Algorithm fixes (#174)

small algorithm fixes * export strip_names_2 * for genus-resolution names where the genus is outdated, the genus-column was being updated, but the new genus wasn't being "inserted" into the suggested name due to a typo * leave suggested_name blank if not even genus/family can be aligned
traitecoevo · Jan 22, 2024 · 3173a85 · 3173a85
1 parent fd20bdd
commit 3173a85
Show file tree

Hide file tree

Showing 7 changed files with 52 additions and 34 deletions.
diff --git a/NAMESPACE b/NAMESPACE
@@ -8,6 +8,7 @@ export(native_anywhere_in_australia)
 export(standardise_names)
 export(state_diversity_counts)
 export(strip_names)
+export(strip_names_2)
 export(update_taxonomy)
 import(dplyr)
 import(stringr)

diff --git a/R/match_taxa.R b/R/match_taxa.R
@@ -519,10 +519,7 @@ match_taxa <- function(
       taxonomic_dataset = NA_character_,
       taxon_rank = "genus",
       aligned_name_tmp = paste0(stringr::word(cleaned_name,1), " sp. [", cleaned_name),
-      aligned_name = ifelse(is.na(identifier_string2),
-                            paste0(aligned_name_tmp, "]"),
-                            paste0(aligned_name_tmp, identifier_string2, "]")
-      ),
+      aligned_name = NA,
       aligned_reason = paste0(
         "Taxon name includes '--' (double dash) indicating an intergrade between two taxa, but exact and fuzzy matches fail to align to a genus in the APC or APNI (",
         Sys.Date(),
@@ -724,10 +721,7 @@ match_taxa <- function(
       taxonomic_dataset = NA_character_,
       taxon_rank = "genus",
       aligned_name_tmp = paste0(stringr::word(cleaned_name,1), " sp. [", cleaned_name),
-      aligned_name = ifelse(is.na(identifier_string2),
-                            paste0(aligned_name_tmp, "]"),
-                            paste0(aligned_name_tmp, identifier_string2, "]")
-      ),
+      aligned_name = NA,
       aligned_reason = paste0(
         "Taxon name includes '/' (slash) indicating an uncertain species identification but an accepted genus and taxon can only be aligned to genus-rank. Exact and fuzzy matches fail to align to a genus in the APC or APNI (",
         Sys.Date(),
@@ -1032,10 +1026,7 @@ match_taxa <- function(
       taxonomic_dataset = NA_character_,
       taxon_rank = "genus",
       aligned_name_tmp = paste0(stringr::word(cleaned_name,1), " sp. [", cleaned_name),
-      aligned_name = ifelse(is.na(identifier_string2),
-                            paste0(aligned_name_tmp, "]"),
-                            paste0(aligned_name_tmp, identifier_string2, "]")
-      ),
+      aligned_name = NA,
       aligned_reason = paste0(
         "Taxon name includes 'affinis' or 'aff' indicating an unknown taxon that bears an affinity to a different taxon in the same genus and taxon can only be aligned to genus-rank. Exact and fuzzy matches fail to align to a genus in the APC or APNI ",
         Sys.Date(),
@@ -1299,10 +1290,7 @@ match_taxa <- function(
       taxonomic_dataset = NA_character_,
       taxon_rank = "genus",
       aligned_name_tmp = paste0(stringr::word(cleaned_name,1), " x [", cleaned_name),
-      aligned_name = ifelse(is.na(identifier_string2),
-                            paste0(aligned_name_tmp, "]"),
-                            paste0(aligned_name_tmp, identifier_string2, "]")
-      ),
+      aligned_name = NA,
       aligned_reason = paste0(
         "Taxon name includes ' x ' indicating a hybrid taxon and taxon can only be aligned to genus-rank. Exact and fuzzy matches fail to align to a genus in the APC or APNI (",
         Sys.Date(),

