Implementation:Datahub project Datahub Pip Install Acryl Datahub

Field	Value
Implementation Name	Pip Install Acryl Datahub
Overview	Concrete tool for installing the DataHub CLI and Python SDK via pip, including optional connector extras.
Type	External Tool Doc
Implements	Datahub_project_Datahub_CLI_Package_Installation
Status	Active
Domains	Data_Integration, Metadata_Management
Source	DataHub Repository -- metadata-ingestion/setup.py (lines 954-1108 for entry points, lines 1111-1149 for package setup)
Last Updated	2026-02-10
Knowledge Sources	DataHub Repository

Description

The acryl-datahub package is the primary distribution artifact for the DataHub CLI and Python ingestion framework. It is installed via pip and provides the datahub command-line tool along with the full SDK for programmatic metadata emission.

The package uses a modular extras system where each data source connector, sink, or feature is declared as an optional dependency group. This allows users to install only the libraries required for their specific use case.

Installation

Basic Installation

pip install acryl-datahub

Installation with a Connector Extra

# Install with Snowflake connector
pip install 'acryl-datahub[snowflake]'

# Install with BigQuery connector
pip install 'acryl-datahub[bigquery]'

# Install with multiple connectors
pip install 'acryl-datahub[snowflake,bigquery,mysql]'

Entry Point

The CLI is registered as a console script entry point in setup.py:

entry_points = {
    "console_scripts": ["datahub = datahub.entrypoints:main"],
    ...
}

After installation, the datahub command is available on the system PATH.

Key Parameters

Parameter	Type	Description
connector extra	string	Name of the optional connector to install (e.g., snowflake, bigquery, mysql, kafka, postgres, hive, looker, tableau, dbt, etc.)

I/O Contract

Inputs

• Python >= 3.10 environment
• pip package manager
• Network access to PyPI (or a configured private index)

Outputs

• datahub CLI command available on PATH
• Python package datahub importable in Python code
• All plugin entry points registered for the installed extras

Available Connector Extras (Partial List)

Extra Name	Source Class
snowflake	datahub.ingestion.source.snowflake.snowflake_v2:SnowflakeV2Source
bigquery	datahub.ingestion.source.bigquery_v2.bigquery:BigqueryV2Source
mysql	datahub.ingestion.source.sql.mysql:MySQLSource
postgres	datahub.ingestion.source.sql.postgres:PostgresSource
kafka	datahub.ingestion.source.kafka.kafka:KafkaSource
hive	datahub.ingestion.source.sql.hive.hive_source:HiveSource
looker	datahub.ingestion.source.looker.looker_source:LookerDashboardSource
tableau	datahub.ingestion.source.tableau.tableau:TableauSource
dbt	datahub.ingestion.source.dbt.dbt_core:DBTCoreSource
redshift	datahub.ingestion.source.redshift.redshift:RedshiftSource

Available Sink Plugins

Sink Name	Sink Class
datahub-rest	datahub.ingestion.sink.datahub_rest:DatahubRestSink
datahub-kafka	datahub.ingestion.sink.datahub_kafka:DatahubKafkaSink
file	datahub.ingestion.sink.file:FileSink
console	datahub.ingestion.sink.console:ConsoleSink

Package Metadata

From setup.py (lines 1111-1149):

setuptools.setup(
    name=package_metadata["__package_name__"],
    version=_version,
    url="https://docs.datahub.com/",
    license="Apache-2.0",
    description="A CLI to work with DataHub metadata",
    python_requires=">=3.10",
    package_dir={"": "src"},
    packages=setuptools.find_namespace_packages(where="./src"),
    ...
)

Usage Examples

Verify Installation

datahub version

Check Available Plugins

datahub check plugins

Related Pages

• Implements: Datahub_project_Datahub_CLI_Package_Installation
• Related: Datahub_project_Datahub_PipelineConfig
• Related: Datahub_project_Datahub_Ingest_CLI_Run
• Environment: Environment:Datahub_project_Datahub_Python_3_10_Ingestion_Environment
• Heuristic: Heuristic:Datahub_project_Datahub_Gradle_Formatting_Over_Direct_Tools