Hi,

[email protected] wrote:
>> If it were a line drawing. But there are cylinders and spheres
>> representing bonds and atoms. And then there is the question of sorting
>
> How are those primitives described? As polygon meshes or centerpoint(s)
> with radius?

In general as centerpoint/radius combination, merged with union.

> It would be great to have a look at an example...

I work in small molecule crystallography and we are often interested in
packing of small organic molecules. There are several software packages
that would write pov. The most prominent is mercury by the CCDC. It is
not open source (I hope you continue to read) but free to academia.

http://www.ccdc.cam.ac.uk/products/mercury/

There are others, that are open and do write pov files, but their user
interface is more awkward.

The pov output is scripted and contains calls like:

ccdc_draw_solid_sphere( < 0.0270702, 2.5544, 0.703669 >, 0.1, rgb < 1,
1, 1 > )
ccdc_draw_closed_cylinder( < -1.37106, 3.46269, -1.3188 >, < -1.53974,
3.85874, -1.23142 >, 0.1, rgb < 0.941176, 0, 0 > )

which will resolve to something like:

#macro ccdc_draw_solid_sphere( position, sphere_radius, sphere_color )
    sphere {
        position, sphere_radius
        ccdc_set_solid_material_properties( sphere_color )
    }
#end

There are some transformations going on to use the crystallographic
symmetry.

>> these elements into a sequence that the overlapping bonds do look ok.
>
> Off the top of my hat: Ascending order by z and then use the Painter's
> algorithm. But for a clean 2D layout there is probably more to do, so that
> all bonds can be seen clearly in 2D space.

Yes, I guess.

> Can your program export those structures in some chemical format, like
> V2000 molfile? Then it might be easier to use e.g. the CDK to generate
> real 2D stuctures from that...

Well, I do not know the V2000 molfile. Do you have a link?
The problem is usually that I am interested in figures not of single
molecules but of symmetry derived arrangements of 5-20 molecules linked
by hydrogen bonds. Most "intelligent descriptions" do not cover this.

Regards,
-- 
Dr. Michael T. Kirchner

Institut für Anorganische Chemie
Universität Duisburg-Essen
Universitätsstr. 5-7
45117 Essen

[email protected]
Tel +49 201 183 4139
Fax +49 201 183 7342
Raum S07 S00 C24

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Blueobelisk-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/blueobelisk-discuss

Reply via email to