#2563 no_list_columns: split files (and clean up tests)

pharmaverse · Dec 13, 2024 · c5d8b2c · c5d8b2c
1 parent 88d6223
commit c5d8b2c
Show file tree

Hide file tree

Showing 5 changed files with 224 additions and 268 deletions.
diff --git a/R/derive_vars_atc.R b/R/derive_vars_atc.R
@@ -0,0 +1,108 @@
+#' Derive ATC Class Variables
+#'
+#' @description Add Anatomical Therapeutic Chemical class variables from `FACM` to `ADCM`.
+#'
+#' **Note:** This is a wrapper function for the more generic `derive_vars_transposed()`.
+#'
+#' @param dataset
+#'   `r roxygen_param_dataset(expected_vars = c("by_vars"))`
+#'
+#' @param dataset_facm FACM dataset
+#'
+#'   The variables specified by the `by_vars`, `id_vars`, and `value_var`
+#'   arguments and `FATESTCD` are required. The variables `by_vars`, `id_vars`,
+#'   and `FATESTCD` must be a unique key.
+#'
+#' @param by_vars Grouping variables
+#'
+#'  Keys used to merge `dataset_facm` with `dataset`.
+#'
+#' @param id_vars ID variables
+#'
+#'  Variables (excluding by_vars) that uniquely identify each observation in `dataset_merge`.
+#'
+#' `r roxygen_param_by_vars()`
+#'
+#' @param value_var The variable of `dataset_facm` containing the values of the
+#'   transposed variables
+#'
+#' @return The input dataset with ATC variables added
+#'
+#' @seealso [derive_vars_transposed()]
+#'
+#' @family der_occds
+#' @keywords der_occds
+#'
+#' @export
+#'
+#' @examples
+#' library(tibble)
+#'
+#' cm <- tribble(
+#'   ~STUDYID,  ~USUBJID,       ~CMGRPID, ~CMREFID,  ~CMDECOD,
+#'   "STUDY01", "BP40257-1001", "14",     "1192056", "PARACETAMOL",
+#'   "STUDY01", "BP40257-1001", "18",     "2007001", "SOLUMEDROL",
+#'   "STUDY01", "BP40257-1002", "19",     "2791596", "SPIRONOLACTONE"
+#' )
+#' facm <- tribble(
+#'   ~STUDYID,  ~USUBJID,       ~FAGRPID, ~FAREFID,  ~FATESTCD,  ~FASTRESC,
+#'   "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC1CD", "N",
+#'   "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC2CD", "N02",
+#'   "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC3CD", "N02B",
+#'   "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC4CD", "N02BE",
+#'   "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC1CD", "D",
+#'   "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC2CD", "D10",
+#'   "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC3CD", "D10A",
+#'   "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC4CD", "D10AA",
+#'   "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC1CD", "D",
+#'   "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC2CD", "D07",
+#'   "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC3CD", "D07A",
+#'   "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC4CD", "D07AA",
+#'   "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC1CD", "H",
+#'   "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC2CD", "H02",
+#'   "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC3CD", "H02A",
+#'   "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC4CD", "H02AB",
+#'   "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC1CD", "C",
+#'   "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC2CD", "C03",
+#'   "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC3CD", "C03D",
+#'   "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC4CD", "C03DA"
+#' )
+#'
+#' derive_vars_atc(cm, facm, id_vars = exprs(FAGRPID))
+derive_vars_atc <- function(dataset,
+                            dataset_facm,
+                            by_vars = exprs(
+                              !!!get_admiral_option("subject_keys"),
+                              CMREFID = FAREFID
+                            ),
+                            id_vars = NULL,
+                            value_var = FASTRESC) {
+  value_var <- assert_symbol(enexpr(value_var))
+  assert_vars(by_vars)
+  assert_vars(id_vars, optional = TRUE)
+  assert_data_frame(dataset, required_vars = replace_values_by_names(by_vars))
+  assert_data_frame(
+    dataset_facm,
+    required_vars = exprs(!!!by_vars, !!value_var, !!!id_vars, FATESTCD)
+  )
+
+  tryCatch(
+    data_transposed <- derive_vars_transposed(
+      dataset,
+      select(dataset_facm, !!!unname(by_vars), !!value_var, !!!id_vars, FATESTCD),
+      by_vars = by_vars,
+      id_vars = id_vars,
+      key_var = FATESTCD,
+      value_var = !!value_var,
+      filter = str_detect(FATESTCD, "^CMATC[1-4](CD)?$")
+    ),
+    merge_duplicates = function(cnd) {
+      cnd$message <- str_replace(cnd$message, "dataset_merge", "dataset_facm")
+      cnd$body[[1]] <- "Please check data and `by_vars` and `id_vars` arguments."
