[ssax-sxml] Re: testing against eof-object value believed non-portable SSAX code

[<ol...@po...>]

Hello!

> 2. R5RS provides an eof-object? predicate, not a eof-object constant, 
> and says "The precise set of end of objects will vary among 
> implementations, [...]".
> I.e. I believe code to be in error which invokes
>   (equal? token expected-token)) ; where expect-token is the above 
> eof-object
> because it may not work in the presence of multiple EOF objects.
> [This is not a problem in Kawa which has a unique such object: #!eof]

It seems the corresponding R5RS passage

  "The precise set of end of file objects will vary among
   implementations, but in any case no end of file object will ever be an
   object that can be read in using read."

is somewhat ambiguous. I have mustered all my boldness and asked a
R5RS author for a clarification. He has very kindly replied that the
ambiguity is indeed present, and in his interpretation, each Scheme
system will likely to choose _one_ EOF object (but different
implementations will generally choose different EOF objects). It's
been my impression that all R5RS systems I'm aware of chose only one
EOF marker.

	Cheers,
	Oleg