test_api.yaml

name: TEST Predict with API

on: [workflow_dispatch]

jobs:
  predict-cost-api:
    runs-on: ubuntu-latest

    steps:
      - uses: actions/checkout@v2
        with:
          path: ./

      # Sometimes kubectl takes a second to forward.
      # Sleeping is bad but I'm not sure how else to do this,
      # maybe repeatedly curl (up to max N times) until 200
      # https://dev.to/richicoder1/how-we-connect-to-kubernetes-pods-from-github-actions-1mg
      # - name: Forward the kubecost service
      #   run: |
      #     kubectl port-forward --namespace kubecost service/kubecost-cost-analyzer 9090 &
      #     sleep 5

      # - name: Curl the kubecost service
      #   run: curl http://localhost:9090

      - name: local action with raw yaml
        id: raw-prediction
        uses: ./
        with:
          log_level: "debug"
          kubecost_api_path: "http://nightly.kubecost.io:9090/model"
          path: ./test

      - name: output raw yaml prediction
        run: |
          echo "${{ steps.raw-prediction.outputs.PREDICTION_TABLE }}"