Reading a binary file and using Response.BinaryWri

Question

I have an app that needs to read a PDF file from the file system and then write it out to the user. The PDF is 183KB and seems to work perfectly. When I use the code at the bottom the browser gets a file 224KB and I get a message from Acrobat Reader saying the file is damaged and cannot be repaired.

Here is my code (I've also tried using File.ReadAllBytes(), but I get the same thing):

using (FileStream fs = File.OpenRead(path))
{
    int length = (int)fs.Length;
    byte[] buffer;

    using (BinaryReader br = new BinaryReader(fs))
    {
        buffer = br.ReadBytes(length);
    }

    Response.Clear();
    Response.Buffer = true;
    Response.AddHeader("content-disposition", String.Format("attachment;filename={0}", Path.GetFileName(path)));
    Response.ContentType = "application/" + Path.GetExtension(path).Substring(1);
    Response.BinaryWrite(buffer);
}

Answer 1

Just for future reference, as stated in this blog post: http://blogs.msdn.com/b/aspnetue/archive/2010/05/25/response-end-response-close-and-how-customer-feedback-helps-us-improve-msdn-documentation.aspx

It is not recommended to call Response.Close() or Response.End() - instead use CompleteRequest().

Your code would look somewhat like this:

    byte[] bytes = {};

    bytes = GetBytesFromDB();  // I use a similar way to get pdf data from my DB

    Response.Clear();
    Response.ClearHeaders();
    Response.Buffer = true;
    Response.Cache.SetCacheability(HttpCacheability.NoCache);
    Response.ContentType = "application/pdf";
    Response.AppendHeader("Content-Disposition", "attachment; filename=" + anhangTitel);
    Response.AppendHeader("Content-Length", bytes.Length.ToString());
    this.Context.ApplicationInstance.CompleteRequest();

Answer 2

In my MVC application, I have enabled gzip compression for all responses. If you are reading this binary write from an ajax call with gzipped responses, you are getting the gzipped bytearray rather than original bytearray that you need to work with.

//c# controller is compressing the result after the response.binarywrite

[compress]
public ActionResult Print(int id)       
{
... 
var byteArray=someService.BuildPdf(id);
return  return this.PDF(byteArray, "test.pdf");
}

//where PDF is a custom actionresult that eventually does this:
 public class PDFResult : ActionResult
{
...
    public override void ExecuteResult(ControllerContext context)
    {
        //Set the HTTP header to excel for download
        HttpContext.Current.Response.Clear();
        //HttpContext.Current.Response.ContentType = "application/vnd.ms-excel";
        HttpContext.Current.Response.ContentType = "application/pdf";
        HttpContext.Current.Response.AddHeader("content-disposition", string.Concat("attachment; filename=", fileName));
        HttpContext.Current.Response.AddHeader("Content-Length", pdfBytes.Length.ToString());
        //Write the pdf file as a byte array to the page
        HttpContext.Current.Response.BinaryWrite(byteArray);
        HttpContext.Current.Response.End();
    }
}

//javascript

function pdf(mySearchObject) {
    return $http({
    method: 'Post',
    url: '/api/print/',
    data: mySearchObject,
    responseType: 'arraybuffer',
    headers: {
    'Accept': 'application/pdf',
    }
    }).then(function (response) {

var type = response.headers('Content-Type');
//if response.data is gzipped, this blob will be incorrect.  you have to uncompress it first.
var blob = new Blob([response.data], { type: type });
var fileName = response.headers('content-disposition').split('=').pop();

if (window.navigator.msSaveOrOpenBlob) { // for IE and Edge
    window.navigator.msSaveBlob(blob, fileName);
} else {

    var anchor = angular.element('<a/>');
    anchor.css({ display: 'none' }); // Make sure it's not visible
    angular.element(document.body).append(anchor); // Attach to document

    anchor.attr({
    href: URL.createObjectURL(blob),
    target: '_blank',
    download: fileName
    })[0].click();

    anchor.remove();
}
});

}

" var blob = new Blob([response.data], { type: type }); " This will give you that invalid/corrupt file that you are trying to open when you turn that byte array into a file in your javascript if you don't uncompress it first.

To fix this, you have a choice to either prevent gzipping this binary data so that you can properly turn it into the file that you are downloading, or you have to decompress that gzipped data in your javascript code before you turn it into a file.

Answer 3

        Response.BinaryWrite(bytes);
        Response.Flush();
        Response.Close();
        Response.End();

This works for us. We create PDFs from SQL Reporting Services.

Answer 4

Try adding

Response.End();

after the call to Response.BinaryWrite().

You may inadvertently be sending other content back after Response.BinaryWrite which may confuse the browser. Response.End will ensure that that the browser only gets what you really intend.