From: "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>
Subject: Re: Partial type specifications and their ordering
Date: Tue, 18 Mar 2008 09:40:54 +0100
Date: 2008-03-18T09:40:53+01:00 [thread overview]
Message-ID: <4yntq9j6q827.kusbb0olb916.dlg@40tude.net> (raw)
In-Reply-To: f2c01e91-8795-44af-869c-17c936553907@e10g2000prf.googlegroups.com
On Mon, 17 Mar 2008 16:14:42 -0700 (PDT), Eric Hughes wrote:
> My comment earlier today got my mind in a buzz on the topic of partial
> types, so as a form of personal exorcism I wrote a skeleton draft.
[...]
Your draft does not explain why certain sets of types (called partial here)
cannot form a proper class.
My guess is that any set of types can be associated with a class. The
procedure is a follows. You construct the intersection of the interfaces of
the types from the set. (The set is countable infinite, so it should be
possible to do) The result is the interface of the root. The relation "S
derived from T" is obviously preserved on the class.
Consequently, generic types (not Ada term, but the meaning is obvious) are
fully equivalent to classes. The only difference is that the former do not
have T'Class and thus lack corresponding polymorphic values (class-wides).
IMO the difference is not in the semantics.
--
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de
next prev parent reply other threads:[~2008-03-18 8:40 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-03-17 23:14 Partial type specifications and their ordering Eric Hughes
2008-03-17 23:20 ` Eric Hughes
2008-03-18 8:40 ` Dmitry A. Kazakov [this message]
2008-03-18 14:30 ` Eric Hughes
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox