Merge branch 'devel' into 1940_reevaluate_package_runtime_dependencie…

…s@devel

.Rprofile

@@ -1,7 +1,7 @@
 # Set renv profile base on R version.
 renv_profile <- paste(R.version$major, substr(R.version$minor, 1, 1), sep = ".")
 if (file.exists("./renv/profile")) {
-   message("Using renv profile from `renv/profile` file.")
+  message("Using renv profile from `renv/profile` file.")
 } else if (renv_profile %in% c("4.1", "4.2", "4.3")) {
   message("Set renv profile to `", renv_profile, "`")
   Sys.setenv("RENV_PROFILE" = renv_profile)
@@ -11,6 +11,6 @@ if (file.exists("./renv/profile")) {
 
 if ((Sys.getenv("GITHUB_ACTIONS") != "") || (Sys.getenv("DOCKER_CONTAINER_CONTEXT") != "")) {
   options(repos = c(CRAN = "https://cran.rstudio.com"))
-  Sys.setenv(RENV_AUTOLOADER_ENABLED=FALSE)
+  Sys.setenv(RENV_AUTOLOADER_ENABLED = FALSE)
 }
 source("renv/activate.R")

.github/workflows/cran-status.yml

@@ -17,7 +17,7 @@ jobs:
     with:
       # Whom should the issue be assigned to if errors are encountered
       # in the CRAN status checks?
-      issue-assignees: "bundfussr,esimms999-gsk,thomas-neitmann,bms63"
+      issue-assignees: "bundfussr,esimms999-gsk,manciniedoardo,bms63"
       # Create an issue if one or more of the following
       # statuses are reported on the check report.
       statuses: "WARN,ERROR,NOTE"

NEWS.md

@@ -8,6 +8,9 @@
 
 - The function `derive_var_extreme_flag()` has a new function argument, `flag_all` that additionally flags all records if the first or last record is not unique. (#1979)
 
+- The function `derive_vars_dy()` is updated to avoid potential error when the input `dataset` with columns ending with `temp`. (#2012)
+
+
 ## Breaking Changes
 - The following functions, which were deprecated in previous `{admiral}` versions, have been removed: (#1950)
 
@@ -34,13 +37,21 @@
 - The `filter` argument in `derive_extreme_records()` was deprecated in favor of
 the `filter_add` using the next phase of the deprecation process. (#1950)
 
+- The `na_val` argument in `derive_var_shift()` has been deprecated in favor of 
+`missing_value` using the first phase of the deprecation process. (#2014)
+
+- The `dataset_expected_obs` argument in `derive_expected_records()` and `derive_locf_records()`
+has been deprecated in favor of `dataset_ref`. (#2037)
+
 ## Documentation
 
 
 ## Various
 
 - The list of package authors/contributors has been reformatted so that those who are actively maintaining the code base are now marked as *authors*, whereas those who made a significant contribution in the past are now down as *contributors*. All other acknowledgements have been moved to README section (#1941).
 
+- `derive_vars_joined()` had two bugs with regards to duplicates messaging and when `new_vars` was set to `NULL` that have now been addressed (#1966). 
+
 # admiral 0.11.1
 
 - Fix bug in `derive_param_tte()`. (#1962)

