This whole issue seems so obvious and trivial to me that I've in fact expected it to be resolved by itself
Obvious - arguable Trivial - I think so too, but probably not in the way you mean
The obviousness is arguable because the differences between __getattr__ and __getattribute__ are genuinely subtle and changing the name of the latter won't alter that. A single attribute access may actually pass through both methods at different points in the process. (It could be said that the similarity in names is actually advantageous, as it corresponds to the similarity in function)
Python makes it fairly easy to override attribute access, but it can't eliminate the basic complexity of the idea - if you want to override builtin behaviour, it's necessary to understand what the builtin behaviour *is* first.
On the triviality aspect, the amount of work involved in changing the name far exceeds any slight aesthetic improvement from a new name. And that's without even getting into the issues of backwards compatibility. I'd prefer to see the developers making speed improvements or adding to the standard library, rather than working on changing the names of special methods.
Cheers, Nick.
--
Nick Coghlan | [EMAIL PROTECTED] | Brisbane, Australia
---------------------------------------------------------------
http://boredomandlaziness.skystorm.net
--
http://mail.python.org/mailman/listinfo/python-list
