update.yml

name: Refresh data

on:
  push:
    branches:
      - main
  workflow_dispatch:
  schedule:
    - cron: "21 11 * * *"

jobs:
  scrape_fresh_data:
    runs-on: macos-latest
    steps:
      - name: Checkout
        uses: actions/checkout@v4.1.1
      - name: Setup Python
        uses: actions/setup-python@v5.0.0
        with:
          python-version: "3.12.x"
      - name: Install Python dependencies
        run: |
          pip install uv
          uv pip install -r requirements.txt --system
      - name: Fetch latest data
        run: |
          cp data/Street_Tree_List.csv Street_Tree_List-old.csv
          curl -o Street_Tree_List-unsorted.csv "https://data.sfgov.org/api/views/tkzw-k3nq/rows.csv?accessType=DOWNLOAD"
          # Remove heading line and use it to start a new file
          head -n 1 Street_Tree_List-unsorted.csv > data/Street_Tree_List.csv
          # Sort all but the first line and append to that file
          tail -n +2 "Street_Tree_List-unsorted.csv" | sort >> data/Street_Tree_List.csv
          # Generate commit message using csv-diff
          csv-diff Street_Tree_List-old.csv data/Street_Tree_List.csv --key=TreeID --singular=tree --plural=trees > message.txt
      - name: Commit and push if it changed
        run: |-
          git config user.name "Automated"
          git config user.email "actions@users.noreply.github.com"
          git add data/Street_Tree_List.csv
          timestamp=$(date -u)
          git commit -F message.txt || exit 0
          git push
  build_to_motherduck:
    uses: ./.github/workflows/build.yml
    needs: scrape_fresh_data
    secrets:
      MOTHERDUCK_TOKEN: ${{ secrets.MOTHERDUCK_TOKEN }}