Re: Declaring super types

From: David BL <davidbl_at_iinet.net.au>
Date: Thu, 6 May 2010 16:51:47 -0700 (PDT)
Message-ID: <dbf2f5dd-0866-4bc3-95fc-6261bfce2747_at_t26g2000prt.googlegroups.com>



On May 7, 1:28 am, Erwin <e.sm..._at_myonline.be> wrote:
> On 20 apr, 04:13, David BL <davi..._at_iinet.net.au> wrote:
>
> > So they can
> > *explain* what implicit coercions should be allowed and what aren't.
>
> My understanding was that they opposed to coercions alltogether.

I meant "coercion" in the context of a statically typed imperative language.

  e : ELLIPSE;
  e := CIRCLE( POINT(0,0), 1);

Coercion has to do with static type analysis, and is not meant to suggest for example that a variable can be assigned a different value to what is evaluated in the right hand expression. Received on Thu May 06 2010 - 18:51:47 CDT

Original text of this message