caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Jacques GARRIGUE <garrigue@kurims.kyoto-u.ac.jp>
To: jon@jdh30.plus.com
Cc: caml-list@inria.fr
Subject: Re: [Caml-list] Phantom types
Date: Fri, 27 Aug 2004 18:03:28 +0900 (JST)	[thread overview]
Message-ID: <20040827.180328.128873320.garrigue@kurims.kyoto-u.ac.jp> (raw)
In-Reply-To: <200408231616.58139.jon@jdh30.plus.com>

From: Jon Harrop <jon@jdh30.plus.com>

> I should add, using:
> 
>   val add_even_even : [> `Even ] t -> [> `Even ] t -> [> `Even ] t
>   val add_even_odd : [> `Even ] t -> [> `Odd ] t -> [> `Odd ] t
> 
> allows:
> 
> # PhantomInt.add_even_odd i j;;
> val k : _[> `Odd ] PhantomInt.t = <abstr>

This part is OK.

> which can then be misused:
> 
> # PhantomInt.add_even_even i k;;
> - : [ `Even ] PhantomInt.t = <abstr>

But this one is wrong.

> I think this is because [> `Even] means any superset of [ `Even ] whereas [< 
> `Even], which was probably intended, means any subset of [ `Even ]. Indeed, 
> the latter appears to work.

Indeed, one must be careful about variance.
So the right signature would be:
   type +'a t
   type any = [`Even|`Odd]
   val add_even_even : [ `Even ] t -> [ `Even ] t -> [> `Even ] t
   val add_even_odd : [ `Even ] t -> [ `Odd ] t -> [> `Odd ] t
   val add_unknown_unknown : any t -> any t -> any t
(Note the + indicating covariance)

Then you have
# let k = PhantomInt.add_even_odd i j;;
val k : [> `Odd ] PhantomInt.t = <abstr>
(polymorphic result!)
# PhantomInt.add_even_even i k;;
** type error
# PhantomInt.add_unknown_unknown i k;;
- : any t = <abstr>

Jacques Garrigue

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners


  reply	other threads:[~2004-08-27  9:03 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-07-31  8:56 [Caml-list] const equivalent for mutable types? Christopher A. Gorski
2004-07-31  9:24 ` Jean-Marie Gaillourdert
2004-07-31 10:24   ` Jean-Marie Gaillourdert
2004-07-31 10:50   ` Markus Mottl
2004-07-31 14:31     ` Brian Hurt
2004-07-31 15:51       ` Markus Mottl
2004-07-31 17:05       ` skaller
2004-07-31 10:34 ` Markus Mottl
2004-07-31 13:44   ` Jon Harrop
2004-07-31 16:31     ` [Caml-list] Phantom types Markus Mottl
2004-08-23  9:49       ` Jon Harrop
2004-08-23 12:25         ` [Caml-list] Why does ocaml use custom buffering? Daan Leijen
2004-08-23 15:16         ` [Caml-list] Phantom types Jon Harrop
2004-08-27  9:03           ` Jacques GARRIGUE [this message]
2004-08-25 21:03         ` brogoff
2004-07-31 16:35     ` [Caml-list] const equivalent for mutable types? skaller
2004-07-31 17:23       ` [Caml-list] Functional arrays Jon Harrop
2004-07-31 18:45         ` skaller
2004-08-02  5:07           ` brogoff
2004-08-02  7:45         ` Diego Olivier Fernandez Pons
2004-08-05 16:42           ` Daniel Ortmann
2004-08-05 17:02             ` Diego Olivier Fernandez Pons
2004-08-05 17:16             ` Diego Olivier Fernandez Pons
2004-07-31 17:45   ` [Caml-list] const equivalent for mutable types? Chris Gorski
2004-07-31 14:11 ` Brian Hurt
2010-05-17 14:59 Phantom types Thomas Braibant
2010-05-17 16:37 ` [Caml-list] " Goswin von Brederlow

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=20040827.180328.128873320.garrigue@kurims.kyoto-u.ac.jp \
    --to=garrigue@kurims.kyoto-u.ac.jp \
    --cc=caml-list@inria.fr \
    --cc=jon@jdh30.plus.com \
    /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).