This repository has been archived by the owner on Feb 26, 2024. It is now read-only.

Update datagouv #42

	name: Update datagouv

	on:
	workflow_dispatch:
	schedule:
	- cron: "0 0 * * 1" # every monday

	concurrency:
	cancel-in-progress: true
	group: update-datagouv

	jobs:
	update-datagouv:
	runs-on: ubuntu-latest
	steps:
	- name: Checkout repository
	uses: actions/checkout@v3
	with:
	token: ${{ secrets.SOCIALGROOVYBOT_BOTO_PAT }}

	- name: Download data
	id: download
	continue-on-error: true
	run: \|
	sudo apt-get update
	sudo apt-get install -y curl git
	# download https://www.data.gouv.fr/fr/datasets/liste-des-conventions-collectives-par-entreprise-siret/
	echo "Download from data.gouv"
	curl -L https://www.data.gouv.fr/fr/datasets/r/bfc3a658-c054-4ecc-ba4b-22f3f5789dc7 -o "./WEEZ.csv"
	# ensure we have an entry for "renault SAS"
	echo "Check validity"
	cat ./WEEZ.csv \| grep 78012998704037
	# check if its a new file
	gunzip ./data/WEEZ.csv.gz
	if [[ $(diff ./WEEZ.csv ./data/WEEZ.csv) ]]; then
	echo "File changed"
	gzip -v ./WEEZ.csv
	gzip -t ./WEEZ.csv.gz
	echo "move"
	mv ./WEEZ.csv.gz ./data/WEEZ.csv.gz
	exit 0
	else
	echo "File not changed"
	rm ./WEEZ.csv
	exit 1
	fi;

	- name: Commit changes
	uses: stefanzweifel/git-auto-commit-action@v4
	if: steps.download.outcome == 'success'
	with:
	file_pattern: "data/*"
	commit_message: "fix: Update datagouv data"
	commit_author: "GitHub bot <[email protected]>"

Provide feedback