Am Mi., 16. Nov. 2022 um 13:04 Uhr schrieb Lassi Kortela <la...@lassi.io>: > > > > - make-product-comparator and make-sum-comparator are defined to take > > zero or more comparators. It'd be better that the behavior when no > > comparators are explicitly stated. Or we may define them to take at > > least one argument. > +1
There is no reason to restrict to >= 1 comparator. It would be an arbitrary restriction. > > - I'm implementing srfi-228 for Gauche from the srfi text and find the > > description of above two procedures a bit confusing, because the > > elements of comparators (type test, equality, ordering and hashing) > > are all written in one paragraph. How about itemize each element? > +1 > > The names make-product-comparator and make-sum-comparator are quite > long. Could the "make-" prefix be dropped? I think Daphne's names are consistent with R7RS and the other relevant SRFIs. > The terms "sum" and "product" are obtuse. I assume they refer to math > (category theory?), as in sum and product types. Are there everyday > words that would say something close enough? The description is also > hard to understand. Instead of "product", "intersection" looks like a good choice because this is what happens with the type governed by this comparator. Then, one could replace "sum" with "union". > Anyway, the ability to compose comparators is a good one. I know that I would have to blame myself as well, but it is funny how often "last calls" elicit first comments by people!
