
On 3/21/19 6:46 PM, Steven D'Aprano wrote:
Antoine and Serhiy seem to worry that there are existing uses of + which are currently easy to understand but will become less so if dict.__add__ is added. I respect that worry, even if I doubt that they are correct.
If someone can demonstrate that their fear is well-founded, that would be an excellent counter-argument to the PEP's proposal to use +.
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, but arguably no worse than c.update(d). Yes, it's "just" polymorphism, but IMO it violates the Principle of Least Surprise. My apologies if this is covered elsewhere in this thread, or it doesn't meet the bar Steven set.