VMD-L Mailing List

From: Axel Kohlmeyer (akohlmey_at_cmm.chem.upenn.edu)
Date: Wed Aug 08 2007 - 09:40:14 CDT

On Wed, 8 Aug 2007, Ondrej Sramek wrote:

hi ondrej,

OS> Hi everyone,
OS> I'm working on molfile plugin for CIF, but I'm stil experiencing few problems:

OS> - I'm doing it under WIN XP, so I'm compilling my molfile into .dll,
OS> but VMD only react on .so , so I have rename the suffix. Is it OK?

it would probably be best to do the development inside the
compiling environment that is used by VMD directly. if you
copy/modify the from some other, similar plugin, it will work
right out of the box. also, it is strongly advisable to do
a cross-check on a unix-like machine (linux would be the easiest),
as this will help to make the code more portable.

OS> - does any reasonable documentation exists on how the plugin API of
OS> VMD communicate with the plugin?

have you looked at?
http://www.ks.uiuc.edu/Research/vmd/plugins/doxygen/molfileplugins.html

also, reading a similar plugin source (e.g. xsfplugin) would give you
some ideas. since you are dealing with crystal structures, i would like
to turn your attention to the required rotation of atom coordinates to
make PBC display working with cells that are not aligned the way VMD
expects it.

OS> - I have already completed that plugin, but I have still problems
OS> with VMD - I can see that plugin, I can find a .cif file and try to
OS> load it, but during loading, it crashes after loading timesteps.
OS> What is the next calling? What am I dong wrong?

how did you test the plugin? again, have a look at the other
plugin sources. not all of them are examples of great programming,
but you should get the general idea. to know what is going wrong,
usually running VMD under a debugger can help to locate the
origin of the failure. based on my own experience, i doubt, that
your plugin is completed (rather the contrary). writing code that
will work correctly on all supported platforms of VMD can offer
some unexpected challenges.

OS> Can anyone help, please?

the easiest way to assist you would be, if you register for the
VMD biocore project and upload your sources and example inputs
(have you implemented the full CIF spec? that is a quite complex
file type, if i remember correctly) to the biofs. this way you
can communicate more directly with developers and have others
make suggestions on improvements.

cheers,
   axel.

OS> Looking forward to hearing from you, guys.
OS>
OS> K.B.
OS> 

-- 
=======================================================================
Axel Kohlmeyer   akohlmey_at_cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.