connect.external.databricks

connect.external.databricks

Databricks SDK integration.

Databricks SDK credentials implementations which support interacting with Posit OAuth integrations on Connect.

Notes

These APIs are provided as a convenience and are subject to breaking changes: https://github.com/databricks/databricks-sdk-py#interface-stability

Attributes

Name	Description
POSIT_OAUTH_INTEGRATION_AUTH_TYPE
logger

Classes

Name	Description
ConnectStrategy	A CredentialsStrategy implementation which supports content running on Posit Connect.

ConnectStrategy

connect.external.databricks.ConnectStrategy(
    self
    client=None
    user_session_token=None
)

A CredentialsStrategy implementation which supports content running on Posit Connect.

This strategy can be used as a valid credentials_strategy when constructing a databricks.sdk.core.Config.

It should be used when content running on a Posit Connect server needs to access a Databricks token that is managed by a Posit Connect OAuth integration. This strategy is only valid when running on Posit Connect. If you need to author content that can run in multiple environments (local content, Posit Workbench, and Posit Connect), consider using the databricks_config() helper method in this module.

This strategy can be used for both Viewer authentication and Service Account authentication (sometimes referred to as Content credentials). When user_session_token is provided then Viewer authentication is used, otherwise the strategy attempts to fall back to Service Account authentication by reading the CONNECT_CONTENT_SESSION_TOKEN environment var.

See Also

• https://docs.posit.co/connect/admin/integrations/oauth-integrations/databricks/

Examples

This example shows how to do Viewer authentication with a Viewer Databricks OAuth integration.

import os

from databricks.sdk.core import ApiClient, Config
from databricks.sdk.service.iam import CurrentUserAPI
from shiny import reactive
from shiny.express import render, session

from posit.connect.external.databricks import ConnectStrategy


@reactive.calc
def cfg():
    session_token = session.http_conn.headers.get("Posit-Connect-User-Session-Token")
    return Config(
        credentials_strategy=ConnectStrategy(user_session_token=session_token),
        host=os.getenv("DATABRICKS_HOST"),
    )


@render.text
def text():
    current_user_api = CurrentUserAPI(ApiClient(cfg()))
    databricks_user_info = current_user_api.me()
    return f"Hello, {databricks_user_info.display_name}!"

This example shows how to do Service Account authentication with a Service Account Databricks OAuth integration.

import os

from databricks.sdk.core import ApiClient, Config
from databricks.sdk.service.iam import CurrentUserAPI
from shiny import reactive
from shiny.express import render

from posit.connect.external.databricks import ConnectStrategy


@reactive.calc
def cfg():
    return Config(
        credentials_strategy=ConnectStrategy(),
        host=os.getenv("DATABRICKS_HOST"),
    )


@render.text
def text():
    current_user_api = CurrentUserAPI(ApiClient(cfg()))
    databricks_user_info = current_user_api.me()
    return f"Hello, {databricks_user_info.display_name}!"

Methods

Name	Description
auth_type

auth_type

connect.external.databricks.ConnectStrategy.auth_type()

Functions

Name	Description
databricks_config	A helper to aid with constructing a valid databricks.sdk.core.Config object.
sql_credentials	A helper method for making a databricks.sdk.core.Config object compatible with databricks.sql.connect.

databricks_config

connect.external.databricks.databricks_config(
    posit_default_strategy=None
    posit_workbench_strategy=None
    posit_connect_strategy=None
    **kwargs
)

A helper to aid with constructing a valid databricks.sdk.core.Config object.

The returned Config is configured with a credentials_strategy which should be compatible with the current execution environment. This is useful when authoring content that needs to run in multiple environments without changing code. This constructor allows you mix-and-match Databricks SDK credentials_strategies. For example, content executing on a developer laptop can use the databricks_cli strategy, while in Workbench the same content can use the pat_auth strategy, and in Connect the content uses the custom ConnectStrategy() with a Databricks OAuth integration.

Parameters

Name	Type	Description	Default
posit_default_strategy	databricks.sdk.credentials_provider.CredentialsStrategy	The default credentials_strategy to use when neither Posit Connect nor Posit Workbench are detected	None
posit_workbench_strategy	databricks.sdk.credentials_provider.CredentialsStrategy	The credentials_strategy to use when Posit Workbench is detected	None
posit_connect_strategy	databricks.sdk.credentials_provider.CredentialsStrategy	The credentials_strategy to use when Posit Connect is detected	None
kwargs	Dict[str, Any]	Additional keyword arguments. kwargs is passed directly to the databricks.sdk.core.Config constructor.	{}

Returns

Name	Type	Description
	databricks.sdk.core.Config

See Also

• https://docs.posit.co/connect/admin/integrations/oauth-integrations/databricks/

