Weekly Supabase → BigQuery mirror #12

	name: Weekly Supabase → BigQuery mirror

	# Mirrors Supabase tables (users, recommendation_log, feedback, groups,
	# group_members, group_votes) into the `plot_supabase_mirror` BigQuery
	# dataset. Full refresh, every Monday 08:00 UTC (after the retrain run
	# at 07:00).
	#
	# Why: gives us a queryable BQ copy for analytics + a belt-and-suspenders
	# backup against accidental drops on Supabase.
	#
	# Required GH Actions secrets (you already have them):
	# GCP_SA_KEY — service-account JSON with bigquery.dataEditor + jobUser
	# on the project
	# GCP_PROJECT — mlops-project-491402
	# DATABASE_URL — Supabase pooler connection string

	on:
	schedule:
	- cron: "0 8 * * 1"
	workflow_dispatch:

	jobs:
	mirror:
	runs-on: ubuntu-latest
	timeout-minutes: 15
	steps:
	- uses: actions/checkout@v4

	- name: Set up Python
	uses: actions/setup-python@v5
	with:
	python-version: "3.11"
	cache: pip
	cache-dependency-path: requirements.txt

	- name: Install deps
	run: \|
	python -m pip install --upgrade pip
	pip install -r requirements.txt

	- name: Auth to GCP
	uses: google-github-actions/auth@v2
	with:
	credentials_json: ${{ secrets.GCP_SA_KEY }}

	- name: Run mirror
	env:
	GCP_PROJECT: ${{ secrets.GCP_PROJECT }}
	DATABASE_URL: ${{ secrets.DATABASE_URL }}
	BQ_BACKUP_DATASET: plot_supabase_mirror
	BQ_BACKUP_LOCATION: US
	run: python scripts/supabase_to_bq.py \| tee mirror_log.txt

	- name: Summarize
	if: always()
	run: \|
	{
	echo "## Supabase → BigQuery mirror"
	echo ""
	echo '```'
	grep -E "rows mirrored\|Done\." mirror_log.txt \|\| true
	echo '```'
	} >> "$GITHUB_STEP_SUMMARY"

Navigation Menu