metpo-content-lint.yml

name: METPO content lint

# Decisive guardrail on METPO ontology content. The Google Sheet is the source of
# truth (the src/templates/*.tsv are a derived build cache), so this job lints a
# FRESH Sheet download rather than the committed TSV. The committed baselines under
# lint-baselines/ record the currently-accepted debt; only findings BEYOND the
# baseline fail the build. Tighten the baseline (regenerate it) as that debt is paid
# down -- it can only go down.
#
# Runs on every PR/push (so a rebuild that pulls in a bad Sheet edit is caught) AND
# on a weekly schedule (so a bad Sheet edit is caught even with no PR open), without
# anyone having to summon it.

on:
  push:
    branches: [ main ]
  pull_request:
    branches: [ main ]
  schedule:
    - cron: '0 13 * * 1'   # Mondays 13:00 UTC
  workflow_dispatch:

jobs:
  content-lint:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v6

      - name: Set up Python
        uses: actions/setup-python@v6
        with:
          python-version: "3.11"

      - name: Install uv
        run: |
          curl -LsSf https://astral.sh/uv/install.sh | sh
          echo "$HOME/.local/bin" >> $GITHUB_PATH

      - name: Install dependencies
        run: uv sync

      - name: Download live Sheet tabs (source of truth)
        run: |
          set -euo pipefail
          curl -sL "$(uv run --quiet python metpo/sheets_config.py classes)"    -o /tmp/classes.tsv
          curl -sL "$(uv run --quiet python metpo/sheets_config.py properties)" -o /tmp/properties.tsv
          # Guard: a login/HTML page instead of TSV means the Sheet stopped being
          # link-shareable; fail loudly rather than silently passing on garbage.
          head -1 /tmp/classes.tsv    | grep -q $'^ID\t' || { echo "::error::classes download is not a TSV (Sheet sharing changed?)";    exit 1; }
          head -1 /tmp/properties.tsv | grep -q $'^ID\t' || { echo "::error::properties download is not a TSV (Sheet sharing changed?)"; exit 1; }

      - name: Lint classes against baseline (fail only on NEW findings)
        run: |
          uv run metpo-proposal-lint /tmp/classes.tsv \
            --known /tmp/properties.tsv --mode submit \
            --baseline lint-baselines/classes.json

      - name: Lint properties against baseline (fail only on NEW findings)
        run: |
          uv run metpo-proposal-lint /tmp/properties.tsv \
            --known /tmp/classes.tsv --mode submit \
            --baseline lint-baselines/properties.json