mypy-boto3-sagemaker-featurestore-runtime Stubs

Warnings

• breaking Beginning with `mypy-boto3-builder` version 8.12.0 (which generates `mypy-boto3-*` packages, including this one, corresponding to `boto3` versions around 1.34.x and above), support for Python 3.8 has been removed. Projects using these stubs must upgrade to Python 3.9 or newer.
  Fix: Upgrade your Python environment to 3.9 or later. Update your `pyproject.toml` or `setup.cfg` to reflect `python_requires >=3.9`.
• breaking Stubs generated by `mypy-boto3-builder` version 8.12.0+ are now PEP 561 compliant, changing the package structure to include the `py.typed` marker. While largely transparent, this *may* affect custom `mypy` configurations or build systems that rely on specific stub discovery paths (e.g., if you manually managed `MYPYPATH`).
  Fix: Ensure `mypy` is up-to-date. If encountering issues, review `mypy`'s documentation on stub discovery and package layout. No changes are typically required for standard setups, as `mypy` handles PEP 561 packages automatically.
• gotcha `mypy-boto3-sagemaker-featurestore-runtime` provides type hints for `boto3`, but it is not a runtime library itself. Do not `import mypy_boto3_sagemaker_featurestore-runtime` in your application code for runtime execution. Its purpose is solely for static analysis tools like `mypy`.
  Fix: Only import symbols from the `mypy_boto3_sagemaker_featurestore_runtime` package for type annotations (e.g., in a `from typing import TYPE_CHECKING` block or as explicit type hints). Your application code should continue to `import boto3`.
• gotcha These stubs are a complement to `boto3`. You must still install `boto3` itself for your application to run. The `mypy-boto3-sagemaker-featurestore-runtime` package does not bundle or install `boto3` for you.
  Fix: Ensure `pip install boto3` is part of your project's runtime dependencies, in addition to installing these stubs for type checking.
• breaking In `mypy-boto3-builder` version 8.9.0+, some `TypedDict` argument types for API calls (e.g., `CreateDistributionRequestRequestTypeDef` from CloudFront) were shortened to `CreateDistributionRequestTypeDef`. Additionally, conflicting `TypedDict` `Extra` postfixes moved to the end (e.g., `CreateDistributionExtraRequestTypeDef` became `CreateDistributionRequestExtraTypeDef`). While this specific service may not be heavily impacted, it's a breaking change for the `mypy-boto3` ecosystem.
  Fix: Review your `TypedDict` usage and update names according to the latest stubs. `mypy` will guide you with type errors if you are using affected `TypedDict`s.

Install

• pip install mypy-boto3-sagemaker-featurestore-runtime Install stubs
• pip install boto3 mypy Install runtime and type checker

Imports

• SagemakerFeatureStoreRuntimeClient

  from mypy_boto3_sagemaker_featurestore_runtime.client import SagemakerFeatureStoreRuntimeClient

  The primary client type for type hints. The `boto3.client` import is for runtime, not type-checking the client instance itself.
• GetRecordRequestRequestTypeDef

  from mypy_boto3_sagemaker_featurestore_runtime.type_defs import GetRecordRequestRequestTypeDef

  Type definition for input parameters of the `get_record` API call.
• GetRecordResponseTypeDef

  from mypy_boto3_sagemaker_featurestore_runtime.type_defs import GetRecordResponseTypeDef

  Type definition for the response of the `get_record` API call.

Quickstart

This quickstart demonstrates how to use `mypy-boto3-sagemaker-featurestore-runtime` for type-checking a `boto3` client interaction. It defines a function to fetch a record from SageMaker Feature Store, explicitly typing the `boto3` client and the request/response payloads. This allows `mypy` to validate your API calls at static analysis time, catching potential errors before runtime.

import boto3
from mypy_boto3_sagemaker_featurestore_runtime.client import SagemakerFeatureStoreRuntimeClient
from mypy_boto3_sagemaker_featurestore_runtime.type_defs import GetRecordRequestRequestTypeDef, GetRecordResponseTypeDef
import os

def get_feature_store_record(
    feature_group_name: str,
    record_identifier_value_as_string: str
) -> GetRecordResponseTypeDef:
    """Fetches a record from SageMaker Feature Store with type hints."""
    # boto3.client returns an untyped client by default. 
    # The type hint on 'client' variable applies the stubs.
    client: SagemakerFeatureStoreRuntimeClient = boto3.client(
        "sagemaker-featurestore-runtime",
        region_name=os.environ.get("AWS_REGION", "us-east-1")
    )

    request_params: GetRecordRequestRequestTypeDef = {
        "FeatureGroupName": feature_group_name,
        "RecordIdentifierValueAsString": record_identifier_value_as_string,
        "FeatureNames": ["example_feature_name"]
    }

    response: GetRecordResponseTypeDef = client.get_record(**request_params)
    return response

# Example usage (ensure AWS credentials are set up)
if __name__ == "__main__":
    try:
        # Use environment variables for sensitive or dynamic values
        # For a runnable example, provide dummy values if env vars are not set
        example_feature_group_name = os.environ.get("SAGE_FEATURE_GROUP_NAME", "my-test-feature-group")
        example_record_id = os.environ.get("SAGE_RECORD_ID", "12345")

        if not os.environ.get("AWS_ACCESS_KEY_ID") or not os.environ.get("AWS_SECRET_ACCESS_KEY"):
            print("WARNING: AWS credentials not found in environment. Example might fail at runtime.")
            print("Please set AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY.")

        print(f"Attempting to get record for Feature Group: {example_feature_group_name}, Record ID: {example_record_id}")
        
        # This call is type-checked by mypy thanks to the stubs
        record_data = get_feature_store_record(example_feature_group_name, example_record_id)
        
        print("Successfully retrieved record (type-checked):")
        if "Record" in record_data:
            for feature in record_data["Record"]:
                print(f"  {feature['FeatureName']}: {feature['ValueAsString']}")
        else:
            print("Record key not found in response, or response is empty.")

    except Exception as e:
        print(f"An error occurred: {e}")
        print("Ensure AWS credentials are configured and the feature group/record exist/are accessible.")