Merge branch 'main' into feature/my_first_fcn

.lintr

@@ -3,7 +3,7 @@ linters: linters_with_defaults(
     object_usage_linter=NULL,
     cyclocomp_linter(complexity_limit = 22),
     indentation_linter=NULL,
-    undesirable_function_linter = undesirable_function_linter()
+    undesirable_function_linter = undesirable_function_linter(symbol_is_undesirable = FALSE)
   )
 exclusions: list(
     "R/data.R" = Inf,

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: admiral
 Title: ADaM in R Asset Library
-Version: 1.1.1.9032
+Version: 1.1.1.9042
 Authors@R: c(
     person("Ben", "Straub", , "[email protected]", role = c("aut", "cre")),
     person("Stefan", "Bundfuss", role = "aut",
@@ -17,6 +17,8 @@ Authors@R: c(
     person("Sadchla", "Mascary", role = "aut"),
     person("Gordon", "Miller", role = "aut"),
     person("Lina", "Patil", role = "aut"),
+    person("Jim", "Rothstein", role = "aut",
+           comment = c(ORCID = "0009-0009-8659-6071")),
     person("Sophie", "Shapcott", role = "aut"),
     person("Eric", "Simms", role = "aut"),
     person("Daniel", "Sjoberg", role = "aut",
@@ -44,7 +46,7 @@ BugReports: https://github.com/pharmaverse/admiral/issues
 Depends: 
     R (>= 4.0)
 Imports:
-    admiraldev (>= 1.1.0),
+    admiraldev (>= 1.1.0.9007),
     cli (>= 3.6.2),
     dplyr (>= 1.0.5),
     hms (>= 0.5.3),

NAMESPACE

@@ -156,6 +156,7 @@ export(restrict_derivation)
 export(set_admiral_options)
 export(signal_duplicate_records)
 export(slice_derivation)
+export(transform_range)
 export(use_ad_template)
 export(yn_to_numeric)
 import(admiraldev)
@@ -167,6 +168,7 @@ importFrom(cli,cli_text)
 importFrom(cli,cli_warn)
 importFrom(dplyr,across)
 importFrom(dplyr,arrange)
+importFrom(dplyr,between)
 importFrom(dplyr,bind_cols)
 importFrom(dplyr,bind_rows)
 importFrom(dplyr,case_when)
@@ -180,6 +182,7 @@ importFrom(dplyr,first)
 importFrom(dplyr,full_join)
 importFrom(dplyr,group_by)
 importFrom(dplyr,group_by_at)
+importFrom(dplyr,group_split)
 importFrom(dplyr,if_else)
 importFrom(dplyr,lag)
 importFrom(dplyr,mutate)
@@ -256,6 +259,7 @@ importFrom(rlang,call2)
 importFrom(rlang,call_name)
 importFrom(rlang,caller_env)
 importFrom(rlang,cnd_muffle)
+importFrom(rlang,cnd_signal)
 importFrom(rlang,current_env)
 importFrom(rlang,enexpr)
 importFrom(rlang,enexprs)
@@ -319,4 +323,5 @@ importFrom(tidyselect,matches)
 importFrom(tidyselect,vars_select)
 importFrom(utils,capture.output)
 importFrom(utils,file.edit)
+importFrom(utils,object.size)
 importFrom(utils,str)

NEWS.md

@@ -3,9 +3,11 @@
 ## New Features
 
 - New function `derive_vars_cat()` for deriving pairs of variables or more, e.g. 
-`AVALCATx` & `AVALCAxN`. (#2480)
+`AVALCATy` & `AVALCAyN`. (#2480)
 - New function `derive_vars_crit_flag()` for deriving criterion flag variables
 (`CRITy`, `CRITyFL`, `CRITyFLN`). (#2468)
+- New function `transform_range()`  to transform values from a source range to a
+target range. (#2571)
 - Replace use of `data("sdtm")` with `sdtm <- pharmaverse::sdtm` in templates and vignettes. (#2498)
 
 - Remove `dthcaus_source()` calls in `ADSL` template because they are deprecated. (#2517)
@@ -39,8 +41,18 @@ or that the queries dataset contains duplicates. (#2543)
 
