Tobias Furuholm

Maybe this could be solved by implementing #99?

It's to delay verification until after the execution. All possible interleavings have to be tested which means that run is invoked once wheras verify (typically) is invoked many times... On...

I think what you are saying is that even if there exists a valid serial command sequence, there may be interleavings produced by a parallel sequence that will violate the...

I see your point. In the end this will boil down to what tradeoffs that you are willing to make in order to add support for parallel testing. A couple...

After discussing this in person we decided on creating separate sub classes for commands that parallalizable and for those that are not. I have however experimented with another solution. I...

There are details left to be hammered out but I think this solution is more elegant than having separate command types and it doesn't complicate the sequential case too much....

There is a pull request with support for parallell testing. Unfortunately there have been major refactorings of the code base after the version that the parallell testing PR was forked...

First of all, of course nothing should be printed when the test succeeds. I don't have a single use case in mind, it's more a combination: - When tests are...

@fire: The race condition testing should be feature complete, but I need to finish writing the documentation first, and some more tests are needed as well. Been away for a...

I see. But how does groups differ from the other properties that are propagated in the current version (username, email, email verified)? Does those never change?