R/derive_expected_records.R

@@ -5,17 +5,19 @@
 #'
 #' @param dataset Input dataset
 #'
-#'   A data frame, the columns from `dataset_expected_obs` and specified by the
+#'   A data frame, the columns from `dataset_ref` and specified by the
 #'   `by_vars` parameter are expected.
 #'
-#' @param dataset_expected_obs Expected observations dataset
+#' @param dataset_expected_obs *Deprecated*, please use `dataset_ref` instead.
+#'
+#' @param dataset_ref Expected observations dataset
 #'
 #'   Data frame with the expected observations, e.g., all the expected
 #'   combinations of `PARAMCD`, `PARAM`, `AVISIT`, `AVISITN`, ...
 #'
 #' @param by_vars Grouping variables
 #'
-#'   For each group defined by `by_vars` those observations from `dataset_expected_obs`
+#'   For each group defined by `by_vars` those observations from `dataset_ref`
 #'   are added to the output dataset which do not have a corresponding observation
 #'   in the input dataset.
 #'
@@ -31,7 +33,7 @@
 #'   "TDOSE", PARCAT1 = "OVERALL")`.
 #'
 #' @details For each group (the variables specified in the `by_vars` parameter),
-#' those records from `dataset_expected_obs` that are missing in the input
+#' those records from `dataset_ref` that are missing in the input
 #' dataset are added to the output dataset.
 #'
 #' @return The input dataset with the missed expected observations added for each
@@ -63,7 +65,7 @@
 #'
 #' derive_expected_records(
 #'   dataset = adqs,
-#'   dataset_expected_obs = parm_visit_ref,
+#'   dataset_ref = parm_visit_ref,
 #'   by_vars = exprs(USUBJID, PARAMCD),
 #'   set_values_to = exprs(DTYPE = "DERIVED")
 #' )
@@ -78,35 +80,46 @@
 #'
 #' derive_expected_records(
 #'   dataset = adqs,
-#'   dataset_expected_obs = parm_visit_ref,
+#'   dataset_ref = parm_visit_ref,
 #'   by_vars = exprs(USUBJID, PARAMCD),
 #'   set_values_to = exprs(DTYPE = "DERIVED")
 #' )
 #'
 derive_expected_records <- function(dataset,
                                     dataset_expected_obs,
+                                    dataset_ref,
                                     by_vars = NULL,
                                     set_values_to = NULL) {
+  if (!missing(dataset_expected_obs)) {
+    deprecate_warn(
+      "0.12.0",
+      "derive_expected_records(dataset_expected_obs = )",
+      "derive_expected_records(dataset_ref = )"
+    )
+    assert_data_frame(dataset_expected_obs)
+    dataset_ref <- dataset_expected_obs
+  }
+
   # Check input parameters
   assert_vars(by_vars, optional = TRUE)
-  assert_data_frame(dataset_expected_obs)
+  assert_data_frame(dataset_ref)
   assert_data_frame(
     dataset,
-    required_vars = expr_c(by_vars, chr2vars(colnames(dataset_expected_obs)))
+    required_vars = expr_c(by_vars, chr2vars(colnames(dataset_ref)))
   )
   assert_varval_list(set_values_to, optional = TRUE)
 
   # Derive expected records
-  ## ids: Variables from by_vars but not in dataset_expected_obs
+  ## ids: Variables from by_vars but not in dataset_ref
   ids <- dataset %>%
-    select(!!!setdiff(by_vars, chr2vars(colnames(dataset_expected_obs)))) %>%
+    select(!!!setdiff(by_vars, chr2vars(colnames(dataset_ref)))) %>%
     distinct()
 
   if (ncol(ids) > 0) {
     exp_obsv <- ids %>%
-      crossing(dataset_expected_obs)
+      crossing(dataset_ref)
   } else {
-    exp_obsv <- dataset_expected_obs
+    exp_obsv <- dataset_ref
   } # tmp workaround, update after using tidyr 1.2.0
 
   exp_obs_vars <- exp_obsv %>%

R/derive_joined.R

@@ -353,6 +353,19 @@ derive_vars_joined <- function(dataset,
   if (is.null(new_vars)) {
     new_vars <- chr2vars(colnames(dataset_add))
   }
+  preexisting_vars <- chr2vars(colnames(dataset))
+  preexisting_vars_no_by_vars <- preexisting_vars[which(!(preexisting_vars %in% by_vars))]
+  duplicates <- intersect(replace_values_by_names(new_vars), preexisting_vars_no_by_vars)
+  if (length(duplicates) > 0) {
+    err_msg <- sprintf(
+      paste(
+        "The following columns in `dataset_add` have naming conflicts with `dataset`,\n",
+        "please make the appropriate modifications to `new_vars`, with respect to:\n%s"
+      ),
+      enumerate(vars2chr(duplicates))
+    )
+    abort(err_msg)
+  }
 