 - test scripts, R, and markdown files for `create_single_dose_dataset` and `occds.Rmd` updated to include a `STUDYID` column because of `get_admiral_option("subject_keys")` update above. (#2501)
 
+- Update `derive_vars_period()` to make it work when there is only one new variable. (#2582)
+
 - In `get_summary_records()`, previously deprecated formal arguments `analysis_var` and `summary_fun` now removed from function, documentation, tests etc. (#2521)
 
+- A check was added to `derive_vars_transposed()` and `derive_vars_atc()` which
+stops execution if the records in `dataset_merge` or `dataset_facm` respectively
+are not unique. (#2563)
+
+- The functions `derive_vars_joined()`, `derive_var_joined_exist_flag()`,
+`derive_extreme_event()`, and `filter_joined()` were updated to reduce their
+memory consumption. (#2590)
+
 ## Breaking Changes
 
 - The following function arguments are entering the next phase of the deprecation process: (#2487)

R/admiral-package.R

@@ -2,11 +2,11 @@
 #' @family internal
 #' @import admiraldev
 #' @importFrom cli cli_abort ansi_collapse cli_div cli_inform cli_text cli_warn
-#' @importFrom dplyr across arrange bind_cols bind_rows case_when coalesce
-#'             desc distinct ends_with everything filter first full_join
-#'             group_by group_by_at if_else mutate n n_distinct na_if pull
-#'             rename rename_with row_number select semi_join slice starts_with
-#'             summarise summarise_all tibble tribble ungroup union lag
+#' @importFrom dplyr across arrange between bind_cols bind_rows case_when
+#'   coalesce desc distinct ends_with everything filter first full_join group_by
+#'   group_by_at group_split if_else lag mutate n n_distinct na_if pull rename
+#'   rename_with row_number select semi_join slice starts_with summarise
+#'   summarise_all tibble tribble ungroup union
 #' @importFrom hms as_hms
 #' @importFrom lifecycle deprecate_warn deprecate_stop deprecated
 #' @importFrom lubridate %--% as_datetime ceiling_date date days duration
@@ -17,11 +17,11 @@
 #'             map_if map_lgl map2 modify_at modify_if pmap reduce transpose
 #'             walk
 #' @importFrom rlang := abort arg_match as_data_mask as_function as_label
-#'   as_name as_string call2 call_name caller_env cnd_muffle current_env .data
-#'   enexpr enexprs eval_bare eval_tidy expr expr_interp exec expr_label exprs
-#'   f_lhs f_rhs  inform is_call is_expression is_missing is_named list2
-#'   new_environment new_formula parse_expr parse_exprs set_names sym syms
-#'   type_of warn
+#'   as_name as_string call2 call_name caller_env cnd_muffle cnd_signal
+#'   current_env .data enexpr enexprs eval_bare eval_tidy expr expr_interp exec
+#'   expr_label exprs f_lhs f_rhs  inform is_call is_expression is_missing
+#'   is_named list2 new_environment new_formula parse_expr parse_exprs set_names
+#'   sym syms type_of warn
 #' @importFrom stats setNames
 #' @importFrom stringr str_c str_count str_detect str_extract str_glue
 #'             str_length str_locate str_locate_all str_match str_remove
@@ -30,6 +30,6 @@
 #'             str_trim
 #' @importFrom tidyr crossing drop_na fill nest pivot_longer pivot_wider unnest
 #' @importFrom tidyselect all_of any_of contains matches vars_select
-#' @importFrom utils capture.output file.edit str
+#' @importFrom utils capture.output file.edit object.size str
 #'
 "_PACKAGE"

R/compute_scale.R

@@ -74,8 +74,8 @@ compute_scale <- function(source,
                           flip_direction = FALSE,
                           min_n = 1) {
   # Function argument checks
-  assert_numeric_vector(source) # nolint: undesirable_function_linter
-  assert_numeric_vector(source_range, optional = TRUE)
+  assert_numeric_vector(source)
+  assert_numeric_vector(source_range, length = 2, optional = TRUE)
   if (!is.null(target_range) && is.null(source_range)) {
     cli_abort(
       c("Argument {.arg source_range} is missing with no default
@@ -84,7 +84,7 @@ compute_scale <- function(source,
       )
     )
   }
-  assert_numeric_vector(target_range, optional = TRUE)
+  assert_numeric_vector(target_range, length = 2, optional = TRUE)
   if (!is.null(source_range) && is.null(target_range)) {
     cli_abort(
       c("Argument {.arg target_range} is missing with no default
@@ -97,19 +97,16 @@ compute_scale <- function(source,
   assert_integer_scalar(min_n, subset = "positive")
 
