No, the code works fine. I just wish the docs weren't so muddled.

I wish the docs started by saying:

> The graph is a dict of {'start_node': ['end_nodes',]}
> The topological sorter puts the end_nodes before their start_nodes.
   [note: this is what the code currently does]

Then the docs could introduce and use different terminology: "tasks" and their "dependencies".
Everyone understands what those are.


And nowhere does it need to say the word "predecessors".


But honestly, I think I'm going to throw in the towel and give up on this endeavor.