diff --git a/R/strip_names.R b/R/strip_names.R
@@ -37,7 +37,7 @@ strip_names <- function(taxon_names) {
 #' Strip taxonomic names of subtaxa designations, filled words and special characters
 #'
 #' Given a vector of taxonomic names, this function removes subtaxa designations ("subsp.", "var.", "f.", and "ser"),
-#' additional filler words and characters (" x " [hybrid taxa], "sp.", "cf"), 
+#' additional filler words and characters (" x " for hybrid taxa, "sp.", "cf"), 
 #' special characters (e.g., "-", ".", "(", ")", "?"), and extra whitespace. The resulting vector
 #' of names is also converted to lowercase.
 #'
@@ -54,7 +54,7 @@ strip_names <- function(taxon_names) {
 #'               "Acacia sp.",
 #'               "Lepidium sp. Tanguin Hill (K.R.Newbey 10501)"))
 #'
-#' @noRd
+#' @export
 strip_names_2 <- function(taxon_names) {
   taxon_names %>%
     stringr::str_replace_all("\\.", "") %>%

diff --git a/R/update_taxonomy.R b/R/update_taxonomy.R
@@ -148,14 +148,13 @@ update_taxonomy <- function(aligned_data,
     dplyr::bind_rows(taxa_blank) %>%
     dplyr::mutate(
       suggested_name = ifelse(is.na(suggested_name), aligned_name, suggested_name),
-      suggested_name = ifelse(is.na(suggested_name), original_name, suggested_name),
       update_reason = ifelse(taxonomic_status_aligned == "accepted", "aligned name accepted by APC", update_reason),
       taxonomic_status = ifelse(is.na(taxonomic_status), "unknown", taxonomic_status),
       taxonomic_dataset = ifelse(stringr::str_detect(taxonomic_dataset, "APC"), "APC", taxonomic_dataset),
       ## `genus` was the first word of the `aligned_name` in the input table; now needs to be set to NA for unknown taxa
       genus = ifelse(taxonomic_status == "unknown", NA_character_, genus),
       taxon_rank = ifelse(taxonomic_status == "unknown", NA_character_, taxon_rank),
-      # the next line makes everythign incosistent. If we want low, should do on loading APC
+      # the next line makes everything inconsistent. If we want low, should do on loading APC
       taxon_rank = stringr::str_to_lower(taxon_rank),
       canonical_name = suggested_name,
       taxonomic_status_aligned = ifelse(is.na(taxonomic_status_aligned), NA_character_, taxonomic_status_aligned)
@@ -244,14 +243,14 @@ relevel_taxonomic_status_preferred_order <- function(taxonomic_status) {
 update_taxonomy_APC_genus <- function(data, resources) {
 
   if(is.null(data)) return(NULL)
-
+  
   data %>% 
   # merge in columns from APC, at the genus-level
   dplyr::left_join(
     by = "genus",
     resources$genera_all %>%
       dplyr::filter(stringr::str_detect(taxonomic_dataset, "APC")) %>%
-      dplyr::arrange(canonical_name, taxonomic_status) %>% ### how do I specify that I want to arrange by `preferred order`
+      dplyr::arrange(canonical_name, taxonomic_status) %>%
       dplyr::distinct(canonical_name, .keep_all = TRUE) %>%
       dplyr::mutate(
         genus = canonical_name,
@@ -276,8 +275,8 @@ update_taxonomy_APC_genus <- function(data, resources) {
       taxon_ID_genus = resources$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, 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),
+      # if there is an APC-accepted genus, replace whatever the initial genus was with the accepted genus, otherwise the suggested name is the aligned name
+      suggested_name = ifelse(taxonomic_status == "genus accepted", paste0(genus_accepted, aligned_minus_genus), aligned_name),
       # indicate taxonomic_status of the genus name in `aligned_name` and why it needed to be updated for the `suggested_name`
       genus_update_reason = as.character(my_order),
       genus = genus_accepted,

diff --git a/_pkgdown.yml b/_pkgdown.yml
@@ -36,6 +36,7 @@ reference:
   - update_taxonomy
   - standardise_names
   - strip_names
+  - strip_names_2
 - subtitle: Established status across states/territories
 - contents:
   - create_species_state_origin_matrix

