Correlating test failures across runs could help improve code quality

Bug #608334 reported by James Westby on 2010-07-21
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Testrepository
Wishlist
Unassigned

Bug Description

Hi,

Writing good tests is a difficult art.

It would be great if one day I could look at some stats from testrepository and
find the tests that fail most frequently, then look at the changes that were made
that caused them to fail.

This would help me determine if they were catching bugs, too tightly bound, or
just preventing changes rather than being useful.

Thanks,

James

Changed in testrepository:
status: New → Triaged
importance: Undecided → Wishlist
summary: - Should have a way to record the changes made between test runs
+ Correlating test failures across runs could help improve code quality
Robert Collins (lifeless) wrote :

A related thing would be tests that fail together, or tests that only fail when another test is run before them.

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers