1. I copletely agree it isn't exactly the best design decision, on the contrary, but the developpers are still too blame as well in my opinion. They chose the d

1. I copletely agree it isn't exactly the best design decision, on the contrary, but the developpers are still too blame as well in my opinion. They chose the datatype so they should have read the documentation, especially the caveats like this, and they should have done something to deal with this. This applies more generally. There are a ton of programming languages, standard libraries and APIs out there with weird/confusing/exceptional/undefined-behavior-invoking types and functions depedning on how they are used so a sane developper should alwyas check this and handle accordingly, even though it might lead to more code to handle edge cases than to the actual expected main code path.