The problem
Recording a user journey in the browser is easy. Turning it into an automated API test by hand is slow.
How it works
- 01Import HAR
- 02Find values
- 03Chain requests
- 04Replay
- 05Assert
What was hard
- The same ID can appear in many places, so each extracted value has to be mapped to the right later request.
- HAR files can hold passwords and tokens, which must be stripped before anything is saved.
- Values from one response feed the next request, so the chain order matters.
The result
A recorded browser session becomes a repeatable, automated API test.