publish_registry.yml

name: Rebuild & Publish Registry

# Trigger manually or weekly (Sunday 03:00 UTC).
# A schema submission (schema_submission.yml) handles per-PR incremental updates;
# this workflow does a full clean rebuild on a schedule.
on:
  workflow_dispatch:
    inputs:
      bump:
        description: "Version bump (patch / minor / major)"
        required: false
        default: "patch"
        type: choice
        options: [patch, minor, major]
      skip_converters:
        description: "Skip fetching external schemas (BIDS/NWB/DANDI/…)"
        required: false
        default: false
        type: boolean
  schedule:
    - cron: "0 3 * * 0"   # weekly, Sunday 03:00 UTC

jobs:
  rebuild:
    runs-on: ubuntu-latest
    permissions:
      contents: write   # needed to push data/ back to main

    steps:
      # -----------------------------------------------------------------------
      - name: Checkout
        uses: actions/checkout@v4
        with:
          # Fetch full history so the bump commit doesn't squash provenance
          fetch-depth: 0

      # -----------------------------------------------------------------------
      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: "3.12"

      # -----------------------------------------------------------------------
      - name: Cache pip
        uses: actions/cache@v4
        with:
          path: ~/.cache/pip
          key: ${{ runner.os }}-pip-${{ hashFiles('requirements.txt') }}
          restore-keys: ${{ runner.os }}-pip-

      # Sentence-transformers downloads all-MiniLM-L6-v2 (~90 MB) on first use.
      # Cache it so rebuilds don't re-download from Hugging Face every week.
      - name: Cache sentence-transformers model
        uses: actions/cache@v4
        with:
          path: ~/.cache/huggingface/hub
          key: hf-minilm-v2
          restore-keys: hf-

      # Embeddings are computed once per unique set of schema content and stored
      # in data/embeddings.parquet.  Cache keyed on schema file hashes so a
      # schema change forces a recompute while an unchanged run reuses it.
      - name: Cache embeddings
        uses: actions/cache@v4
        with:
          path: data/embeddings.parquet
          key: embeddings-${{ hashFiles('schemas/*.yml') }}
          restore-keys: embeddings-

      # -----------------------------------------------------------------------
      - name: Install dependencies
        run: pip install -r requirements.txt

      # -----------------------------------------------------------------------
      - name: Determine pipeline flags
        id: flags
        run: |
          BUMP="${{ inputs.bump || 'patch' }}"
          SKIP=""
          if [ "${{ inputs.skip_converters }}" = "true" ]; then
            SKIP="--skip-converters"
          fi
          echo "bump=$BUMP"   >> $GITHUB_OUTPUT
          echo "skip=$SKIP"   >> $GITHUB_OUTPUT

      # -----------------------------------------------------------------------
      # Full pipeline: wipe → seed → (converters) → ingest all → align → export
      - name: Seed, ingest, align, export
        run: |
          python neuro_ghost/pipeline.py \
            --fresh \
            ${{ steps.flags.outputs.skip }} \
            --bump  ${{ steps.flags.outputs.bump }} \
            --agent "github-actions"

      # -----------------------------------------------------------------------
      - name: Commit & push updated registry
        run: |
          git config user.name  "github-actions[bot]"
          git config user.email "github-actions[bot]@users.noreply.github.com"

          # Stage registry outputs + any new converted schemas
          git add data/registry.json data/versions/ data/provenance.json
          git add schemas/ --ignore-errors

          if git diff --cached --quiet; then
            echo "Registry unchanged — nothing to commit."
          else
            VERSION=$(python - <<'PY'
          import json, pathlib
          d = pathlib.Path("data/registry.json")
          if d.exists():
              print(json.loads(d.read_text()).get("registry_version", ""))
          PY
            )
            git commit -m "chore(registry): rebuild v${VERSION} [skip ci]"
            git push origin HEAD:main
          fi