Definitions of numeric identifiers

[vaitkus]

Similar to issue #13 data items _atom_sites_axes.matrix_seq_id and _atom_site_fourier_wave_vector.seq_id are defined as integers in the range of 0: while some of the items that reference them are instead defined as having the Code type. What should the data type of these items be (e.g. Integer, Code, Word)?

[brantonc]

I can't imagine why _atom_site_fourier_wave_vector.seq_id would be of any type other than Integer.

[vaitkus]

Yes, I can now see that an integer value of _atom_site_Fourier_wave_vector.seq_id is expected by some items from the CIF_MAG dictionary (see PR COMCIFS/magnetic_dic#69). If/when that PR is merged, I will update the _atom_site_Fourier_wave_vector.seq_id and linked items accordingly if needed.

[vaitkus]

PR #22 restricts the values of _atom_site_fourier_wave_vector.seq_id to positive integers as discussed in COMCIFS/magnetic_dic#69 (comment).

The preferred data type and value range of _atom_sites_axes.matrix_seq_id still needs to be agreed upon.

[vaitkus]

Merging of PR #30 fixed a lot of things. Two minor inconsistencies remain:

• _atom_site_fourier_wave_vector.seq_id is defined as having the range of 0: while in the magnetic CIF dictionary related items have the range of 1:0 (see discussion PR Restrict *_Fourier_wave_vector.seq_id data items to positive integers #22).
• _atom_sites_axes.matrix_seq_id is defined as an integer with the range of 0:, however, all of the examples given in the dictionary start at 1. Should the range be changed to 1?

[brantonc]

I have no opinion on the seq_id ranges. I always start at 1. Perhaps others have reasons to start indexing at zero. I don’t remember what _atom_sites_axes.matrix is used for. Best wishes, Branton From: Antanas Vaitkus ***@***.***> Sent: Friday, November 7, 2025 3:39 AM To: COMCIFS/Modulated_Structures ***@***.***> Cc: Branton Campbell ***@***.***>; Comment ***@***.***> Subject: Re: [COMCIFS/Modulated_Structures] Definitions of numeric identifiers (Issue #14) [https://avatars.githubusercontent.com/u/5329163?s=20&v=4]vaitkus left a comment (COMCIFS/Modulated_Structures#14)<#14 (comment)> Merging of PR #30<#30> fixed a lot of things. Two minor inconsistencies remain: * _atom_site_fourier_wave_vector.seq_id is defined as having the range of 0: while in the magnetic CIF dictionary related items have the range of 1:0 (see discussion PR #22<#22>). * _atom_sites_axes.matrix_seq_id is defined as an integer with the range of 0:, however, all of the examples given in the dictionary start at 1. Should the range be changed to 1? — Reply to this email directly, view it on GitHub<#14 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACZAIQTS6RJOIN7TUR23ZXD33RZE7AVCNFSM6AAAAACLNXWGBSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTKMBRG43TSMZRGA>. You are receiving this because you commented.Message ID: ***@***.******@***.***>>

[vaitkus]

The ATOM_SITES_ORTHO and ATOM_SITE_DISPLACE_ORTHO category usage examples in the dictionary currently do start the numbering of _atom_site_Fourier_wave_vector.seq_id from 0, however, in all other examples the numbering starts from 0. Just noting this as a reminder to update the examples in case the range is changed.

[brantonc]

Interesting. The ATOM_SITES_ORTHO and ATOM_SITE_DISPLACE_ORTHO category examples indicate that the zeroth wavevector is the zero wavevector (i.e. a constant). I would think that any constant term should be part of the commensurate basic structure rather than the modulated structure. Who knows? Maybe someone had a good reason for doing this. From: Antanas Vaitkus ***@***.***> Sent: Monday, November 10, 2025 2:44 AM To: COMCIFS/Modulated_Structures ***@***.***> Cc: Branton Campbell ***@***.***>; Comment ***@***.***> Subject: Re: [COMCIFS/Modulated_Structures] Definitions of numeric identifiers (Issue #14) [https://avatars.githubusercontent.com/u/5329163?s=20&v=4]vaitkus left a comment (COMCIFS/Modulated_Structures#14)<#14 (comment)> The ATOM_SITES_ORTHO and ATOM_SITE_DISPLACE_ORTHO category usage examples in the dictionary currently do start the numbering of _atom_site_Fourier_wave_vector.seq_id from 0, however, in all other examples the numbering starts from 0. Just noting this as a reminder to update the examples in case the range is changed. — Reply to this email directly, view it on GitHub<#14 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACZAIQQ4R5DSC2J3XOAQVJT34BM4RAVCNFSM6AAAAACLNXWGBSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTKMJQGQ4TCNBZHA>. You are receiving this because you commented.Message ID: ***@***.******@***.***>>