publish-mock-server

Publish Mock Server

Build and publish the mock-server Docker image to ghcr.io/vfarcic/dot-ai-mock-server:latest.

Prerequisites

• Docker is installed and running (use sudo docker if your user is not in the docker group)
• Authenticated to ghcr.io. The simplest path is to reuse the gh token (needs write:packages scope):

  gh auth token | docker login ghcr.io -u vfarcic --password-stdin
  

Steps

Run all commands from the repository root. The build context is the mock-server/ directory (relative — do NOT hardcode an absolute path).

1. Set up a multi-arch builder (one-time per machine)

   The default docker driver cannot push a multi-platform manifest, and building arm64 on an amd64 host needs QEMU. Register the emulator and create a container-driver builder:

   docker run --privileged --rm tonistiigi/binfmt --install arm64
   docker buildx create --name mockbuilder --driver docker-container --use --bootstrap
   

2. Build and push multi-arch Docker image

   Tag both latest and the current dot-ai release version (so downstream repos can pin to a contract version instead of a digest). Derive the version from the root package.json:

   VERSION=$(node -p "require('./package.json').version")
   docker buildx build --builder mockbuilder \
     --platform linux/amd64,linux/arm64 \
     -t ghcr.io/vfarcic/dot-ai-mock-server:latest \
     -t ghcr.io/vfarcic/dot-ai-mock-server:$VERSION \
     --push mock-server
   

3. Verify the push

   Confirm both tags resolve to the same digest and carry both platforms:

   docker buildx imagetools inspect ghcr.io/vfarcic/dot-ai-mock-server:latest
   

   Note the manifest-list digest (sha256:…) — downstream repos pin to it.

Usage

After publishing, consumers (like dot-ai-ui) can use:

services:
  mock-api:
    image: ghcr.io/vfarcic/dot-ai-mock-server:latest
    ports:
      - "3001:3001"

When to Publish