Precompute matching keys at build time — precompute

Used by the taxifydb build pipeline and by taxify's own test fixtures. Adds key_ci, key_normalized, key_species, and fuzzy_block columns to the backbone data.frame for direct lookup at query time. For a syntaxon backbone (code = "syntaxon") it adds key_ci, key_normalized (from normalize_syntaxon_name()), block_key, the name's first word, which the fuzzy pass groups candidates by, and key_stem, the name's plant-name stems without their epithets, which resolves a unit written in short form; a syntaxon has no genus or epithet.

Usage

precompute_keys(
  df,
  name_col,
  genus_col = NULL,
  epithet_col = NULL,
  code = "taxon"
)

Arguments

df: The backbone data.frame.
name_col: Name of the canonical name column.
genus_col: Name of the genus column (taxon code only).
epithet_col: Name of the specific epithet column (taxon code only).
code: Nomenclature code of the backbone: "taxon" (default) or "syntaxon".

Value

The data.frame with added key columns.