alexandrepasc / locust-load-starter Goto Github PK

A load test must sometimes replicate a client's behaviour (using the GUI). If the client uses a webpage to interact with the service we could retrieve from the browser an HAR file with the list of network communication.

So if we need to replicate a client behaviour using a webpage, and map that to a load test, it would be a nice addition to give the ability to do the behaviour using the webpage, export the HAR file, and import it to the load testing tool automatically.

• Investigate how to read the HAR file
• Find a way to integrate the flow with the existing generated artifacts
• Create a new script to read, and create the flow artifacts

The script run_tests.py has some flags that it supports and that are used to give the name to the reports, so the tester can run multiple tests in sequence without the need to in each run change the names.

But there are more options that the tester can use when executing Locust directly, add a way to pass the Locust options that the script doesn't support directly to the test call.

When developing the scripts, some prints were added to help troubleshoot some behaviors, that are not needed now, and that create some clutter in the execution.

Look at all the scripts and remove all the unnecessary prints.
After removing the prints look if there is some information that could be helpful to the user that is not being shown and add it.

The OpenAPI can support the yaml and the json file formats, at the moment the endpoints generation script only supports the yaml.

Since there are multiple file formats and the idea of this project was to be a helpful tool for multiple projects support to this format needs to be added.

ref: https://swagger.io/docs/specification/basic-structure/