scrabble-game/.gitea/workflows/ci.yaml

name: CI

# Single gated pipeline for the test contour. Gitea cannot express
# cross-workflow `needs`, so the full test suite and the auto test-deploy live in
# one workflow.
#
# Branch model (CLAUDE.md): feature branches are cut from `development`; a commit
# to a feature branch triggers nothing. The pipeline runs on a PR into
# `development` or `master` (the full test suite — the merge gate) and on a push
# to `development` (after a merge). The deploy job runs only for `development`
# (PR or merge), so a PR into `master` is test-only; the prod deploy is a manual
# workflow.
#
# Path-conditional jobs: `unit`/`integration`/`ui` run only when their
# code changed (the `changes` job decides). Because a skipped required check would
# block a merge under branch protection, the always-running `gate` job aggregates
# their results and is the ONLY required status check; it passes when every
# upstream job either succeeded or was skipped.
#
# Console output is kept plain (NO_COLOR + `docker compose --ansi never` +
# `--progress plain`) so the Gitea logs stay readable.

on:
  pull_request:
    branches: [development, master]
  push:
    branches: [development]

# The dictionary release. One Gitea variable is the single source of truth: the
# test suite validates against it here (inherited by the unit/integration jobs) and
# both contours' deploy jobs seed a fresh volume with the same value. A release bump
# is one edit (the variable). See deploy/README.md.
env:
  DICT_VERSION: ${{ vars.DICT_VERSION }}

jobs:
  # changes detects which areas a PR/push touched, so the test jobs can skip when
  # irrelevant. It defaults to running everything when the diff cannot be computed.
  changes:
    runs-on: ubuntu-latest
    defaults:
      run:
        shell: bash
    outputs:
      go: ${{ steps.filter.outputs.go }}
      ui: ${{ steps.filter.outputs.ui }}
    steps:
      - name: Checkout
        uses: actions/checkout@v4
        with:
          fetch-depth: 0

      - name: Detect changed paths
        id: filter
        run: |
          if [ "${{ github.event_name }}" = "pull_request" ]; then
            git fetch -q origin "${{ github.base_ref }}" || true
            range="origin/${{ github.base_ref }}...HEAD"
          else
            before="${{ github.event.before }}"
            if [ -z "$before" ] || [ "$before" = "0000000000000000000000000000000000000000" ] || ! git cat-file -e "${before}^{commit}" 2>/dev/null; then
              range="HEAD~1...HEAD"
            else
              range="${before}...HEAD"
            fi
          fi
          echo "comparison range: $range"
          # Default to running everything; narrow only when the diff is computable.
          go=true; ui=true
          files="$(git diff --name-only "$range" 2>/dev/null || echo __DIFF_FAILED__)"
          if [ "$files" != "__DIFF_FAILED__" ]; then
            echo "changed files:"; echo "$files"
            go=false; ui=false
            if echo "$files" | grep -qE '^(backend/|pkg/|gateway/|platform/|loadtest/|go\.work)'; then go=true; fi
            if echo "$files" | grep -qE '^ui/'; then ui=true; fi
            # The render sidecar bundles ui/src/lib, so its dir rides the ui lane (the
            # deploy's compose build picks it up either way).
            if echo "$files" | grep -qE '^renderer/'; then ui=true; fi
            # A workflow or deploy change re-runs everything as a safety net.
            if echo "$files" | grep -qE '^(\.gitea/workflows/|deploy/)'; then go=true; ui=true; fi
          else
            echo "diff failed; running all jobs"
          fi
          echo "selected: go=$go ui=$ui"
          echo "go=$go" >> "$GITHUB_OUTPUT"
          echo "ui=$ui" >> "$GITHUB_OUTPUT"

  unit:
    needs: changes
    if: ${{ needs.changes.outputs.go == 'true' }}
    runs-on: ubuntu-latest
    defaults:
      run:
        shell: bash
    env:
      # The engine consumes the published scrabble-solver module from this Gitea;
      # GOPRIVATE makes go fetch it directly (skipping the public proxy/checksum DB).
      GOPRIVATE: gitea.iliadenisov.ru/*
    steps:
      - name: Checkout
        uses: actions/checkout@v4

