I am sure others have experienced frustration with how updates to test suites are not reflected in active test runs that split test cases manually, even when a selection is for an entire section or sub-section.
The general idea for how to build around this limitation seems to require taking where you would normally split a suite into a section (or even just splitting based on where assignment workload) and instead splitting it into its own suite where the “select all” option can be used and updates with each change.
That presents two issues: the main issue I’m raising and the inverse where you might want to freeze suites in active runs so that new test cases aren’t suddenly inserted into a testers To Do list in the middle of a run. The primary use case that comes to mind for me would be when I’m adding/updating cases in advance of new features actually being implemented.
The solution that I imagined is to automatically detect if changes are being made to a suite* that is also within an active run or plan, and present a popup modal with few options to the user:
- Push changes to all active runs
- Self-explanatory. This would only apply the push to runs that trigger the suite/section/sub-section level selections
- Push changes to the selected active runs
- This option could potentially also be used to push the new/updated cases into a run that previously did not include any of the “original” sections. Useful in a situation like doing a specific re-run of “retest” status results but there are also new additional changes that should’ve been tested at the same time but were missed
- Save changes without pushing to active runs
- This would be identical to making updates to a suite that is not involved in any active runs in any way (current behavior, assuming no active runs)
- To go back and make more edits before attempting to save/push again
- “Discard changes”
- Reverts suite back to before any updates or changes were made. This may be too destructive to include, but I’m trying to not miss anything
This would only apply the logic to suite/section/sub-section level detection, so if you had an active run that was only testing a manual selection of individual cases without ever selecting a full section, etc. those would not trigger the options.
Does this feature request make sense? I think it handles most cases for
*either when attempting to navigate away, or introducing “save changes” and “discard” buttons at the top and bottom once the first changes are made, or both.