Re: Design questions for a small s-expression library

[Philipp Janda <siffiejoe@...>]

Am 19.01.2014 05:35 schröbte H. Conrad Cunningham:
> On Sat, Jan 18, 2014 at 10:38 AM, Philipp Janda <siffiejoe@gmx.net> wrote:
> > `cons`, `car`, `cdr`, `setcar`, and `setcdr` are the only functions that
> > need to know the difference, so you can even change the implementation
> > later ...
>
>
> I have a set of example modules I created for the course I taught in my
> university's Fall semester (late August to mid-December) on my website at
> http://www.cs.olemiss.edu/~hcc/csci658/notes/658lectureNotes.html#cellList
>
> However, I intentionally did not include anything like a setcar or setcdr,
> instead forcing creation or a new Cons cell. I wanted immutability of the
> cells.

I started my experiments that way too. And then I tried to implement 
some functions from SRFI-1[1] using only `cons`, `car`, `cdr`, and 
recursion. And I got stack overflows for large input lists. Some 
functions I could fix easily (e.g. length), but others are difficult to 
implement efficiently without mutation. So I added `setcar` and `setcdr` 
(and started using explicit loops in the implementation) ...

> - Conrad Cunningham

Philipp

  [1]: http://srfi.schemers.org/srfi-1/srfi-1.html