VMD-L Mailing List

From: Justin Gullingsrud (justin_at_ks.uiuc.edu)
Date: Sat Aug 17 2002 - 12:01:07 CDT

Hi,

On Sat, Aug 17, 2002 at 11:36:24AM -0500, Pawel Kedzierski wrote:
> On Thu, Aug 15, 2002 at 04:05:55PM -0500, Justin Gullingsrud wrote:
> > Sounds like a cool app! Unfortunately, it looks like the only workaround
> > I have at the moment for adding graphics data is to load a real molecule
> > instead of a 'graphics' molecule. You could create a pdb file containing
>
> This does not work. The graphics.xxx() functions refuse to accept an id
> which is not of a 'graphics' type object. It looks like a dead-end
> -- it is no longer possible to create 'graphics' objects, but the
> Python interface refuses to work on anything else... Do You have
> another idea, what could I try?

There's nothing else I can suggest at the moment through the Python interface.
I'll get that fixed in a future version.

>
> There is a molecule.import_graphics() function but the user guide
> suggests that it also do not create 'graphics' objects. The
> info, how to use it, is missing anyway -- what file types does it
> accept?
>
> What was really the reason behind disabling the old functionality
> of molecule.load('graphics', name)? Apart from it should not complain
> about nonexistant file, I was rather useful, as was the concept
> of 'graphics'-only objects. Oh well, maybe it would be more consistent
> to have graphics.create().

We're doing a lot of work behind the scenes to generalize what kinds
of things VMD is able to treat as a molecule. VMD 1.7.1 and earlier
treated molecules containing atoms differently than molecules containing
graphics, and had no way of dealing with volumetric data. Now there
is no distinction: you can load graphics, atoms, and volumetric data in
any order into a molecule and it should happily accept it. In the Tcl
interface we kept backward compatibility with the "mol load graphics"
command by having VMD create an "empty" molecule for the user; we
forgot to do this for the Python interface. As you can see, it no
longer really makes sense to create something called a "graphics"
molecule, since there's no longer any such thing as far as the user
is concerned. So we probably need a new command, something along the
lines of "molecule.new()".

I'll let you know when we've something for you to try.

Cheers,

Justin

>
> Greetings,
>
> Pawel Kedzierski
>
> --
> Pawel Kedzierski, kedziers_at_ccmsi.us
> Chemistry/CREST, Jacson State University 

-- 
  Justin Gullingsrud        3111 Beckman Institute        217-244-8946
  I been dropping the new science, and I be kicking the new knowledge,
  and I'm seeing to a degree that you can't get in college.  -- b.boys