storywrangler-schemas 0.2.0

Shared Pydantic schemas for the Storywrangler platform

storywrangler-schemas

Shared Pydantic schemas for the Storywrangler platform. Both the backend API and the storywrangler SDK import from this package — neither owns a copy of the models.

Implements the Storywrangler Specification v0.0.3.

Installation

pip install storywrangler-schemas

Most users should install storywrangler instead, which includes this package as a dependency.

What's inside

Registry models (storywrangler_schemas.registry)

Model	Purpose
DatasetCreate	Full registration payload (domain, format, transform, entity mapping, lineage, ...)
EndpointSchemaConfig	Output shape declaration (types-counts, time-series)
TransformConfig	Query slice axes (time dimension, filter dimensions, hash bucket)
EntityMappingConfig	Entity ID resolution config (local column + namespace)
EntityRow	One row in the entity mapping table
ManifestConfig	Coverage index (availability, partition index)
OwnershipConfig	Owner group, contact, lifecycle status
LineageConfig	Sources, upstream datasets, pipeline repo, archival DOI

Hash bucket assignment (storywrangler_schemas.hashing)

Canonical murmur3_32 hash function shared between the backend query layer and data pipelines:

from storywrangler_schemas.hashing import assign_bucket

bucket = assign_bucket("hello", num_buckets=16)  # deterministic int in [0, 16)

Standards (storywrangler_schemas.standards)

Entity ID validation, namespace registry, and spec URL helpers.