fix how genus determined (#166)

Fix how a "genus" is identified within a taxon name

- For hybrid genera the genus name is "x genus" - this was causing problems in AusTraits, because I'd only included an ifelse to hunt for "x" in some places

- As part of this fix, package `extract_genus` into its own function

R/load_taxonomic_resources.R

@@ -56,7 +56,7 @@ load_taxonomic_resources <-
       dataset_name = .data$datasetName
     ) %>%
     mutate(
-      genus = stringr::word(canonical_name, 1),
+      genus = extract_genus(canonical_name),
       taxon_rank = stringr::str_to_lower(taxon_rank),
       taxon_rank = stringr::str_replace(taxon_rank, "regnum", "kingdom"),
       taxon_rank = stringr::str_replace(taxon_rank, "classis", "class"),
@@ -82,7 +82,7 @@ load_taxonomic_resources <-
       name_element = .data$nameElement
     )  %>%
     mutate(
-      genus = stringr::word(canonical_name, 1),
+      genus = extract_genus(canonical_name),
       taxon_rank = stringr::str_to_lower(taxon_rank),
       taxon_rank = stringr::str_replace(taxon_rank, "regnum", "kingdom"),
       taxon_rank = stringr::str_replace(taxon_rank, "classis", "class"),
@@ -121,7 +121,7 @@ load_taxonomic_resources <-
         ),
         binomial = ifelse(is.na(binomial), zzz, binomial),
         binomial = base::replace(binomial, duplicated(binomial), zzz),
-        genus = stringr::word(stripped_canonical, 1),
+        genus = extract_genus(stripped_canonical),
         trinomial = stringr::word(stripped_canonical2, start = 1, end = 3),
         trinomial = ifelse(is.na(trinomial), zzz, trinomial),
         trinomial = base::replace(trinomial, duplicated(trinomial), zzz),
@@ -159,7 +159,7 @@ load_taxonomic_resources <-
         trinomial = stringr::word(stripped_canonical2, start = 1, end = 3),
         trinomial = ifelse(is.na(trinomial), "zzzz zzzz", trinomial),
         trinomial = base::replace(trinomial, duplicated(trinomial), "zzzz zzzz"),
-        genus = stringr::word(stripped_canonical, 1),
+        genus = extract_genus(stripped_canonical),
         taxonomic_dataset = "APNI"
       ) %>%
       dplyr::distinct() %>%

R/match_taxa.R

@@ -84,7 +84,7 @@ match_taxa <- function(
         update_na_with(strip_names_2(cleaned_name)),
       trinomial = stringr::word(stripped_name2, start = 1, end = 3),
       binomial = stringr::word(stripped_name2, start = 1, end = 2),
-      genus = stringr::word(original_name, start = 1, end = 1),
+      genus = extract_genus(original_name),
       fuzzy_match_genus =
         fuzzy_match_genera(genus, resources$genera_accepted$canonical_name),
       fuzzy_match_genus_known =

R/standardise_names.R

@@ -98,3 +98,27 @@ standardise_names <- function(taxon_names) {
     stringr::str_squish()
 }
 
+#' Extract Genus
+#' 
+#' This function extracts the genus component of a scientific name. 
+#' It identifies if the genus is/is not a hybrid. For a hybrid genus,
+#' the first two words of the taxon name are extracted (e.g. "x Cynochloris"),
+#' while for a non-hybrid genus just the first word is extracted (e.g. "Banksia").
+#'
+#' @param taxon_name 
+#'
+#' @return The genus for a scientific name.
+#'
+#' @examples
+#' genus = extract_genus(stripped_name)
+#' 
+#' 
+extract_genus <- function(taxon_name) {
+  genus <- 
+    ifelse(
+      stringr::word(taxon_name, 1) == "x",
+      stringr::word(taxon_name, start = 1, end = 2),
+      stringr::word(taxon_name, 1)
+    )
+  genus
+}

R/update_taxonomy.R

@@ -97,7 +97,7 @@ update_taxonomy <- function(aligned_data,
     aligned_data %>%
     dplyr::select(original_name, aligned_name, taxon_rank, taxonomic_dataset, aligned_reason) %>%
     dplyr::mutate(
-      genus = stringr::word(aligned_name, 1),
+      genus = extract_genus(aligned_name),
       row_number = dplyr::row_number()
     )
 
@@ -300,7 +300,7 @@ update_taxonomy_APC_genus <- function(data, resources) {
       taxonomic_status = ifelse(is.na(accepted_name_usage_ID), as.character(my_order), paste("genus", taxonomic_status_genus)),
       taxon_ID_genus = resources$genera_all$taxon_ID[match(accepted_name_usage_ID, resources$genera_all$accepted_name_usage_ID)],
       # genus names in `aligned_name` that are not APC-accepted need to be updated to their current name in `suggested_name`
-      aligned_minus_genus = ifelse(is.na(genus_accepted), NA, stringr::str_replace(aligned_name, stringr::word(aligned_name, 1), "")),
+      aligned_minus_genus = ifelse(is.na(genus_accepted), NA, stringr::str_replace(aligned_name, extract_genus(aligned_name), "")),
       suggested_name = ifelse(taxonomic_status == "accepted", paste0(genus_accepted, aligned_minus_genus), NA),
       suggested_name = ifelse(taxonomic_status != "accepted", aligned_name, suggested_name),
       # indicate taxonomic_status of the genus name in `aligned_name` and why it needed to be updated for the `suggested_name`
@@ -379,8 +379,7 @@ update_taxonomy_APC_species_and_infraspecific_taxa <- function(data, resources,
     dplyr::filter(species_and_infraspecific(taxon_rank)) %>%
     dplyr::distinct(canonical_name, .keep_all = TRUE) %>%
     dplyr::select(canonical_name, accepted_name_usage_ID)
-
-
+
   split_taxa_table <- 
     resources$APC %>%
     dplyr::filter(species_and_infraspecific(taxon_rank)) %>%
@@ -548,7 +547,7 @@ update_taxonomy_APC_species_and_infraspecific_taxa <- function(data, resources,
       suggested_name = ifelse(is.na(suggested_name), aligned_name, suggested_name),
       taxonomic_status = ifelse(is.na(accepted_name),  taxonomic_status_aligned, "accepted"),
       # for APC-accepted species, the `genus` is the first word of the `accepted_name`
-      genus_accepted = stringr::word(suggested_name, 1),
+      genus_accepted = extract_genus(suggested_name),
       taxon_ID_genus = resources$genera_all$taxon_ID[match(genus_accepted, resources$genera_all$canonical_name)],
       update_reason =  taxonomic_status_aligned,
       taxonomic_dataset = "APC",