python fork

changes:

• copy input to user to avoid confusion when user mutates (this is why I am touching it to begin with)
• remove redundant describes (only contained one it, describe is for grouping)
• custom formatted error messages to include what the input was
• replace parts of test generation with equivalent calls to the random module

Javascript translation posted

Rust translation

Javascript translation

I think this is terrific graph problem but I'm not sure it belongs on the current difficulty it is at :) (currently 6). Anyone else feel the same?

what happens for cases like

I wanna eat and then drink some S ?

1000 tests seems like overkill. you could instead test for edge cases that are easily missed

Fix The Top Left

Random tests are weak. My solution should have never passed.

Why is top left measured differently than the other edges?