For the next release (2.2), I’m inclined to use the gemmi library to read CIF files and handle unit cell manipulations. It also handles mmCIF / PDBx files as well as grid data (e.g. electron density maps)
In the meantime, I’ve been pushing a new branch that should handle a bunch of cases in which Avogadro wouldn’t know the full space group setting and ask the user to pick. Through some automated testing, this was happening ~300 / 500 example files in the crystals directory.
This happens for a few reasons - inconsistent space group naming (e.g., evolving rules) and the space groups don’t necessarily specify the setting for a particular file.
The new branch uses the symmetry transforms in the CIF file to deduce the space group and setting. It looks good, but I could use some feedback and testing from folks like @Thomas because I’d like to finish it and merge for v2.1.
To amend automatic checks against the set of crystals on GitHub, I’ll add a couple of .cif about small (organic) molecule structure models (monoclinic P 2_1/c (and special setting P2_1/a), orthorhombic P2_12_12_1, some R3 and R\bar{3}, etc) preferentially without complications by disorder, nor twinning. In line with the established pattern, they will copy files of the COD.
That would be great. If you want, you can send me the files or post here as a ZIP.
At the moment for v2.1, I’m simply aiming for “better than 2.0” - for v2.2, as I think I mentioned to you, I’m intending to adapt the gemmi library, which should significantly improve our support for CIF / mmCIF.
