Thanks for posting the conclusion!
Posting the resolution only takes a few minutes, but its so helpful for
future readers, and those of us just keeping tabs on the thread.
Thanks
On Fri, May 6, 2011 at 8:31 AM, batate <br...@rapidred.com> wrote:
> Here's the solution for posterity. There were two problems to solve:
>
> Recall that I'm using CarrierWave.
> 1) Heroku allows permissions only to the tmp directory. I needed to
> specify the cache dir on the attachment uploader.
> 2) I needed to set content headers for s3. You can specify them when
> you upload a file, and Amazon will provide them in the content headers
> when you download a file.
>
> So in my model (Mongoid), I have
>
> class Report
> include Mongoid::Document
>
> ...
>
> mount_uploader :attachment, AttachmentUploader
>
>
>
>
>
> And in my uploader, I have:
>
> class AttachmentUploader < CarrierWave::Uploader::Base
>
>
> def store_dir
> "uploads/#{model.class.to_s.underscore}/#{mounted_as}/#{model.id}"
> end
>
> def s3_headers
> { "Content-Disposition" => "attachment; filename=export.csv;" }
> end
>
> # Use Heroku's temp folder for uploads
> def cache_dir
> "#{Rails.root}/tmp/uploads"
> end
>
> end
>
>
>
> The s3 headers above set the content disposition type to attachment,
> so it will download as a file instead of a page. The filename will set
> the filename upon download.
>
> The cache_dir sets the temp holding area for Heroku, since the
> uploads/ directory is not public.
>
>
> To upload an attachment, I say report.attach. That does this:
>
> def attach(filename)
> self.attachment = File.open(filename)
> self.save
> end
>
>
>
> Hope this helps.
>
> -bt
>
> On May 5, 10:51 am, batate <br...@rapidred.com> wrote:
> > I am uploading reports, via Carrierwave, to Amazon s3, but with
> > protected urls. I'm able to process the upload fine, and I can see the
> > file on Amazon, but it renders as text. I want it to trigger a browser
> > download instead of just render the text.
> >
> > When I was generating the file via a template, I could control the
> > headers and force the right download like this:
> >
> > helper_method :render_csv
> > def render_csv(file, view=nil)
> > filename = "export"
> > filename = "user_export" if params[:action] == "process_users"
> > filename ||= params[:action]
> > filename += '.csv'
> >
> > (request.env['HTTP_USER_AGENT'] =~ /msie/i) ?
> > set_ie_headers( filename ) : set_headers( filename )
> > render :layout => false, :action => view
> > end
> >
> > But now, I'm downloading from s3.
> >
> > Any suggestions?
>
> --
> You received this message because you are subscribed to the Google Groups
> "Heroku" group.
> To post to this group, send email to heroku@googlegroups.com.
> To unsubscribe from this group, send email to
> heroku+unsubscr...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/heroku?hl=en.
>
>
--
-John
--
You received this message because you are subscribed to the Google Groups
"Heroku" group.
To post to this group, send email to heroku@googlegroups.com.
To unsubscribe from this group, send email to
heroku+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/heroku?hl=en.
