Launch the Data Update App

Launches an interactive Shiny app for correcting plot metadata and individual data one record at a time. It is the user-friendly counterpart to update_records, which is powerful but expects the caller to already know which table a value lives in.

Usage

launch_data_update_app(lang = "fr")

Arguments

lang

Character. Initial UI language: "en" or "fr". Default: "fr".

Value

Launches a Shiny app (does not return until the app closes).

Details

The app has two sections:

• Plot metadata - pick a plot, edit the columns stored directly in data_liste_plots (including the method and country lookups, offered as dropdowns), and edit its features.

• Individual data - find an individual by plot and tag or by id_n, edit the columns of data_individuals, change its identification through an embedded taxonomic search, and edit its trait measurements.

Why features need care. Many columns of an extracted plot or individual table are not columns of that record at all. Plot features are rows of data_liste_sub_plots; individual features are rows of data_traits_measures. Worse, one extracted column can be the aggregate of several such rows - the mean of a trait measured at three censuses, or the concatenated names of everyone recorded as additional_people. Writing back to that single value is meaningless, which is why update_records() refuses it.

The app therefore never edits an aggregate. For every feature it shows how many records back it, what the extracted table would display, and how that display was computed; the editable inputs are the underlying records, each labelled with its own id and its census or subplot context.

Why an identification is not just idtax_n. merge_individuals_taxa() resolves the individual's idtax_n through table_idtax synonymy into idtax_f, resolves the identification of the specimen linked to the individual the same way into idtax_specimen_f, and uses idtax_individual_f = coalesce(idtax_specimen_f, idtax_f) everywhere downstream. The identification section shows that whole cascade. While a specimen is linked, its identification wins: editing idtax_n is stored but changes nothing an extraction shows, and the app says so both in the section and in the preview. Re-identifying the specimen is done with launch_specimen_identification_app.

Only existing records can be edited. Adding or deleting measurements is done with the feature wizard (launch_feature_wizard) and the safe_delete_* functions.

Every write goes through detect_direct_changes() and execute_direct_updates(), so stored values are re-read immediately before writing, only genuine differences are written, and records are backed up to their follow-up table where one exists.

See also

update_records, query_plot_features, query_individual_features, launch_specimen_identification_app, launch_feature_wizard

Examples

if (FALSE) { # \dontrun{
launch_data_update_app()
launch_data_update_app(lang = "en")
} # }