      - name: Fetch dictionary DAWGs
        run: |
          mkdir -p "${GITHUB_WORKSPACE}/dawg"
          curl -fsSL -o /tmp/dawg.tar.gz "https://gitea.iliadenisov.ru/developer/scrabble-dictionary/releases/download/${DICT_VERSION}/scrabble-dawg-${DICT_VERSION}.tar.gz"
          tar xzf /tmp/dawg.tar.gz -C "${GITHUB_WORKSPACE}/dawg"
          ls -la "${GITHUB_WORKSPACE}/dawg"

      - name: Set up Go
        uses: actions/setup-go@v5
        with:
          go-version-file: go.work
          cache: true

      - name: gofmt
        run: |
          unformatted="$(gofmt -l .)"
          if [ -n "$unformatted" ]; then
            echo "gofmt needed on:"; echo "$unformatted"; exit 1
          fi

      - name: vet
        run: go vet ./backend/... ./pkg/... ./gateway/... ./platform/telegram/... ./loadtest/...

      - name: build
        run: go build ./backend/... ./pkg/... ./gateway/... ./platform/telegram/... ./loadtest/...

      - name: test
        env:
          BACKEND_DICT_DIR: ${{ github.workspace }}/dawg
        run: go test -count=1 ./backend/... ./pkg/... ./gateway/... ./platform/telegram/... ./loadtest/...

  integration:
    needs: changes
    if: ${{ needs.changes.outputs.go == 'true' }}
    runs-on: ubuntu-latest
    defaults:
      run:
        shell: bash
    env:
      # Ryuk (testcontainers' reaper) does not start cleanly on every runner; the
      # suite's TestMain terminates its own container, so disable it.
      TESTCONTAINERS_RYUK_DISABLED: "true"
      GOPRIVATE: gitea.iliadenisov.ru/*
    steps:
      - name: Checkout
        uses: actions/checkout@v4

      - name: Fetch dictionary DAWGs
        run: |
          mkdir -p "${GITHUB_WORKSPACE}/dawg"
          curl -fsSL -o /tmp/dawg.tar.gz "https://gitea.iliadenisov.ru/developer/scrabble-dictionary/releases/download/${DICT_VERSION}/scrabble-dawg-${DICT_VERSION}.tar.gz"
          tar xzf /tmp/dawg.tar.gz -C "${GITHUB_WORKSPACE}/dawg"
          ls -la "${GITHUB_WORKSPACE}/dawg"

      - name: Set up Go
        uses: actions/setup-go@v5
        with:
          go-version-file: go.work
          cache: true

      - name: Integration tests
        # -count=1 disables the cache; -p=1 -parallel=1 keeps the container-backed
        # tests serial; the 15-minute timeout bounds a stuck container pull.
        env:
          BACKEND_DICT_DIR: ${{ github.workspace }}/dawg
        run: go test -tags=integration -count=1 -p=1 -parallel=1 -timeout=15m ./backend/...

  ui:
    needs: changes
    if: ${{ needs.changes.outputs.ui == 'true' }}
    runs-on: ubuntu-latest
    defaults:
      run:
        shell: bash
        working-directory: ui
    steps:
      - name: Checkout
        uses: actions/checkout@v4

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

      - name: Install pnpm
        run: npm install -g pnpm@11.0.9

      - name: Install deps
        run: pnpm install --frozen-lockfile

      - name: Type-check
        run: pnpm run check

      - name: Unit tests
        run: pnpm run test:unit

      - name: Build
        run: pnpm run build

      - name: Bundle-size budget
        run: node scripts/bundle-size.mjs

      # The render sidecar executes the shared ui/src/lib/gameimage.ts on skia-canvas;
      # its smoke test guards the bundling + skia seam (docs/TESTING.md).
      - name: Render sidecar test
        working-directory: renderer
        run: |
          pnpm install --frozen-lockfile
          pnpm test

      - name: Install Playwright browsers
        run: pnpm exec playwright install chromium webkit
        timeout-minutes: 5

      # The offline e2e plays a real local vs_ai game, so it needs the per-variant dawgs; fetch the
      # release the same way the Go jobs do and point the mock preview's copy step (scripts/
      # e2e-dict.mjs, via E2E_DICT_DIR below) at it.
      - name: Fetch dictionary DAWGs
        run: |
          mkdir -p "${GITHUB_WORKSPACE}/dawg"
          curl -fsSL -o /tmp/dawg.tar.gz "https://gitea.iliadenisov.ru/developer/scrabble-dictionary/releases/download/${DICT_VERSION}/scrabble-dawg-${DICT_VERSION}.tar.gz"
          tar xzf /tmp/dawg.tar.gz -C "${GITHUB_WORKSPACE}/dawg"
          ls -la "${GITHUB_WORKSPACE}/dawg"

