mypy-boto3-bedrock-runtime Type Annotations

Warnings

• breaking Starting with `mypy-boto3-builder` version 8.12.0, support for Python 3.8 has been removed. All generated packages, including `mypy-boto3-bedrock-runtime`, now require Python 3.9 or newer.
  Fix: Upgrade your Python environment to 3.9 or a later version.
• breaking Version 8.9.0 of `mypy-boto3-builder` introduced breaking changes in TypeDef naming conventions. Specifically, `RequestRequestTypeDef` suffixes were shortened to `RequestTypeDef`, and `Extra` postfixes were moved to the end of TypeDef names (e.g., `CreateDistributionExtraRequestTypeDef` became `CreateDistributionRequestExtraTypeDef`). While `mypy-boto3-bedrock-runtime` specific changes might be rare, these builder-wide changes affect how you reference TypeDefs across all `mypy-boto3` stubs.
  Fix: Review your TypeDef imports and usages, adjusting names according to the new conventions.
• gotcha It is best practice to wrap type stub imports within an `if TYPE_CHECKING:` block. This ensures that the type stubs are only loaded during static analysis by tools like Mypy and Pyright, preventing them from becoming runtime dependencies in your production environment.
  Fix: Encapsulate `mypy-boto3-*` imports within `from typing import TYPE_CHECKING` and `if TYPE_CHECKING:` blocks.
• gotcha The `mypy-boto3-bedrock-runtime` package is generated for a specific `boto3` version (1.42.82 in this case). Mismatching `boto3` runtime versions with the installed stub version can lead to incorrect type hints or unresolved attributes, as the API might have changed between `boto3` versions.
  Fix: Ensure that your `boto3` installation closely matches the version indicated in the `mypy-boto3-bedrock-runtime` package name (e.g., `mypy-boto3-bedrock-runtime==1.42.82` for `boto3==1.42.82`).
• gotcha While `mypy` can often infer types, explicitly annotating the return type of `boto3.client("bedrock-runtime")` with `BedrockRuntimeClient` (and similarly for paginators and waiters) provides the most robust type checking and significantly improves auto-completion in IDEs like VSCode and PyCharm, especially for `boto3`'s dynamically generated clients.
  Fix: Add explicit type annotations when initializing `boto3` clients, paginators, and waiters, e.g., `client: BedrockRuntimeClient = boto3.client("bedrock-runtime")`.

Install

• pip install mypy-boto3-bedrock-runtime boto3 Install with pip

Imports

• BedrockRuntimeClient

  from mypy_boto3_bedrock_runtime import BedrockRuntimeClient

  Client types are provided by the `mypy-boto3` stub package, not directly by `boto3` or `botocore`.
• ListAsyncInvokesPaginator

  from mypy_boto3_bedrock_runtime.paginator import ListAsyncInvokesPaginator

  Paginator types are located in the `paginator` submodule.
• AsyncInvokeStatusType

  from mypy_boto3_bedrock_runtime.literals import AsyncInvokeStatusType

  Literal types (e.g., specific string values) are in the `literals` submodule.
• AppliedGuardrailDetailsTypeDef

  from mypy_boto3_bedrock_runtime.type_defs import AppliedGuardrailDetailsTypeDef

  TypedDicts for API request/response structures are in the `type_defs` submodule.

Quickstart

This quickstart demonstrates how to obtain a type-annotated BedrockRuntime client using `mypy-boto3-bedrock-runtime`. It shows how to call a client method (`invoke_model`) and how to use a paginator (`list_async_invokes`), leveraging `TYPE_CHECKING` for conditional imports. Ensure `boto3` is installed and AWS credentials are configured in your environment or via standard AWS configuration methods.

import boto3
from typing import TYPE_CHECKING
import os

if TYPE_CHECKING:
    from mypy_boto3_bedrock_runtime import BedrockRuntimeClient
    from mypy_boto3_bedrock_runtime.type_defs import InvokeModelResponseTypeDef
    from mypy_boto3_bedrock_runtime.paginator import ListAsyncInvokesPaginator

def get_bedrock_runtime_client() -> "BedrockRuntimeClient":
    """Returns a typed BedrockRuntime client."""
    return boto3.client(
        "bedrock-runtime",
        region_name=os.environ.get('AWS_REGION', 'us-east-1'),
        aws_access_key_id=os.environ.get('AWS_ACCESS_KEY_ID', ''),
        aws_secret_access_key=os.environ.get('AWS_SECRET_ACCESS_KEY', ''),
        aws_session_token=os.environ.get('AWS_SESSION_TOKEN', '')
    )

def invoke_model_example():
    client: BedrockRuntimeClient = get_bedrock_runtime_client()
    # Example: Invoke a model (replace with actual model_id and body)
    response: InvokeModelResponseTypeDef = client.invoke_model(
        modelId="anthropic.claude-v2",
        contentType="application/json",
        accept="application/json",
        body='{"prompt": "Human: Hello, how are you? Assistant:", "max_tokens_to_sample": 200}'
    )
    print(f"Model invocation response: {response['body'].read().decode('utf-8')}")

def list_async_invokes_paginator_example():
    client: BedrockRuntimeClient = get_bedrock_runtime_client()
    paginator: ListAsyncInvokesPaginator = client.get_paginator("list_async_invokes")
    for page in paginator.paginate():
        print(f"Async invokes page: {page.get('asyncInvokes')}")

if __name__ == "__main__":
    invoke_model_example()
    list_async_invokes_paginator_example()