index-developer-docs.yml

name: Index Developer Docs
on:
  schedule:
    - cron: '0 2 * * *'
  workflow_dispatch:
    inputs:
      reason:
        description: 'Reason for running (optional)'
        required: false
        default: 'Manual trigger'
      dry_run:
        description: 'Dry run (validate extraction without uploading)'
        required: false
        type: boolean
        default: false
      force_reindex:
        description: 'Force full reindex (calls processalldocuments after upload)'
        required: false
        type: boolean
        default: false
      disable_stale_deletion_check:
        description: 'Bypass 20% stale-deletion safeguard (use for one-off cleanup runs only)'
        required: false
        type: boolean
        default: false

jobs:
  run-indexer:
    runs-on: ubuntu-latest
    timeout-minutes: 45
    steps:
      - name: Checkout repository
        uses: actions/checkout@v4
        with:
          fetch-depth: 0
          token: ${{ secrets.GITHUB_TOKEN }}

      - name: Set up mise
        uses: jdx/mise-action@v4
        with:
          cache: true
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

      - name: Install site dependencies
        run: pnpm install --frozen-lockfile

      - name: Build site
        run: pnpm build

      - name: Install indexing dependencies
        working-directory: scripts/indexing
        run: uv sync

      - name: Run indexing script
        working-directory: scripts/indexing
        env:
          GLEAN_INDEXING_API_TOKEN: ${{ secrets.GLEAN_INDEXING_API_TOKEN }}
          GLEAN_INSTANCE: ${{ secrets.GLEAN_INSTANCE }}
          DRY_RUN: ${{ inputs.dry_run }}
          DISABLE_STALE_DELETION_CHECK: ${{ inputs.disable_stale_deletion_check }}
        run: uv run main.py

      - name: Force reindex
        if: inputs.force_reindex && inputs.dry_run != true
        working-directory: scripts/indexing
        env:
          GLEAN_INDEXING_API_TOKEN: ${{ secrets.GLEAN_INDEXING_API_TOKEN }}
          GLEAN_INSTANCE: ${{ secrets.GLEAN_INSTANCE }}
        run: |
          uv run python3 -c "
          from glean.indexing.common import api_client
          from glean.api_client.models import ProcessAllDocumentsRequest
          with api_client() as client:
              client.indexing.documents.process_all(request=ProcessAllDocumentsRequest(datasource='devdocs'))
          print('processalldocuments called for devdocs')
          "