Ziparchive 解压多部分 zip 文件

Question

I'm looking to unzip a multipart zipfile through ZipArchive. Unzipping one 'standalone' zipfile works but if I try to unzip the first file it just throws an error. It works like this in windows explorer (Just unzip the first file), so I guessed it might work since I can't find any other ways online.

using (ZipArchive archive = new ZipArchive(fs)) {
    archive.ExtractToDirectory(tempLocation, true);
}

But, maybe this is an XY Problem , so I'm using it as a simplified way of chunkedfile-downloads. We're using this for an updateservice, which downloads the contents of the update. To be a bit flexible we would use (multipart)zipfiles. If this is the wrong approach, please do tell me.

EDIT: After figuring out that the first error was PEBCAK I got the error that "Splitted or spread archives aren't supported".

Are there any other Zipfile libraries that do support it?

Answer 1

I recommend using DotNetZip to accomplish what you want.

First you must combine all your ".zip.001" ".zip.002" ".zip.003" files into one huge file and then follow up on decompressing as normal:

        string destFileName = @"C:\Temp\zip\finalHugeFile.zip";
        List<string> myMultiPartZip00xFiles = System.IO.Directory.GetFiles(@"C:\Temp\zip", "*.zip.*").ToList();

        using (Stream destStream = File.OpenWrite(destFileName))
        {
            foreach (string srcFileName in myMultiPartZip00xFiles)
            {
                using (Stream srcStream = File.OpenRead(srcFileName))
                {
                    srcStream.CopyTo(destStream);
                }
            }
        }

        // we extract here
        using (ZipFile zip = ZipFile.Read(destFileName))
        {
            foreach (ZipEntry entry in zip)
            {
                entry.Extract(@"C:\Temp\zip", ExtractExistingFileAction.OverwriteSilently);
            }
        }

Answer 2

Multipart zip is just a regular zip split into files. So we need a Stream that combines multiple files into one:

var multiFileStream = new MultiFileStream(new[] { "a.001", "a.002" });
using var zip = new ZipArchive(multiFileStream);

// ...

public class MultiFileStream : Stream
{
    private readonly IList<string> _files;
    private int _fileIdx;
    private FileStream? _file;

    public MultiFileStream(IList<string> files)
    {
        _files = files;
    }

    public override int Read(byte[] buffer, int offset, int count)
    {
        while (true)
        {
            _file ??= File.OpenRead(_files[_fileIdx]);

            var res = _file.Read(buffer, offset, count);

            if (res > 0)
            {
                return res;
            }

            _file.Dispose();
            _file = null;

            if (_fileIdx < _files.Count - 1)
            {
                _fileIdx++;
            }
            else
            {
                return 0;
            }
        }
    }

    public override void Flush() => throw new NotSupportedException();
    public override long Seek(long offset, SeekOrigin origin) => throw new NotSupportedException();
    public override void SetLength(long value) => throw new NotSupportedException();
    public override void Write(byte[] buffer, int offset, int count) => throw new NotSupportedException();
    public override bool CanRead => true;
    public override bool CanSeek => false;
    public override bool CanWrite => false;
    public override long Length => throw new NotSupportedException();
    public override long Position
    {
        get => throw new NotSupportedException();
        set => throw new NotSupportedException();
    }
}

The other answer works, but a lot slower because of the temp file and all the copying (and with big archives you can even run out of disk space).