Examples

This example shows how to construct a databricks Config that is compatible with: - Databricks CLI authentication for local development - Workbench-managed Databricks Credentials in Posit Workbench - Viewer OAuth integration authentication in Posit Connect

import os

from databricks.sdk.core import ApiClient, databricks_cli
from databricks.sdk.service.iam import CurrentUserAPI
from shiny import reactive
from shiny.express import render, session

from posit.connect.external.databricks import (
    ConnectStrategy,
    databricks_config,
)
from posit.workbench.external.databricks import WorkbenchStrategy


@reactive.calc
def cfg():
    session_token = session.http_conn.headers.get("Posit-Connect-User-Session-Token")
    return databricks_config(
        posit_default_strategy=databricks_cli,
        posit_workbench_strategy=WorkbenchStrategy(),
        posit_connect_strategy=ConnectStrategy(user_session_token=session_token),
        host=os.getenv("DATABRICKS_HOST"),
    )


@render.text
def text():
    current_user_api = CurrentUserAPI(ApiClient(cfg()))
    databricks_user_info = current_user_api.me()
    return f"Hello, {databricks_user_info.display_name}!"

This example shows how to construct a databricks Config that is compatible with: - OAuth Service Principal (oauth-m2m) in Posit Workbench - Viewer OAuth integration authentication in Posit Connect

This examples uses environment variables to load the correct configurations at runtime. Make sure to set the DATABRICKS_CLIENT_ID and DATABRICKS_CLIENT_SECRET as environment variables in Posit Workbench, but do not set them inside of Posit Connect.

import os

from databricks.sdk.core import ApiClient, oauth_service_principal
from databricks.sdk.service.iam import CurrentUserAPI
from shiny import reactive
from shiny.express import render, session

from posit.connect.external.databricks import (
    ConnectStrategy,
    databricks_config,
)


@reactive.calc
def cfg():
    session_token = session.http_conn.headers.get("Posit-Connect-User-Session-Token")
    return databricks_config(
        posit_workbench_strategy=oauth_service_principal,
        posit_connect_strategy=ConnectStrategy(user_session_token=session_token),
        host=os.getenv("DATABRICKS_HOST"),
        client_id=os.getenv("DATABRICKS_CLIENT_ID"),
        client_secret=os.getenv("DATABRICKS_CLIENT_SECRET"),
    )


@render.text
def text():
    current_user_api = CurrentUserAPI(ApiClient(cfg()))
    databricks_user_info = current_user_api.me()
    return f"Hello, {databricks_user_info.display_name}!"

This example shows how to construct a databricks Config that is compatible with: - Azure Service Principal credentials in Posit Workbench - Service Account OAuth integration authentication in Posit Connect

This examples uses environment variables to load the correct configurations at runtime. Make sure to set the ARM_CLIENT_ID, ARM_CLIENT_SECRET, and ARM_TENANT_ID as environment variables in Posit Workbench, but do not set them inside of Posit Connect.

import os

from databricks.sdk.core import ApiClient, azure_service_principal
from databricks.sdk.service.iam import CurrentUserAPI
from shiny import reactive
from shiny.express import render

from posit.connect.external.databricks import (
    ConnectStrategy,
    databricks_config,
)


@reactive.calc
def cfg():
    return databricks_config(
        posit_workbench_strategy=azure_service_principal,
        posit_connect_strategy=ConnectStrategy(),
        host=os.getenv("DATABRICKS_HOST"),
        azure_client_id=os.getenv("ARM_CLIENT_ID"),
        azure_client_secret=os.getenv("ARM_CLIENT_SECRET"),
        azure_tenant_id=os.getenv("ARM_TENANT_ID"),
    )


@render.text
def text():
    current_user_api = CurrentUserAPI(ApiClient(cfg()))
    databricks_user_info = current_user_api.me()
    return f"Hello, {databricks_user_info.display_name}!"

sql_credentials

connect.external.databricks.sql_credentials(cfg)

A helper method for making a databricks.sdk.core.Config object compatible with databricks.sql.connect.

The SQL client’s ExternalAuthProvider is not compatible w/ the SDK’s implementation of CredentialsProvider, so create a no-arg lambda that wraps the args defined by the real caller. This way we can pass in a databricks Config object required by most of the SDK’s CredentialsProvider implementations from where sql.connect is called.

See Also

• https://github.com/databricks/databricks-sql-python/issues/148#issuecomment-2271561365

Examples

import os

from databricks import sql
from databricks.sdk.core import ApiClient, Config
from databricks.sdk.service.iam import CurrentUserAPI

from posit.connect.external.databricks import sql_credentials

cfg = Config(
    host=os.getenv("DATABRICKS_HOST"),
    warehouse_id=os.getenv("DATABRICKS_WAREHOUSE_ID"),
)

with sql.connect(
    server_hostname=cfg.host,
    http_path=cfg.sql_http_path,
    credentials_provider=sql_credentials(cfg),
) as connection:
    with connection.cursor() as cursor:
        cursor.execute("SELECT * FROM samples.nyctaxi.trips LIMIT 10;")
        print(cursor.fetchall())