On 3/31/12 7:29 AM, Barton Willis wrote:
> ----maxima-bounces at math.utexas.edu wrote: -----
>
>> My copy of the documentation doesn't seem to have any information about
>> what a statement like
>> (%i1) a/b : c/d;
>> should do. The results are quite surprising to me, at least. Is there
>> some logic behind them?
> The assignment (a+b) : (c+d) gives an error. Surely a/b : c/d should also give
> an error. The user documentation doesn't say what happens for assignments to
> expressions such as a/b.
>
> --Barton
> _______________________________________________
>
a/b:c/d parses as a/ (b:c/d) and so what happens is that b is
assigned c/d, and then
the value a/(c/d) or a*d/c is returned.
(a/b):(c/d) gives an error message.
note that a+b:c+d does not give an error.
RJF