Skip to content

HDE/python-lambda-local

Repository files navigation

python-lambda-local

Join the chat at https://gitter.im/HDE/python-lambda-local Github Actions status PyPI version

Run lambda function on local machine

Prepare development environment

Please use a newly created virtualenv of Python 3.7+.

Installation

Within virtualenv, run the following command.

$ pip install python-lambda-local

This will install the package with name python-lambda-local in the virtualenv. Now you can use the command python-lambda-local to run your AWS Lambda function written in Python on your own machine.

Usage as a shell command

Run python-lambda-local -h to see the help.

usage: python-lambda-local [-h] [-l LIBRARY_PATH] [-f HANDLER_FUNCTION]
                           [-t TIMEOUT] [-a ARN_STRING] [-v VERSION_NAME]
                           [-e ENVIRONMENT_VARIABLES] [--version]
                           FILE EVENT

Run AWS Lambda function written in Python on local machine.

positional arguments:
  FILE                  lambda function file name
  EVENT                 event data file name

optional arguments:
  -h, --help            show this help message and exit
  -l LIBRARY_PATH, --library LIBRARY_PATH
                        path of 3rd party libraries
  -f HANDLER_FUNCTION, --function HANDLER_FUNCTION
                        lambda function handler name, default: "handler"
  -t TIMEOUT, --timeout TIMEOUT
                        seconds until lambda function timeout, default: 3
  -a ARN_STRING, --arn-string ARN_STRING
                        ARN string for lambda function
  -v VERSION_NAME, --version-name VERSION_NAME
                        lambda function version name
  -e ENVIRONMENT_VARIABLES, --environment-variables ENVIRONMENT_VARIABLES
                        path to flat json file with environment variables
  --version             print the version of python-lambda-local and exit

Prepare development directory

Project directory structure

Suppose your project directory is like this:

├── event.json
├── lib
│   ├── rx
│   │   ├── abstractobserver.py
│   │   ├── ... (package content of rx)
...
│   │       └── testscheduler.py
│   └── Rx-1.6.1.dist-info
│       ├── DESCRIPTION.rst
│       ├── METADATA
│       ├── metadata.json
│       ├── pbr.json
│       ├── RECORD
│       ├── top_level.txt
│       ├── WHEEL
│       └── zip-safe
└── test.py

The handler's code is in test.py and the function name of the handler is handler. The source depends on 3rd party library rx and it is installed in the directory lib. The test event in json format is in event.json file.

Installing rx library in lib/

pip install --target lib rx==1.6.1

Content of test.py:

from __future__ import print_function
from rx import Observable


def handler(event, context):
    xs = Observable.from_(range(event['answer']))
    ys = xs.to_blocking()
    zs = (x*x for x in ys if x % 7 == 0)
    for x in zs:
        print(x)

Content of event.json:

{
  "answer": 42
}

Run the lambda function

Within the project root directory, you can run the lambda function with the following command

python-lambda-local -l lib/ -f handler -t 5 test.py event.json

The output will be like:

[root - INFO - 2018-11-20 17:10:53,352] Event: {'answer': 42}
[root - INFO - 2018-11-20 17:10:53,352] START RequestId: 3c8e6db4-886a-43da-a1c7-5e6f715de531 Version: 
0
49
196
441
784
1225
[root - INFO - 2018-11-20 17:10:53,359] END RequestId: 3c8e6db4-886a-43da-a1c7-5e6f715de531
[root - INFO - 2018-11-20 17:10:53,360] REPORT RequestId: 3c8e6db4-886a-43da-a1c7-5e6f715de531	Duration: 2.17 ms
[root - INFO - 2018-11-20 17:10:53,360] RESULT:
None

Usage as a library

API signature

call(func, event, context, environment_variables={})

Call a handler function func with given event, context and custom environment_variables.

Sample

  1. Make sure the 3rd party libraries used in the AWS Lambda function can be imported.
pip install rx==1.6.1
  1. To call the lambda function above with your python code:
from lambda_local.main import call
from lambda_local.context import Context

import test

event = {
    "answer": 42
}
context = Context(5)

if __name__ == '__main__':
    call(test.handler, event, context)