      - name: E2E smoke (mock)
        run: pnpm run test:e2e
        timeout-minutes: 5
        env:
          E2E_DICT_DIR: ${{ github.workspace }}/dawg

  # conformance proves the client's local move preview (the ported dawg reader +
  # validator, ui/src/lib/dict) byte-for-byte against the authoritative Go engine:
  # a Go step generates golden parity vectors from the release dictionaries, then the
  # gated Vitest suite replays them. It spans both toolchains, so it runs whenever the
  # Go engine side or the UI side changed.
  conformance:
    needs: changes
    if: ${{ needs.changes.outputs.go == 'true' || needs.changes.outputs.ui == 'true' }}
    runs-on: ubuntu-latest
    defaults:
      run:
        shell: bash
    env:
      GOPRIVATE: gitea.iliadenisov.ru/*
    steps:
      - name: Checkout
        uses: actions/checkout@v4

      - name: Fetch dictionary DAWGs
        run: |
          mkdir -p "${GITHUB_WORKSPACE}/dawg"
          curl -fsSL -o /tmp/dawg.tar.gz "https://gitea.iliadenisov.ru/developer/scrabble-dictionary/releases/download/${DICT_VERSION}/scrabble-dawg-${DICT_VERSION}.tar.gz"
          tar xzf /tmp/dawg.tar.gz -C "${GITHUB_WORKSPACE}/dawg"

      - name: Set up Go
        uses: actions/setup-go@v5
        with:
          go-version-file: go.work
          cache: true

      - name: Generate golden parity vectors
        run: |
          go run ./backend/cmd/dictgen -dawg-dir "${GITHUB_WORKSPACE}/dawg" -out /tmp/dictgold
          go run ./backend/cmd/validategen -dawg-dir "${GITHUB_WORKSPACE}/dawg" -out /tmp/validgold
          go run ./backend/cmd/movegen -dawg-dir "${GITHUB_WORKSPACE}/dawg" -out "${GITHUB_WORKSPACE}/movegold"

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

      - name: Install pnpm
        run: npm install -g pnpm@11.0.9

      - name: Install deps
        working-directory: ui
        run: pnpm install --frozen-lockfile

      - name: Local-eval conformance (reader + validator vs the Go engine)
        working-directory: ui
        env:
          DICT_DAWG_DIR: ${{ github.workspace }}/dawg
          DICT_GOLD_DIR: /tmp/dictgold
          DICT_VALID_DIR: /tmp/validgold
          DICT_MOVEGEN_DIR: ${{ github.workspace }}/movegold
        run: pnpm exec vitest run src/lib/dict/

  # gate is the single branch-protection required check. It always runs and passes
  # only when each upstream job succeeded or was skipped (a path-filtered no-op),
  # failing the merge if any actually failed or was cancelled.
  gate:
    needs: [unit, integration, ui, conformance]
    if: always()
    runs-on: ubuntu-latest
    defaults:
      run:
        shell: bash
    steps:
      - name: Aggregate required checks
        run: |
          fail=
          for r in "unit:${{ needs.unit.result }}" "integration:${{ needs.integration.result }}" "ui:${{ needs.ui.result }}" "conformance:${{ needs.conformance.result }}"; do
            name="${r%%:*}"; res="${r#*:}"
            echo "$name = $res"
            case "$res" in
              success|skipped) ;;
              *) echo "::error::$name=$res"; fail=1 ;;
            esac
          done
          [ -z "$fail" ] || { echo "one or more required jobs failed"; exit 1; }
          echo "all required jobs passed or were skipped"

  deploy:
    # Auto test-deploy on a PR into development and on the push that merges it.
    # A PR into master is test-only (this job is skipped); prod deploy is manual.
    # Gates on `gate` (so a real test failure blocks the deploy) but runs even when
    # some test jobs were path-skipped. Skipped entirely when neither the Go nor the
    # UI side changed (e.g. a docs-only change): the contour image is unchanged, so
    # there is nothing to redeploy. `changes` still defaults both to true when the
    # diff is uncomputable, and a workflow/deploy edit forces both true, so an
    # ambiguous or infra change still deploys as a safety net.
    needs: [changes, gate]
    if: ${{ (needs.changes.outputs.go == 'true' || needs.changes.outputs.ui == 'true') && ((github.event_name == 'push' && github.ref == 'refs/heads/development') || (github.event_name == 'pull_request' && github.base_ref == 'development')) }}
    runs-on: ubuntu-latest
    defaults:
      run:
        shell: bash
    env:
      NO_COLOR: "1"
      DOCKER_CLI_HINTS: "false"
    steps:
      - name: Checkout
        uses: actions/checkout@v4

