Change method of checking if a key is in the dictionary.

[budude2]

The original method would return false if the key contained the boolean "false" causing the value to not be deleted even if it existed.

[russellballestrini]

Could you add a failing test that this fixed to prevent future regressions?

[budude2]

Sure, let me put one together really quick.

[budude2]

Here's the JSON code:

{
    "test1":
    {
        "integerTest": 10,
        "floatTest": 3.4,
        "boolTrue": true,
        "boolFalse": false,
        "arrayTest": [10, 20, 30, 40],
        "arrayTest2": [2.2, 3.14, 5.23, 6.28]
    },

    "test2":
    {
        "integerTest2": 24,
        "floatTest2": 6.28,
        "boolTrue2": true,
        "boolFalse2": false,
        "arrayTest3": [60, 70, 80, 90],
        "arrayTest4": [1.24, 6.78, 3.42, 8.90]
    }
}

When I run it before hand:

>>> pprint.pprint(nested_lookup.nested_delete(configDict, 'boolFalse'))
{'test1': {'arrayTest': [10, 20, 30, 40],
           'arrayTest2': [2.2, 3.14, 5.23, 6.28],
           'boolFalse': False,
           'boolTrue': True,
           'floatTest': 3.4,
           'integerTest': 10},
 'test2': {'arrayTest3': [60, 70, 80, 90],
           'arrayTest4': [1.24, 6.78, 3.42, 8.9],
           'boolFalse2': False,
           'boolTrue2': True,
           'floatTest2': 6.28,
           'integerTest2': 24}}

When I run it after:

>>> pprint.pprint(nested_lookup.nested_delete(configDict, 'boolFalse'))
{'test1': {'arrayTest': [10, 20, 30, 40],
           'arrayTest2': [2.2, 3.14, 5.23, 6.28],
           'boolTrue': True,
           'floatTest': 3.4,
           'integerTest': 10},
 'test2': {'arrayTest3': [60, 70, 80, 90],
           'arrayTest4': [1.24, 6.78, 3.42, 8.9],
           'boolFalse2': False,
           'boolTrue2': True,
           'floatTest2': 6.28,
           'integerTest2': 24}}

[grenzi]

just to chime in here, the same issue exists (and would be resolved by this PR) if the keys value is falseish, e.g., 0

>>> d = {"x":0, "y": 12, "z":3}
>>> nested_delete(d, 'x')
{'x': 0, 'y': 12, 'z': 3}
>>> nested_update(d, 'x', 1, in_place=True)
>>> nested_delete(d, 'x')
{'y': 12, 'z': 3}

[russellballestrini]

@grenzi could you take a pass at adding this as a unit test?