[R6RS] Unicode SRFI - responses needed

[Michael Sperber]

Thanks for the exceptional job summarizing the discussion!

On a general note, some people have noted the desire to make very
small implementations of Scheme (on embedded systems, say) and still
be standards-compliant.  I support that in the sense that these
implementations should be able to say what subset of the standard they
support in the terminology provided by R6RS---possibly just by saying
"we implement the libraries such and such" given a suitable
partitioning of what's in R6RS into libraries.

Concretely, this means that I think subsets should be supported, and
that the -ci procedures and string case mapping should live in a
distinct library.

On everything else, I agree with you.

Details:

Matthew Flatt <mflatt at cs.utah.edu> writes:

>  * Does anyone doubt that we really want to pin down the definition of
>    character as "Unicode scalar value"? (I still don't.)

No.

>  * Does anyone want to argue that supporting a subset of Unicode might
>    count as standard-compliant? (I think that it's not necessary to
>    allow this in the standard.)

Yes.  This would a) enable (well, not disable at least) very small
implementations, and b) allow implementations to provide a high degree
of interoperability with, say, a Java environment.  (I don't know
about the CLR, but I suspect the same is the case there.)

We may not want to allow arbitrary subsets, but could just specify
what happens on a Latin-1-only or a UCS-2-only subset.

>  * Is anyone unhappy with slightly more complex string operations that
>    take into account non-1-to-1 conversions? (I think I'm happy with
>    this, and I'll implement it today to be sure.)

I'm not unhappy---I was meaning to implement this stuff as well this
week.  Any work we can share?

>  * Who wants to keep character-based comparison and conversion
>    operations? (I do.)

I do, too.

>  * How many editors want to keep here strings? How many would prefer to
>    see them go? (I'm now inclined to get rid of them.)

I'm neutral personally, but vote for getting rid of them to simplify
the standard and the discussion.

-- 
Cheers =8-} Mike
Friede, V?lkerverst?ndigung und ?berhaupt blabla