Archiving Multiple Files Using ZipOutputStream gives me an Empty Archived File

Question

I am attempting to download a bunch of files that I am zipping up(archiving) via ZipoutputStream.

using (var zipStream = new ZipOutputStream(outputMemStream))
{
    foreach (var documentIdString in documentUniqueIdentifiers)
    {   
        ...
        var blockBlob = container.GetBlockBlobReference(documentId.ToString());

        var fileMemoryStream = new MemoryStream();

        blockBlob.DownloadToStream(fileMemoryStream);

        zipStream.SetLevel(3); 

        fileMemoryStream.Position = 0;

        ZipEntry newEntry = new ZipEntry(document.FileName);
        newEntry.DateTime = DateTime.Now;

        zipStream.PutNextEntry(newEntry);

        fileMemoryStream.Seek(0, SeekOrigin.Begin);
        StreamUtils.Copy(fileMemoryStream, zipStream, new byte[4096]);

        zipStream.IsStreamOwner = false; // False stops the Close also Closing the underlying stream.

    }

    outputMemStream.Seek(0, SeekOrigin.Begin);
    return outputMemStream;
}

In my controller I am returning the following code that should download the Zip file i created in the previous example. The controller actions downloads the file as it should in the browser, but the Archived File is empty. I can see the content length populated returning from the method above...

file.Seek(0, SeekOrigin.Begin);
return File(file, "application/octet-stream", "Archive.zip");

Does anyone have any idea why my file that is returned by my controller is empty or corrupt?

Haney · Accepted Answer · 2015-03-27T17:30:09.690

2

I believe you need to close your entries and your final zip stream. You should also using and dispose all of your streams. Try this:

using (var zipStream = new ZipOutputStream(outputMemStream))
{
    zipStream.IsStreamOwner = false;
    // Set compression level
    zipStream.SetLevel(3); 

    foreach (var documentIdString in documentUniqueIdentifiers)
    {   
        ...
        var blockBlob = container.GetBlockBlobReference(documentId.ToString());

        using (var fileMemoryStream = new MemoryStream())
        {
            // Populate stream with bytes
            blockBlob.DownloadToStream(fileMemoryStream);

            // Create zip entry and set date
            ZipEntry newEntry = new ZipEntry(document.FileName);
            newEntry.DateTime = DateTime.Now;
            // Put entry RECORD, not actual data
            zipStream.PutNextEntry(newEntry);
            // Copy date to zip RECORD
            StreamUtils.Copy(fileMemoryStream, zipStream, new byte[4096]);
            // Mark this RECORD closed in the zip
            zipStream.CloseEntry();
        }
    }

    // Close the zip stream, parent stays open due to !IsStreamOwner
    zipStream.Close();

    outputMemStream.Seek(0, SeekOrigin.Begin);
    return outputMemStream;
}

EDIT - you should remove:

// Reset position of stream
fileMemoryStream.Position = 0;

Pretty sure that's the problem.

edited Mar 27 '15 at 17:30

answered Mar 27 '15 at 15:16

Haney

32,775
8
59
68

So I updated my code, but when I download from the controller, it is still giving me an error that it is invalid. ie: The Compressed (zipped) folder "..." is invalid. – jimfromthegym - Jim Mackin Mar 27 '15 at 16:48
@jimfromthegym-JimMackin does it have size? Or is it 0 bytes? – Haney Mar 27 '15 at 17:08
@jimfromthegym-JimMackin actually don't reset the position of the stream. See updated code. – Haney Mar 27 '15 at 17:10
I reset the position to 0 and it still does not work. – jimfromthegym - Jim Mackin Mar 27 '15 at 17:29
@jimfromthegym-JimMackin I meant to *remove* the reset to 0 – Haney Mar 27 '15 at 17:29
1

Let us [continue this discussion in chat](http://chat.stackoverflow.com/rooms/73956/discussion-between-haney-and-jimfromthegym-jim-mackin). – Haney Mar 27 '15 at 17:34

Archiving Multiple Files Using ZipOutputStream gives me an Empty Archived File

1 Answers1