diff --git a/man/strip_names_2.Rd b/man/strip_names_2.Rd
diff --git a/tests/testthat/benchmarks/test_matches_alignments_updates.csv b/tests/testthat/benchmarks/test_matches_alignments_updates.csv
@@ -31,8 +31,8 @@ Aporuelliaa abc--def,match_03c,Aporuellia sp. [Aporuelliaa abc--def; test_all_ma
 Driandra abc--def,match_03c,Dryandra sp. [Driandra abc--def; test_all_matches_TRUE],APC,genus,Banksia,FALSE,https://id.biodiversity.org.au/instance/apni/865048,https://id.biodiversity.org.au/name/apni/77744,Dryandra R.Br.
 Xyystidium abc--def,match_03d,Xystidium sp. [Xyystidium abc--def; test_all_matches_TRUE],APNI,genus,Xystidium,FALSE,NA,https://id.biodiversity.org.au/name/apni/244613,Xystidium Trin.
 Zygiaa abc--def,match_03d,Zygia sp. [Zygiaa abc--def; test_all_matches_TRUE],APNI,genus,Zygia,FALSE,NA,https://id.biodiversity.org.au/name/apni/65077,Zygia P.Browne
-Abcde fgh -- ijk,match_03e,Abcde sp. [Abcde fgh -- ijk; test_all_matches_TRUE],NA,genus,NA,FALSE,NA,NA,NA
-Ryandra abc--def,match_03e,Ryandra sp. [Ryandra abc--def; test_all_matches_TRUE],NA,genus,NA,FALSE,NA,NA,NA
+Abcde fgh -- ijk,match_03e,NA,NA,genus,NA,TRUE,NA,NA,NA
+Ryandra abc--def,match_03e,NA,NA,genus,NA,TRUE,NA,NA,NA
 Abildgaardia odontocarpa / Abildgaardia oxystachya,match_04a,Abildgaardia sp. [Abildgaardia odontocarpa / Abildgaardia oxystachya; test_all_matches_TRUE],APC,genus,Abildgaardia,FALSE,https://id.biodiversity.org.au/node/apni/2905759,https://id.biodiversity.org.au/name/apni/55984,Abildgaardia Vahl
 Acanthocarpus fimbriatus / Acanthocarpus mucronatus,match_04a,Acanthocarpus sp. [Acanthocarpus fimbriatus / Acanthocarpus mucronatus; test_all_matches_TRUE],APC,genus,Acanthocarpus,FALSE,https://id.biodiversity.org.au/node/apni/2899190,https://id.biodiversity.org.au/name/apni/72610,Acanthocarpus Lehm.
 Acanthocarpus fimbriatus / mucronatus,match_04a,Acanthocarpus sp. [Acanthocarpus fimbriatus / mucronatus; test_all_matches_TRUE],APC,genus,Acanthocarpus,FALSE,https://id.biodiversity.org.au/node/apni/2899190,https://id.biodiversity.org.au/name/apni/72610,Acanthocarpus Lehm.
@@ -51,8 +51,8 @@ Aporuelliaa abc / def,match_04c,Aporuellia sp. [Aporuelliaa abc / def; test_all_
 Drrandra abc / def,match_04c,Dryandra sp. [Drrandra abc / def; test_all_matches_TRUE],APC,genus,Banksia,FALSE,https://id.biodiversity.org.au/instance/apni/865048,https://id.biodiversity.org.au/name/apni/77744,Dryandra R.Br.
 Xyystidium abc/def,match_04d,Xystidium sp. [Xyystidium abc/def; test_all_matches_TRUE],APNI,genus,Xystidium,FALSE,NA,https://id.biodiversity.org.au/name/apni/244613,Xystidium Trin.
 Zygiaa abc / def,match_04d,Zygia sp. [Zygiaa abc / def; test_all_matches_TRUE],APNI,genus,Zygia,FALSE,NA,https://id.biodiversity.org.au/name/apni/65077,Zygia P.Browne
