Compressing pdf files with DotNetZip returns damaged archive

Question

I'm trying to compress 2 pdf files using DotNetZip with this code:

void Main()
{
    byte[] file1 = File.ReadAllBytes(@"C:\temp\a.pdf");
    byte[] file2 = File.ReadAllBytes(@"C:\temp\b.pdf");
    //byte[] file3 = File.ReadAllBytes(@"C:\temp\c.pdf");

    byte[] zip = Zipper.CreateZipFromFileContentList(new List<Tuple<string, byte[]>> { 
        new Tuple<string, byte[]>(@"a.pdf", file1),
        new Tuple<string, byte[]>(@"b.pdf", file2)//,
        //new Tuple<string, byte[]>(@"c.pdf", file3) 
    });
    File.WriteAllBytes(@"C:\temp\c.zip", zip);

    using(Ionic.Zip.ZipFile zipFile = Ionic.Zip.ZipFile.Read(@"C:\temp\c.zip"))
    {
        foreach(ZipEntry entry in zipFile)
        {
            entry.Extract(@"C:\temp\t");
        }
    }
}

// Define other methods and classes here
static class Zipper
{
    public static byte[] CreateZipFromFileContentList(IList<Tuple<string, byte[]>> fileContentList)
    {
        try
        {
            using (ZipFile zip = new ZipFile())
            {
                int i = 0;

                foreach (var item in fileContentList)
                {
                    ZipEntry entry = null;
                    entry = zip.AddEntry(item.Item1, item.Item2);
                    ++i;

                }
                MemoryStream ms = new MemoryStream();
                zip.Save(ms);
                return ms.GetBuffer();
            }
        }
        catch (Exception)
        {
            throw;
        }
    }
}

Actually everything works, because the extract process works after the archive is created. But if I try to open the zip file using winRar, I get a message that the archive is damaged. If I try with 7Zip, I get a message saying that there are data beyond the end of the useful block (translated, I don't know if the english version gives the exact same message).

If I zip 1 or 3 files, I have no problems at all. How can I fix that?

Answer 1

Your zip output probably has extra bytes because .GetBuffer() returns the streams internal buffer - only part of which contains valid data.

Use .ToArray() instead which will return only the used part of the buffer.