Skip to contents

Validate the five measurement-face registries against Appendix A

Source: R/registry_measurement_face.R
validate_measurement_faces.Rd

Returns findings rather than stopping (mirrors check_measurement_bounds()): a workstream fills the registries, runs this, and resolves every row it reports before calling itself done. Checks, one row of output each:

Usage

validate_measurement_faces(
  chem_path,
  method_path,
  scale_path,
  why_path,
  face_path,
  measurement_type_path = NULL
)

Arguments

chem_path, method_path, scale_path, why_path, face_path

paths to the five registries

measurement_type_path

path to metadata/measurement_type.csv, used to check measurement_face keys exist. NULL skips that check.

Value

a tibble of findings (registry, key, rule, detail); zero rows means clean.

Details

  • a measurement_chem / measurement_method / measurement_scale / measurement_face row with an empty source

  • a value outside its column's Appendix A vocabulary (chem role/via, method platform, scale kind, why kind, face face_kind)

  • a measurement_why key with zero, or more than one, rank == 1 row

  • a measurement_face key absent from measurement_type.csv

  • a measurement_face standsin key whose measurement_type.csv row has its own nerc_p01: a stand-in borrows the lender's chain, so a key with its own P01 takes scale (or its own face) instead

  • a measurement_face short_label that is set but blank, longer than 48 characters, ends in punctuation, or is shared by another key (ignoring case); an NA short_label is allowed

  • a nerc_l22 (method) that is not an exact-form L22 concept URI