Re: [ESPResSo-users] Copy_particles and the right numeration

[Axel Arnold]

Hi Aleksei,

for technical reasons, I prefer copy_particles to use consecutive 
identities.  Because the particle ids that you give the routine might 
very well not be consecutive, a simple "index + shift" would not work.

Now, if you add the parameter "-integer" to the lsort in copy_particles 
(scripts/auxiliary.tcl), then the particles are at least in order of 
increasing id. And in the special case you copy a single, consecutive 
range, the given ids will be consecutive as well, so that would give the 
behavior you desire.

For a general solution, I prefer to just return the mapping of the old 
ids to the new ones, which is the solution I put into the git. 
copy_particles then just returns a list of old-new pairs, which can be 
converted to an array, so that you can find the new identities simply by 
asking array(old id).

Best regards,
Axel

On 17.02.14 14:01, Aleksei Kabedev wrote:
> Dear all,
>
> I experience another problem with copy_particles procedure.
> It seems to me, that if I copy a big structure with bonds between 
> internal particles, the second molecule's particles won't have numbers 
> "as originals  +  constant shift". Perhaps it happens when I copy set 
> of particles once, then add some new particles, bond them and finally 
> copy all the range once again. In all examples I experienced this 
> problem, there were branched structures, where I mixed numbers, but 
> they are not matches respectively anyway.
> I need numbers, for example, if there are few branches from one point 
> and I have to know what are the id's of particles closest to the core 
> particle and what are the id's of next level, so that I could use 
> angle bonds after creating them. Here is a simple sample (sorry, it's 
> not really good). As you can see, particles with different types 
> should be at the same positions.
>
> --
>
> Kind regards,
> Aleksei Kabedev


--
JP Dr. Axel Arnold
ICP, Universität Stuttgart
Allmandring 3
70569 Stuttgart, Germany
Email: address@hidden
Tel: +49 711 685 67609

auxiliary.tcl
Description: Tcl script