Video processing and YouTube deployment after Configuring Video Upload.

[Ahmed Jazzar]

Hello everyone:

I've been working days ago on configuring Video Upload Pipeline as described here: Developing-on-the-edX-Developer-Stack, and I reached a point that all the videos have been uploaded successfully to the configured S3 Bucket, but as the documentation stated and I quote:

 These steps will only get you through the upload stage. Video processing and YouTube deployment are not covered here

So any body has any idea where should I go next, or what can I do to enable the video processing and youtube deployment? I reached dead ends in the every edX documentation and I could find nothing.

 P.S. This is stated earlier in this post with no answer: https://groups.google.com/forum/#!msg/edx-code/nretG2PL7Sc/DMDa9UVyoHMJ

[Ahmed Jazzar]

Further investigation lead me to this wiki: Video Pipeline 1.0

What I understood is that edx-val and S3 are communicating with a module named VEDA which I could not find any evidence of its existence in edX Github repos, configuration repo, and the edx-platform repo itself, so the next questions raised here are: what is VEDA? How can I run it? And where can I find it?

[Alexander Kryklia]

Hi Ahmed

Unfortunately, VEDA is closed source 3rd party proprietary software. So it is not possible to use it.

However, Appsembler and Raccoon Gang are thinking about staring works toward creating possibilities to use 3rd party video processing services like Wistia in Open edX.

We can elaborate a bit more if you interested in.

Alexander Kryklia

CTO@Raccoon Gang

[Ahmed Jazzar]

Hi Alexander

Thanks for elaborating, that was really helpful! I think that edX should document that so no one spends days looking after something not exists later. 

Again, thank you. 

[Régis Behmo]

Hi,

FYI, at FUN (https://fun-mooc.fr) we are currently working on an open source implementation of a video transcoding/storage/streaming pipeline. For our needs, this pipeline will be based on AWS, but it will feature pluggable modules that will allow people to use other backends (e.g: local server). The software components of the stack will be : Python 3, Django 1.10, Django REST framework, Celery, SQL (Postgresql, Mysql or sqlite...). It should be a relatively simple project to install and will be completely separate from Open edX, but easily pluggable. This is the continuation of our work on building an uploader integrated into the Studio : https://github.com/openfun/fun-apps/blob/dev/fun/templates/cms/videoupload/js/videoupload.js

The expected delivery date is mid-october 2016.

--
Régis
Senior software engineer @ FUN-MOOC