      - name: Build and (re)deploy the test contour
        working-directory: deploy
        env:
          # Sensitive values -> secrets; non-sensitive -> variables. The compose
          # interpolates these unprefixed names (see deploy/.env.example).
          POSTGRES_PASSWORD: ${{ secrets.TEST_POSTGRES_PASSWORD }}
          AWG_CONF: ${{ secrets.TEST_AWG_CONF }}
          GM_BASICAUTH_HASH: ${{ secrets.TEST_GM_BASICAUTH_HASH }}
          GRAFANA_ADMIN_PASSWORD: ${{ secrets.TEST_GRAFANA_ADMIN_PASSWORD }}
          TELEGRAM_BOT_TOKEN: ${{ secrets.TEST_TELEGRAM_BOT_TOKEN }}
          TELEGRAM_PROMO_BOT_TOKEN: ${{ secrets.TEST_TELEGRAM_PROMO_BOT_TOKEN }}
          # VK Mini App protected key (offline HMAC for the launch-param signature); empty
          # leaves the VK auth path (auth.vk) disabled until the operator sets the secret.
          # One VK Mini App serves every contour -> unprefixed secret.
          GATEWAY_VK_APP_SECRET: ${{ secrets.GATEWAY_VK_APP_SECRET }}
          # VK ID web login (browser VK-identity linking): the VK ID "Web" app's protected key
          # for the server-side confidential code exchange — a SEPARATE VK app from the Mini
          # App above. One VK ID "Web" app serves every contour -> unprefixed secret.
          GATEWAY_VK_ID_CLIENT_SECRET: ${{ secrets.GATEWAY_VK_ID_CLIENT_SECRET }}
          # Client-version gate (ARCHITECTURE.md §2): one plain (unprefixed) variable serves every
          # contour. In the test contour the stamped client version is a commit hash (unparseable ⇒
          # fail-open), so setting these only enforces on real semver prod builds. Empty ⇒ dormant.
          GATEWAY_MIN_CLIENT_VERSION: ${{ vars.GATEWAY_MIN_CLIENT_VERSION }}
          GATEWAY_RECOMMENDED_CLIENT_VERSION: ${{ vars.GATEWAY_RECOMMENDED_CLIENT_VERSION }}
          # Planted honeytoken bearer: presenting it flags the caller (logs + a ban metric on
          # test where the IP ban is off; a 24h IP ban on prod). Per-contour secret; empty = trap off.
          GATEWAY_HONEYTOKEN: ${{ secrets.TEST_GATEWAY_HONEYTOKEN }}
          # Signs the finished-game export download URLs (backend + compose interpolation).
          EXPORT_SIGN_KEY: ${{ secrets.TEST_EXPORT_SIGN_KEY }}
          # Transactional email via the shared Selectel relay: one account for every
          # contour -> unprefixed host/port/tls/user/pass. Empty host leaves the backend
          # on the log mailer (email disabled) but the contour still boots.
          SMTP_RELAY_USER: ${{ secrets.SMTP_RELAY_USER }}
          SMTP_RELAY_PASS: ${{ secrets.SMTP_RELAY_PASS }}
          SMTP_RELAY_HOST: ${{ vars.SMTP_RELAY_HOST }}
          SMTP_RELAY_PORT: ${{ vars.SMTP_RELAY_PORT }}
          SMTP_RELAY_TLS: ${{ vars.SMTP_RELAY_TLS }}
          # Direct-rail (YooKassa) intake on the contour: the credentials of a YooKassa TEST shop,
          # which takes test cards and moves no real money. YOOKASSA_WEB_TEST is forced to 1 so the
          # contour refuses to credit anything but a test payment, whatever credentials are set.
          # Empty credentials leave the direct rail off.
          YOOKASSA_WEB_SHOP_ID: ${{ secrets.TEST_BACKEND_YOOKASSA_WEB_SHOP_ID }}
          YOOKASSA_WEB_SECRET_KEY: ${{ secrets.TEST_BACKEND_YOOKASSA_WEB_SECRET_KEY }}
          YOOKASSA_WEB_TEST: "1"
          YOOKASSA_VAT_CODE: ${{ vars.TEST_BACKEND_YOOKASSA_VAT_CODE }}
          MYNALOG_KEY: ${{ secrets.TEST_MYNALOG_KEY }}
          MYNALOG_TZ: ${{ vars.TEST_MYNALOG_TZ }}
          SMTP_RELAY_FROM: ${{ vars.TEST_SMTP_RELAY_FROM }}
          # Operator alerts: backend admin emails (new feedback / complaints) + Grafana
          # infra alerts. Distinct senders + recipients; Grafana uses the relay's STARTTLS
          # host:port. Empty leaves the alert worker off and Grafana SMTP disabled.
          SMTP_RELAY_ADMIN_FROM: ${{ vars.TEST_SMTP_RELAY_ADMIN_FROM }}
          ADMIN_EMAIL: ${{ vars.TEST_ADMIN_EMAIL }}
          SMTP_RELAY_SERVICE_FROM: ${{ vars.TEST_SMTP_RELAY_SERVICE_FROM }}
          SERVICE_EMAIL: ${{ vars.TEST_SERVICE_EMAIL }}
          GRAFANA_SMTP_PORT: ${{ vars.GRAFANA_SMTP_PORT }}
          GF_SMTP_ENABLED: ${{ vars.TEST_GF_SMTP_ENABLED }}
          # Canonical public origin for links in the email (this contour's URL);
          # required by the backend whenever SMTP_RELAY_HOST is set.
          PUBLIC_BASE_URL: ${{ vars.TEST_PUBLIC_BASE_URL }}
          GM_BASICAUTH_USER: ${{ vars.TEST_GM_BASICAUTH_USER }}
          CADDY_SITE_ADDRESS: ${{ vars.TEST_CADDY_SITE_ADDRESS }}
          # TELEGRAM_MINIAPP_URL, GRAFANA_ROOT_URL and VITE_VK_ID_REDIRECT_URL are derived
          # from PUBLIC_BASE_URL in the run step below, not stored as their own variables.
          TELEGRAM_GAME_CHANNEL_ID: ${{ vars.TEST_TELEGRAM_GAME_CHANNEL_ID }}
          TELEGRAM_CHAT_ID: ${{ vars.TEST_TELEGRAM_CHAT_ID }}
          TELEGRAM_SUPPORT_CHAT_ID: ${{ vars.TEST_TELEGRAM_SUPPORT_CHAT_ID }}
          TELEGRAM_BOT_USERNAME: ${{ vars.TEST_TELEGRAM_BOT_USERNAME }}
          # The promo button reuses the UI's Mini App link variable.
          TELEGRAM_BOT_LINK: ${{ vars.TEST_VITE_TELEGRAM_LINK }}
          # The test contour always uses Telegram's test environment — pinned here,
          # not an operator variable. The prod workflow leaves it false.
          TELEGRAM_TEST_ENV: "true"
          VITE_TELEGRAM_BOT_ID: ${{ vars.TEST_VITE_TELEGRAM_BOT_ID }}
          VITE_TELEGRAM_LINK: ${{ vars.TEST_VITE_TELEGRAM_LINK }}
          VITE_TELEGRAM_GAME_CHANNEL_NAME: ${{ vars.TEST_VITE_TELEGRAM_GAME_CHANNEL_NAME }}
          # VK Mini App landing link + VK ID "Web" app id: one value each serves every
          # contour -> unprefixed. VITE_VK_APP_ID also feeds the gateway (GATEWAY_VK_ID_APP_ID);
