Skip to contents

Vectorized Latin orthographic normalization

Source: R/clean.R
normalize_epithets.Rd

Reduces common Latin spelling alternations to a canonical form so that e.g. hirtaeformis and hirtiformis produce the same normalized key. Applied identically to both query names and backbone names so the keys line up on either side of the join.

Usage

normalize_epithets(names)

Arguments

names

Character vector of cleaned taxonomic names (genus + epithet).

Value

Character vector of normalized forms.

Details

Pipeline:

  1. Lowercase.

  2. Strip Latin-1 diacritics and ligatures (e-acute to e, ae-ligature to ae, sharp-s to ss, etc.), applied to genus and epithet.

  3. Orthographic alternation on the epithet only: ae/oe -> i, trailing ii -> i, y -> i, ph -> f, rh -> r, th -> t.

Step 2 runs before step 3, so ae-ligature -> ae -> i and oe-ligature -> oe -> i fold into the same key as the de-ligatured forms.