William D Clinger wrote:
I am posting this as an individual member of the Scheme
community. I am not speaking for the R6RS editors, and
this message should not be confused with the editors'
eventual formal response.
Thomas Lord wrote:
I'd given some thought to that. You could describe it by saying
that CHAR->INTEGER *must* be total over CHAR and *may* be
a many-to-one function while INTEGER->CHAR *must* be
non-divergent only for numeric scalar values and, over that
domain, *must* be an injection into CHAR.
I did not describe what you just described. I described
an alternative in which "an implementation could add bucky
bits to characters while making those bits invisible to
all of the standard operations on characters and strings."
What you are describing would make those bits visible to
the standard operations.
???
I don't think so. Aren't you saying that you would want
CHAR->INTEGER to be unable to distinguish #\A from
a hypothetical, permitted-not-required, #\M-H-A? That
would leave CHAR->INTEGER total over CHAR and,
in the presence of those extensions, many-to-one.
If bucky-bits are to be "invisible to standard operations" then
I gather you'd have INTEGER->CHAR remain restricted
to scalar value integers, making it an injection into the
extended CHAR type.
The language
of 5.92 seems to me to explicitly forbid this interpretation but
you have brought into consideration so let's see where it goes.
I did not propose anything that would violate the language
of the 5.92 draft.
Perhaps you could unpack things a bit, then. 5.92 seems to me to
prohibit implementations which have more characters than there are
Unicode scalar values:
/ Characters/ are objects that represent Unicode scalar values [...].
[....]
Given a character, char->integer returns its Unicode scalar value as
an exact integer.
[....]
These procedures impose a total ordering on the set of characters
according to their Unicode scalar values.
All three but especially the last statement are particularly damning.
A total ordering according to scalar values clearly implies that the
CHAR type contains no more values than there are scalar value
integers.
Finally, I think there is a debate implicit in this explicit debate.
It seems to me that you and Cowan would like Scheme programs
to be type-safe in an unprecedented way that goes beyond the
type safety goals of the report:
I appreciate the joke.
It's funny, yes, but not simply a joke. Perhaps it failed but I'm
trying to figure out a concise principle on which we can agree that
transcends but also resolves the debate.
-t
Will
_______________________________________________
r6rs-discuss mailing list
[email protected]
http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss
