dagster-rocky

Dagster integration for Rocky, the typed graph between your code and your warehouse.

dagster-rocky wraps the rocky CLI as a Dagster ConfigurableResource and exposes Rocky-managed tables as materializable Dagster assets. Compile-time contracts, column-level lineage, schema-drift detection, quality check results, and per-model cost surface as native Dagster events, so the guarantees Rocky enforces at compile time appear directly in the Dagster asset graph.

Two behaviors worth knowing. Rocky's per-check severity maps to Dagster's, so a failing advisory check emits WARN instead of paging anyone. And when the engine's failure containment is enabled, the models Rocky withholds behind a failed upstream appear as AssetObservation events naming the blocking model, so a partial run reads honestly in the asset graph.

Install

uv add dagster-rocky

You'll also need the Rocky CLI on your $PATH: https://github.com/rocky-data/rocky/releases?q=engine

Quick start (component)

Add a defs.yaml next to your other Dagster definitions:

type: dagster_rocky.RockyComponent
attributes:
  binary_path: rocky
  config_path: config/rocky.toml
  models_dir: models

Dagster's component loader will:

Run rocky discover (and rocky compile, when models are present) and cache the result.
Build one subset-aware multi_asset per Rocky group, with declared row_count / column_match / freshness checks per table.
On materialization, shell out to rocky run --filter <key>=<value> for the selected subset and yield MaterializeResult + AssetCheckResult events with rich metadata.

Quick start (resource)

import dagster as dg
from dagster_rocky import RockyResource

rocky = RockyResource(
    binary_path="rocky",
    config_path="config/rocky.toml",
    timeout_seconds=3600,
)

defs = dg.Definitions(resources={"rocky": rocky})

Then in an asset:

@dg.asset
def acme_orders(rocky: RockyResource) -> dg.MaterializeResult:
    result = rocky.run(filter="tenant=acme")
    return dg.MaterializeResult(
        metadata={"tables_copied": result.tables_copied, "duration_ms": result.duration_ms},
    )

Public API

Symbol	Purpose
`RockyResource`	`ConfigurableResource` wrapping the Rocky CLI
`RockyComponent`	State-backed Dagster component that loads Rocky tables as assets
`RockyDagsterTranslator`	Subclass to customize asset key / group / tag mapping
`RockyMetadataSet`	Namespaced metadata (`source_id`, `strategy`, `watermark`, …)
`load_rocky_assets()`	Functional helper that returns `AssetSpec` for each Rocky table
`emit_materializations()` / `emit_check_results()`	Convert a `RunResult` into Dagster events
`check_metadata()`	Build a metadata mapping for a single Rocky `CheckResult`
`cost_metadata_from_optimize()`	Extract per-model cost recommendations from `OptimizeResult`
`parse_rocky_output()`	Auto-detect a Rocky JSON payload and return the matching Pydantic model

Documentation

Dagster Integration docs: resource, component, translator, schedules, sensors, pipes, and more
DEVELOPMENT.md: local setup, architecture, testing
CHANGELOG.md: release notes

Related projects

Rocky: the Rust SQL transformation engine
Rocky VS Code extension: VS Code extension with LSP and AI features

License

Apache 2.0

Release files for dagster-rocky 1.62.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for dagster-rocky 1.62.0
File	Size	Uploaded
dagster_rocky-1.62.0.tar.gz	375.1 kB	Aug 11, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for dagster-rocky 1.62.0
File	Interpreter	ABI	Platform	Reset
dagster_rocky-1.62.0-py3-none-any.whl 129.7 kB Aug 11, 2026	Python 3	none	any	Details

Total release size: 504.8 kB

Release files / dagster_rocky-1.62.0.tar.gz

Download URL	dagster_rocky-1.62.0.tar.gz
Size	375.1 kB
Tags	Source
SHA-256 checksum How to use checksums	`d2d30765cf5154c8bf1ff8a91e50bfd4bd1888385fff8ecbd005a04068d28f35`
BLAKE2b-256 checksum How to use checksums	`7454552bbaa274422843e0f6395fa10111b696b6f4689069487de2b02315ba6c`
Upload date	Aug 11, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/6.1.0 CPython/3.13.14`

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/rocky-data/rocky
Publishing commit: github.com/rocky-data/rocky/tree/96d95b0b2706cc62193c760d1b3981b8abb8f0e1

Workflow

Publishing configuration: github.com/rocky-data/rocky/blob/96d95b0b2706cc62193c760d1b3981b8abb8f0e1/.github/workflows/dagster-release.yml
Publishing logs: github.com/rocky-data/rocky/actions/runs/31542469835/attempts/1

Artifact

Subject: dagster_rocky-1.62.0.tar.gz
SHA-256 checksum: d2d30765cf5154c8bf1ff8a91e50bfd4bd1888385fff8ecbd005a04068d28f35 Verifying attestations

Release files / dagster_rocky-1.62.0-py3-none-any.whl

Download URL	dagster_rocky-1.62.0-py3-none-any.whl
Size	129.7 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`9d9704982e64e0e2a52897489ffb0c04f80fabe22f89f2a985f632d4a7c2f83c`
BLAKE2b-256 checksum How to use checksums	`b37b4152e1eadbf592fbfc088cf9970fb41ca95f7357daf9ed297352c43b26bf`
Upload date	Aug 11, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/6.1.0 CPython/3.13.14`

Provenance

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/rocky-data/rocky
Publishing commit: github.com/rocky-data/rocky/tree/96d95b0b2706cc62193c760d1b3981b8abb8f0e1

Workflow

Publishing configuration: github.com/rocky-data/rocky/blob/96d95b0b2706cc62193c760d1b3981b8abb8f0e1/.github/workflows/dagster-release.yml
Publishing logs: github.com/rocky-data/rocky/actions/runs/31542469835/attempts/1

Artifact

Subject: dagster_rocky-1.62.0-py3-none-any.whl
SHA-256 checksum: 9d9704982e64e0e2a52897489ffb0c04f80fabe22f89f2a985f632d4a7c2f83c Verifying attestations