On Tue, 2009-10-13 at 23:42 -0400, John Cowan wrote:
> Ray Dillinger scripsit:
>
> > This approach (import as a runtime binding form) completely
> > eliminates the need for additional machinery such as cond-expand
> > etc, which can be WG2-only. All WG1 needs is a general, runtime
> > import procedure (not syntax) and a module-available? predicate.
>
> That's not import, it's load. Nobody (or any rate not me) is proposing
> to remove load from R5RS, though I have put it in a feature group^W^Wmagic
> module. But people can ignore magic modules.
Hm. The distinction I had been drawing between the desired
functionality of "load" vs. "import" is that "import" is aware
of and can be used for names defined as libraries/modules
(whose implementation may be binary or FASL or whatever)
whereas "load" is aware of and can be used for filenames
(implicitly files of scheme source code) only.
If "load" gets specified such that both library names and
filenames are acceptable targets, rather than just filenames,
then I will be happy with that answer and declare categorically
that WG1 scheme needs no separate "import" form so WG2 can
restrict "import" in whatever way makes the greatest contribution
to its users' happiness.
Bear
_______________________________________________
r6rs-discuss mailing list
[email protected]
http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss
