- Executor and Pool API now take function directly instead of the function name Eg: pool.map(fibonacci, list(range(10))) instead of pool.map('algorithms.fibonacci', list(range(10)))
- Makes the project and module naming consistent
- Removes the dependency of setup.py on lambdapool/agent.py
- Modifies the default aws endpoint timeout and adds the support to specify the same
- Fixes some naming inconsistencies
- Changes Author Info and restructures
setup.py
- Adds
cloudpickleto the requirements
- Adds support for data serializing using cloudpickle
- Fixes multithreading quirks with boto3 client
- Minor bug fixes
- Improves the CLI with help text and better formatting
- Adds test and coverage infrastructure
- Adds tests for most of the codebase
- Adds option to specify memory and timeout while creating or updating functions
- Adds support for specifying layers to LambdaFunction's
- Implements LambdaExecutor
- Implements functionality to propagate and raise appropriate errors
- Changes the lambdapool function identification protocol to environment variables from tags
- Makes lambdapool list an O(1) operation from the existing O(n)
- Fixes an issue with AWS IAM role
- Fixes bugs with return result not being json decoded
- Implements lambdapool cli
- Adds
create,update,list,deletecli command functionalities
- Fixes typo in lambdapool.agent.load_function
- Reverts back to the old style of lambda_handler
- Fixes typo in LambdaHandler
- Finalizes LambdaHandler API
- Finalizes LambdaPool API
- Changes the behavior of AWS Lambda handler function. It returns the function result as it is now.
- Minor changes to
lambdapool/__init__.py
- Refactors the LambdaPool.apply API
- Minor refactors
- Initial code with the AWS Lambda agent and LambdaPool