On 3/21/19 9:19 PM, Christopher Barker wrote:
https://docs.python.org/3.8/library/collections.html has some examples using collections.Counter, which is clearly described as being a subclass of dict. Amongst the examples:
c + d # add two counters together: c[x] + d[x]
That's the + operator operating on two dicts (don't make me quote the Liskov Substitution Principle), but doing something really different than the base operator.
So if I know that c and d (or worse, that one of them) is a dict, then interpreting c + d becomes much more interesting,
Killing a use of a common operator with a very common built in data type because the operator is used in a different way by a specialized object in the stdlib seems a bit backwards to me.
Perhaps. Note that Counter also uses | and & for other operations that probably wouldn't make much sense on base dicts.
Frankly, I think considering Counter as a dict subclass is the mistake here, even if it is true.
I had the same thought that Counter is misdesigned in one way or another, but (a) that ship has long sailed, and (b) I didn't want to run off on that tangent.
My point remains: because Counter is a subclass of dict, and Counter uses the + operator for something that doesn't apply to base dicts, adding + to dicts *may* cause confusion that wasn't there before.
Presently, +, -, |, and & all raise an exception when given a Counter and a dict. This (raising an exception) is probably still the Right Thing to do in that case, even with a + operator on dicts, but that violates the LSP and IMO the PLS.