[r6rs-discuss] [Formal] allow port position to be "magic cookie"

[Per Bothner]

Jon Wilson wrote:
> Perhaps I've missed something, but it doesn't seem immediately obvious 
> to me that a straight-forward wrapping of C's fgetpos/fsetpos is 
> desirable.  Not that it is a bad thing, but I don't see why this 
> criterion should motivate any particular design decision.  Does having a 
> straightforward wrapping further any of the guiding principles listed in 
> the introduction of the report?  Furthermore, does having a 
> straightforward wrapping further those guiding principles better than or 
> more than a design decision which does not result in a straightforward 
> wrapping of fgetpos/fsetpos?

If you want to provide the functionality of being able to seek,
and if you accept there are some systems that require the fpos_t
magic cookies (as used by fgetpos/fsetpos), and if you want a
portable implementations, then using fgetpos/fsetpos gets you
the required functionality in a portable manner.

I don't know if that answers your question.
-- 
	--Per Bothner
per at bothner.com   http://per.bothner.com/