Relationship between _diffrn_data_frame.id and _diffrn_scan.id

[jamesrhester]

I believe that I have been informed in the past that a frame id is unique only within the scan that it belongs to. In this case, both scan id and frame id must be specified together in order to uniquely identify a particular frame of data. This not done for categories diffrn_scan_frame_axis, diffrn_refln or diffrn_data_frame itself.

If a frame_id is in fact required to be globally unique (for all diffrn_ids and scans in a data block) this issue can be closed.

[jamesrhester]

@yayahjb , can you comment on this issue? Is it true that a frame_id is unique only within a scan?

[yayahjb]

We are dealing with relational databases, so the uniqueness of any column in any table is determined by the keys we have designed and the parent-child relationships among the tables. If we look at, say, the 1.7.10, imgCIF dictionary, the parent of all the instances of 'frame_id' is the column 'id' in the 'diffrn_data_frame' category and the key for that category is composite, containing the columns 'id', 'detector_element_id' and 'variant'. The relationships among scans and particular frames are described in the category 'diffrn_scan_frame'. That key is also composite, containing the columns 'scan_id', 'frame_id' and 'variant", so in that context, if you are globally trying to identify a particular frame in a data set, you really need to specify at least the variant, the detector element, the scan, and the frame id together. The variant has an implicit default. If you don't have multiple streams of frames for different detector elements, then having the scan id and the frame id are sufficient to identify the particular frame in question, but there is no strict requirement that the frame_id be globally unique.

…

On Wed, Apr 4, 2018 at 11:05 PM jamesrhester ***@***.***> wrote: I believe that I have been informed in the past that a frame id is unique only within the scan that it belongs to. In this case, both scan id and frame id must be specified together in order to uniquely identify a particular frame of data. This not done for categories diffrn_scan_frame_axis, diffrn_refln or diffrn_data_frame itself. If a frame_id is in fact required to be globally unique (for all diffrn_ids and scans in a data block) this issue can be closed. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#4>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AEPiAfr2oW6UyUDfRPG0jtLCyn6SMQGVks5tlYnggaJpZM4THynF> .

[jamesrhester]

I take Herbert's reply to confirm that _diffrn_data_frame, diffrn_refln and diffrn_scan_frame_axis all require that a child data name of scan_id is defined to be a part of the key in order to uniquely identify a particular frame in the most general case of multiple scans with overlapping frame identifiers.