selfpost/.github/workflows/release.yml

name: release

# Publishes an immutable, version-tagged image on ghcr.io (spec 10.1).
# Ordinary commits and bare git tag pushes do not publish anything. A published
# GitHub Release starts release-on-publish.yml, which dispatches this workflow
# with the release version. You can also run it manually via workflow_dispatch
# with an explicit SemVer X.Y.Z version.
#
# Native per-architecture builds (see docs/development.md), not qemu:
# running the full Postfix/OpenDKIM stack under emulation for the e2e gate
# below is impractically slow. Each arch builds, e2e-gates and pushes its own
# tag on its own native runner; a merge job then combines them into the one
# manifest tag documented in deploy/docker-compose.yml. "test, then push" (not
# push-by-digest then test) is deliberate: it means the bytes that get tagged
# are exactly the bytes that passed e2e, at the cost of per-arch tags lingering
# in the registry as a side effect (harmless — the version tag's immutability,
# spec 10.1, is about that tag, not these).
on:
  workflow_dispatch:
    inputs:
      version:
        description: "Image version as X.Y.Z (no v prefix)."
        required: true
        type: string

permissions:
  contents: read
  packages: write

jobs:
  prepare:
    runs-on: ubuntu-latest
    outputs:
      version: ${{ steps.version.outputs.version }}
    steps:
      - name: Derive a SemVer X.Y.Z version
        id: version
        env:
          INPUT_VERSION: ${{ github.event.inputs.version }}
          EVENT_NAME: ${{ github.event_name }}
        run: |
          set -euo pipefail
          raw="${INPUT_VERSION:-}"
          raw="${raw#v}"
          if ! [[ "$raw" =~ ^[0-9]+\.[0-9]+\.[0-9]+$ ]]; then
            echo "::error::release version is not SemVer X.Y.Z (event=${EVENT_NAME}; pass inputs.version on workflow_dispatch). Refusing to publish ghcr.io/${{ github.repository }}:${raw:-?}"
            exit 1
          fi
          #region agent log
          echo "::group::agent-log prepare version"
          echo "event=${EVENT_NAME} version=${raw} checkout_ref=v${raw}"
          echo "::endgroup::"
          #endregion
          echo "version=${raw}" >> "$GITHUB_OUTPUT"

  build:
    needs: prepare
    strategy:
      fail-fast: false
      matrix:
        include:
          - runner: ubuntu-latest
            arch: amd64
          - runner: ubuntu-24.04-arm
            arch: arm64
    runs-on: ${{ matrix.runner }}
    steps:
      - uses: actions/checkout@v4
        with:
          ref: v${{ needs.prepare.outputs.version }}

      - uses: docker/setup-buildx-action@v3

      - name: Build image (native, loaded locally for the e2e gate)
        uses: docker/build-push-action@v6
        with:
          context: .
          file: build/Dockerfile
          load: true
          provenance: false
          build-args: |
            VERSION=${{ needs.prepare.outputs.version }}
          tags: selfpost:e2e

      - uses: actions/setup-go@v5
        with:
          go-version: "1.26"
          cache-dependency-path: test/e2e/go.sum

      - name: e2e (gates publishing — see docs/development.md)
        run: cd test/e2e && go test -v -timeout 20m ./...

      - name: Log in to ghcr.io
        uses: docker/login-action@v3
        with:
          registry: ghcr.io
          username: ${{ github.actor }}
          password: ${{ secrets.GITHUB_TOKEN }}

