Process video stored in Supabase
Connect a Supabase Storage bucket by signing in with Supabase or with S3 keys. Any job then reads its inputs there and writes its outputs back to the same project.
Overview
The files your app uploads become job inputs, read as storage:// paths, and results land back in the same project. Sign in with Supabase from the dashboard or paste S3 keys, and the Supabase docs show what each one stores. An Edge Function can send the job and get the MP4 back in the bucket.
Questions
Does the Supabase connection follow Row Level Security?
No. Both sign-in and S3 keys reach every bucket in the project and bypass Row Level Security. Sign-in stores the project's service_role key, which also reaches the database, so S3 keys are the narrower choice.
Why does Supabase sign-in need the legacy API keys?
Supabase Storage's S3 endpoint accepts only the legacy anon and service_role keys. If they're turned off, turn them on under Project Settings > API Keys, or connect with S3 keys from Storage > S3.
Does removing Rendobar's Supabase authorization stop the connection?
No. The connection uses the project's keys, not the sign-in. Deactivate the legacy API keys under Project Settings > API Keys to stop it, which also stops every other client on those keys. For an S3 key, delete it under Storage > S3.
Why did a Supabase output replace a file with the same name?
Supabase Storage doesn't refuse an overwrite, so Name conflicts is fixed to Replace on a Supabase connection. Put {job_id} in the output path to keep every output.
Related
- Amazon S3 Read inputs from your S3 bucket and write every output back to it.
- Cloudflare R2 Keep your media in R2 and process it without moving it first.
- S3-compatible Point Rendobar at MinIO, Wasabi, Backblaze B2 or any bucket that speaks S3.