+      cnd_signal(cnd)
+    }
+  )
+  data_transposed %>%
+    select(-starts_with("FA")) %>%
+    rename_with(.fn = ~ str_remove(.x, "^CM"), .cols = starts_with("CMATC"))
+}
diff --git a/R/derive_vars_transposed.R b/R/derive_vars_transposed.R
@@ -181,132 +181,6 @@ derive_vars_transposed <- function(dataset,
         ),
         call = parent.frame(n = 4)
       )
-    },
-    "dplyr_error_join_relationship_one_to_many" = function(cnd) {
-      cli_abort(
-        message = c(
-          str_replace(
-            str_replace(
-              cnd$message, "`x`", "`dataset`"
-            ), "`y`", "the transposed `dataset_merge`"
-          ),
-          i = str_replace(
-            str_replace(
-              cnd$body, "`x`", "`dataset`"
-            ), "`y`", "the transposed `dataset_merge`"
-          )
-        ),
-        call = parent.frame(n = 4)
-      )
-    }
-  )
-}
-
-#' Derive ATC Class Variables
-#'
-#' @description Add Anatomical Therapeutic Chemical class variables from `FACM` to `ADCM`.
-#'
-#' **Note:** This is a wrapper function for the more generic `derive_vars_transposed()`.
-#'
-#' @param dataset
-#'   `r roxygen_param_dataset(expected_vars = c("by_vars"))`
-#'
-#' @param dataset_facm FACM dataset
-#'
-#'   The variables specified by the `by_vars`, `id_vars`, and `value_var`
-#'   arguments and `FATESTCD` are required. The variables `by_vars`, `id_vars`,
-#'   and `FATESTCD` must be a unique key.
-#'
-#' @param by_vars Grouping variables
-#'
-#'  Keys used to merge `dataset_facm` with `dataset`.
-#'
-#' @param id_vars ID variables
-#'
-#'  Variables (excluding by_vars) that uniquely identify each observation in `dataset_merge`.
-#'
-#' `r roxygen_param_by_vars()`
-#'
-#' @param value_var The variable of `dataset_facm` containing the values of the
-#'   transposed variables
-#'
-#' @return The input dataset with ATC variables added
-#'
-#' @seealso [derive_vars_transposed()]
-#'
-#' @family der_occds
-#' @keywords der_occds
-#'
-#' @export
-#'
-#' @examples
-#' library(tibble)
-#'
-#' cm <- tribble(
-#'   ~STUDYID,  ~USUBJID,       ~CMGRPID, ~CMREFID,  ~CMDECOD,
-#'   "STUDY01", "BP40257-1001", "14",     "1192056", "PARACETAMOL",
-#'   "STUDY01", "BP40257-1001", "18",     "2007001", "SOLUMEDROL",
-#'   "STUDY01", "BP40257-1002", "19",     "2791596", "SPIRONOLACTONE"
-#' )
-#' facm <- tribble(
-#'   ~STUDYID,  ~USUBJID,       ~FAGRPID, ~FAREFID,  ~FATESTCD,  ~FASTRESC,
-#'   "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC1CD", "N",
-#'   "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC2CD", "N02",
-#'   "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC3CD", "N02B",
-#'   "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC4CD", "N02BE",
-#'   "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC1CD", "D",
-#'   "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC2CD", "D10",
-#'   "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC3CD", "D10A",
-#'   "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC4CD", "D10AA",
-#'   "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC1CD", "D",
-#'   "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC2CD", "D07",
-#'   "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC3CD", "D07A",
-#'   "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC4CD", "D07AA",
-#'   "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC1CD", "H",
-#'   "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC2CD", "H02",