      - name: Push per-arch tag
        # GHCR sometimes rejects the final manifest with "unknown blob" after
        # every layer reports Pushed (registry race / overwrite of the same
        # version tag). Retries are idempotent for identical local bytes.
        run: |
          set -euo pipefail
          image="ghcr.io/${{ github.repository }}:${{ needs.prepare.outputs.version }}-${{ matrix.arch }}"
          docker tag selfpost:e2e "$image"
          attempt=1
          max=4
          backoff=15
          while true; do
            set +e
            out=$(docker push "$image" 2>&1)
            rc=$?
            set -e
            if [ "$rc" -eq 0 ]; then
              printf '%s\n' "$out"
              exit 0
            fi
            printf '%s\n' "$out" >&2
            if [ "$attempt" -ge "$max" ] || ! grep -qiE 'unknown blob|blob unknown|blob upload invalid|manifest unknown|received unexpected HTTP status: 5[0-9]{2}|429 Too Many Requests|temporarily unavailable' <<<"$out"; then
              echo "::error::docker push failed for ${image} (attempt ${attempt}/${max}, exit ${rc})" >&2
              exit "$rc"
            fi
            echo "::warning::transient GHCR error pushing ${image} (attempt ${attempt}/${max}); retry in ${backoff}s" >&2
            sleep "$backoff"
            attempt=$((attempt + 1))
            backoff=$((backoff * 2))
          done

  merge:
    needs: [prepare, build]
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          ref: v${{ needs.prepare.outputs.version }}

      - name: Debug — snapshot GitHub Release state (agent)
        env:
          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          VERSION: ${{ needs.prepare.outputs.version }}
        run: |
          #region agent log
          set -euo pipefail
          tag="v${VERSION}"
          echo "::group::agent-log release snapshot for ${tag}"
          if out=$(gh release view "$tag" --repo "${{ github.repository }}" --json isDraft,tagName,publishedAt,url 2>&1); then
            echo "release-exists: ${out}"
          else
            echo "release-missing: ${out}"
          fi
          echo "event=${{ github.event_name }} ref=${GITHUB_REF_NAME}"
          echo "::endgroup::"
          #endregion

      - uses: docker/setup-buildx-action@v3

      - name: Log in to ghcr.io
        uses: docker/login-action@v3
        with:
          registry: ghcr.io
          username: ${{ github.actor }}
          password: ${{ secrets.GITHUB_TOKEN }}

      - name: Merge per-arch tags into the version manifest
        # Same GHCR "unknown blob" class of failure can hit imagetools create
        # when assembling the multi-arch version tag.
        run: |
          set -euo pipefail
          version_tag="ghcr.io/${{ github.repository }}:${{ needs.prepare.outputs.version }}"
          amd64="ghcr.io/${{ github.repository }}:${{ needs.prepare.outputs.version }}-amd64"
          arm64="ghcr.io/${{ github.repository }}:${{ needs.prepare.outputs.version }}-arm64"
          attempt=1
          max=4
          backoff=15
          while true; do
            set +e
            out=$(docker buildx imagetools create -t "$version_tag" "$amd64" "$arm64" 2>&1)
            rc=$?
            set -e
            if [ "$rc" -eq 0 ]; then
              printf '%s\n' "$out"
              #region agent log
              echo "::group::agent-log ghcr tags published"
              echo "version=${{ needs.prepare.outputs.version }} event=${{ github.event_name }} ref=${GITHUB_REF_NAME}"
              echo "tags_pushed=${{ needs.prepare.outputs.version }},${{ needs.prepare.outputs.version }}-amd64,${{ needs.prepare.outputs.version }}-arm64"
              echo "::endgroup::"
              #endregion
              exit 0
            fi
            printf '%s\n' "$out" >&2
            if [ "$attempt" -ge "$max" ] || ! grep -qiE 'unknown blob|blob unknown|blob upload invalid|manifest unknown|received unexpected HTTP status: 5[0-9]{2}|429 Too Many Requests|temporarily unavailable' <<<"$out"; then
              echo "::error::imagetools create failed for ${version_tag} (attempt ${attempt}/${max}, exit ${rc})" >&2
              exit "$rc"
            fi
            echo "::warning::transient GHCR error merging ${version_tag} (attempt ${attempt}/${max}); retry in ${backoff}s" >&2
            sleep "$backoff"
            attempt=$((attempt + 1))
            backoff=$((backoff * 2))
          done