-Abcde fgh / ijk,match_04e,Abcde sp. [Abcde fgh / ijk; test_all_matches_TRUE],NA,genus,NA,FALSE,NA,NA,NA
-Ryandra abc / def,match_04e,Ryandra sp. [Ryandra abc / def; test_all_matches_TRUE],NA,genus,NA,FALSE,NA,NA,NA
+Abcde fgh / ijk,match_04e,NA,NA,genus,NA,TRUE,NA,NA,NA
+Ryandra abc / def,match_04e,NA,NA,genus,NA,TRUE,NA,NA,NA
 Cycas candida K.D.Hill,match_05a,Cycas candida,APC,species,Cycas candida,TRUE,https://id.biodiversity.org.au/node/apni/2893335,https://id.biodiversity.org.au/name/apni/188177,Cycas candida K.D.Hill
 Eremophila papillata Chinnock,match_05a,Eremophila papillata,APC,species,Eremophila papillata,TRUE,https://id.biodiversity.org.au/node/apni/2910890,https://id.biodiversity.org.au/name/apni/207453,Eremophila papillata Chinnock
 Acalypha indica var. australis F.M.Bailey,match_05b,Acalypha indica var. australis,APC,variety,Acalypha lanceolata,TRUE,https://id.biodiversity.org.au/instance/apni/889946,https://id.biodiversity.org.au/name/apni/72588,Acalypha indica var. australis F.M.Bailey
@@ -125,8 +125,8 @@ Aporuelliaa aff def,match_09c,Aporuellia sp. [Aporuelliaa aff. def; test_all_mat
 Drrandra affinis def,match_09c,Dryandra sp. [Drrandra aff. def; test_all_matches_TRUE],APC,genus,Banksia,FALSE,https://id.biodiversity.org.au/instance/apni/865048,https://id.biodiversity.org.au/name/apni/77744,Dryandra R.Br.
 Xyystidium aff. abc,match_09d,Xystidium sp. [Xyystidium aff. abc; test_all_matches_TRUE],APNI,genus,Xystidium,FALSE,NA,https://id.biodiversity.org.au/name/apni/244613,Xystidium Trin.
 Zygiaa aff. abc,match_09d,Zygia sp. [Zygiaa aff. abc; test_all_matches_TRUE],APNI,genus,Zygia,FALSE,NA,https://id.biodiversity.org.au/name/apni/65077,Zygia P.Browne
-Abcde affinis fgh,match_09e,Abcde sp. [Abcde aff. fgh; test_all_matches_TRUE],NA,genus,NA,FALSE,NA,NA,NA
-Rryandra aff def,match_09e,Rryandra sp. [Rryandra aff. def; test_all_matches_TRUE],NA,genus,NA,FALSE,NA,NA,NA
+Abcde affinis fgh,match_09e,NA,NA,genus,NA,TRUE,NA,NA,NA
+Rryandra aff def,match_09e,NA,NA,genus,NA,TRUE,NA,NA,NA
 Aceeena x ovinaaa,match_10a,Acaena x ovina,APC,species,Acaena x ovina,FALSE,https://id.biodiversity.org.au/taxon/apni/51446291,https://id.biodiversity.org.au/name/apni/72209,Acaena x ovina A.Cunn.
 Banksiia  serrratte,match_10a,Banksia serrata,APC,species,Banksia serrata,TRUE,https://id.biodiversity.org.au/taxon/apni/51293610,https://id.biodiversity.org.au/name/apni/109014,Banksia serrata L.f.
 Eremoophila opppositifolia ssp. rubraaa,match_10a,Eremophila oppositifolia subsp. rubra,APC,subspecies,Eremophila oppositifolia subsp. rubra,TRUE,https://id.biodiversity.org.au/node/apni/7951458,https://id.biodiversity.org.au/name/apni/117903,Eremophila oppositifolia subsp. rubra (C.T.White & W.D.Francis) Chinnock
@@ -154,8 +154,8 @@ Aporuelliaa abc x def,match_11c,Aporuellia x [Aporuelliaa abc x def; test_all_ma
