Kinesis Quality Toolkit (KiQT)

Quality assurance toolkit for testing Kinesis Data Analytics applications.

Overview

Amazon Kinesis is a powerful, real-time streaming and analytics platform where a single streaming pipeline is composed of multiple AWS components representing the application as well as its inputs and outputs. The goal of this project is to provide a single cohesive interface for:

Inspecting a running application to dynamically create input writers and output readers
Defining deterministic and repeatable functional test cases by:
- serializing and writing data to the application input
- reading from the application output(s), deserializing contents, and performing assertions on the output
- managing application lifecycle

Project Structure

This repository is currently divided in to 2 main projects:

kiqt-core is the set of resources needed to interact with Kinesis Analytics and Kinesis Streams without taking on any dependencies specific to testing
kiqt-junit provides a behavior driven development style API for writing JUnit functional tests for an application

Example

Given a list of inputRecords and an expected list of outputRecords, an example test may look something like this:

// creates the test harness
KinesisQualityTool kiqt = new KinesisQualityTool(new ApplicationIOProvider("MyApplication"));
// gets the input kinesis stream
kiqt.theInputStream()
        // specifies a response handler
        .withResponseHandler(response -> Assert.assertEquals(0, response.failedRecordCount().longValue()))
        // writes the inputRecords to the input stream
        .given(inputRecords);
// gets the output stream named OUTPUT_STREAM_NAME and deserializes its records to type OutputRecord
kiqt.theOutput("OUTPUT_STREAM_NAME", OutputRecord.class)
        // sets up a timeout of 30 seconds for all assertions on this output
        .within(30, TimeUnit.SECONDS)
        // executes the assertion
        .should(Matchers.equalTo(expectedOutput));

For more sample code and use cases, see the samples.

Roadmap

Publishing

We're working on setting up processes for publishing our artifacts to Maven Central.

Supported Destinations

Currently only Kinesis Streams are supported as inputs and outputs, with the other integrations to come. Following the project structure decribed above, our highest priorities are creating projects:

kiqt-firehose for parsing resource information from a Kinesis Firehose destination
kiqt-s3 for reading data that has been written (via a firehose) to an S3 bucket

Kinesis Best Practices Assertions

We'd like to add some assertions specific to ensuring applications under test follow the guidelines set in the Best Practices documentation for SQL applications.

Supported Test Frameworks

This isn't a high priority for us currently, we may be open to accepting PRs specialized to frameworks other than JUnit if there is demand.

Contributing

See the contributing guidelines for more information.

Nuvalence/kinesis-quality-toolkit