   # Computation
-  if (sum(!is.na(source)) >= min_n) { # nolint: undesirable_function_linter
-    target <- mean(source, na.rm = TRUE) # nolint: undesirable_function_linter
+  if (sum(!is.na(source)) >= min_n) {
+    target <- mean(source, na.rm = TRUE)
 
     if (!is.null(source_range) && !is.null(target_range)) {
-      scale_constant <- min(target_range) - min(source_range)
-      scale_coefficient <- (max(target_range) - min(target_range)) /
-        (max(source_range) - min(source_range))
-
-      target <- (target + scale_constant) * scale_coefficient
-
-      if (flip_direction == TRUE) {
-        target <- max(target_range) - target
-      }
+      target <- transform_range(
+        target,
+        source_range = source_range,
+        target_range = target_range,
+        flip_direction = flip_direction
+      )
     }
   } else {
     target <- NA

R/derive_basetype_records.R

@@ -39,7 +39,6 @@
 #' @examples
 #' library(tibble)
 #' library(dplyr, warn.conflicts = FALSE)
-#' library(rlang)
 #'
 #' bds <- tribble(
 #'   ~USUBJID, ~EPOCH,         ~PARAMCD,  ~ASEQ, ~AVAL,

R/derive_joined.R

@@ -806,17 +806,92 @@ get_joined_data <- function(dataset,
 
   # join the input dataset with itself such that to each observation of the
   # input dataset all following observations are joined
+  data_add_to_join <- select(
+    data_add,
+    !!!by_vars,
+    !!!replace_values_by_names(extract_vars(order)),
+    !!!replace_values_by_names(join_vars),
+    !!tmp_obs_nr_var
+  )
+
+  # split input dataset into smaller pieces and process them separately
+  # This reduces the memory consumption.
+  if (is.null(by_vars_left)) {
+    # create batches of about 1MB input data
+    obs_per_batch <- floor(1000000 / as.numeric(object.size(data) / nrow(data)))
+    tmp_batch_nr <- get_new_tmp_var(data, prefix = "tmp_batch_nr")
+    data_list <- data %>%
+      mutate(!!tmp_batch_nr := ceiling(row_number() / obs_per_batch)) %>%
+      group_by(!!tmp_batch_nr) %>%
+      group_split(.keep = FALSE)
+    data_add_list <- list(data_add_to_join)
+  } else {
+    data_nest <- nest(data, data = everything(), .by = vars2chr(unname(by_vars_left)))
+    data_add_nest <- nest(data_add, data_add = everything(), .by = vars2chr(unname(by_vars_left)))
+    data_all_nest <- inner_join(data_nest, data_add_nest, by = vars2chr(by_vars_left))
+    data_list <- data_all_nest$data
+    data_add_list <- data_all_nest$data_add
+  }
+
+  joined_data <- map2(
+    data_list,
+    data_add_list,
+    function(x, y) {
+      get_joined_sub_data(
+        x,
+        y,
+        by_vars = by_vars_left,
+        tmp_obs_nr_var = tmp_obs_nr_var,
+        tmp_obs_nr_left = tmp_obs_nr_left,
+        join_type = join_type,
+        first_cond_upper = first_cond_upper,
+        first_cond_lower = first_cond_lower,
+        filter_join = filter_join
+      )
+    }
+  )
+
+  bind_rows(joined_data) %>%
+    remove_tmp_vars() %>%
+    select(-!!tmp_obs_nr_var_join)
+}
+
+#' Join Data for "joined" functions
+#'
+#' The helper function joins the data for the "joined" functions. All `.join`
+#' variables are included in the output dataset. It is called by
+#' `get_joined_data()` to process each by group separately. This reduces the
+#' memory consumption.
+#'
+#' @inheritParams get_joined_data
+#'
+#' @details
+#'
+#' 1. The input dataset (`dataset`) and the additional dataset (`dataset_add`)
+#' are left joined by the grouping variables (`by_vars`). If no grouping
+#' variables are specified, a full join is performed.
+#'
+#' 1. The joined dataset is restricted as specified by arguments `join_type`,
+#' `first_cond_upper`, and `first_cond_lower`. See argument descriptions for
+#' details.
+#'
+#' 1. The joined dataset is restricted by the `filter_join` condition.
+#'
+#' @keywords internal
+get_joined_sub_data <- function(dataset,
+                                dataset_add,
+                                by_vars,
+                                tmp_obs_nr_var,
+                                tmp_obs_nr_left,
+                                join_type,
+                                first_cond_upper,
+                                first_cond_lower,
+                                filter_join) {
   data_joined <-
     left_join(
-      data,
-      select(
-        data_add,
-        !!!by_vars,
-        !!!replace_values_by_names(extract_vars(order)),
-        !!!replace_values_by_names(join_vars),
-        !!tmp_obs_nr_var
-      ),
-      by = vars2chr(by_vars_left),
+      dataset,
+      dataset_add,
+      by = vars2chr(by_vars),
       suffix = c("", ".join")
     )
 
