[Random] currently supports several primitive types (mostly numeric ones), but it does not work with Guid. To provide different values to a Guid argument, one has to manually provide them either via [TestCaseSource] or via [ValueSource].
In my experience, being able to easily provide different Guid values to a test is fairly useful. Currently, I'm doing this by using a ValueSource, which caused a different problem, forcing me to resort to hardcoded Guids instead of using Guid.NewGuid.
I'm aware that Guid randomization is different than the standard one using the Random class, so I'm not sure if it would be possible to have the same "seed" in discovery and execution phases to build the unique test names with the built-in values (probably not).
[Random]currently supports several primitive types (mostly numeric ones), but it does not work withGuid. To provide different values to aGuidargument, one has to manually provide them either via[TestCaseSource]or via[ValueSource].In my experience, being able to easily provide different
Guidvalues to a test is fairly useful. Currently, I'm doing this by using aValueSource, which caused a different problem, forcing me to resort to hardcodedGuids instead of usingGuid.NewGuid.I'm aware that
Guidrandomization is different than the standard one using theRandomclass, so I'm not sure if it would be possible to have the same "seed" in discovery and execution phases to build the unique test names with the built-in values (probably not).