Skip to contents

Validate semantics with graceful gap reporting

Source: R/validation_helpers.R
validate_semantics.Rd

Ensures structural requirements, adds a required column if missing, runs validate_dictionary(), and reports measurement rows missing term_iri. Also flags non-canonical Salmon ontology IRIs so source boundaries stay explicit (smn under https://w3id.org/smn/, gcdfo under https://w3id.org/gcdfo/salmon#). In non-strict mode (require_iris = FALSE), semantic gaps emit warnings but do not fail the overall call.

Usage

validate_semantics(
  dict,
  require_iris = FALSE,
  entity_defaults = NULL,
  vocab_priority = NULL
)

Arguments

dict

Dictionary tibble/data frame, a package directory, or a path to column_dictionary.csv.

require_iris

Logical; if TRUE, require non-empty semantic fields (term_iri, property_iri, entity_iri, unit_iri) for measurement rows.

entity_defaults

Deprecated and ignored. Previously reserved for future default entity mapping.

vocab_priority

Deprecated and ignored. Previously reserved for future vocabulary ordering.

Value

A list with elements:

  • dict: normalized dictionary with required column.

  • issues: tibble of structural issues (empty if none).

  • missing_terms: tibble of measurement rows missing term_iri.

Examples

dictionary <- system.file(
  "extdata", "column_dictionary.csv", package = "metasalmon"
)
result <- validate_semantics(dictionary, require_iris = FALSE)
#> ✔ Dictionary validation passed
nrow(result$missing_terms)
#> [1] 0