Daniel Russel wrote: > Since I think such a new module is a good idea, I'll make it more > concrete. Let's have a new module called IMP.helper. Its primary purpose > is to contain functions that perform common tasks that > - involve creation of more than one restraint or score state > - creation of a restraint or score state followed by non-trivial > manipulation of it > - functions to create standard particle sets > > Examples would include existing code such as: > - creating a rigid body and associated score states > - restraining the diameter of a set of particles > - creating a hard sphere excluded volume restraint which intelligently > handles rigid bodies > - creating a coarse grained model of a protein with known number of > amino acids, but no known structure > > Having it in a module means that there is a clear first place to look > for such things. The low level functionality in core is clearly > differentiated from functions that use it.
I completely agree that a separate module is the best place for such functionality. I'm just not sure I like the name. ;) Can't think of a better one though...
Ben