Adding category examples

[rowlesmr]

Adding examples of use at the category level.

Not updating update dates yet.

Categories:

• PD_AMORPHOUS
• PD_BACKGROUND
• PD_BLOCK
  • example is in _pd_block.id, but it's the only member.
• PD_CALC_COMPONENT
• PD_CALC_OVERALL
• PD_CALIB_D_TO_TOF
• PD_CALIB_DETECTED_INTENSITY
• PD_CALIB_INCIDENT_INTENSITY
• PD_CALIB_WAVELENGTH
• PD_CALIB_XCOORD
• PD_CALIBRATION
• PD_CHAR
• PD_DATA (containing PD_CALC, PD_MEAS, PD_PROC)
• PD_DIFFRACTOGRAM
  • example is in _pd_diffractogram.id, but there're only two members.
• PD_INSTR
• PD_INSTR_DETECTOR
• PD_MEAS_OVERALL
• PD_PEAK
• PD_PEAK_OVERALL
• PD_PHASE
  • examples are in the individual members
• PD_PHASE_MASS
• PD_PREF_ORIENT_MARCH_DOLLASE
• PD_PREF_ORIENT_SPHERICAL_HARMONICS
• PD_PREP
• PD_PROC_LS
• PD_PROC_OVERALL
• PD_QPA_CALIB_FACTOR
• PD_QPA_EXTERNAL_STD
• PD_QPA_INTENSITY_FACTOR
• PD_QPA_INTERNAL_STD
• PD_QPA_OVERALL
• PD_SPEC

Not (currently) going to, as other categories are taking over, or have taken over:

• PD_BLOCK_DIFFRACTOGRAM
• PD_CALIB
• PD_CALIB_OFFSET
• PD_CALIB_STD
• PD_MEAS_INFO_AUTHOR
• PD_PHASE_BLOCK
• PD_PREF_ORIENT
• PD_PROC_INFO_AUTHOR

[rowlesmr]

I don't think that _pd_calib_detected_intensity.id is a necessary category key, but also:

Apart from that I think including an identifier data name in the key is fine, e.g. you just keep calibrating the instrument every week and assign an identifier to each calibration #92 (comment)

[rowlesmr]

I don't think that _pd_calib_detected_intensity.id is a necessary category key, but also:

Apart from that I think including an identifier data name in the key is fine, e.g. you just keep calibrating the instrument every week and assign an identifier to each calibration #92 (comment)

I think we should keep _pd_calib_detected_intensity.id (due to being able to record multiple calibrations on the same instrument over time*), but I am now unsure that _pd_calib_detected_intensity.diffractogram_id is doing it's job.

If I have

data_unknown_diffractogram
_pd_diffractogram.id UNKNOWN_SPECIMEN
    
loop_
_pd_calib_detected_intensity.id
_pd_calib_detected_intensity.detector_id
_pd_calib_detected_intensity.detector_response
_pd_calib_detected_intensity.special_details
1   1_4913c6ed   1       'Scanned through direct beam.'
1   2_4913c6ed   0.973   'Scanned through direct beam.'
1   3_4913c6ed   0.997   'Scanned through direct beam.'
1   4_4913c6ed   1.039   'Scanned through direct beam.'
#...

then _pd_calib_detected_intensity.diffractogram_id gets the value 'UNKNOWN_SPECIMEN', when the intention is to record the diffractogram which was used to make the calibration, if such a diffractogram exists (which doesn't, in this case).

I think this ties in to #163.

This whole argument also applies to _pd_calib_incident_intensity.diffractogram_id

* I need to add a key to PD_CALIB_INCIDENT_INTENSITY.

[rowlesmr]

These examples need to be checked in light of changes over time in the rest of the dictionary.

[vaitkus]

Thank you for the work, I noted a few minor discrepancies.

However, do you think it would be possible to split the PR into several more manageable parts? There are several example that could be merged right away while others may need some more time to digest.

[rowlesmr]

However, do you think it would be possible to split the PR into several more manageable parts?

Boo. I was coming to the same conclusion...

I'll see if I can get it done tonight.

[rowlesmr]

PD_CALIB_WAVELENGTH needs to be fixed before updating the examples has been deleted
PD_CALIB_XCOORD needs to be rejigged before updating the examples
PD_QPA_EXTERNAL_STD needs to be updated before doing the examples PD_DIFFRACTOGRAM now has link to _pd_instr.id, so no update needed.

[vaitkus]

Thanks! This will make the reviewing and merging much simpler. Let's leave this PR open so once we merge all we others, we could resync just to see if we missed anything.

[vaitkus]

I think that all of the PR that splintered of this one have now been merged. @rowlesmr would probably be the best one to decide how to continue with this PR (e.g. close it outright or try to sync it with the main branch to see if nothing important got accidentally left out). As this branch will not get merged, fully resolving all of the sync conflict might not be the best use of our time, but starting a merge and quickly going through the diff list might still provide some insights.