caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Gabriel Scherer <gabriel.scherer@gmail.com>
To: Reed Wilson <cedilla@gmail.com>
Cc: caml users <caml-list@inria.fr>
Subject: Re: [Caml-list] Type generalization confusion
Date: Mon, 8 May 2017 20:49:49 -0700	[thread overview]
Message-ID: <CAPFanBHJsRy1YXsHJpL66BDr4pq-BSc=Pdk=8Tg+uQ98ogKnfA@mail.gmail.com> (raw)
In-Reply-To: <CALLFq5SVR_mLGkof-CF4T1ud_qZn0-YKU4O-4OpeWiGNnRGiRQ@mail.gmail.com>

This is the value restriction at work: only values are generalized
(often of the form "fun x -> ..."), and while (Int 3) is a value, (Int
(1 + 2)) is not.

On Mon, May 8, 2017 at 11:32 AM, Reed Wilson <cedilla@gmail.com> wrote:
> I've been working with some of the new features introduced with GADTs, and
> ran into a confusing instance of the "type variables that cannot be
> generalized" error.
>
> The simplified program is as follows:
> type _ field = Int : int -> 'a field
> let a = Int 3
> let b = Int (1 + 2)
> let inside = 1 + 2
> let c = Int inside
>
> ocamlc -i returns:
> type _ field = Int : int -> 'a field
> val a : 'a field
> val b : '_a field
> val inside : int
> val c : 'a field
>
> with b remaining non-generalized. The problem I'm having with it is that the
> type variable doesn't depend on the value at all, so I don't see how that
> can prevent generalization.
>
> Also, the manual says the reason some types aren't generalized is due to
> "polymorphic mutable data structures". Nothing I created is mutable, so why
> was generalization turned off in the first place?
>
> Finally, I'm confused why separating the function from the definition is
> enough to fix this; c is generalized simply by defining 1+2 in a separate
> value (which must be global, apparently).
>
> Thanks,
> Reed Wilson
>
> --
> ç

  reply	other threads:[~2017-05-09  3:50 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-08 18:32 Reed Wilson
2017-05-09  3:49 ` Gabriel Scherer [this message]
2017-05-09  5:43   ` Reed Wilson
2017-05-09  8:54     ` Leo White
2017-05-09 19:56       ` Reed Wilson
2017-05-09 20:54         ` Jeremy Yallop
2017-05-10 18:14           ` Reed Wilson
2017-05-10 13:29         ` Mikhail Mandrykin

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='CAPFanBHJsRy1YXsHJpL66BDr4pq-BSc=Pdk=8Tg+uQ98ogKnfA@mail.gmail.com' \
    --to=gabriel.scherer@gmail.com \
    --cc=caml-list@inria.fr \
    --cc=cedilla@gmail.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).