Splitting documents in heroku

Hello,

I am working on a slide sharing application which lets people upload
presentations and pdfs and shows them in the browser. The website is
hosted
on heroku and slides are being uploaded to AWS S3 using CarrierWave gem.

Link: SlideSquare http://slidesquare.herokuapp.com

http://slidesquare.herokuapp.com

Problem: I am using Google Docs viewer to view presentations online
right
now, which fails to display .odp and .ppt, .pptx and related file
formats.
I have come across DocSplit Gem
https://github.com/documentcloud/docsplit (
GitHub - documentcloud/docsplit: Break Apart Documents into Images, Text, Pages and PDFs ) to extract images from ppt,
pptx but unfortunately I have failed to implement it because Heroku does
not allow write access to its system, and images can hence not be
extracted
in production.

Please describe a suitable way to resolve the problem, while using
Heroku.
If not so, please provide good alternatives.

The website is live here: http://slidesquare.herokuapp.com
Source code: http: GitHub - exagil/SlideSquare: SlideSquare is an open source online utility to share and explore cool presentations with the world.

On 2015-Feb-4, at 11:40 , Chirag A. [email protected] wrote:

Hello,

I am working on a slide sharing application which lets people upload
presentations and pdfs and shows them in the browser. The website is hosted on
heroku and slides are being uploaded to AWS S3 using CarrierWave gem.

Link: SlideSquare

http://slidesquare.herokuapp.com

Problem: I am using Google Docs viewer to view presentations online right now,
which fails to display .odp and .ppt, .pptx and related file formats. I have come
across DocSplit Gem ( GitHub - documentcloud/docsplit: Break Apart Documents into Images, Text, Pages and PDFs ) to extract
images from ppt, pptx but unfortunately I have failed to implement it because
Heroku does not allow write access to its system, and images can hence not be
extracted in production.

Actually, you can write to a temporary directory on Heroku. You just
can’t rely on that storage being available past the end of the request.
I’ve relied on this behavior for building PDFs and then delivering them
to the browser with a Rails send_file

Depending on whether you pull images from PPT only on demand, this might
be enough for you.

-Rob