)0);
        foreach (string file in Files)
        {

            //Write the files name
            output_s.Write(file);
            long start = temp_fs.Position;

            //Write the starting point
            output_s.Write(start);

            //Compress the file to the payload
            using (GZipStream gzip = new GZipStream(temp_fs, CompressionMode.Compress, true))
            {
                using (FileStream fs = new FileStream(file, FileMode.Open))
                {
                    fs.CopyTo(gzip);
                }
            }

            //Write the length
            output_s.Write(temp_fs.Position - start);
        }

        //When all files are written
        //Get the size of our header
        long headersize = output_s.BaseStream.Length - 8;

        //Copy the temp file data to the end
        temp_fs.CopyTo(output_s.BaseStream);

        //Reset to the start of the stream
        output_s.BaseStream.Position = 0;

        //override our zero
        output_s.Write(headersize);
    }

}
File.Delete(payload);

When you read your file will a binary read you will be able to grab the file by name then the next to long after the filename will be its size and start. then you would decompress section header+pos to length. to retrieve your file.