caches_cron_job.yml

name: Create caches for gin ecephys data and virtual env

on: 
  workflow_dispatch:
  push:  # When someting is pushed into main this checks if caches need to re-created
    branches:
      - main
  schedule:
    - cron: "0 12 * * *"  # Daily at noon UTC

jobs:
  
  create-virtual-env-cache-if-missing:
    name: Caching virtual env
    runs-on: "ubuntu-latest"
    steps:
      - uses: actions/checkout@v3
      - uses: actions/setup-python@v4
        with:
          python-version: '3.9'
      - name: Get current year-month
        id: date
        run: |
          echo "date=$(date +'%Y-%m')" >> $GITHUB_OUTPUT
      - name: Get current dependencies hash
        id: dependencies
        run: |
          echo "hash=${{hashFiles('**/pyproject.toml')}}" >> $GITHUB_OUTPUT
      - uses: actions/cache@v3
        id: cache-venv
        with:
          path: ~/test_env
          key: ${{ runner.os }}-venv-${{ steps.dependencies.outputs.hash }}-${{ steps.date.outputs.date }}
      - name: Cache found?
        run: echo "Cache-hit == ${{steps.cache-venv.outputs.cache-hit == 'true'}} "
      - name: Create the virtual environment to be cached 
        if: steps.cache-venv.outputs.cache-hit != 'true'
        run: |
          git config --global user.email "CI@example.com"
          git config --global user.name "CI Almighty"
          python -m venv ~/test_env  # Environment used in the caching step
          source ~/test_env/bin/activate
          python -m pip install -U pip  # Official recommended way
          # herdingspikes need numpy to installed first (!)
          pip install numpy==1.22
          pip install -e .[test,extractors,full]
  
  create-data-cache-if-missing:
    name: Caching data env
    runs-on: "ubuntu-latest"
    steps:
      - uses: actions/setup-python@v4
        with:
          python-version: '3.9'
      - name: Get ephy_testing_data current head hash
        id: repo_hash
        run: |
          echo "dataset_hash=$(git ls-remote https://gin.g-node.org/NeuralEnsemble/ephy_testing_data.git HEAD | cut -f1)" >> $GITHUB_OUTPUT
      - uses: actions/cache@v3
        id: cache-datasets
        with:
          path: ~/spikeinterface_datasets
          key: ${{ runner.os }}-datasets-${{ steps.repo_hash.outputs.dataset_hash }}
      - name: Cache found?
        run: echo "Cache-hit == ${{steps.cache-datasets.outputs.cache-hit == 'true'}} "
      - name: Installing datalad and git-annex
        if: steps.cache-datasets.outputs.cache-hit != 'true'
        run: |
          git config --global user.email "CI@example.com"
          git config --global user.name "CI Almighty"
          python -m pip install -U pip  # Official recommended way
          pip install datalad-installer
          datalad-installer --sudo ok git-annex -m datalad/packages
          pip install datalad
      - name: Download dataset
        if: steps.cache-datasets.outputs.cache-hit != 'true'
        run: |
          datalad install -rg https://gin.g-node.org/NeuralEnsemble/ephy_testing_data
          mv --force ./ephy_testing_data ~/spikeinterface_datasets
          cd ~
          du -hs spikeinterface_datasets