          # the VK ID redirect URL is derived from PUBLIC_BASE_URL in the run step below.
          VITE_VK_APP_LINK: ${{ vars.VITE_VK_APP_LINK }}
          VITE_VK_APP_ID: ${{ vars.VITE_VK_APP_ID }}
          # Rewarded-ad test stub: set TEST_VITE_ADS_STUB=1 to swap real ads for a toast on the
          # contour (empty = real ads, for capturing the real VK ad result). Prod never sets it.
          VITE_ADS_STUB: ${{ vars.TEST_VITE_ADS_STUB }}
          # VITE_GATEWAY_URL omitted: the SPA is served same-origin, so it stays the
          # compose ":-" empty default. Other unset vars likewise fall to their defaults.
          POSTGRES_DB: ${{ vars.TEST_POSTGRES_DB }}
          POSTGRES_USER: ${{ vars.TEST_POSTGRES_USER }}
          DICT_VERSION: ${{ vars.DICT_VERSION }}
          LOG_LEVEL: ${{ vars.TEST_LOG_LEVEL }}
        run: |
          # Seed the config files to a stable host path. The runner checks out into
          # an ephemeral act workspace that is removed after the job, which would
          # dangle the compose config bind mounts in the long-lived containers
          # (e.g. Grafana then logs "no such file or directory"). Bind from a stable
          # dir instead (mirrors ../galaxy-game's $HOME/.galaxy-dev/monitoring).
          conf="$HOME/.scrabble-deploy"
          rm -rf "$conf"
          mkdir -p "$conf"
          cp -r caddy otelcol prometheus tempo grafana blackbox "$conf"/
          export SCRABBLE_CONFIG_DIR="$conf"
          # Maintenance page for the redeploy window, mirroring prod-deploy.sh so the SPA
          # overlay is exercised on the test contour too (not only prod). Raised just before
          # the recreate and lowered once caddy is back (below); the trap clears it if the
          # step fails so the contour never sticks in maintenance (and the reseed above wipes a
          # stale flag anyway). The caddy-routed probes run in the NEXT step, after it is lowered.
          maint_flag="$conf/caddy/on"
          trap 'rm -f "$maint_flag"' EXIT
          # Derive the public URLs from the one canonical origin instead of storing each as
          # its own variable (paths are structural SPA routes / the Caddy /_gm sub-path).
          # Exported before build so the VK ID redirect is baked into the SPA.
          base="${PUBLIC_BASE_URL%/}"
          export TELEGRAM_MINIAPP_URL="$base/telegram/"
          export GRAFANA_ROOT_URL="$base/_gm/grafana/"
          export VITE_VK_ID_REDIRECT_URL="$base/app/"
          # Grafana's SMTP from_address must be a BARE address (it rejects the "Name" <addr>
          # form the backend go-mail accepts) and validates it even when SMTP is disabled — a
          # bad value crash-loops Grafana. Split the display-format SERVICE From into a bare
          # address + name for Grafana; the backend keeps the full form.
          svc_from="${SMTP_RELAY_SERVICE_FROM:-}"
          case "$svc_from" in
            *"<"*">"*)
              export GRAFANA_SMTP_FROM_ADDRESS="$(printf '%s' "$svc_from" | sed -E 's/.*<([^>]+)>.*/\1/')"
              export GRAFANA_SMTP_FROM_NAME="$(printf '%s' "$svc_from" | sed -E 's/[[:space:]]*<[^>]*>.*$//; s/^"//; s/"$//')" ;;
            *)
              export GRAFANA_SMTP_FROM_ADDRESS="$svc_from" ;;
          esac
          # Bot-link mTLS material for the test contour: a private CA + gateway/bot
          # leaves (CN=gateway, the service name the bot dials). Prod supplies these
          # from PROD_ secrets instead. Regenerated each deploy; both ends redeploy
          # together so they always share the fresh CA (see deploy/gen-certs.sh).
          bash "$GITHUB_WORKSPACE/deploy/gen-certs.sh" "$conf/certs"
          # App version for the About screen: the git tag if present, else the short SHA
          # (the test checkout is shallow/untagged, so this is the SHA here — fine).
          export APP_VERSION="$(git -C "$GITHUB_WORKSPACE" describe --tags --always 2>/dev/null || echo dev)"
          # The telegram-local profile brings the bot + its VPN sidecar; prod runs the
          # bot on its own host instead (deploy/docker-compose.bot.yml), and the prod
          # main host omits both. Without the profile they would not start here.
          docker compose --ansi never --profile telegram-local build --progress plain
          # Raise the maintenance page, THEN bring caddy onto the reseeded config mount so it
          # actually carries the flag: the running caddy sits on the stale pre-reseed mount (the
          # dir was rm'd + recreated — see the force-recreate note below), so a flag written to
          # the new dir is invisible until caddy is recreated. With the fresh caddy up, an open
          # SPA sees the 503 marker + overlay for the whole recreate window, not a bare reconnect.
          : > "$maint_flag"
          docker compose --ansi never up -d --force-recreate --no-deps caddy
          docker compose --ansi never --profile telegram-local up -d --remove-orphans
          # The config-only services bind-mount the reseeded config dir. A plain `up -d`
          # leaves them on the previous bind mount (the dir was rm'd + recreated), so a
          # changed Grafana dashboard is ignored — force-recreate them to pick up the fresh
          # config. (Caddy was already recreated above so it would carry the maintenance flag.)
          docker compose --ansi never up -d --force-recreate --no-deps otelcol prometheus tempo grafana
          # Lower the maintenance page: services are back. An open SPA's poll now gets through
          # (once the gateway finishes booting) and reloads into the fresh client; the caddy
          # probes in the next step see 200. The EXIT trap is a backstop if we failed earlier.
          rm -f "$maint_flag"

