(v3.0.1.9097) Never remove released taxa; link orthographic variants; registry line in NEWS

- taxonomy build: released taxa (since v2.0.0) that are missing are restored from their
  last release with their old code and last known status, and are exempt from later
  removals (decision by Matthijs S. Berends)
- exceptions are retired on purpose in data-raw/microorganisms_files/mo_code_retirements.csv
  with a reason: other organisms with the same genus name (e.g. the Graphium butterflies,
  wrongly in the Fungi until v3.0.1) and taxa whose genus is now only known in another
  domain; as.mo() translates these codes to NA with that reason
- synonyms without a current name are linked to an accepted name in the same genus that
  only differs by a Latin gender ending (e.g. Candida glabratus -> Nakaseomyces glabratus)
- placeholders such as "(unknown class)" never become records; codes of manually added
  records are matched by domain and name
- integrity rule released_taxon_missing; known defects of the data on main updated

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
dr. M.S. (Matthijs) BerendsandClaude Opus 5.5 committed 2026-10-03 14:15:08 +02:00
1 parent d683b9e7c6
commit 83a4853936
12 files changed
+5326 -23

No files matched your search

+20 -1
View File
@@ -61,6 +61,14 @@ read_mo_renames <- function(root = mo_repository_root()) {
# only approved renames count
renames[renames$approved_by != "" & renames$approved_date != "", , drop = FALSE]
}
# MO codes that are retired on purpose, because their taxon was another organism (see the README)
read_mo_retirements <- function(root = mo_repository_root()) {
file <- file.path(root, "data-raw", "microorganisms_files", "mo_code_retirements.csv")
if (is.null(root) || !file.exists(file)) {
return(NULL)
}
utils::read.csv(file, colClasses = "character", na.strings = character(0))
}
mo_repository_root <- function() {
# works from the root of the repository and from tests/testthat
for (path in c(".", "../..", "..")) {
@@ -78,7 +86,7 @@ mo_name_without_suffix <- function(x) {
# Returns a named list with one character vector per rule, each describing the records that break the rule.
# All vectors must be empty.
mo_integrity_issues <- function(df, registry = NULL, renames = NULL) {
mo_integrity_issues <- function(df, registry = NULL, renames = NULL, retirements = NULL) {
df <- as.data.frame(df, stringsAsFactors = FALSE)
df$mo <- as.character(df$mo)
describe <- function(rows) {
@@ -161,6 +169,17 @@ mo_integrity_issues <- function(df, registry = NULL, renames = NULL) {
issues$registered_code_with_other_taxon <- describe(
!is.na(registered_name) & mo_name_without_suffix(registered_name) != mo_name_without_suffix(df$fullname)
)
# a released taxon is never removed, unless its code was retired on purpose
expected <- registry$fullname
if (!is.null(renames) && nrow(renames) > 0) {
renamed <- match(registry$mo, renames$mo)
expected[!is.na(renamed)] <- renames$new_name[renamed[!is.na(renamed)]]
}
missing <- !mo_name_without_suffix(expected) %in% mo_name_without_suffix(df$fullname)
if (!is.null(retirements)) {
missing <- missing & !registry$mo %in% retirements$mo
}
issues$released_taxon_missing <- if (any(missing)) paste0(registry$mo[missing], " (", expected[missing], ")") else character(0)
}
issues
File diff suppressed because it is too large. Load diff
+10 -4
View File
@@ -97,7 +97,7 @@ sentinel_defects <- function() {
test_that("microorganisms: integrity rules", {
skip_on_cran()
issues <- mo_integrity_issues(microorganisms, registry = read_mo_registry(), renames = read_mo_renames())
issues <- mo_integrity_issues(microorganisms, registry = read_mo_registry(), renames = read_mo_renames(), retirements = read_mo_retirements())
for (rule in names(issues)) {
defects <- new_defects(rule, issues[[rule]])
expect_true(length(defects) == 0,
@@ -111,7 +111,7 @@ test_that("microorganisms: known defects are no longer than needed", {
# a known defect that is solved must be removed from the list, so that it cannot come back unnoticed
kd <- known_defects()
skip_if(nrow(kd) == 0)
issues <- mo_integrity_issues(microorganisms, registry = read_mo_registry(), renames = read_mo_renames())
issues <- mo_integrity_issues(microorganisms, registry = read_mo_registry(), renames = read_mo_renames(), retirements = read_mo_retirements())
current <- c(
unlist(lapply(names(issues), function(rule) paste(rule, issues[[rule]], sep = "|"))),
paste("sentinel", sentinel_defects(), sep = "|")
@@ -145,6 +145,12 @@ test_that("MO codes: every code of every release since v2.0.0 still means the sa
expected_name[!is.na(renamed)] <- renames$new_name[renamed[!is.na(renamed)]]
# deliberate in as.mo() since v2.0.0: the kingdom/domain Fungi returns the code for an unknown fungus
expected_name[registry$mo == "F_[KNG]_FUNGI"] <- "(unknown fungus)"
# codes retired on purpose (their taxon was another organism) must lead to NA
retirements <- read_mo_retirements()
retired_on_purpose <- registry$mo %in% retirements$mo
expect_true(all(is.na(suppressWarnings(as.mo(registry$mo[retired_on_purpose], info = FALSE)))))
registry <- registry[!retired_on_purpose, , drop = FALSE]
expected_name <- expected_name[!retired_on_purpose]
result <- suppressWarnings(as.mo(registry$mo, keep_synonyms = TRUE, info = FALSE))
result_name <- microorganisms$fullname[match(result, microorganisms$mo)]
wrong <- !is.na(result) & mo_name_without_suffix(result_name) != mo_name_without_suffix(expected_name)
@@ -154,8 +160,8 @@ test_that("MO codes: every code of every release since v2.0.0 still means the sa
)
# the internal lookup table must be up to date with the registry
retired <- registry$mo[!registry$mo %in% microorganisms$mo]
expect_setequal(AMR:::MO_RETIRED_CODES$old_mo, retired)
full_registry <- read_mo_registry()
expect_setequal(AMR:::MO_RETIRED_CODES$old_mo, full_registry$mo[!full_registry$mo %in% microorganisms$mo])
})
test_that("MO codes: codes of earlier releases are translated, never guessed", {