CST branch -- March 25, 2010 - 07:57

Hi,
I looked for the possible simplification of the code construction of the in the CST branch. Because you emphasize several times, that everything must be discussed I would first suggest a general idea of the changes and then I would start to make them step by step. 

Just to recall, Edward didn't like the fact, that I kept the code which was already written and just add the functions dealing with asymmetric CST and other features . That means that many similar mathematical functions were abundantly repeated within the code.

Lets start with the first step in the calculation procedure: direction_cosine.py file
the "standard" version operates just with one dipole-dipole interaction to attached proton and axially symmetric CST collinear with the XH bond . Our idea was to add the possibility to calculate also with other nuclei in proximity. Hence there would be necessary to change the scalar variable defined by functions within the direction_cosine.py to vectors of the dimensionality equal to all nuclei which dipole-dipole interaction should be taken into the account. Additionally +2 if the asymmetric CS tensor should be considered. Following the philosophy of this file the analogy of xh_unit_vector for two eigenvectors of CST tensor should be calculated outside this file.
Any comment? Do you agree?
Pavel