epycloud build¶

Build Docker images for pipeline execution.

Usage¶

epycloud build [MODE] [OPTIONS]

Description¶

Creates Docker images containing the pipeline execution environment. Supports three build modes:

cloud - Asynchronous builds using Google Cloud Build (production)
local - Local builds with push to Artifact Registry
dev - Local-only development builds (no push)

Build Modes¶

cloud¶

Build using Google Cloud Build service (default, asynchronous).

epycloud build cloud [CONTEXT] [OPTIONS]
epycloud build  # cloud is default

Arguments: - CONTEXT - Build context directory (default: project root)

Options: - --cache - Enable build cache (disabled by default) - --tag TAG - Custom image tag - --wait - Wait for build completion - -f DOCKERFILE, --file DOCKERFILE - Path to Dockerfile (default: docker/Dockerfile)

Returns build ID immediately. Monitor with epycloud build status.

local¶

Build locally using Docker and push to Artifact Registry.

epycloud build local [CONTEXT] [OPTIONS]

Arguments: - CONTEXT - Build context directory (default: docker/)

Options: - --no-cache - Disable build cache - --tag TAG - Custom image tag - --no-push - Don't push to registry - -f DOCKERFILE, --file DOCKERFILE - Path to Dockerfile (default: docker/Dockerfile)

Synchronous build using local Docker daemon.

dev¶

Build locally for development (no push by default).

epycloud build dev [CONTEXT] [OPTIONS]

Arguments: - CONTEXT - Build context directory (default: docker/)

Options: - --cache - Enable build cache (disabled by default) - --tag TAG - Custom image tag - --push - Push to registry (optional) - -f DOCKERFILE, --file DOCKERFILE - Path to Dockerfile (default: docker/Dockerfile)

Fast iteration for local testing.

Build Status¶

Check build status:

epycloud build status [OPTIONS]

Options: - --limit LIMIT - Maximum number of builds to display (default: 10) - --ongoing - Show only active builds (QUEUED, WORKING)

Using Environments¶

The global --env flag selects an environment configuration that overrides build settings like image_tag and modeling_suite_ref. This is the standard way to build images for different branches or purposes.

Create an environment file at ~/.config/epymodelingsuite-cloud/environments/<name>.yaml:

# environments/feature1.yaml
docker:
  image_tag: feature1
github:
  modeling_suite_ref: feature1

Build with that environment:

epycloud --env feature1 build cloud    # Cloud build with feature1 settings
epycloud --env feature1 build dev      # Dev build with feature1 settings

This builds the Docker image using the feature1 branch of epymodelingsuite and tags the image as feature1. Without --env, the base config values are used.

See Environments for details on environment configuration.

Examples¶

# Cloud build (default, asynchronous)
epycloud build
epycloud build cloud

# Cloud build and wait for completion
epycloud build cloud --wait

# Cloud build with cache enabled
epycloud build cloud --cache

# Cloud build with custom tag
epycloud build cloud --tag v1.2.3

# Cloud build with custom Dockerfile
epycloud build cloud -f custom/Dockerfile

# Local build and push
epycloud build local

# Local build without pushing
epycloud build local --no-push

# Dev build (local only, no push)
epycloud build dev

# Dev build with cache enabled
epycloud build dev --cache

# Build with environment overrides
epycloud --env feature1 build cloud
epycloud --env prod build cloud
epycloud --env dev build dev

# Check build status
epycloud build status
epycloud build status --limit 20
epycloud build status --ongoing

Exit Codes¶

Code	Description
`0`	Success
`1`	Build failed

epycloud run - Run workflows with built images
epycloud config - Configure Docker and Artifact Registry settings