   # number observations of the input dataset to get a unique key
   # (by_vars and tmp_obs_nr)
@@ -371,7 +384,7 @@ derive_vars_joined <- function(dataset,
     filter_if(filter_add) %>%
     select(
       !!!by_vars,
-      !!!chr2vars(names(order)),
+      !!!replace_values_by_names(extract_vars(order)),
       !!!replace_values_by_names(join_vars),
       !!!intersect(unname(extract_vars(new_vars)), chr2vars(colnames(dataset_add)))
     )
@@ -410,6 +423,7 @@ derive_vars_joined <- function(dataset,
       by_vars = exprs(!!!by_vars_left, !!tmp_obs_nr),
       new_vars = add_suffix_to_vars(new_vars, vars = common_vars, suffix = ".join"),
       missing_values = missing_values,
+      check_type = check_type,
       duplicate_msg = paste(
         paste(
           "After applying `filter_join` the joined dataset contains more",

R/derive_locf_records.R

@@ -8,14 +8,17 @@
 #'   The columns specified by the `by_vars`, `analysis_var`, `order`,
 #'   `keep_vars` parameters are expected.
 #'
-#' @param dataset_expected_obs Expected observations dataset
+#' @param dataset_expected_obs *Deprecated*, please use `dataset_ref` instead.
+#'
+#' @param dataset_ref Expected observations dataset
 #'
 #'   Data frame with all the combinations of `PARAMCD`, `PARAM`, `AVISIT`,
 #'   `AVISITN`, ... which are expected in the dataset is expected.
 #'
+#'
 #' @param by_vars Grouping variables
 #'
-#'   For each group defined by `by_vars` those observations from `dataset_expected_obs`
+#'   For each group defined by `by_vars` those observations from `dataset_ref`
 #'   are added to the output dataset which do not have a corresponding observation
 #'   in the input dataset or for which `analysis_var` is `NA` for the corresponding observation
 #'   in the input dataset.
@@ -40,7 +43,7 @@
 #' @author G Gayatri
 #'
 #' @details For each group (with respect to the variables specified for the
-#' by_vars parameter) those observations from dataset_expected_obs are added to
+#' by_vars parameter) those observations from `dataset_ref` are added to
 #' the output dataset
 #' - which do not have a corresponding observation in the input dataset or
 #' - for which `analysis_var` is NA for the corresponding observation in the input dataset.
@@ -101,19 +104,30 @@
 #' )
 #'
 #' derive_locf_records(
-#'   data = advs,
-#'   dataset_expected_obs = advs_expected_obsv,
+#'   dataset = advs,
+#'   dataset_ref = advs_expected_obsv,
 #'   by_vars = exprs(STUDYID, USUBJID, PARAMCD),
 #'   order = exprs(AVISITN, AVISIT),
 #'   keep_vars = exprs(PARAMN)
 #' )
 #'
 derive_locf_records <- function(dataset,
                                 dataset_expected_obs,
+                                dataset_ref,
                                 by_vars,
                                 analysis_var = AVAL,
                                 order,
                                 keep_vars = NULL) {
+  if (!missing(dataset_expected_obs)) {
+    deprecate_warn(
+      "0.12.0",
+      "derive_locf_records(dataset_expected_obs = )",
+      "derive_locf_records(dataset_ref = )"
+    )
+    assert_data_frame(dataset_expected_obs)
+    dataset_ref <- dataset_expected_obs
+  }
+
   #### Input Checking ####
   analysis_var <- assert_symbol(enexpr(analysis_var))
 
@@ -123,27 +137,27 @@ derive_locf_records <- function(dataset,
   assert_expr_list(order)
 
   # Check by_vars and order variables in input datasets
-  assert_data_frame(dataset_expected_obs)
+  assert_data_frame(dataset_ref)
   assert_data_frame(
     dataset,
     required_vars = expr_c(
       by_vars, analysis_var, extract_vars(order), keep_vars,
-      chr2vars(colnames(dataset_expected_obs))
+      chr2vars(colnames(dataset_ref))
     )
   )
 
 
-  #### Prepping 'dataset_expected_obs' ####
+  #### Prepping 'dataset_ref' ####
 
 
   # Get the IDs from input dataset for which the expected observations are to be added
 
   ids <- dataset %>%
-    select(!!!setdiff(by_vars, chr2vars(colnames(dataset_expected_obs)))) %>%
+    select(!!!setdiff(by_vars, chr2vars(colnames(dataset_ref)))) %>%
     distinct()
 
   exp_obsv <- ids %>%
-    crossing(dataset_expected_obs)
+    crossing(dataset_ref)
 
