This is a template to conveniently get a serverless application model started with Python 3.8. In order to use the template and get started with AWS Lambda functions, you'll need both the AWS and SAM command line interfaces installed.

Once installed, the template can be downloaded from this repository directly into a specified directory on your local machine.

The file structure that will be generated for your SAM-initialized project based on this template is below.

.
├── Makefile
├── README.md
├── events
│   └── event.json
├── requirements.txt
├── src
│   └── service
│       ├── __init__.py
│       └── app.py
├── template.yaml
└── tests
    └── unit
        ├── __init__.py
        └── test_handler.py

It's important to understand that the main goal of the template is to make life easier by bypassing some of the nitty-gritty and often tedious steps to create, build, and deploy a successful Lambda application.

To that end, the first step is to initialize your project with this command:

sam init -l gh:aridyckovsky/sam-template-python3.8

It will prompt you for a project name, defaulting to "My App", and that name is used to create a project slug, defaulting to "my-app". You should almost never need to specify a project slug, and simply press enter to the suggested default (which is dynamically produced based on your project name choice).

To build your first Lambda function, run

make build

This is a convenient abstraction over a SAM command. If you'd like to see what that command is, open the Makefile in your project directory.

To see if the Lambda function works as expected, run

make run

Again, this overlays the more complicated SAM command, and will invoke the lambda_handler function found at ./src/service/app.py locally. When you use this command, you should see the underlying SAM invokation (i.e., sam local invoke ...), and then a console message stating

Invoking service.app.lambda_handler (python3.8)

If the handler runs successfully, you will see a a few more messages, and finally a JSON data payload starting with a status code of 200.

To deploy the Lambda function, in this case called "Service", to AWS, simply run

make deploy

It will ask a series of questions, and if your AWS credentials are up to date and configured, the command will lead to a new resource being pushed to S3 named after the project and a respective Lambda function.