@@ -837,7 +912,7 @@ get_joined_data <- function(dataset,
     # select all observations up to the first confirmation observation
     data_joined <- filter_relative(
       data_joined,
-      by_vars = expr_c(by_vars_left, tmp_obs_nr_var),
+      by_vars = expr_c(by_vars, tmp_obs_nr_var),
       condition = !!first_cond_upper,
       order = exprs(!!parse_expr(paste0(as_name(tmp_obs_nr_var), ".join"))),
       mode = "first",
@@ -851,7 +926,7 @@ get_joined_data <- function(dataset,
     # select all observations up to the first confirmation observation
     data_joined <- filter_relative(
       data_joined,
-      by_vars = expr_c(by_vars_left, tmp_obs_nr_var),
+      by_vars = expr_c(by_vars, tmp_obs_nr_var),
       condition = !!first_cond_lower,
       order = exprs(!!parse_expr(paste0("desc(", as_name(tmp_obs_nr_var), ".join)"))),
       mode = "first",
@@ -862,9 +937,7 @@ get_joined_data <- function(dataset,
   }
   # apply confirmation condition, which may include summary functions
   data_joined %>%
-    group_by(!!!by_vars_left, !!tmp_obs_nr_left) %>%
+    group_by(!!!by_vars, !!tmp_obs_nr_left) %>%
     filter_if(filter_join) %>%
-    ungroup() %>%
-    remove_tmp_vars() %>%
-    select(-!!tmp_obs_nr_var_join)
+    ungroup()
 }

R/derive_param_computed.R

@@ -2,9 +2,12 @@
 #'
 #' Adds a parameter computed from the analysis value of other parameters. It is
 #' expected that the analysis value of the new parameter is defined by an
-#' expression using the analysis values of other parameters. For example mean
-#' arterial pressure (MAP) can be derived from systolic (SYSBP) and diastolic
-#' blood pressure (DIABP) with the formula
+#' expression using the analysis values of other parameters, such as addition/sum,
+#' subtraction/difference, multiplication/product, division/ratio,
+#' exponentiation/logarithm, or by formula.
+#' <br/><br/>
+#' For example mean arterial pressure (MAP) can be derived from systolic (SYSBP)
+#' and diastolic blood pressure (DIABP) with the formula
 #' \deqn{MAP = \frac{SYSBP + 2 DIABP}{3}}{MAP = (SYSBP + 2 DIABP) / 3}
 #'
 #' @param dataset

