Stefan Monnier <
address@hidden> schrieb am So., 18. Dez. 2016 um 21:15 Uhr:
> For some of them definitely. I'd draw the line between pure functions like
> car and eq, where those facilities never make sense and would be
> ineffective anyway as the functions are compiled away or called directly,
> and impure functions like call-process, where fset is necessary for
> mocking. As a rule of thumb, I'd suggest to ban fset on all symbols that
> have a byte-code equivalent, and on constant symbols.
I still very doubt that the potential benefit is worth the added cost
(more specifically, as a maintainer I would strongly oppose such
measure).
Why is then the potential benefit for the value cells worth the added cost, i.e. why not also allow (set t 5)?
Are you also going to try and prevent the user from using all
the other ways he can shoot himself in the foot?
To the extent that it's feasible, yes. Emacs Lisp isn't C. If the user wants to shoot themselves in the foot, they can write a C module.