Skip to contents

Match Internal Taxa to WCVP Names

match_taxa_to_wcvp.Rd

Matches taxa from the internal table_taxa to WCVP names already uploaded in the database, using exact and optionally fuzzy matching.

Usage

match_taxa_to_wcvp(
  con_taxa = NULL,
  tax_ids = NULL,
  methods = c("exact", "fuzzy"),
  fuzzy_threshold = 0.9,
  author_match = c("none", "exact", "fuzzy"),
  author_threshold = 0.6,
  n_cores = 1L,
  verbose = TRUE
)

Arguments

con_taxa

Connection to the taxa database. If NULL, calls call.mydb.taxa().

tax_ids

Optional integer vector of idtax_n to match. If NULL, matches all accepted taxa.

methods

Character vector of matching methods to use. Default c("exact", "fuzzy").

fuzzy_threshold

Numeric (0-1). Minimum similarity for fuzzy matches. Default 0.9.

author_match

Character. How to use author strings during exact name matching.

  • "none" (default): ignore authors entirely.

  • "exact": authors must match character-for-character. Reduces false positives but misses any formatting difference.

  • "fuzzy": exact name match first, then Jaro-Winkler author similarity to select among homonyms and filter below author_threshold. More tolerant of abbreviation/spacing differences.

Authors are built from author1 (basionym) and author2 (combination) columns in table_taxa: "(author1) author2". Not applied to fuzzy name matching (author disambiguation is not meaningful when the name itself is inexact).

author_threshold

Numeric (0-1). Minimum Jaro-Winkler similarity required to keep a match when author_match = "fuzzy" and author info is present on both sides. Default 0.6.

n_cores

Integer. Number of parallel workers for fuzzy matching. Uses forking on Unix and a PSOCK cluster on Windows. Default 1 (sequential). Set to parallel::detectCores() - 1 to use all available cores.

verbose

Logical. Show progress. Default TRUE.

Value

A tibble with columns: idtax_n, taxon_name_internal, plant_name_id, wcvp_taxon_name, match_type, match_score.

Details

Returns a tibble for review. Does NOT write to the database automatically. Use save_wcvp_links() to persist reviewed matches.

Examples

if (FALSE) { # \dontrun{
con_taxa <- call.mydb.taxa()
matches <- match_taxa_to_wcvp(con_taxa)
# Review matches, then save
save_wcvp_links(matches, con_taxa)
} # }