Non-ASCII characters not supported

[CasperWA]

As laid out in emmo-repo/CIF-ontology#182 by @vaitkus, CIF2 now supports non-ASCII characters, which trip up the PyCIFRW parser(s).
See the referenced issue for an example.

[CasperWA]

I tried using characterset="unicode" and standard="CIF2" without any luck, unfortunately.

[jamesrhester]

I've fixed this in the latest commit to the development branch, and I'll bundle up a new release once I've figured out the new python packaging approach.

The problem was that nested blocks (e.g. dictionary definitions) had default characterset setting (ascii). Plain CIF2 data files are fine.

[jamesrhester]

Please try version 4.4.6. Source distribution and linux wheels provided.

[CasperWA]

I've tested v4.4.6 and it works beautifully!
I now get unicode characters nicely in the parsed CIF and dic files.

Just to show - here is a greek kappa that has gone all the way from dic file to Turtle ontology (see κ-geometry diffractometer in the circa center of the image):