This is of course a symptom of the more general problem that we fail to have a mechanism that specializes a module to a particular language context. In *SL, the restriction is absolutely great; in Racket, the restriction is a pain. Note that units wouldn't solve the problem cleanly either -- Matthias
On Oct 22, 2013, at 8:14 AM, Robby Findler <ro...@eecs.northwestern.edu> wrote: > I did this explicitly because (I believe) it helps with parenthesis errors > when beginning students make simple mistakes using the library. For more > advanced programmers, it seems straightforward to define things like: > > (define (above* . is) (cond [(null? is) (blank)][(null? (cdr is)) (car > is)][else (apply above is)]) > > and you can even abstract that as a function that takes something like above > and builds an above* from it. > > Robby > > > On Mon, Oct 21, 2013 at 10:58 PM, Daniel Prager <daniel.a.pra...@gmail.com> > wrote: > Currently above from 2htdp/image requires at least two arguments. I.e. > > procedure > (above i1 i2 is ...) → image? > > Recently I have been using an idiom where I stack a list of one or more > images. > I.e. > > (apply above list-of-images) > > I have a workaround, but it seems to me that a more tolerant "above" -- which > acts as an identity when passed a single image -- would be an improvement to > the > library. > > Similarly for beside, etc. > > > Dan > > ____________________ > Racket Users list: > http://lists.racket-lang.org/users > > > ____________________ > Racket Users list: > http://lists.racket-lang.org/users ____________________ Racket Users list: http://lists.racket-lang.org/users
