v0.24.0
Migration-path fix release on top of 0.23.0.
Highlights
chapkit mlproject migrate: Dockerfile now installs chapkit viauv sync(#53). After the chapkit-images split into base +-clivariants, thechapkit-{py,r,r-inla}:latesttags no longer shipchapkitpre-installed - migrated projects need to install it themselves. The generated Dockerfile now mirrors thechapkit initscaffold:uv sync --frozen --no-dev --no-install-projectagainst the generatedpyproject.toml+uv.lock.ARG BASE_PLATFORM=linux/amd64+FROM --platform=...forchapkit-r-inla(INLA is amd64-only);chapkit-py/chapkit-rstay multi-arch.CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]instead ofpython main.py, so server flags are visible in the Dockerfile.- Floor-pin in the generated
pyproject.tomlbumped tochapkit>=0.23.0.
make testfor migrated projects now goes through Docker (#53).make testbuilds the image, runs the container, driveschapkit test --urlagainst it, then tears down. Required for R models (the host doesn't have R + the renv packages) and exercises the same image that ships downstream for Python ones too.- Scaffolded Dockerfile uses
BASE_PLATFORMbuild arg (#52). Apple Silicon / arm64 hosts can override--build-arg BASE_PLATFORM=linux/arm64forchapkit-rinstead of being pinned to amd64. - chapkit-images base/
-clirename documented across CHAPKIT.md / READMEs (#51). bench-memoryscript for scaffolded templates, with progress reporting andKeyboardInterrupthandling. Ports the previous bash version to Python and surfaces results viarich.Table.
Compatibility
- No breaking API changes.
- Migrated projects from 0.22.x / 0.23.x re-migrate cleanly. Existing migrated projects can adopt the new Dockerfile by copying the updated
Dockerfile_migrate.jinja2output (or re-runningchapkit mlproject migrateagainst the original MLproject).