Compatibility of code base with previous version not obvious #57
Description
@ghickman in https://github.com/py3minepi/py3minepi/pull/54#issuecomment-59644353 wrote
If you have qualms about areas of the code base which you believe are not backwards compatible then please raise them in a new issue. We are not looking to improve legacy code for the 1.0 release.
This is that issue, but not in exact form requested. Tony Hoare wrote
There are two ways of constructing a software design: One way is to make it so simple that there are obviously no deficiencies, and the other way is to make it so complicated that there are no obvious deficiencies. The first method is far more difficult.
The diff between the two codebases is full of noise, and has many changes that are not necessary for the purpose of backwards compatibility. As a result, I would say, there are no obvious deficiencies.
Or put another way, any deficiencies in the backwards compatibility are not obvious. There is evidence for this. Just 10 days ago @doismellburning added f7468b1, which was a reverted a change to broke the previously working intFloor.
Someone, it seems, decided to make a change to intFloor and it got committed. This change was only discovered after testing. It should have been obvious from a diff, and spotted. Are you, @ghickman, sure that all changes made are backwards compatible?
Tony Hoare wrote
The first method is far more difficult.
I believe in our case the first method is easier. It does however require discipline. I know we're itching to revise or rewrite the Javesque code.