Hello, all.

I'm currently working on a project that involves users uploading fairly large(~50MB) files to Amazon S3 at least once a week. We've currently built something where users upload to our server, we store an entry into the database with all the associated data for the object in question, and we upload the file they gave us to S3.

However, this is hitting our server a little too hard - if too many users are uploading at once, the entire service is crippled. With that in mind, I was doing some reading and discovered that you can upload directly to S3 with a simple form.

However, there's a problem here - I can't see anything in that tutorial(or the supporting documentation) that explains(or gives an example) on how to associate extra metadata with an upload. I'd like to be able to associate something like a row's ID field with the file that is getting uploaded - but can't seem to find any documentation on how to do this.

Has anyone done this before? How did you do it?

Thanks,

girasquid

After doing some tests and building a small sample app, I managed to get this figured out - when you pass in the success_action_redirect argument in your policy document, you can also pass in GET parameters - which will be preserved in the redirect.