Re: [cclib-devel] orbitals for correlated methods

[Karol L. <kar...@kn...>]

On Thursday 12 July 2007 11:45, Karol Langner wrote:
> At some point I'd like cclib to also parse the MO eigenvectors for
> correlated methods - natural orbitals from MP2 or CI, and optimized
> orbitals after MCSCF calcualtions. They all have the format of mocoeffs,
> with different values. I'd like to have your opinion - should these be
> parsed into the same attribute as the SCF MOs (mocoeffs) or into a new one
> (mpcoeffs?)?

I'm still not in my normal workspace, so It'll be a few days before I read 
through all the mail from the past few weeks, but I now did one thing I am 
very eager to see in cclib in the next release - parsing natural orbitals.

I wrote about this some time ago (see the above), and some thinking made me 
confident to create a new attribute - nocoeffs. Natural orbitals are 
mathematically totally different numbers than MO coefficients, although their 
print-out usually looks very similar. I added code for them in Gaussian and 
GAMESS for CIS and MP2 - see my commit for details.

Cheers,
Karol

-- 
written by Karol Langner
Wed Aug  8 10:43:03 EDT 2007