Kent Budge wrote:
> Jim Edwards wrote:
>
> > A final thought on this is that in generic programming
> > the code is there for all to see. I wonder how we
> > would think about the math library if we had access to
> > the code? For this reason, I would like to see a
> > library of template code that has easily understood
> > headers and a pre compile object that the compiler
> > understands but is not read by the user. Not that I
> > want to keep my code a secret, but I would like to
> > present the end user with only the data they need to
> > understand. In line functions should also be included.
>
> I've seen at least one implementation of STL where the nonpublic
> portions of the headers files were encrypted in a way that the
> compiler recognized and could decrypt. My recollection is probably
> off, but it seemed to me this was the Rogue Wave headers with the
> Borland Visual C++ compiler. So it's been done at least once. I
> will be interested also to see how the export keyword is implemented
> by different vendors.
The Rogue Wave encrypted files are actually fairly annoying for those
of us who tend to use the STL in a lot of our code. Whenever something
goes wrong or the compiler cannot find some function that you expected
to be provided, you cannot look at the file and figure out the problem.
I do not know of any compiler vendors that are even considering an
implementation of the "export" keyword anytime soon. I am not sure
if it is really possible, though it sounds great in theory. But with some
care, I think you can design codes such that the headers are intelligible
and the implementation details are out of plain sight.
Julian C.
-- Julian C. Cummings Advanced Computing Laboratory Los Alamos National Laboratory (505) 667-6064 julianc@acl.lanl.gov--------------------- Object Oriented Numerics List -------------------------- * To subscribe/unsubscribe: use the handy web form at http://oonumerics.org/oon/ * If this doesn't work, please send a note to owner-oon-list@oonumerics.org
This archive was generated by hypermail 2b29 : Mon Apr 10 2000 - 12:47:56 EST