Snowflake Native App

The Orchestrate SDK is available as a Snowflake Native App for use in the Snowflake Ecosystem. This allows you to leverage the Orchestrate APIs in your Snowflake environment.

Installation

To install it in your Snowflake account, search for “Orchestrate” in the Snowflake Marketplace. In order to interface with the Orchestrate API, you will need to create a secret for your API key and network integrations.

For specific installation instructions for the version you’re installing, refer to package documentation in the marketplace.

Function Signatures

Functions within the Native App follow the exact same signatures as those in the Python SDK.

For example, the terminology.standardize_medication function in the Python SDK has the same signature as the terminology.standardize_medication function in the Snowflake Native App.

In the Python SDK:

from orchestrate import OrchestrateApi
import os

os.environ["ORCHESTRATE_API_KEY"] = "<your-api-key>"
api = OrchestrateApi()

api.terminology.standardize_medication("861004", "RxNorm")

Because the display argument is optional, it can be omitted in Python. In the Snowflake Native App, it must be explicitly null:

select orchestrate.terminology.standardize_medication('861004', 'RxNorm', null);

Type Differences

The Native App uses the same function names as the Python SDK, but request and response types expecting a dict or Bundle in Python expect a string in the Native App. This is because the Native App uses SQL to interact with the Orchestrate API.

For example, the convert.fhir_stu3_to_fhir_r4 function in the Python SDK:

api.convert.fhir_stu3_to_fhir_r4({"resourceType": "Bundle", "type": "searchset", ... })

select orchestrate.convert.fhir_stu3_to_fhir_r4('{"resourceType": "Bundle", "type": "searchset", ... }');

OMOP Differences

When invoking convert.fhir_r4_to_omop in the Native App, the response is a table function containing three columns:

• file_name: Name of the CSV file in the response zip.
• row_type: data or header. Indicates whether this row is a header row in the CSV or a data row.
• content: The content of the CSV row.