[r6rs-discuss] Issues with section 8.2. Port I/O, a.k.a. (rnrs io ports (6))

[John Cowan]

Michael Sperber scripsit:

> Does "line for flushing upon line endings or other
> implementation-defined separators and reading up to line endings or
> other implementation-defined separators" work?

I'd rather go for "line for flushing upon line endings and reading up
to line endings, or other implementation-dependent behavior.  On some
ports there simply will be no concept of a line ending; in other cases
variable-sized buffering will be no more efficient than no buffering; in
still other cases it will be impractical to change hardwired underlying
assumptions like "line mode = LF only".

> > I think it would be within editorial discretion to extend the license
> > in 8.2.4 to provide implementation-dependent results on nonstandard
> > eol-symbols to buffer-mode-symbols as well.
> 
> That I don't think I can do.  I wish the possibility of other buffer
> modes had been suggested prior to the ratification candidate.

I suppose you are right.  In which case buffer-mode? is indeed just
(lambda (s) (if (memq s '(none line block)) #t #f), which is rather
foolish.

Bah.

-- 
There is / One art                      John Cowan <cowan at ccil.org>
No more / No less                       http://www.ccil.org/~cowan
To do / All things
With art- / Lessness                     -- Piet Hein