#82 Resolve Bug: Duplicate observation for group USUBJID and VISIT

pharmaverse · Feb 12, 2024 · 6b66b15 · 6b66b15
1 parent f901b2a
commit 6b66b15
Show file tree

Hide file tree

Showing 4 changed files with 16 additions and 2 deletions.
diff --git a/R/data.R b/R/data.R
@@ -218,7 +218,8 @@
 
 #' Subject Visits Dataset
 #'
-#' A SDTM SV dataset from the CDISC pilot project
+#' A SDTM SV dataset from the CDISC pilot project. Duplicate observation for group
+#' variable USUBJID and VISIT is corrected.
 #'
 #' @source \url{https://github.com/cdisc-org/sdtm-adam-pilot-project/blob/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/tabulations/sdtm/sv.xpt?raw=true} # nolint
 "sv"

diff --git a/data-raw/sv.R b/data-raw/sv.R
@@ -4,9 +4,21 @@
 # Load libraries -----
 library(haven)
 library(admiral)
+library(dplyr)
 raw_sv <- read_xpt("https://github.com/cdisc-org/sdtm-adam-pilot-project/blob/master/updated-pilot-submission-package/900172/m5/datasets/cdiscpilot01/tabulations/sdtm/sv.xpt?raw=true") # nolint
 sv <- convert_blanks_to_na(raw_sv)
 
+sv <- sv %>%
+  group_by(USUBJID, VISITNUM) %>%
+  mutate(
+    ASEQ = row_number(),
+    n=n()) %>%
+  mutate(
+    VISIT = ifelse(n > 1, paste0("UNSCHEDULED ", floor(VISITNUM), ".", ASEQ), VISIT),
+    VISITNUM = ifelse(n > 1, as.numeric(paste0(floor(VISITNUM), ".", ASEQ)), VISITNUM)) %>%
+  ungroup() %>%
+  select(-ASEQ,-n)
+
 # Label dataset ----
 attr(sv, "label") <- "Subject Visits"
 

diff --git a/data/sv.rda b/data/sv.rda
diff --git a/man/sv.Rd b/man/sv.Rd