caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Aaron Bohannon <bohannon@cis.upenn.edu>
To: Mark Shinwell <mshinwell@janestcapital.com>
Cc: caml-list@inria.fr
Subject: Re: [Caml-list] annotations and type-checking
Date: Wed, 29 Jul 2009 09:41:34 -0400	[thread overview]
Message-ID: <c413fcb70907290641v4e040978v227d4fd3a8531370@mail.gmail.com> (raw)
In-Reply-To: <20090729074050.GK19609@janestcapital.com>

Thank you for that link.  To boil it down, it seems (1) type variables
annotating top-level declarations are ignored, and (2) type variables
annotating local bindings are treated existentially (as if one had
written '_a, although that name itself is considered syntactically
ill-formed).

So if OCaml cannot do anything better than this, then why are type
variables even syntactically legal in annotations?  If backwards
compatibility is the issue, could it not at the very very least give a
compiler warning when they are used?

 - Aaron

On Wed, Jul 29, 2009 at 3:40 AM, Mark
Shinwell<mshinwell@janestcapital.com> wrote:
> On Tue, Jul 28, 2009 at 05:47:25PM -0400, Aaron Bohannon wrote:
>> Why do the first two programs type-check but the thrid one does not?
>
> Dark corners of the type system.
>
>> let f (x : 'a) : 'a = x in (f true, f 3);;
>
> Explicit type variables in this situation are considered "global".  They are
> not generalized until the type of the whole toplevel declaration has been
> determined.  Consequentially, during type-checking of the body of your
> let expression, 'a is not a generalized variable.
>
> There is more detail on similar situations here:
>
> http://caml.inria.fr/pub/ml-archives/caml-list/2002/06/a03da53be62c12671a891708c51e85f9.en.html
>
> Mark
>


  reply	other threads:[~2009-07-29 13:41 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-07-28 21:47 Aaron Bohannon
2009-07-28 23:28 ` [Caml-list] " Jon Harrop
2009-07-28 23:04   ` Philippe
2009-07-29  5:38     ` Johannes Kanig
2009-07-29  7:40 ` Mark Shinwell
2009-07-29 13:41   ` Aaron Bohannon [this message]
2009-07-29 14:39     ` Jacques Garrigue
2009-07-29 16:50       ` Aaron Bohannon
2009-07-29 19:01       ` Jon Harrop

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=c413fcb70907290641v4e040978v227d4fd3a8531370@mail.gmail.com \
    --to=bohannon@cis.upenn.edu \
    --cc=caml-list@inria.fr \
    --cc=mshinwell@janestcapital.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).