[feature]: Create a Testing Framework for KEPLOY Functionality Validation

[gouravkrosx]

[feature]: Create a Testing Framework for KEPLOY Functionality Validation

[gouravkrosx]

Basic Testing:

• Start by testing the core functions of Keploy, like verifying if the provided commands are correct and detecting errors in the user's command, such as port conflicts or non-zero exit commands.

Advanced Testing:

• Move on to more advanced testing, focusing on the "record and test" feature, which is the heart of Keploy.

Method:

• Begin by recording test cases and mocks for your application.
• Test the "keploy record" command with the "keploy test" command. It looks like this: keploy record -c "keploy test -c './sampleApp'". The "keploy record" tool records the same APIs that "keploy test" uses with the sample application.
• When your sample app makes an external call, "keploy record" captures the mocks sent by "keploy test." "Keploy test" reads the recorded mocks and sends them back. This way, no database is needed while testing the record feature of Keploy.

Initial Challenges:

• "Keploy record" couldn't run the "keploy test" command initially due to issues with environment variables. This was fixed by explicitly setting the environment for the command.
• Managing test epbf hooks was a bit tricky, but it's crucial to note that the hooks are executed in the order they were loaded, with record hooks running before test hooks.
• It's important to filter out the PID of the "keploy test" binary in "keploy record" to avoid capturing external calls made by "keploy test".
• Adjust the redirect mechanism so that the flow goes through "keploy record proxy" before "keploy test proxy." Initially, it was going directly to "keploy test proxy," preventing proper recording of tests and mocks. This issue was resolved by adding checks in the ebpf hooks.

This approach allows us to make significant advancements in our testing framework