You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Actually, there is nothing random in the sequential write/read procedure for now. Besides this, it is a good feature for any tests to keep the determined results.
So I think we should fix these random things rather than support the proposed the ignore interceptor.
And I find out it that the random table id results from that fact the data generated by the previous run is not reset, so the id for the created-dropped-created-again table will be incremented finally, which leads to the test failure.
Description
Some cases are not expected to run due to unstable features, temporary changes or anything else. It should be supported to just ignore them.
Proposal
Implement the
ignore
interceptor. Proposed syntax:Follows the K-V format. Value for this interceptor is the reason why this case got ignored. It won't take effort and is a comment in fact.
Additional context
TBD: should we count/record those ignored cases and report it in the end?
ref #154
The text was updated successfully, but these errors were encountered: