Link to home
Start Free TrialLog in
Avatar of VerticalOne
VerticalOneFlag for United States of America

asked on

How to stream from varbinary(max) field to browser

I have a SQL database with fairly large varbinary(max) fields that are binary documents. I need to read that data from the SQL database and stream it down to a user's browser. I can open the whole field into memory and stream from there but some of the files are going to tax the server so I really want to sequentially access the data. I have found code on Microsoft that shows how to sequentially pull the data from the SQL database but I can't find any code to the sequentially push that down to the browser. The binary data is different document types such as pdf, doc files, xls files etc which I know at the time I pull them so I can set the content type. I have a simple aspx page that the user calls with a querystring identifying the ID of the document to pull. Can someone help me figure out how to pull this data out sequentially and not kill the server? thanks
Terry
Avatar of deighton
deighton
Flag of United Kingdom of Great Britain and Northern Ireland image

how about writing the document to a file then forwarding the user to the url of the file?
Avatar of VerticalOne

ASKER

I really don't want to waste the resources on the server to make another copy of the file that is already in database. What I am searching for is an ability to stream the binary content directly out of the SQL database to a browser. Microsoft has some sample code how to use a SQLDataReader in sequential access mode to pull data in chunks rather than having to pull the whole file into memory. This is what I am looking for but I have been unable to find any direction of how to hook that to a stream down to the browser.
what I've done before in asp.net for images is stuff like

in page_load()

Response.Clear()
dim b() as Byte = GenerateImage(....etc)
Response.ContentType = "image/png"
Response.BinaryWrite(b)
Response.end()


you'd need to change ContentType to the correct one for pdf

instead of GenerateImage, you'd read in your pdf and convert it to binary data, you might need memoryStream etc to do that - I forget how to do it exactly
the whole file would end up in memory though, for a short time.
I still really need a solution to this issue but have not been able to find any sample code that allows me to sequentially pull from the db and send down to a browser.
so you want to obtain one block of data, then send to the browser, then the next block and send to the browser etc?  I don't know how to do that - sorry.
Avatar of lcohan
"Can someone help me figure out how to pull this data out sequentially and not kill the server?"

How big are these files and which server is killed? I think you have already options from comments above and here's one more suggestion for large files - use a file storage to store your download files (we use Samba fileserver for exactly that purpose and reason behind a busy ecommerce website with SQL backend) and store only the path to the file in SQL table then when user clicks on link to download the file (pdf, image, ISO, etc...) will get it from your file server not killing your DB or Web servers.
As much as you want a magic solution I don't think there's such thing and even though my suggestion means code changes, sometimes you must do whatever you must do instead of killing servers.
http://www.samba.org/
http://www.melbpc.org.au/pcupdate/2403/2403article7.htm
ASKER CERTIFIED SOLUTION
Avatar of lcohan
lcohan
Flag of Canada image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Thanks, this got me working just fine.