As a student, I want a key_held() alias for keys_held()

[AlphaRLee]

Currently we have keys_held(), but the most popular use case would be for just a single key

AC:

• Add a key_held(str, pattern=None) with the @ignite_global, where str is a string representing if a single key is held down.
  • If pattern is None, then check if the key is held.
  • If pattern is a boolean, then check that the key matches the boolean state (True means key is held, False means key is not held).
  • Return a boolean indicating if the key is held or not
  • Add validation for a single string and an optional pattern=boolean (validating the pattern param is low priority)
• Add a unit test
• Update docs

[lawruble13]

Personally, I don't think this is a particularly useful addition, because

• for the presumed most popular use case it's a single additional character (i.e. keys_held('a') vs key_held('a'));
• the conceptual difficulty of the method as-is is almost nil;
• using the pattern here requires more characters and reduces clarity vs the pythonic way (i.e. key_held('a', False) vs not key_held('a')); and
• even if we remade key_held to do a more efficient check against the internal dictionary, the performance improvement would almost certainly be countered by the additional load of the validation function

[AlphaRLee]

for the presumed most popular use case it's a single additional character (i.e. keys_held('a') vs key_held('a'));
the conceptual difficulty of the method as-is is almost nil;

I'm not a fan of the lack of clarity (why "keys held" instead of "key held")? I agree it's trivial to handle this either way. Even when you do want to do multiple key presses, it's not often that you want to do key1 AND key2, it's more common for key1 or key2 (e.g. moving a character around), which is usually done in 2 separate if statements

using the pattern here requires more characters and reduces clarity vs the pythonic way (i.e. key_held('a', False) vs not key_held('a'));

I'm ok tossing out this feature, I agree there's not much value in this one.

even if we remade key_held to do a more efficient check against the internal dictionary, the performance improvement would almost certainly be countered by the additional load of the validation function

Tbh, I'm not concerned about efficiency on this check. Correct me if I'm wrong, but I don't see this bogging down performance significantly.

[lawruble13]

Merged PR #74, does not include documentation or unit test