On Nov 6, 2007 10:45 AM, Robert Dodier <robert.dodier at gmail.com> wrote:
> On 11/6/07, Stavros Macrakis <macrakis at alum.mit.edu> wrote:
>
> > quad_qags is sloppily written, and treats non-numerical endpoints as
> > zero rather than throwing an error:...
> It turns out this behavior is implementation-specific....
> GCL happily treats them as 0. You have GCL + Maxima, yes?
Correct!
> The 4th argument is a diagnostic, which can be interpreted by
> looking up the comments in the original Fortran ...
> I agree that's obscure at best, but whatever we do (e.g. throw
> something if the return code is anything except "OK") I think we
> want to preserve the diagnostic in some way.
I agree entirely that the information in the return should be preserved
somehow. Probably the cleanest way is to have quad_qags throw different
errors in the different cases. Are the other parts of the return value ever
useful when the error code is non-zero? That is harder to handle....
-s