Add an observation in trait measurement table

Add a trait measure in trait measurement table

Usage

add_traits_measures(
  new_data,
  col_names_select = NULL,
  col_names_corresp = NULL,
  collector_field = NULL,
  plot_name_col = NULL,
  tag_col = NULL,
  id_individual_col = NULL,
  traits_field,
  features_field = NULL,
  allow_multiple_value = FALSE,
  add_data = FALSE,
  census_col = NULL,
  id_sub_plots_col = NULL
)

Arguments

new_data

tibble

col_names_select

string vector

col_names_corresp

string vector

collector_field

string column name which contain the collector name

plot_name_col

string column name containing plot names for linking to id_liste_plots. Use together with tag_col to resolve individuals by plot name + tag combination.

tag_col

string column name containing individual tag numbers. Used together with plot_name_col to resolve individuals by plot + tag.

id_individual_col

string column name containing individual IDs (id_n from data_individuals). Self-sufficient — no plot ID argument is required when this is provided; the plot ID is resolved internally only when needed for census lookup.

traits_field

string vector listing trait columns names in new_data

features_field

string vector listing features (column names) to link to measurements in new_data

allow_multiple_value

if multiple values linked to one individual can be uploaded at once

add_data

logical whether or not data should be added - by default FALSE

census_col

string. Optional column name in new_data containing the census typevalue (integer, e.g. 1, 2, 3). The corresponding id_sub_plots is resolved automatically from the database per plot × census. Rows with different census values are handled correctly within a single call. Takes precedence over the interactive census prompt but is overridden by id_sub_plots_col.

id_sub_plots_col

string. Optional column name in new_data containing the id_sub_plots value directly. When provided, no database lookup or interactive prompt is performed. Takes precedence over census_col.

Value

list of tibbles that should be/have been added

Details

This function now uses database transactions to ensure atomic operations. When features_field is provided, both measurements and their features are added together in a single transaction. If any error occurs, all changes are rolled back automatically.

The function uses PostgreSQL's RETURNING clause for efficient ID retrieval, eliminating the need for separate queries to fetch generated IDs. This improves reliability and prevents race conditions from concurrent operations.

Author

Gilles Dauby, gilles.dauby@ird.fr