Skip to main content
Software Engineering

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

Required fields*

4
  • 20
    The received wisdom is that any change can introduce defects. I prefer your teams strategy of adding tests first on that basis alone. Commented May 30 at 13:20
  • 2
    If I understand you correctly, the premise of the question is that unit tests do not prevent regressions, and that well-structured code does instead? Commented May 31 at 0:26
  • 1
    Consider that unit tests probably aren't your only tool; aside from the fact that refactoring to make code more unit testable can add risk, software bugs are commonly rooted in issues that can only manifest in a production-like deployed environment too. For example, unexpected behaviour in 3rd-party dependencies, hardware, infrastructure, user load/volume, environment-specific configuration, etc. With that in mind, I'd also look for opportunities to add integration tests and e2e tests around the legacy code. (Ideally running as part of the PR/merge process too) Commented May 31 at 7:32
  • The first part of the question seems to distinguish between "design improvements" and "new code", but that distinction gets muddled further down. In my mind, "design improvements" are behavior preserving refactorizations. "New code" is new code or code changes that intentionally change or extend the behavior. We typically do refactoring to improve the design in order to make it easier to extend the functionality. Adding tests for the existing code you plan to refactor is the best tool we have to ensure the refactorizations do not change behavior. Commented Jun 2 at 16:54