From: John Cowan <[email protected]> Subject: Re: [Chicken-users] sequences egg Date: Sat, 20 Nov 2010 14:39:58 -0500
> Felix scripsit: > >> > I don't see much point in the constructors, really. Schemers already >> > understand how to make lists and vectors. >> >> Sorry, I don't understand this remark. > > Ah. I just assumed that random access sequences and linear access > sequences were vectors and lists respectively, but looking at the code > I see they aren't. What advantages do they have? They are intended to allow the user to define other types of sequences (make-linear-sequence and make-random-access-sequence take a number of procedures and so abstract away (hopefully) all direct access, making them usable by the sequence operators). > >> I know all these by heart. > > I figured that, but a sequence library is useful in proportion to its > richness, so I'd add more functions. Ah, ok. I somehow would like to avoid a huge pile of procedures, though. I find it very difficult to decide what is essential and what is trivial (or what could be defined by simple combination of existing operators). But you are right - the advantage of using such a library is both being able to use generic operators and also have a large collection of "standard" operations readily available. Thanks for the suggestion. cheers, felix _______________________________________________ Chicken-users mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/chicken-users
