SDK 10-minute tutorial¶

Status: Available in ETLantic 0.41.0. Secondary path — complete the CLI Quickstart (Ada/Grace in data/out.json) first. This page is for SDK muscle memory after that green path. No clone required.

Use the curated root (import etlantic as etl) for a tiny in-process pipeline.

Install¶

Same install as Installation:

python -m pip install 'etlantic==0.41.0'

Author, validate, plan, run¶

import etlantic as etl


class Row(etl.Data):
    id: int
    name: str


class Identity(etl.Transformation):
    rows: etl.Input[Row]
    result: etl.Output[Row]


@Identity.implementation("local")
def identity_local(rows: list[Row]) -> list[Row]:
    return list(rows)


class Demo(etl.Pipeline):
    raw: etl.Extract[Row] = etl.Extract(asset="rows")
    step = Identity.step(rows=raw)
    out: etl.Load[Row] = etl.Load(input=step.result, asset="out")


profile = etl.Profile(name="sdk-demo", assets={"rows": "memory", "out": "memory"})
runtime = etl.PipelineRuntime()
runtime.memory.seed(
    "rows",
    [Row(id=1, name="Ada"), Row(id=2, name="Grace")],
)

Demo.validate(profile=profile).raise_for_errors()
Demo.plan(profile=profile)
run = Demo.run(profile=profile, runtime=runtime)
assert run.status.value == "succeeded"
print(runtime.memory.get("out"))

Notes¶

Curated symbols match _CURATED — see API reference.
Primary path remains CLI init → validate → run. Use this page only after first file-backed success.
Next: First Pipeline (if you skipped it), Engine selection.