Hey @eliangcs, thanks for writing this!
Have you discussed the possibility of merging this upstream with @wbolster by any chance? Having a library with some fakes is great, but if they're not covered under some overlapping tests with upstream, it's hard to have confidence that the fakes both stay in sync with upstream, but more importantly, that they have behavior that matches what real HBases will do rather than just being internally consistent which is what an isolated test suite will tell you. @wbolster have you seen this library / would you possibly have interest in that?
Hey @eliangcs, thanks for writing this!
Have you discussed the possibility of merging this upstream with @wbolster by any chance? Having a library with some fakes is great, but if they're not covered under some overlapping tests with upstream, it's hard to have confidence that the fakes both stay in sync with upstream, but more importantly, that they have behavior that matches what real HBases will do rather than just being internally consistent which is what an isolated test suite will tell you. @wbolster have you seen this library / would you possibly have interest in that?