Rows that repeat a measurement the database already holds
dot-existing_measurement_rows.RdCompares each row of prepared measurement data against what is already recorded, and answers on the terms the row itself sets:
Arguments
- data
Data frame of prepared measurements, with `tag`, `traitid` and `id_liste_plots`, and optionally `id_sub_plots`.
- individuals
Data frame of the individuals recorded for the selected plots, with `tag`, `id_table_liste_plots_n` and `id_n`.
- measures
Data frame of the measurements already recorded for those individuals, with `id_data_individuals`, `traitid` and `id_sub_plots`.
Value
List of two integer vectors of row numbers into `data`: `with_census` (matched on individual, feature and census) and `without_census` (matched on individual and feature alone).
Details
* A row carrying a census is a repeat only of a measurement recorded **during that same census**. The same feature measured during another campaign is a new measurement, not a duplicate, which is the whole point of a census. * A row carrying no census - a position, a quadrat, anything the census link policy keeps off a campaign - is a repeat of **any** recorded value of that feature for that individual. There is no campaign to narrow the comparison to, and the tree has one position, not one per census.
The two are returned separately because they are not the same claim, and a single count would misreport one of them.
Values are not compared, only the existence of a measurement: whether re-recording the same feature is a mistake or a deliberate second reading is the user's call, not this function's.