DFE Schemas

Shared schema definitions for the Data Fusion Engine (DFE) platform - the single source of truth for the data-structure definitions that multiple DFE components must agree on (dfe-engine, dfe-loader, dfe-receiver, dfe-archiver).

Usage

Install the Python package. The schema trees ship as package data under dfe_schemas/data/, alongside the dfe_schemas.clickhouse engine resolver and the declared deploy defaults:

pip install dfe-schemas
python -c "import dfe_schemas; print(dfe_schemas.schemas_root())"

Point DFE_SCHEMAS_DIR at your own directory to override the shipped trees:

export DFE_SCHEMAS_DIR=/opt/dfe/schemas

Validate locally (needs dfe-engine importable - point PY at an interpreter that has it):

make validate PY=../dfe-engine/.venv/bin/python

Structure

dfe-schemas/
|-- common-header/     # header profiles: timeseries (9 col, default),
|                      #   minimal (5 col), passthrough (4 col)
|-- meta/              # source meta schemas, by provider (aws/ azure/ gcp/ m365/)
|-- additional/        # extra-field overlays (aws/)
|-- hunts/             # hunt output (results.yaml) + runner checkpoint schema
|-- tables/            # tables in exact ClickHouse types: otel/ + engine internal/
|-- sources/           # engine-owned source definitions (main, the landing table)
|-- scripts/           # validate_schemas / annotate_meta_schemas
|-- docs/meta-schema.md  # the YAML format reference (version tree, columns, types)
|-- docs/tables.md     # the tables/ format reference (clauses, exact CH columns)
'-- Makefile           # validate

Every schema YAML carries its own version tree - each version entry is a complete column snapshot with SchemaVer semantics (model / addition / revision), published versions are immutable, and consumers pin versions independently. Full format reference, column fields, the 13-primitive type system, and the @directive expression language: docs/meta-schema.md.

How these reach ClickHouse

One applier, three callers. The dfe-schema entry point in the dfe-engine image reads this tree and reconciles ClickHouse against it: dfe-infra runs it as a Job before the data plane starts, dfe-docker runs it as a compose init service, and the engine repeats it idempotently at boot. Rendering SQL here and applying it separately was a SECOND definition of the same tables, so it is gone -- there is one DDL path and it reads these files.

Consumers

Rust services slave from the DEPLOYED ClickHouse schema at runtime (system.columns) - they never read this YAML directly.

Updating schemas

1. Branch here, add a NEW version entry (complete column snapshot - never modify a published version), update current, commit, PR to main.
2. Cut a release (CI workflow_dispatch, from-head=true) so the wheel carrying the new trees reaches PyPI.
3. Raise the dfe-schemas floor in each consumer and relock (dfe-engine: pyproject.toml + uv.lock).
4. Copy changed common-header profiles to the consumers' bundled fallback locations (dfe-engine: src/dfe_engine/schema/profiles/), which is what answers when neither DFE_SCHEMAS_DIR nor the package is available.

Shipped files here are read-only defaults - customise by pointing DFE_SCHEMAS_DIR at your own directory with only the profiles you override.