Yitzchak Gale schrieb:
When using it in practice, it would be very useful to have an analogue to the mconcat method of Monoid. It has the obvious default implementation, but allows for an optimized implementation for specific instances. That turns out to be something that comes up all the time (at least for me) in real life.
Btw. has someone an idea how to design 'mconcat' for pair types? I mean, it is certainly sensible to define: instance (Monoid a, Monoid b) => Monoid (a,b) where mappend (a0,b0) (a1,b1) = (mappend a0 a1, mappend b0 b1) mconcat pairs = (mconcat (map fst pairs), mconcat (map snd pairs)) but the mconcat definition would be inefficient for the type (Sum a, Sum b). E.g. if I evaluate the first sum in (mconcat pairs) first and the second sum last, then 'pairs' must be stored until we evaluate the second sum. Without the Monoid instance I would certainly compute both sums simultaneously in one left fold. Since a left fold can be expressed in terms of a right fold, it would certainly work to map left-fold-types like Sum to an Endo type, but this leads us to functional dependent types or type functions.