Describe the issue:
If a run is manually judged, or rejudged, and assigned a different judgment than recommended by the system, the final judgment presented on the event feed may not match what is indicated by the runs sent on the event feed. Reported on #1279 during testing.
To Reproduce:
This comment pretty much summarizes how to duplicate it
Expected behavior:
It is unclear what should be done. The runs are presented on the event feed as the judging system judged them. Simply because the human judge decided to override the recommended judgment should not change the disposition of any of the runs.
Actual behavior:
The possibly overridden judgment is presented on the event feed, followed by the runs (Test cases). They may be inconsistent in that, for example, all the runs may indicate accepted, but the human judge overrode the judgment with TLE, for example.
Environment:
All
Log Info:
Screenshots:
See This comment
Additional context:
Describe the issue:
If a run is manually judged, or rejudged, and assigned a different judgment than recommended by the system, the final judgment presented on the event feed may not match what is indicated by the runs sent on the event feed. Reported on #1279 during testing.
To Reproduce:
This comment pretty much summarizes how to duplicate it
Expected behavior:
It is unclear what should be done. The runs are presented on the event feed as the judging system judged them. Simply because the human judge decided to override the recommended judgment should not change the disposition of any of the runs.
Actual behavior:
The possibly overridden judgment is presented on the event feed, followed by the runs (Test cases). They may be inconsistent in that, for example, all the runs may indicate accepted, but the human judge overrode the judgment with TLE, for example.
Environment:
All
Log Info:
Screenshots:
See This comment
Additional context: