Getting started¶

Install¶

The package requires Python 3.11 or newer.

pip install biomero-schema

For development, clone the repository and install it in editable mode:

pip install -e .

The repository also provides Pixi tasks:

pixi run test
pixi run json-schema

Validate a normalized workflow descriptor¶

The CLI accepts the native WorkflowSchema representation. The biomero runtime performs BIAFLOWS and BILAYERS conversion before this validation step; workflow providers should validate their source file against its upstream format as well.

biomero-schema validate workflow.json

Parse it and show a summary:

biomero-schema parse workflow.json

Use --pretty for the complete parsed representation or --json for JSON.

Use a cross-service contract¶

Consumers should import the Pydantic model rather than copying its fields:

from biomero_schema.zarr import CanonicalZarrSource

source = CanonicalZarrSource.from_dict(payload)
wire_payload = source.to_dict()  # stable camelCase representation
json_schema = CanonicalZarrSource.model_json_schema()

Models reject unknown fields and are immutable after construction. This makes contract drift visible at the boundary where it occurs.