Re: User define attributes

[Brian Rogoff <bpr@shell5.ba.best.com>]

On Mon, 16 Apr 2001, Martin Dowie wrote:
> been there; done that; got the compiler listing... :-)

:-)

> I suspect there are many companies out there with very
> similar generics.
> 
> I suppose we have to ask the question - if it is that
> easy (and relative to the actual application, it
> usually is!) then why isn't 'Image for composite types
> part of the language?..

Given the desire to make array and function call syntax identical, I
wonder why there is even an attribute syntax at all, rather than just 
having function call syntax and system defined functions. I'd much rather 
just write Image(X) everywhere (but write A[I] for array access :). 

-- Brian
 
> Samuel T. Harris <u61783@gsde.hou.us.ray.com> wrote in message
> news:3ADB033A.F10344A9@gsde.hou.us.ray.com...
> > It is trivial to build a generic which implements the a
> > width, value, and image function for arrays. All it needs
> > are the width, value, and image functions for the indexes
> > and composite types.
> >
> > It is somewhat harder to build a generic for records
> > since the "index" are the field names and the composite
> > function depend upon which field is in use. But it
> > can be done.
> >
> > I routinely write width, value, and image functions
> > for all my important types. Having these generics
> > in place, I get these routines with some simple
> > handler code and a generic instantiations. With them
> > I get the following advantages ...
> >
> > 1. I can easily insert put_line kinds of debug code.
> > 2. If supported, I can call them within my debugger.
> > 3. I can defer aggregrate initialization of variables
> >    to file_io during elaboration, and vice-versa.
> 
> 
> 
>