Type annotations for boto3 CloudWatchObservabilityAccessManager

Warnings

• breaking Support for Python 3.8 has been removed across all `mypy-boto3` packages, including `mypy-boto3-oam`. Users on Python 3.8 will need to upgrade to Python 3.9 or newer.
  Fix: Upgrade your Python environment to 3.9 or a newer version (e.g., `python -m pip install --upgrade pip` then `pyenv install 3.9.18` and switch if using pyenv).
• breaking The `mypy-boto3-builder` (which generates this stub) introduced changes to TypeDef naming conventions in version 8.9.0. Specifically, `CreateDistributionRequestRequestTypeDef` became `CreateDistributionRequestTypeDef`, and `Extra` postfixes moved (`CreateDistributionExtraRequestTypeDef` -> `CreateDistributionRequestExtraTypeDef`). While this package's core Client type is stable, explicit references to generated TypeDefs might require updates if you're using older code with newer stubs.
  Fix: Review your code for direct imports or explicit usage of `mypy_boto3_oam.type_defs` to ensure TypeDef names align with the latest generated stubs. IDEs with `mypy` or `pyright` should help identify these.
• gotcha `mypy-boto3` migrated to PEP 561 packages, which is the standard for distributing type information. While this improves compatibility, users of older `mypy` or `IDE` versions might experience issues recognizing type stubs. Ensure your type checking tools are up-to-date.
  Fix: Update your type checker (e.g., `pip install --upgrade mypy`) and ensure your IDE's Python language server (e.g., Pylance for VSCode) is also up to date and correctly configured to use `mypy` or `pyright`.
• gotcha For optimal IDE auto-completion and type inference (especially in VSCode and PyCharm), it is often recommended to use explicit type annotations when instantiating `boto3` clients, even though `mypy-boto3` aims for automatic type discovery. This can prevent cases where the IDE struggles to infer the correct `boto3` client type.
  Fix: Explicitly type hint your `boto3.client()` calls, e.g., `client: CloudWatchObservabilityAccessManagerClient = boto3.client('oam')`.
• gotcha When using `mypy-boto3-oam` with Pylint, if you conditionally import type stubs using `if TYPE_CHECKING: ... else: ...`, Pylint might report `undefined variable` errors for the runtime `object` assignments. This is a known issue with Pylint's interaction with `TYPE_CHECKING` guards.
  Fix: To fix this, you can configure Pylint to ignore these specific errors (e.g., `disable=no-name-in-module,ungot-typeinfo` in your `.pylintrc`) or use `# type: ignore` comments judiciously.

Install

• pip install mypy-boto3-oam boto3 Install with boto3

Imports

• CloudWatchObservabilityAccessManagerClient

  from mypy_boto3_oam.client import CloudWatchObservabilityAccessManagerClient

  Type stubs are imported from the mypy_boto3_oam package, not directly from boto3's internal structure.

Quickstart

This quickstart demonstrates how to obtain a type-hinted `CloudWatchObservabilityAccessManagerClient` using `boto3` and `mypy-boto3-oam`. It includes a `TYPE_CHECKING` guard for conditional imports, which is a common practice to avoid runtime dependencies on stub packages. Replace placeholder credentials or ensure your environment is configured for AWS authentication.

import boto3
from typing import TYPE_CHECKING

if TYPE_CHECKING:
    from mypy_boto3_oam.client import CloudWatchObservabilityAccessManagerClient

def get_oam_client() -> 'CloudWatchObservabilityAccessManagerClient':
    """Gets a type-hinted CloudWatch Observability Access Manager client."""
    # In a real application, credentials would be managed by boto3's configuration
    # or environment variables.
    client: CloudWatchObservabilityAccessManagerClient = boto3.client(
        "oam",
        region_name="us-east-1",
        aws_access_key_id="{}".format(os.environ.get('AWS_ACCESS_KEY_ID', '')) if os.environ.get('AWS_ACCESS_KEY_ID') else None,
        aws_secret_access_key="{}".format(os.environ.get('AWS_SECRET_ACCESS_KEY', '')) if os.environ.get('AWS_SECRET_ACCESS_KEY') else None
    )
    return client

# Example usage (will not actually call AWS without valid credentials)
oam_client = get_oam_client()
print(f"Client type: {type(oam_client)}")
# mypy will now correctly infer methods like oam_client.list_attachments()
# without requiring a runtime call.