      - name: Probe the landing, gateway and backend
        run: |
          set -u
          # Three probes. "/" is the static landing container and "/app/" the
          # gateway-served SPA shell (both through the contour caddy on the edge net).
          # The backend /readyz is probed on the internal net as well: the caddy probes
          # are blind to a crash-looping backend (the landing is static and the SPA
          # shell is served without it), which let a bad deploy go green while the
          # backend was down — so check it directly here.
          for i in $(seq 1 20); do
            if docker run --rm --network edge alpine:3.20 wget -q -T 5 -O /dev/null http://scrabble/ &&
               docker run --rm --network edge alpine:3.20 wget -q -T 5 -O /dev/null http://scrabble/app/ &&
               docker run --rm --network scrabble-internal alpine:3.20 wget -q -T 5 -O /dev/null http://backend:8080/readyz; then
              echo "healthy: GET / (landing) + /app/ (gateway) + backend /readyz"
              exit 0
            fi
            sleep 3
          done
          echo "probe failed; recent landing + gateway + backend logs:"
          docker logs --tail 50 scrabble-landing || true
          docker logs --tail 50 scrabble-gateway || true
          docker logs --tail 50 scrabble-backend || true
          exit 1

      - name: Probe the /offer/ public offer page is served
        run: |
          set -u
          # /offer/ is rendered by the render sidecar: it splices the live catalog price list
          # (fetched from the backend's internal endpoint) into the committed ui/legal/offer_ru.md.
          # If the @offer caddy route is missing, the request falls to the landing shell (also 200),
          # and if the backend fetch fails the sidecar returns 502 — so assert offer-specific content
          # (the seller INN, §11) AND that the pricing marker was substituted (proves the fetch +
          # splice ran), never just the status. Data-independent: an empty catalog still substitutes
          # the marker with nothing, so "pricing_template" must be absent either way.
          # Full body is needed (the INN is in §11 and the marker check spans the page), so this
          # cannot be a head-only probe like the legal one. Retry with a timeout instead: the offer is
          # now bilingual (RU + EN, larger), and a single-shot fetch can race the renderer's restart in
          # the same deploy.
          ok=
          for i in $(seq 1 20); do
            out="$(docker run --rm --network edge alpine:3.20 wget -q -T 10 -O - http://scrabble/offer/ 2>&1 || true)"
            if echo "$out" | grep -q "290210610742" && ! echo "$out" | grep -q "pricing_template"; then
              echo "ok: /offer/ serves the rendered offer with the price list spliced in"
              ok=1
              break
            fi
            sleep 3
          done
          if [ -z "$ok" ]; then
            echo "FAIL: /offer/ did not serve the rendered offer (route fell through, or the price splice failed)"
            docker logs --tail 50 scrabble-renderer || true
            docker logs --tail 50 scrabble-backend || true
            docker logs --tail 50 scrabble-landing || true
            exit 1
          fi

