ansaurus

Question

Download office document without the web server trying to render it

Answer 1

A:

Have you tried spoofing Internet Explorer's User-Agent?

jleedev 2010-04-01 15:24:57

Answer 2

A:

There is a HTTP response header that makes a HTTP user agent download a file instead of trying to display it:

Content-Disposition: attachment; filename=paper.doc

I understand that you may not have access to the server, but this is one straight-forward way to do this if you can access the server scripts.

See the HTTP/1.1 specification and/or say, Google, for more details on the header.

amn 2010-04-01 15:28:36

Answer 3

A:

This is vb.net, but you should get the point. I've done this with an .aspx page that you pass the filename into, then return the content type of the file and add a header to make it an attachment, which prompts the browser to treat it as such.

Response.AddHeader("Content-Disposition", "attachment;filename=filename.xsn")
Response.ContentType = "application/x-microsoft-InfoPathFormTemplate"
Response.WriteFile(FilePath)  ''//Where FilePath is... the path to your file ;)
Response.Flush()
Response.End()

Shawn Steward 2010-04-01 15:29:11

The difference is, I am the browser in this case. I'm not the server so I don't control the response. All I can do is change my request to try and make the server return the file itself. Except I don't know how to do this.

Dan Revell 2010-04-01 15:37:15

You can try to use a proxy server that you do control, which WILL add the mentioned header to a response, prior to returning it to your agent.

amn 2010-04-01 15:39:59

@amn -- but if the original server has already altered the file before it reaches your proxy, the game is already lost.

James Curran 2010-04-01 15:49:56

James, am I missing something here? A HTTP server that matches a URL to a file does not modify anything, it just sends the file data as the body of the response, along with some headers. A proxy is well, a proxy - when your agent requests the URL through the proxy, the proxy will make request to the origin server, get the response, add the "Content-Disposition" header and return the modified response to the agent. Am i missing something?

amn 2010-04-01 16:02:51

@amn You are missing the fact that the web server does do something to the file. You can't always assume a uri is a physical file on a web server. In my case the difference is that manually the physical file (well a file in the sharepoint database) gets returned and programmatically infopath form services renders the form into html and returns that. I'm trying to discover what the difference is in the two requests that makes the web server do these different things.

Dan Revell 2010-04-01 16:35:58

Hmm, I see. Well then I will join the others in taking a chance on replicating IE request behavior. Obviously, it all has to do with that. IE sends simple plaintext as the request, so there is no magic involved - just send the same request (use Fiddler to spoof it). Sorry, I could not help you otherwise.

amn 2010-04-01 16:47:29

Answer 4

+1 A:

I'd suggest using Fiddler or WireShark, to see exactly how IE is sending the request, then duplicating that.

James Curran 2010-04-01 15:38:23

ansaurus

tags:

views:

answers:

Download office document without the web server trying to render it

related questions