R/derive_var_base.R

@@ -43,19 +43,19 @@
 #' library(tibble)
 #'
 #' dataset <- tribble(
-#'   ~STUDYID, ~USUBJID,   ~PARAMCD, ~AVAL,   ~AVALC,      ~AVISIT, ~ABLFL,  ~ANRIND,
-#'   "TEST01",  "PAT01",  "PARAM01", 10.12,       NA,   "Baseline",    "Y", "NORMAL",
-#'   "TEST01",  "PAT01",  "PARAM01", 9.700,       NA,      "Day 7",    "N",    "LOW",
-#'   "TEST01",  "PAT01",  "PARAM01", 15.01,       NA,     "Day 14",    "N",   "HIGH",
-#'   "TEST01",  "PAT01",  "PARAM02", 8.350,       NA,   "Baseline",    "Y",    "LOW",
-#'   "TEST01",  "PAT01",  "PARAM02",    NA,       NA,      "Day 7",    "N",       NA,
-#'   "TEST01",  "PAT01",  "PARAM02", 8.350,       NA,     "Day 14",    "N",    "LOW",
-#'   "TEST01",  "PAT01",  "PARAM03",    NA,    "LOW",   "Baseline",    "Y",       NA,
-#'   "TEST01",  "PAT01",  "PARAM03",    NA,    "LOW",      "Day 7",    "N",       NA,
-#'   "TEST01",  "PAT01",  "PARAM03",    NA, "MEDIUM",     "Day 14",    "N",       NA,
-#'   "TEST01",  "PAT01",  "PARAM04",    NA,   "HIGH",   "Baseline",    "Y",       NA,
-#'   "TEST01",  "PAT01",  "PARAM04",    NA,   "HIGH",      "Day 7",    "N",       NA,
-#'   "TEST01",  "PAT01",  "PARAM04",    NA, "MEDIUM",     "Day 14",    "N",       NA
+#'   ~STUDYID, ~USUBJID, ~PARAMCD, ~AVAL, ~AVALC,   ~AVISIT,    ~ABLFL, ~ANRIND,
+#'   "TEST01", "PAT01", "PARAM01", 10.12, NA,       "Baseline", "Y",    "NORMAL",
+#'   "TEST01", "PAT01", "PARAM01", 9.700, NA,       "Day 7",    NA,     "LOW",
+#'   "TEST01", "PAT01", "PARAM01", 15.01, NA,       "Day 14",   NA,     "HIGH",
+#'   "TEST01", "PAT01", "PARAM02", 8.350, NA,       "Baseline", "Y",    "LOW",
+#'   "TEST01", "PAT01", "PARAM02",    NA, NA,       "Day 7",    NA,     NA,
+#'   "TEST01", "PAT01", "PARAM02", 8.350, NA,       "Day 14",   NA,     "LOW",
+#'   "TEST01", "PAT01", "PARAM03",    NA, "LOW",    "Baseline", "Y",    NA,
+#'   "TEST01", "PAT01", "PARAM03",    NA, "LOW",    "Day 7",    NA,     NA,
+#'   "TEST01", "PAT01", "PARAM03",    NA, "MEDIUM", "Day 14",   NA,     NA,
+#'   "TEST01", "PAT01", "PARAM04",    NA, "HIGH",   "Baseline", "Y",    NA,
+#'   "TEST01", "PAT01", "PARAM04",    NA, "HIGH",   "Day 7",    NA,     NA,
+#'   "TEST01", "PAT01", "PARAM04",    NA, "MEDIUM", "Day 14",   NA,     NA
 #' )
 #'
 #' ## Derive `BASE` variable from `AVAL`

R/derive_var_chg.R

@@ -25,10 +25,10 @@
 #' advs <- tribble(
 #'   ~USUBJID, ~PARAMCD, ~AVAL, ~ABLFL, ~BASE,
 #'   "P01",    "WEIGHT", 80,    "Y",    80,
-#'   "P01",    "WEIGHT", 80.8,  "",     80,
-#'   "P01",    "WEIGHT", 81.4,  "",     80,
+#'   "P01",    "WEIGHT", 80.8,  NA,     80,
+#'   "P01",    "WEIGHT", 81.4,  NA,     80,
 #'   "P02",    "WEIGHT", 75.3,  "Y",    75.3,
-#'   "P02",    "WEIGHT", 76,    "",     75.3
+#'   "P02",    "WEIGHT", 76,    NA,     75.3
 #' )
 #' derive_var_chg(advs)
 derive_var_chg <- function(dataset) {