[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Scheme-reports] read-error? and file-error? should be part of their respective packages.

If library writers create alternative api for files and the reader, they 
should provide their own error facilities. These 2 procedures are for 
the standardized (scheme file) and (scheme read) libraries. I believe 
that the WG should reconsider having these procedures in (scheme base) 
for facilities that have been manifestly declared optional. It seems 
self-evident that the location of these procedures is contrary to the 
intent of the design of the R7 report.

On 11/13/12 10:29 PM, Aaron W. Hsu wrote:
> Alex Shinn <alexshinn@x> wrote:
>> Since all other exception handling and introspection is
>> in base, and you want to be able to detect these even
>> if not using file/read yourself, I'm not sure this move
>> would even be desirable.  We would need to vote on it.
> In particular, I imagine that library writers would want to be able
> to handle these sorts of errors even if they are writing the library
> on Schemes that do not provide standard read libraries or the like.
> The counter argument to this is that the COND-EXPAND facility would
> make it easy to write versions for either case, regardless as to
> whether these predicates were required and in the base.
> Nonetheless, I agree with Alex that this is not as clear cut as
> I think many of us would want, and as such, I think we can't change
> the draft with this sort of thing at this point. On the other hand,
> it may be worth discussing this some more and spending some time
> working out the kinks on the idea to make an erratum to the draft or
> something to that effect.

Scheme-reports mailing list