>From what I recall, the only class of non-top-level `super-new' calls I've >seen are those from deep within the mred hierarchy that are under lock >protection. I think they were all top-level under that, but that would need to >be double-checked.
Stevie On Jun 10, 2014, at 10:03 AM, Asumu Takikawa <as...@ccs.neu.edu> wrote: > On 2014-06-09 11:22:59 +0200, Laurent wrote: >> Given that it's easy enough to create a /new/ class on Wednesday that has >> 3 arguments instead of 2, doesn't it make sense to constrain a given >> existing class to have a fixed structure? (just asking) >> >> Maybe it could also allow for some optimizations. > > I agree, it would be nice (especially for contracts & Typed Racket) if > `super-new` calls were only allowed in fixed positions such as the > top-level of a class. > > That should also prevent weird dynamic initialization like in my > example. Also, the macro could use that information to record the actual > init arguments in the class so that reflective operations could query > for them. > >> I wonder if such a additional constraint would break anything in the >> existing code. > > It looks like the majority of uses of `super-new` are at the class > top-level in the Racket code base. I bet it would break one or two weird > examples though. > > Cheers, > Asumu > ____________________ > Racket Users list: > http://lists.racket-lang.org/users ____________________ Racket Users list: http://lists.racket-lang.org/users
