From: Markus Mottl <markus.mottl@gmail.com>
To: yminsky@cs.cornell.edu
Cc: Caml Mailing List <caml-list@inria.fr>
Subject: Re: [Caml-list] Phantom types and read-only variables
Date: Sat, 5 Feb 2005 14:21:13 -0500 [thread overview]
Message-ID: <20050205192113.GA26897@mobile> (raw)
In-Reply-To: <891bd33905020508504e272acf@mail.gmail.com>
On Sat, 05 Feb 2005, Yaron Minsky wrote:
> I'm trying to use phantom types to build a "freezable" variable, where
> I can create a version of the variable to which write operations can
> not be applied. Here's my first attempt, which was rejected by the
> compiler:
[snip]
> I do basically understand why the compiler rejects module N. It
> basically complains that the freeze in M is not compatible with the
> constraints on N. In particular:
>
> Values do not match:
> val freeze : 'a -> 'a
> is not included in
> val freeze : 'a t -> ro t
>
> So, what's the right approach here?
Hm, it seems to me that the compiler could do a better job here.
Instead of writing:
type 'a t = { mutable value: int }
write:
type v = { mutable value: int }
type 'a t = v
This should make the code compile.
I think the compiler should be able to infer automatically that 'a isn't
used on the right hand side of the record definition (i.e. that it is
just a phantom type) without using the hint of a separate monomorphic
record definition. I guess this feature should be trivial to add.
Maybe in the next release? :-)
Regards,
Markus
--
Markus Mottl markus.mottl@gmail.com http://www.ocaml.info
next prev parent reply other threads:[~2005-02-05 19:21 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-02-05 16:50 Yaron Minsky
2005-02-05 17:24 ` Yaron Minsky
2005-02-05 20:21 ` Ethan Aubin
2005-02-05 18:34 ` [Caml-list] " Remi Vanicat
2005-02-05 19:21 ` Markus Mottl [this message]
2005-02-06 0:48 ` Jacques Garrigue
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20050205192113.GA26897@mobile \
--to=markus.mottl@gmail.com \
--cc=caml-list@inria.fr \
--cc=yminsky@cs.cornell.edu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).