 
 

R/derive_var_shift.R

@@ -14,7 +14,9 @@
 #'
 #' @param to_var Variable containing value to shift to.
 #'
-#' @param na_val Character string to replace missing values in `from_var` or `to_var`.
+#' @param na_val *Deprecated*, please use `missing_value` instead.
+#'
+#' @param missing_value Character string to replace missing values in `from_var` or `to_var`.
 #'
 #'  Default: "NULL"
 #'
@@ -24,7 +26,7 @@
 #'
 #' @details `new_var` is derived by concatenating the values of `from_var` to values of `to_var`
 #' (e.g. "NORMAL to HIGH"). When `from_var` or `to_var` has missing value, the
-#' missing value is replaced by `na_val` (e.g. "NORMAL to NULL").
+#' missing value is replaced by `missing_value` (e.g. "NORMAL to NULL").
 #'
 #'
 #' @return The input dataset with the character shift variable added
@@ -71,20 +73,28 @@ derive_var_shift <- function(dataset,
                              new_var,
                              from_var,
                              to_var,
-                             na_val = "NULL",
+                             na_val,
+                             missing_value = "NULL",
                              sep_val = " to ") {
+  ### BEGIN DEPRECATION
+  if (!missing(na_val)) {
+    deprecate_warn("0.12.0", "derive_var_shift(na_val = )", "derive_var_shift(missing_value = )")
+    missing_value <- na_val
+  }
+  ### END DEPRECATION
+
   new_var <- assert_symbol(enexpr(new_var))
   from_var <- assert_symbol(enexpr(from_var))
   to_var <- assert_symbol(enexpr(to_var))
-  na_val <- assert_character_scalar(na_val)
+  missing_value <- assert_character_scalar(missing_value)
   sep_val <- assert_character_scalar(sep_val)
   assert_data_frame(dataset, required_vars = exprs(!!from_var, !!to_var))
 
   # Derive shift variable. If from_var or to_var has missing value then set to na_val.
   dataset %>%
     mutate(
-      temp_from_var = if_else(is.na(!!from_var), !!na_val, as.character(!!from_var)),
-      temp_to_var = if_else(is.na(!!to_var), !!na_val, as.character(!!to_var))
+      temp_from_var = if_else(is.na(!!from_var), !!missing_value, as.character(!!from_var)),
+      temp_to_var = if_else(is.na(!!to_var), !!missing_value, as.character(!!to_var))
     ) %>%
     mutate(
       !!new_var := paste(temp_from_var, temp_to_var, sep = !!sep_val)

R/derive_vars_dy.R

@@ -102,31 +102,21 @@ derive_vars_dy <- function(dataset,
     abort(err_msg)
   }
 
-  dy_vars <- if_else(
+  # named vector passed to `.names` in `across()` to derive name of dy_vars
+  dy_vars <- set_names(if_else(
     source_names == "",
     str_replace_all(vars2chr(source_vars), "(DT|DTM)$", "DY"),
     source_names
-  )
+  ), vars2chr(source_vars))
+
   warn_if_vars_exist(dataset, dy_vars)
 
-  if (n_vars > 1L) {
-    dataset %>%
-      mutate(
-        across(
-          .cols = vars2chr(unname(source_vars)),
-          .fns = list(temp = ~
-            compute_duration(start_date = !!reference_date, end_date = .))
-        )
-      ) %>%
-      rename_with(
-        .cols = ends_with("temp"),
-        .fn = ~dy_vars
-      )
-  } else {
-    dataset %>%
-      mutate(
-        !!sym(dy_vars) :=
-          compute_duration(start_date = !!reference_date, end_date = !!source_vars[[1]])
+  dataset %>%
+    mutate(
+      across(
+        .cols = vars2chr(unname(source_vars)),
+        .fns = ~ compute_duration(start_date = !!reference_date, end_date = .x),
+        .names = "{dy_vars}"
       )
-  }
+    )
 }