      - name: Probe the /privacy/ and /eula/ legal pages are served
        run: |
          set -u
          # /privacy/ and /eula/ are static legal markdown rendered by the render sidecar. If the
          # @legal caddy route is missing they fall to the landing shell (also 200, canonical
          # erudit-game.ru/), so assert the page-specific canonical URL — it uniquely identifies the
          # rendered legal doc reaching the edge, not the landing catch-all. Read only the <head>
          # (head -c 4096; the canonical link sits in the first bytes): the check is then
          # size-independent, so the large EULA page does not exceed the timeout while the monitoring
          # stack is still booting post-deploy and starving the CPU-capped renderer. Retry like the
          # landing/gateway/backend probe to ride out that window.
          for route in privacy eula; do
            ok=
            for i in $(seq 1 20); do
              head_out="$(docker run --rm --network edge alpine:3.20 sh -c "wget -q -T 10 -O - http://scrabble/${route}/ 2>/dev/null | head -c 4096" || true)"
              if printf '%s' "$head_out" | grep -qF "erudit-game.ru/${route}/"; then
                echo "ok: /${route}/ serves the rendered legal page"
                ok=1
                break
              fi
              sleep 3
            done
            if [ -z "$ok" ]; then
              echo "FAIL: /${route}/ did not serve the rendered legal page (@legal route missing?)"
              docker logs --tail 50 scrabble-renderer || true
              exit 1
            fi
          done

