Re: [Scheme-reports] Import declarations in programs

[arcfide@x (Aaron W. Hsu)]

Marc Nieper-Wißkirchen <marc.nieper@x> writes:

>> Sorry for being dense here, as I'm a bit out of touch, but which import
>> form are we talking about? I was under the impression that we cleared up
>> any questions of scoping and bindings for the import forms of libraries
>> and the like? I thought that with a library or other program, the import
>> forms were clearly static and outside of the scope of any environments
>> defined by those import forms? I thought that we made it explicit that
>> import forms and the like could appear in any order, as well.
>
> I think this is the main point: the syntax for libraries permit to
> distinguish between the static library language (consisting of import,
> include, begin, cond-expand, etc.) and the Scheme language (what's
> contained inside the begins). The syntax for programs has no way of
> distinguishing between the two different levels after the first import
> declaration.

Not all of us think that the library language and the Scheme language
should be different things. It's also not required that they be
different things in R7RS, though it's not required that they be the
same, either. 

> Therefore I think that more powerful import semantics for programs
> (including cond-expand, etc.) won't look too well until we have a second
> syntax for programs that mimics that for libraries. So for the way programs
> are currently written I see no advantage of more than one import
> declaration even for a future standard.

For a counterpoint to this, see the way that some schemes, including
Chez Scheme, handle imports at any location in the program. 

-- 
Aaron W. Hsu | arcfide@x | http://www.sacrideo.us
לֵ֤ב חֲכָמִים֙ בְּבֵ֣ית אֵ֔בֶל וְלֵ֥ב כְּסִילִ֖ים בְּבֵ֥ית שִׂמְחָֽה׃

_______________________________________________
Scheme-reports mailing list
Scheme-reports@x
http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports