This project contains packages to interact with various Amazon Web Services. Currently the following functionality is provided:
- Build Lambda functions using Pharo Smalltalk
- Writing log events to the CloudWatch Logs service
- Access to the Amazon Simple Storage Service (S3) - only a part of api is implemented yet
- Access to the Amazon Simple Queue Service (SQS)
With AWS Lambda you can run code in the cloud without setting up any kind of server. AWS Lamda has build in support for some common programming languages but also supports custom runtimes. With the Pharo Lambda runtime you can implement Lambda functions in Pharo Smalltalk. It consists of a layer with the Smalltalk VM and a small support library for communicating with the Lambda Runtime API.
Pharo Smalltalk is a pretty efficient environment to implement Lambda functions. Functions implemented in Smalltalk normally don't need more than 256MB of memory to execute. This means that you can execute a million function calls for free each month.
For more details see Pharo Lambda Runtime.
You can load the Pharo AWS Toolbox using Metacello:
Metacello new
repository: 'github://jvdsandt/pharo-aws-toolbox/repository';
baseline: 'AWSToolbox';
load.Pharo AWS Toolbox has the following dependencies:
- NeoJSON - Used for reading and writing JSON data.
- XML-XMLParser - The SQS service uses XML for request and response messages.
To make requests to Amazon Web Services, you must supply AWS credentials. You can use the class AWSEnvironment to access the credentials and region information stored on your computer. It can read your credentials and region name from standard AWS environment variables:
- AWS_ACCESS_KEY_ID / AWS_ACCESS_KEY
- AWS_SECRET_KEY
- AWS_SESSION_TOKEN
- AWS_DEFAULT_REGION
Another possibility is to read the credentails from standard AWS credentials file ~/.aws/credentials and the region name from ~/.aws/config