Re: [ESPResSo-users] Bug in python interface?

[Henri]

On Tue, 2018-01-23 at 13:26 +0100, Andreas Härtel wrote:
> Hello Henri,
> 
> the most simple script for getting the error is just one line:
> import espressomd.System
> 
> It happens using ESPResSo 3.3.1, the most recent release that you can 
> get from the web page.
> 
> The myconfig.hpp file is:
> #define PARTIAL_PERIODIC
> #define CONSTRAINTS
> #define BOND_VIRTUAL
> #define ROTATION
> #define ROTATIONAL_INERTIA
> #define ELECTROSTATICS
> #define VIRTUAL_SITES_RELATIVE
> #define VIRTUAL_SITES_NO_VELOCITY
> #define LENNARD_JONES
> #define LENNARD_JONES_GENERIC

I just tested your configuration and it seems to be resolved in the development 
version, so you
should switch to that.

Note that in the development version the features

    BOND_VIRTUAL VIRTUAL_SITES_NO_VELOCITY

have been removed.

I don't think that

    import espressomd.System

ever worked because System is not a module but a class in the module 
espressomd.  Did you mean

    from espressomd import System

> 
> Best regards,
> Andreas
> 
> 
> On 23.01.2018 02:30, Henri wrote:
> > 
> > Hello Andreas,
> > 
> > The error you are seeing is from the C++ simulation core and seems to be 
> > related to broken
> > memory
> > handling.  There is nothing you can do from a user perspective.
> > 
> > > 
> > >   I start a simple python script
> > Can you please share the script?  It would help us a lot with debugging 
> > with this problem.
> >  Please
> > also share with us which version of ESPResSo you are using and how you 
> > compiled it
> > (myconfig.hpp).
> > 
> > Cheers, Henri
> > 
> > On Mon, 2018-01-22 at 13:13 +0100, Andreas Härtel wrote:
> > > 
> > > Hello list,
> > > when I start a simple python script using pypresso, I always get errors
> > > as the attached one. Can someone tell me what is going wrong here?
> > > TCL scripts run without problems.
> > > Thanks and best regards,
> > > Andreas