-#'   "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC3CD", "H02A",
-#'   "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC4CD", "H02AB",
-#'   "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC1CD", "C",
-#'   "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC2CD", "C03",
-#'   "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC3CD", "C03D",
-#'   "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC4CD", "C03DA"
-#' )
-#'
-#' derive_vars_atc(cm, facm, id_vars = exprs(FAGRPID))
-derive_vars_atc <- function(dataset,
-                            dataset_facm,
-                            by_vars = exprs(
-                              !!!get_admiral_option("subject_keys"),
-                              CMREFID = FAREFID
-                            ),
-                            id_vars = NULL,
-                            value_var = FASTRESC) {
-  value_var <- assert_symbol(enexpr(value_var))
-  assert_vars(by_vars)
-  assert_vars(id_vars, optional = TRUE)
-  assert_data_frame(dataset, required_vars = replace_values_by_names(by_vars))
-  assert_data_frame(
-    dataset_facm,
-    required_vars = exprs(!!!by_vars, !!value_var, !!!id_vars, FATESTCD)
-  )
-
-  tryCatch(
-    data_transposed <- derive_vars_transposed(
-      dataset,
-      select(dataset_facm, !!!unname(by_vars), !!value_var, !!!id_vars, FATESTCD),
-      by_vars = by_vars,
-      id_vars = id_vars,
-      key_var = FATESTCD,
-      value_var = !!value_var,
-      filter = str_detect(FATESTCD, "^CMATC[1-4](CD)?$")
-    ),
-    merge_duplicates = function(cnd) {
-      cnd$message <- str_replace(cnd$message, "dataset_merge", "dataset_facm")
-      cnd$body[[1]] <- "Please check data and `by_vars` and `id_vars` arguments."
-      cnd_signal(cnd)
     }
   )
-  data_transposed %>%
-    select(-starts_with("FA")) %>%
-    rename_with(.fn = ~ str_remove(.x, "^CM"), .cols = starts_with("CMATC"))
 }
diff --git a/tests/testthat/_snaps/derive_vars_transposed.md b/tests/testthat/_snaps/derive_vars_transposed.md
@@ -12,13 +12,25 @@
       2 STUDY01 P02        31     3
       3 STUDY01 P03        42    NA
 
-# derive_vars_atc Test 5: error if facm not unique
+# derive_vars_transposed Test 4: error if `relationship` is unexpected
 
     Code
-      derive_vars_atc(dataset = cm, dataset_facm = facm)
+      cm %>% derive_vars_transposed(facm, by_vars = exprs(USUBJID, CMREFID = FAREFID),
+      id_vars = exprs(FAGRPID), key_var = FATESTCD, value_var = FASTRESC,
+      relationship = "one-to-one")
     Condition
-      Error in `signal_duplicate_records()`:
-      ! Dataset `dataset_facm` contains duplicate records with respect to `STUDYID`, `USUBJID`, `FAREFID`, and `FATESTCD`
-      Please check data and `by_vars` and `id_vars` arguments.
-      i Run `admiral::get_duplicates_dataset()` to access the duplicate records
+      Error in `tryCatch()`:
+      ! Each row in `dataset` must match at most 1 row in the transposed `dataset_merge`.
+      i Row 2 of `dataset` matches multiple rows in the transposed `dataset_merge`.
+
+---
+
+    Code
+      cm %>% derive_vars_transposed(facm, by_vars = exprs(USUBJID, CMREFID = FAREFID),
+      id_vars = exprs(FAGRPID), key_var = FATESTCD, value_var = FASTRESC,
+      relationship = "many-to-one")
+    Condition
+      Error in `derive_vars_transposed()`:
+      ! Each row in `dataset` must match at most 1 row in the transposed `dataset_merge`.
+      i Row 2 of `dataset` matches multiple rows in the transposed `dataset_merge`.
