Layered dictionary (__missing__ hook)

layereddict.py

class layereddict (dict):
    
    def __init__(self, layer, **kwargs):
        super().__init__(**kwargs)
        self._layer = layer
    
    def __missing__(self, key):
        return self._layer[key]

Think of it like class and instance namespaces. The instance "inherits" the class namespace as a (read-only) background layer and can overshadow it.

See also https://docs.python.org/3/reference/datamodel.html#object.__missing__ and https://docs.python.org/3/library/stdtypes.html#dict ¹, and https://docs.python.org/3/library/collections.html#collections.defaultdict

Footnotes

1. non-linkable subsection d[key] in supported operations ↩

I wonder though, is a python layered dict backed by __missing__ as efficient as the instance -> class delegation, or less so? (Or is the former inefficient too?)