From: Olaf Lenz (olenz_at_icp.uni-stuttgart.de)
Date: Tue Feb 04 2014 - 08:32:49 CST

Hi everybody!

Sorry if you get this mailing twice. In the last mailing, I have attached a
DLL file, which caused the mailing to be banned by some systems.

After I have found out that Microsoft does actually have a free compiler
version, I have taken the time to set up a devel environment in windows,
and I have set out to debug the code. The problem was that sscanf does not
really work the same in Windows and Unix, at least when "%n" is involved.

I have fixed the problem, and furthermore I have learned how to use Visual
C++ and how to compile a DLL. I have tested the plugin with VMD on Windows,
and I was able to open your sample file and others that failed as well.

At
  https://github.com/olenz/vtfplugin/releases/tag/wintest
you can get the dynamic library and its source code. Put the DLL into some
directory on your machine and use

  vmd_plugin_scandirectory <dir> *.dll

in vmd.rc to load it. Ming-Tsung, would you please test whether the plugin
works for you as well?

John, I would be happy if you would add the updated plugin to the VMD repo.

Olaf

2014-02-04 Axel Kohlmeyer <akohlmey_at_gmail.com>:

> hi,
>
> here is a binary for win32 cross-compiled from the unmodified latest
> vtfplugin sources on linux. this version does not support the Tcl
> functions and transparent gzip decompression. please uncompress it,
> make a backup copy of the original and replace it with this file and
> give it a try.
>
> axel.
>
> On Mon, Feb 3, 2014 at 5:36 PM, John Stone <johns_at_ks.uiuc.edu> wrote:
> > Hi guys,
> > Most likely this is an issue with uninitialized data, a buffer overrun,
> > or something similar. Have you done tests with the failing input using
> > valgrind on Linux? If not, I'd try that first. If that works out fine,
> > then it's possible we're dealing with something that is indeed specific
> > to Windows, but most things that fail on Windows will also show up in
> > testing under Linux if one makes careful use of valgrind etc. I will
> > soon be making test builds of VMD 1.9.2 beta versions on Windows, so it
> > will be easier to track down anything that is truly unique to Windows
> > once I have those builds prepared.
> >
> > Cheers,
> > John
> >
> > On Mon, Feb 03, 2014 at 07:09:13PM +0100, Olaf Lenz wrote:
> >> Hi!
> >>
> >> The trouble is that I do not have access to a Windows development
> system.
> >> With your message, I can have a look at the code again, but I am
> unable to
> >> compile or debug under Windows.
> >>
> >> Olaf
> >>
> >> Dr. rer. nat. Olaf Lenz
> >> Institut fA 1/4r Computerphysik, Allmandring 3, D-70569 Stuttgart
> >> Phone: +49.711.685.63607
> >>
> >> Am 03.02.2014 18:51 schrieb "Ming-Tsung Lee" <mtlee_at_rci.rutgers.edu
> >:
> >>
> >> Hi Olaf,
> >> http://www.ks.uiuc.edu/Research/vmd/mailing_list/vmd-l/22262.html
> >> I think that I probably find the bug. In the VTF file generated by
> >> DL_MESO (attached, not sure the situation of ESPresSo). In the
> first
> >> frame of the trajectory, box dimensions are defined as follows
> >> timestep indexed
> >> pbc A 15.000000 15.000000 15.000000
> >> A A A A 0 10.867398 12.150888 14.039090
> >> A A A A 1 A 9.467130 10.474655 12.615115
> >> A A A A 2 A ...
> >> I have checked the default syntax of VTF. The box angles should be
> >> optional.
> >> https://github.com/olenz/vtfplugin/wiki/VTF-format
> >> I have played that around. It looks like that VMD is frozen while
> >> loading the line pbc/unitcell, as long as there are no "90 90 90"
> >> assigned. Nevertheless, the same VTF works perfectly well with the
> Linux
> >> version VMD. Would you please check your plugins for both
> platforms?
> >> Probably something is messed up in the Windows version. Thank you.
> >> For those who are using DL_MESO and want a quick solution for this,
> >> simply modify the trajectory output by putting a "90 90 90" after
> your
> >> box dimensions. It will work.
> >> Best,
> >> MT
> >> P.S. I am using Windows 7 x64 enterprise, VMD 1.9.1. Trajectory is
> >> generated from DL_MESO/DEMO/Polyelectrolyte/HISTORY, using
> >> /DL_MESO/DPD/utility/trajectselected.f90A
> >> Ming-Tsung Lee
> >> Department of Chemical & Biochemical Engineering
> >> Rutgers, The State University of New Jersey
> >> 98 Brett Road
> >> Piscataway, NJ 08854
> >> tel: 848-445-6953
> >
> > --
> > NIH Center for Macromolecular Modeling and Bioinformatics
> > Beckman Institute for Advanced Science and Technology
> > University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
> > http://www.ks.uiuc.edu/~johns/ Phone: 217-244-3349
> > http://www.ks.uiuc.edu/Research/vmd/
>
>
>
> --
> Dr. Axel Kohlmeyer akohlmey_at_gmail.com http://goo.gl/1wk0
> College of Science & Technology, Temple University, Philadelphia PA, USA
> International Centre for Theoretical Physics, Trieste. Italy.
>

-- 
Dr. rer. nat. Olaf Lenz
Institut für Computerphysik, Allmandring 3, D-70569 Stuttgart
Phone: +49-711-685-63607