GitLab

removed bug. Total dist will be computed only when the solve() function gets removed from undulator.cpp

check the length of the elements and compare it with the length of a time step; check that the user inputs for DT and MAXSTEPS of the TRACK command and for L of the elements are positive values; replace the length_m member variables in the elements with method call getElementLength;

Implements tracking bunch back in time; replace queues for time step size, num time steps and zstop with class that stores all these informations in a list; let the SOURCE element absorbe all particles in case of back tracking; disables automatic centering of FROMFILE distributions;

Use different chains of random numbers on each core of a parallel run in CollimatorPhysics. Furhtermore fix simulation time after all particles were in material and fix spurious structure in longitudinal position with period length of time step that was present after passage of material.

fix reference for case when all particles are in degrader material; refactor CollimatorPhysics and tried to clean up mix of units and convertions

than zero particles are outside the material. When a single particle was
outside the material OPAL aborted with the message 'h<0, can not build a
mesh'. The bounding box around a single particle has no volume and hence
the volume of the mesh would be zero as well. This is fixed if more than
one particle is outside the material. For consistency with the method
computeSpaceChargeFields I chose minBinEmitted_m as number of particles
that have to be outside. Its default value is 10.

Avoid duplication of member variables for location and momentum of the reference particle and the transformation from the reference to lab coordinate system. Deleted the member variables in the ParallelTTracker and just using the ones in the PartBunch. This is a first step towards #287.

This reverts commit 5fbd4c4b.

 zero but bunch is at e.g. z = 10 m; reintroducing __DBGMSG__ define
which is useful for debugging printing the name of the file without its
path and the line number

renaming Collimator to CCollimator where C stands for cyclotron; all other collimators are implemented using FlexibleCollimator; This also fixes #224

first version of multi-slit/pepperpot/what ever aka flexible collimator. However way to slow to be practical