name: CI

on:
  push:
    branches: [master]
  pull_request:
    branches: [master]
  workflow_dispatch:

jobs:
  ci:
    name: build + unit + e2e
    runs-on: ubuntu-latest
    steps:
      - name: Checkout
        uses: actions/checkout@v4

      # The runner provides a working docker CLI out of the box (full-dind) — no endpoint
      # resolution needed. Job and app containers share the daemon, so put them on one
      # user-defined network and reach the app by service DNS (the job's netns has no
      # default route here, so host-IP URLs are unreliable). The alias `app` is declared
      # in docker-compose.ci.yml.
      - name: Prepare e2e network
        run: |
          set -u
          docker network inspect flight-ci-e2e >/dev/null 2>&1 || docker network create flight-ci-e2e
          if ! docker network connect flight-ci-e2e "$(hostname)"; then
            echo "::error::failed to attach job container '$(hostname)' to flight-ci-e2e — the job must run in a container on the shared daemon (check runner config)" >&2
            exit 1
          fi
          docker network inspect flight-ci-e2e --format 'containers on flight-ci-e2e: {{range .Containers}}{{.Name}} {{end}}'
          echo "E2E_BASE_URL=http://app:3000" >> "$GITHUB_ENV"

      - name: Setup Node
        uses: actions/setup-node@v4
        with:
          node-version: 22

      # bun drives installs (repo ships bun.lock, no package-lock.json); pinned to 1.4.2.
      # Retried: the runner's outbound network/DNS is flaky (transient EAI_AGAIN seen in CI).
      - name: Install bun
        run: |
          set -u
          ok=0
          for i in 1 2 3; do
            if npm i -g bun@1.4.2; then ok=1; break; fi
            echo "::warning::bun install attempt $i/3 failed (likely transient network) — retrying in 5s" >&2
            sleep 5
          done
          [ "$ok" = "1" ] || { echo "::error::bun install failed after 3 attempts" >&2; exit 1; }
          bun --version

      # Job-level dep cache via the runner's built-in cache server. continue-on-error on
      # both steps: an unreachable/broken cache server must never break the run.
      - name: Restore deps cache
        uses: actions/cache@v4
        continue-on-error: true
        with:
          path: |
            node_modules
            $HOME/.bun/install/cache
          key: ${{ runner.os }}-deps-bun-${{ hashFiles('bun.lock') }}

      # Retried: the runner's outbound network/DNS is flaky (transient EAI_AGAIN seen in CI).
      - name: Install dependencies
        run: |
          set -u
          ok=0
          for i in 1 2 3; do
            if bun install --frozen-lockfile; then ok=1; break; fi
            echo "::warning::dependency install attempt $i/3 failed (likely transient network) — retrying in 5s" >&2
            sleep 5
          done
          [ "$ok" = "1" ] || { echo "::error::dependency install failed after 3 attempts" >&2; exit 1; }

      - name: Save deps cache
        uses: actions/cache@v4
        continue-on-error: true
        with:
          path: |
            node_modules
            $HOME/.bun/install/cache
          key: ${{ runner.os }}-deps-bun-${{ hashFiles('bun.lock') }}

      - name: Build
        run: npm run build

      - name: Unit tests
        run: npx vitest run

      # Browsers run locally in CI (PW_REMOTE_BROWSER=0 below): the shared remote browser
      # server at 192.168.1.4 cannot reach a stack inside dind, and local chromium keeps
      # the job self-contained.
      - name: Install browsers (local)
        run: npx playwright install --with-deps chromium

      # Cache the base image on the shared daemon up front: a single loud failure here
      # beats a 6-minute BuildKit "failed to solve" when the daemon's outbound net is down.
      - name: Pull e2e base image
        run: docker pull node:22-slim

      # Pre-build the e2e image with retries (flaky runner network); the E2E step's own
      # build then becomes a fast no-op. Plain progress keeps BuildKit output visible.
      - name: Build e2e image (retries)
        run: |
          set -u
          ok=0
          for i in 1 2 3; do
            if docker compose -f docker-compose.ci.yml -p flight-ci build --progress=plain; then ok=1; break; fi
            echo "::warning::e2e image build attempt $i/3 failed (likely transient network) — retrying in 5s" >&2
            sleep 5
          done
          [ "$ok" = "1" ] || { echo "::error::e2e image build failed after 3 attempts" >&2; exit 1; }

      # E2E against an isolated stack on the shared daemon — never the production instance.
      # docker-compose.ci.yml joins the flight-ci-e2e network prepared above; project name
      # flight-ci keeps it distinct from production. E2E_DEPLOY_CMD makes the config's
      # webServer block deploy + wait for readiness (http://app:3000 via service DNS);
      # E2E_COMPOSE_ARGS points state.spec.ts' mid-run `docker compose restart flight` at
      # the same file + project. E2E_BASE_URL comes from GITHUB_ENV (Prepare e2e network).
      - name: E2E (isolated stack in dind)
        env:
          PW_REMOTE_BROWSER: "0"
          # Build separately with plain progress so BuildKit step output is visible in the
          # job log (auto progress hides it), then start without rebuilding.
          E2E_DEPLOY_CMD: "docker compose -f docker-compose.ci.yml -p flight-ci build --progress=plain && docker compose -f docker-compose.ci.yml -p flight-ci up -d"
          E2E_COMPOSE_ARGS: "-f docker-compose.ci.yml -p flight-ci"
        run: npx playwright test

      # The daemon is shared with production — never leave our stack or network behind.
      # -v drops the named DB volume so every run seeds fresh (no schema drift between PRs).
      - name: Cleanup e2e stack + network
        if: always()
        run: |
          docker compose -f docker-compose.ci.yml -p flight-ci down --remove-orphans -v || true
          docker network remove flight-ci-e2e 2>/dev/null || true