 
diff --git a/tests/testthat/test-derive_vars_atc.R b/tests/testthat/test-derive_vars_atc.R
@@ -0,0 +1,90 @@
+## Test 1: ATC variables are merged properly ----
+test_that("derive_vars_atc Test 1: ATC variables are merged properly", {
+  cm <- tibble::tribble(
+    ~STUDYID,  ~USUBJID,       ~CMGRPID, ~CMREFID,  ~CMDECOD,
+    "STUDY01", "BP40257-1001", "14",     "1192056", "PARACETAMOL",
+    "STUDY01", "BP40257-1001", "18",     "2007001", "SOLUMEDROL",
+    "STUDY01", "BP40257-1002", "19",     "2791596", "SPIRONOLACTONE"
+  )
+  facm <- tibble::tribble(
+    ~STUDYID,  ~USUBJID,       ~FAGRPID, ~FAREFID,  ~FATESTCD,  ~FASTRESC,
+    "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC1CD", "N",
+    "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC2CD", "N02",
+    "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC3CD", "N02B",
+    "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC4CD", "N02BE",
+    "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC1CD", "D",
+    "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC2CD", "D10",
+    "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC3CD", "D10A",
+    "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC4CD", "D10AA",
+    "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC1CD", "D",
+    "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC2CD", "D07",
+    "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC3CD", "D07A",
+    "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC4CD", "D07AA",
+    "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC1CD", "H",
+    "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC2CD", "H02",
+    "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC3CD", "H02A",
+    "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC4CD", "H02AB",
+    "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC1CD", "C",
+    "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC2CD", "C03",
+    "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC3CD", "C03D",
+    "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC4CD", "C03DA"
+  )
+  # nolint start
+  expected_output <- tibble::tribble(
+    ~STUDYID, ~USUBJID, ~CMGRPID, ~CMREFID, ~CMDECOD, ~ATC1CD, ~ATC2CD, ~ATC3CD, ~ATC4CD,
+    "STUDY01", "BP40257-1001", "14", "1192056", "PARACETAMOL", "N", "N02", "N02B", "N02BE",
+    "STUDY01", "BP40257-1001", "18", "2007001", "SOLUMEDROL", "D", "D07", "D07A", "D07AA",
+    "STUDY01", "BP40257-1001", "18", "2007001", "SOLUMEDROL", "D", "D10", "D10A", "D10AA",
+    "STUDY01", "BP40257-1001", "18", "2007001", "SOLUMEDROL", "H", "H02", "H02A", "H02AB",
+    "STUDY01", "BP40257-1002", "19", "2791596", "SPIRONOLACTONE", "C", "C03", "C03D", "C03DA"
+  )
+  # nolint end
+  actual_output <- derive_vars_atc(
+    dataset = cm,
+    dataset_facm = facm,
+    id_vars = exprs(FAGRPID)
+  )
+
+  expect_dfs_equal(expected_output, actual_output, keys = c("USUBJID", "CMDECOD", "ATC4CD"))
+})
+
+## Test 2: error if facm not unique ----
+test_that("derive_vars_atc Test 2: error if facm not unique", {
+  cm <- tibble::tribble(
+    ~STUDYID,  ~USUBJID,       ~CMGRPID, ~CMREFID,  ~CMDECOD,
+    "STUDY01", "BP40257-1001", "14",     "1192056", "PARACETAMOL",
+    "STUDY01", "BP40257-1001", "18",     "2007001", "SOLUMEDROL",
+    "STUDY01", "BP40257-1002", "19",     "2791596", "SPIRONOLACTONE"
+  )
+  facm <- tibble::tribble(
+    ~STUDYID,  ~USUBJID,       ~FAGRPID, ~FAREFID,  ~FATESTCD,  ~FASTRESC,
+    "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC1CD", "N",
+    "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC2CD", "N02",
+    "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC3CD", "N02B",
+    "STUDY01", "BP40257-1001", "1",      "1192056", "CMATC4CD", "N02BE",
+    "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC1CD", "D",
+    "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC2CD", "D10",
+    "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC3CD", "D10A",
+    "STUDY01", "BP40257-1001", "1",      "2007001", "CMATC4CD", "D10AA",
+    "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC1CD", "D",
+    "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC2CD", "D07",
+    "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC3CD", "D07A",
+    "STUDY01", "BP40257-1001", "2",      "2007001", "CMATC4CD", "D07AA",
+    "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC1CD", "H",
+    "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC2CD", "H02",
+    "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC3CD", "H02A",
+    "STUDY01", "BP40257-1001", "3",      "2007001", "CMATC4CD", "H02AB",
+    "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC1CD", "C",
+    "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC2CD", "C03",
+    "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC3CD", "C03D",
+    "STUDY01", "BP40257-1002", "1",      "2791596", "CMATC4CD", "C03DA"
+  )
+
+  expect_snapshot(
+    derive_vars_atc(
+      dataset = cm,
+      dataset_facm = facm
+    ),
+    error = TRUE
+  )
+})