Type annotations for boto3 CloudWatch

Warnings

• breaking Python 3.8 support has been removed since version 8.12.0 of `mypy-boto3-builder` (which generates this package). Ensure your project uses Python 3.9 or higher.
  Fix: Upgrade your Python environment to 3.9+.
• breaking The `mypy-boto3` ecosystem has migrated to PEP 561-compatible packages since builder version 8.12.0. This means you should now directly install `mypy-boto3-cloudwatch` rather than relying on `boto3-stubs` for service-specific types.
  Fix: Ensure you `pip install mypy-boto3-cloudwatch` directly. If you were using `boto3-stubs[cloudwatch]`, switch to the direct package.
• breaking Breaking changes in `mypy-boto3-builder` 8.9.0 introduced shorter TypeDef names (e.g., `CreateDistributionRequestRequestTypeDef` became `CreateDistributionRequestTypeDef`) and moved `Extra` postfixes. If you directly imported and used these TypeDefs, your code may break.
  Fix: Review your TypeDef imports and usages, adjusting names according to the new conventions.
• gotcha This package provides *only* type annotations. You must install `boto3` separately to have the runtime functionality for interacting with AWS services.
  Fix: Ensure `boto3` is installed in your environment: `pip install boto3`.
• gotcha While explicit type annotations are often beneficial, recent `mypy-boto3` versions offer auto-discovery for `boto3.client` and `boto3.resource` calls. Your IDE (e.g., VSCode with Pylance, PyCharm with mypy plugin) should pick up types automatically without explicit imports of `CloudWatchClient` in many cases.
  Fix: Consider simplifying your type hinting where auto-discovery works, but explicit imports provide the most robust type checking.
• gotcha For Pylint compatibility and to avoid runtime dependencies on `mypy-boto3-cloudwatch` in production, it is recommended to guard type imports with `if TYPE_CHECKING:` to set types to `object` at runtime.
  Fix: Wrap type-specific imports with `from typing import TYPE_CHECKING` and define runtime fallbacks: `if TYPE_CHECKING: from mypy_boto3_cloudwatch.client import CloudWatchClient else: CloudWatchClient = object`.

Install

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

Imports

• CloudWatchClient

  from mypy_boto3_cloudwatch.client import CloudWatchClient

  Import the specific client type for detailed type hints.
• PutMetricDataInputRequestTypeDef

  from mypy_boto3_cloudwatch.type_defs import PutMetricDataInputRequestTypeDef

  Import TypeDefs for precise type checking of request/response structures.

Quickstart

This quickstart demonstrates how to instantiate a `boto3` CloudWatch client with explicit type hints from `mypy-boto3-cloudwatch` and how to use a generated TypeDef for a method's input parameters.

import boto3
from mypy_boto3_cloudwatch.client import CloudWatchClient
from mypy_boto3_cloudwatch.type_defs import PutMetricDataInputRequestTypeDef

# Instantiate a CloudWatch client with type annotation
client: CloudWatchClient = boto3.client(
    "cloudwatch",
    region_name="us-east-1",
    aws_access_key_id="AKIATESTKEY", # Use os.environ.get('AWS_ACCESS_KEY_ID', '') in real code
    aws_secret_access_key="YOUR_SECRET_KEY" # Use os.environ.get('AWS_SECRET_ACCESS_KEY', '') in real code
)

# Example of using a TypeDef for a request payload
metric_data: PutMetricDataInputRequestTypeDef = {
    "Namespace": "MyApplication",
    "MetricData": [
        {
            "MetricName": "CustomMetric",
            "Dimensions": [
                {"Name": "Environment", "Value": "Production"}
            ],
            "Value": 10.5,
            "Unit": "Count",
        }
    ],
}

try:
    response = client.put_metric_data(**metric_data)
    print(f"Successfully put metric data: {response['ResponseMetadata']['HTTPStatusCode']}")
except Exception as e:
    print(f"Error putting metric data: {e}")