[cclib-devel] cclib methods/calculations

[Adam T. <a-t...@st...>]

I've been thinking about adapting PyMOlyze to work with cclib's  
parsers. I believe the mocoeffs format is slightly different than  
what PyMOlyze currently parses too, so I need to update the code that  
does population analyses. I figure since I'm making those  
modifications anyways, what if I just committed those changes to the  
cclib tree under a methods of calculations subdirectory? Then cclib  
would provide an interface to parsing files and using that data to do  
population analysis.

The methods I propose are:

1. C-squared PA (not as famous as mulliken, but can be done without  
aooverlap matrices)
2. Mulliken PA
3. Overlap PA
4. Mayer bond orders
5. Density and/or spin-density matrices
6. Charge density analysis (ooh, hard since it depends on multiple  
files!)

I don't think these functions would result in any data other than  
their result, so that makes me think they shouldn't have classes  
associated with them. Unless we want to keep referring to their data  
after the fact (c.result, or something like that).

What do you think? (Yeah, yeah, I know I should be working on the  
parsers, but I also need to update PyMOlyze at some point, and I  
figure it makes sense to start thinking about it now.)

Adam