[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
re: portability of .fas files
Don Cohen writes:
> I had not even realized that there were macros that expanded differently
> on different platforms. Is this really necessary? How many such macros
Well, IMHO you can't have everything a function that can hide system
dependencies. Especially when you need some clean-up forms, macros
come in handy (see user1.lsp:with-keyboard).
> are there and how hard would it be to encapsulate the differences inside
> functions instead? I had definitely hoped for .fas files to be shared
Don't know. I Didn't grep all #+/-UNIX/DOS/AMIGA/... features. I would
not expect it to be trivial.
One could reasonably assume that Common-Lisp code could be portable
(see J. Schrod's post on this). Good system design would give you only
a few files where specific machine dependencies reside (either within
an implementation on different platforms, or for different
implementations, e.g. CMU, Allegro and others). They would need site-
and user-specific adaption and compilation.
> at least between DOS and UNIX machines.
Why just these? Because these are the only ones that you are concerned
about? I wouldn't want a CLISP were there would be the kind of
"full-featured" version and another one for the "poor".