#90 Support for different flavors of GAMESS
Currently, it appears that Avagadro (v0.9.2 running on Windows XP SP3) will only load in an output file for GAMESS-US, with the '.gamout' extension. I've tried to parse output files from the version of GAMESS that runs on nix with the '*.output' extension (the UCLA Beowulf cluster is currently running version 12 JAN 2009 (R1)), and I get an error message saying that my file is not a valid format (though it will read the input files with no problem). I'm not a GAMESS expert, so I don't know what the differences are in the output files of the various flavors of GAMESS, but I would very much like to be able to parse my output files to display the orbitals and other calculated surfaces.
The output files are large, so I won't post them unless requested, but I do have several output files that I can submit as examples if needed.
The problem isn't a "different flavor of GAMESS." The problem is that the file has an unusual extension.
I've seen GAMESS files with ".gamout" or ".out" or ".log" extensions. I've never seen ".output". So the "auto-detect from extension" feature in Avogadro (and Open Babel) has no idea what kind of file that is.
I'll add that to Open Babel's generic output code, but it'll be a few weeks before OB-2.2.2 is released with that change. If you want a source code patch, I can certainly provide you with one.
I changed the extension and that worked...almost. I can read in the molecule, but only the geometry is read in. The orbitals and other computed properties are discarded. When I tried converting the file to the different supported formats using Open Babel 2.2.1, everything but the atomic coordinates are again discarded. In fact, I could not find any output format in OB-2.2.1 that could extract any information other than the molecular geometry.
Am I missing something?
I am afraid you are correct - vibrations are not yet supported by the GAMESS-UK parser. It should not be difficult to add them though, as support for them is there in the GAMESS-US parser.
As for other stuff (besides maybe the dipole moment), that is not yet implemented in OpenBabel, most notably molecular orbitals.
PS: Ian - You didn't by chance do a post-doc at Domcke's group in Munich?
Alas, molecular orbitals are what I'm after. So I'd like to add that as a feature request.
I did not do a post-doc in Munich. I'm currently a grad student at UCLA.