``_atom_site_rot_Fourier.id`` has been removed

[jamesrhester]

_atom_site_rot_Fourier_id exists in the DDL1 version so should be retained in this version, even if not used.

[vaitkus]

Furthermore, the _atom_site_rot_Fourier.id item is referenced by the _atom_site_rot_Fourier_param.id data item so it should definitely be defined.

[vaitkus]

PR #19 restores the _atom_site_rot_Fourier.id item.

Note, however, that according to the human-readable definition and DDL1 usage examples [1], this particular id was actually a composite key of sorts and not an arbitrary identifier (e.g. Ph1_a1_1 where a1 and 1 refer to values of _atom_site_rot_Fourier_axis and _atom_site_rot_Fourier_wave_vector_seq_id respectively).

Also, the ATOM_SITE_ROT_FOURIER category currently seems a bit strange now:

• _atom_site_rot_Fourier.atom_site_label is the category key.
• _atom_site_rot_Fourier.id exists, but is not mandatory.
• _atom_site_rot_Fourier_param.id references _atom_site_rot_Fourier.id while no items seem to explicitly reference _atom_site_rot_Fourier.atom_site_label.

I have a hunch that there was an effort made to move away from the usage of _atom_site_rot_Fourier.id and some definitions were not fully addressed in this regard.

[1] https://github.com/COMCIFS/DDL1-legacy-dictionaries/blob/28e20dc928790dceb889716d2bed435fc10c4c79/dictionaries/cif_ms.dic#L1951

[vaitkus]

I guess this issue has been resolved with the merging of PR #30?

[jamesrhester]

Yes, it appears all issues raised above are fixed.