caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Andrew Lawson <andrew@absentis.com>
To: caml-list@inria.fr
Subject: Re: [Caml-list] Passing self to a new object
Date: Mon, 29 Oct 2001 20:30:36 +0000	[thread overview]
Message-ID: <20011029203036.A1039@alba.sw> (raw)
In-Reply-To: <20011029180207D.garrigue@kurims.kyoto-u.ac.jp>; from garrigue@kurims.kyoto-u.ac.jp on Mon, Oct 29, 2001 at 06:02:07PM +0900

On Mon, Oct 29, 2001 at 06:02:07PM +0900, Jacques Garrigue wrote:
> From: Andrew Lawson <andrew@absentis.com>
> 
> > 	I'm writing a gui program where the callback for a button
> > creates a new object (also a gui) and needs to pass it a reference to
> > itself in order that the new object can contact the original for
> > information. The classes look something like this;
> > 
> > class xyz = 
> > 	let top = ... in
> > 	let ... = ... in
> > 	object (self)
> > 	  method btnNew  = ... ~command:(fun () = new abc self) top in
> > 	...
> > 	...
> > 	end
> > 
> > and abc (parent:xyz) =
> > 	object (self)
> > 	var myparent = parent
> > 	...
> > 	end
> > 
> >My error is;
> > .....
> > Self type cannot escape its class
> 
> You've bumped into a painful quirk of ocaml's object system: self's
> type is not xyz, and must be handled with care.
 
	Ouch, I knew that even such an otherwise nice language would turn
out to have some fuzzy bits someday :)
 
> (By the way, your program seemed strange, since you were using self
> and abc before defining them, but I suppose it was a copy error, and I
> corrected it.)

	I admit this was a quick example of the problem, not the program
itself, which has many more hairy bits.
 
> The general solution to this is to first define a class type:
>   class type xyz_t = object ('self)
>      method 
>        btnNew : ...
>        ...
>      end

	Well this gives me an excuse to start playing with types and
interfaces  anyway.  Is this a 'known problem, will be sorted
eventually' or a 'just live with it' sort of thing?

> Then change the code in btnNew to
>   new abc (self :> xyz_t)

	 Casting! and here was I giving a C++ programmer some abuse
about this last week :)

> Remark that making xyz_t exact may be a bit lengthy, but you may leave
> out methods you don't need to call from abc.

	Well, thats not so bad then ... I suppose good style should ahve
me doing this anyway.

> Hope this helps,

	Most definitely

> Jacques Garrigue

		Thanks very much

			Andrew

-- 

   Andrew Lawson
andrew@absentis.com
-------------------
Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ: http://caml.inria.fr/FAQ/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr


  reply	other threads:[~2001-10-29 19:32 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-10-28 15:56 Andrew Lawson
2001-10-29  9:02 ` Jacques Garrigue
2001-10-29 20:30   ` Andrew Lawson [this message]
2001-10-30  8:09     ` 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=20011029203036.A1039@alba.sw \
    --to=andrew@absentis.com \
    --cc=caml-list@inria.fr \
    /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).