medallantic 0.36.0

Engine-agnostic medallion facade on ETLantic (native authoring + SparkForge IR migrate).

Medallantic

Engine-agnostic medallion facade on ETLantic: native bronze/silver/gold authoring plus a SparkForge IR migrate path. Prefer medallantic over the deprecated etlantic-sparkforge redirect.

Medallantic owns bronze/silver/gold authoring conventions; ETLantic owns the portable contracts, graph, validation, planning, execution lifecycle, and plugin coordination underneath it. ETLantic core never gains medallion types.

The current 0.36 line includes native authoring, portable quality gates, importable transform execution, lifecycle/write semantics, PySpark and Delta differential coverage, SQLAlchemy relational parity, observability providers, durable run history, event consumers, production conformance, and the completed M7 migration plus joint compatibility burn-in. Native PySpark Column and Moltres rules remain explicitly capability-gated.

Documentation:

• Documentation index
• Getting started
• SparkForge migration
• Compatibility
• Architecture

Install

pip install medallantic

Native authoring

from medallantic import MedallionBuilder

defn = (
    MedallionBuilder("ecommerce", schema="demo")
    .bronze("orders", asset="bronze_orders")
    .silver("clean", source="orders", asset="silver_orders")
    .gold("kpis", source="clean", asset="gold_kpis", write_mode="merge")
    .build()
)

Migration bridges

Feed SparkForgePipelineSpec (JSON/YAML fixtures or hand-built dataclasses) via adapt_pipeline or medallantic.migrate.sparkforge. Existing live builders can use:

from medallantic.migrate.sparkforge import from_pipeline_builder
from medallantic.migrate.sql import from_sql_pipeline_builder

The adapted result supplies an ordinary ETLantic pipeline and profile; select execution plugins such as Profile.spark_engine="pyspark" separately. Production profiles must allowlist every trusted execution plugin.

Links

Documentation · Source · Issues