      - name: Probe the /pay/ callback route reaches the gateway
        run: |
          set -u
          # /pay/yookassa/notify must reach the gateway, not fall to the landing catch-all. The probe
          # is a bare GET from an address that is not one of YooKassa's senders, so it is rejected
          # downstream and the gateway answers a 4xx/5xx (never a 200 or a 404 landing.html), which
          # proves the edge route is wired.
          out="$(docker run --rm --network edge alpine:3.20 wget -S -q -O /dev/null http://scrabble/pay/yookassa/notify 2>&1 || true)"
          echo "$out" | grep -E "HTTP/" || true
          if echo "$out" | grep -qE "HTTP/1\.1 (4|5)[0-9][0-9]"; then
            echo "ok: /pay/ reaches the gateway (non-landing response)"
          else
            echo "FAIL: /pay/yookassa/notify did not reach the gateway (landing catch-all?)"
            docker logs --tail 50 scrabble-gateway || true
            exit 1
          fi

      - name: Probe the /dict edge route reaches the gateway
        run: |
          set -u
          # The client fetches each game's dictionary blob at {edge}/dict/{variant}/{version}
          # for the local move preview. If caddy does not route /dict to the gateway the request
          # falls to the static landing and the client silently gets a non-dawg blob. Probed
          # unauthenticated it must be the gateway's 401 (the route reaches the gateway), never a
          # 404/200 from the landing catch-all.
          out="$(docker run --rm --network edge alpine:3.20 wget -S -q -O /dev/null http://scrabble/dict/scrabble_en/v1 2>&1 || true)"
          echo "$out" | grep -E "HTTP/" || true
          if echo "$out" | grep -q " 401"; then
            echo "ok: /dict reaches the gateway (401 unauthenticated)"
          else
            echo "FAIL: /dict did not reach the gateway (expected 401) — caddy route missing?"
            exit 1
          fi

      - name: Probe the Telegram validator and bot liveness
        run: |
          set -u
          # The gateway/backend probes cannot see a crash-looping validator or bot
          # (the validator answers only internal gRPC; the bot long-polls + egresses
          # through the VPN sidecar with no public ingress). Inspect the containers
          # directly: each must be running, not restarting, with a stable restart
          # count. A grace period lets the VPN handshake and the bot-link dial settle.
          sleep 20
          for name in scrabble-telegram-validator scrabble-telegram-bot; do
            ok=
            for i in $(seq 1 20); do
              status="$(docker inspect -f '{{.State.Status}}' "$name" 2>/dev/null || echo missing)"
              restarting="$(docker inspect -f '{{.State.Restarting}}' "$name" 2>/dev/null || echo true)"
              if [ "$status" = "running" ] && [ "$restarting" = "false" ]; then
                c1="$(docker inspect -f '{{.RestartCount}}' "$name")"
                sleep 5
                c2="$(docker inspect -f '{{.RestartCount}}' "$name")"
                if [ "$c1" = "$c2" ]; then
                  echo "$name healthy: status=$status restarts=$c2"
                  ok=1
                  break
                fi
                echo "$name still restarting ($c1 -> $c2); waiting"
              fi
              sleep 3
            done
            if [ -z "$ok" ]; then
              echo "$name not healthy; recent logs:"
              docker logs --tail 80 "$name" || true
              exit 1
            fi
          done

      - name: Prune dangling images
        if: always()
        run: docker image prune -f