On Monday, March 3, 2003, at 04:13 AM, Michael Sperber [Mr.
Preprocessor] wrote:
>
> The behavior upon EOF of the R5RS I/O primitives is already defined
> to be non-exceptional, so that doesn't make sense.  It might be
> feasible to create a subtype of &READ-ERROR, but that doesn't seem
> very useful.
>

Ahh, reading the 'Abstract' section of the SRFI document again I see
that it was
designed specifically and explicitly for R5RS I/O primitives to raise,
and thus
nothing not related to them was in it.  (In some code I was writing a
while ago,
instead of quietly returning an EOF object when it was met while
parsing from a
stream, I intended to raise an EOF-related condition, and thought that
perhaps
I should look in this SRFI for something of that sort.)

> Taylor> or should this SRFI be amended?
>
> Note that there's no way to amend a SRFI after it's been finalized.
>

True, but after noting the existence of a post-finalisation mailing
list for
each SRFI, I was put under the impression that perhaps something
vaguely similar
to that could maybe be done.  Apparently not.

> --
> Cheers =8-} Mike
> Friede, Völkerverständigung und überhaupt blabla
>