 Drrandra x def,match_11c,Dryandra x [Drrandra x def; test_all_matches_TRUE],APC,genus,Banksia,FALSE,https://id.biodiversity.org.au/instance/apni/865048,https://id.biodiversity.org.au/name/apni/77744,Dryandra R.Br.
 Xyystidium x def,match_11d,Xystidium x [Xyystidium x def; test_all_matches_TRUE],APNI,genus,Xystidium,FALSE,NA,https://id.biodiversity.org.au/name/apni/244613,Xystidium Trin.
 Zygiaa abc x Zygia def,match_11d,Zygia x [Zygiaa abc x Zygia def; test_all_matches_TRUE],APNI,genus,Zygia,FALSE,NA,https://id.biodiversity.org.au/name/apni/65077,Zygia P.Browne
-Abcde fgh x ijk,match_11e,Abcde x [Abcde fgh x ijk; test_all_matches_TRUE],NA,genus,NA,FALSE,NA,NA,NA
-Ryandra abc x def,match_11e,Ryandra x [Ryandra abc x def; test_all_matches_TRUE],NA,genus,NA,FALSE,NA,NA,NA
+Abcde fgh x ijk,match_11e,NA,NA,genus,NA,TRUE,NA,NA,NA
+Ryandra abc x def,match_11e,NA,NA,genus,NA,TRUE,NA,NA,NA
 Baeckea sp. murchison river,match_12a,Baeckea sp. Murchison River (M.E.Trudgen 12009),APC,species,Baeckea sp. Murchison River (M.E.Trudgen 12009),TRUE,https://id.biodiversity.org.au/node/apni/2888052,https://id.biodiversity.org.au/name/apni/191267,Baeckea sp. Murchison River (M.E.Trudgen 12009) WA Herbarium
 Eremophila oppositifolia rubra (needle leaves),match_12a,Eremophila oppositifolia subsp. rubra,APC,subspecies,Eremophila oppositifolia subsp. rubra,TRUE,https://id.biodiversity.org.au/node/apni/7951458,https://id.biodiversity.org.au/name/apni/117903,Eremophila oppositifolia subsp. rubra (C.T.White & W.D.Francis) Chinnock
 Eremophila oppositifolia rubra early collection,match_12a,Eremophila oppositifolia subsp. rubra,APC,subspecies,Eremophila oppositifolia subsp. rubra,TRUE,https://id.biodiversity.org.au/node/apni/7951458,https://id.biodiversity.org.au/name/apni/117903,Eremophila oppositifolia subsp. rubra (C.T.White & W.D.Francis) Chinnock
@@ -220,5 +220,5 @@ Achneriia crevifoliaa,match_22b,Achneria sp. [Achneriia crevifoliaa; test_all_ma
 Actinocarpos,match_22b,Actinocarpus sp. [Actinocarpos; test_all_matches_TRUE],APC,genus,Damasonium,FALSE,https://id.biodiversity.org.au/instance/apni/884226,https://id.biodiversity.org.au/name/apni/74816,Actinocarpus R.Br.
 Drryandra,match_22b,Dryandra sp. [Drryandra; test_all_matches_TRUE],APC,genus,Banksia,FALSE,https://id.biodiversity.org.au/instance/apni/865048,https://id.biodiversity.org.au/name/apni/77744,Dryandra R.Br.
 Dryandraa,match_22b,Dryandra sp. [Dryandraa; test_all_matches_TRUE],APC,genus,Banksia,FALSE,https://id.biodiversity.org.au/instance/apni/865048,https://id.biodiversity.org.au/name/apni/77744,Dryandra R.Br.
-Actiniladum sp.,NA,NA,NA,NA,NA,FALSE,NA,NA,NA
-Ecalypha indica australis,NA,NA,NA,NA,NA,FALSE,NA,NA,NA
+Actiniladum sp.,NA,NA,NA,NA,NA,TRUE,NA,NA,NA
+Ecalypha indica australis,NA,NA,NA,NA,NA,TRUE,NA,NA,NA