Merge pull request #87 from pharmaverse/85_resolve_bug_duplicate_reco…

…rds_in_SV_for_USUBIJD_and_VISIT

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

DESCRIPTION

@@ -21,7 +21,8 @@ URL: https://pharmaverse.github.io/pharmaversesdtm/main/, https://github.com/pha
 Depends: R (>= 3.5.0)
 Encoding: UTF-8
 LazyData: true
-RoxygenNote: 7.2.3
+Roxygen: list(markdown = TRUE)
+RoxygenNote: 7.3.1
 Suggests:
     devtools,
     lintr,

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"

data-raw/sv.R

@@ -4,9 +4,29 @@
 # Load libraries -----
 library(haven)
 library(admiral)
+library(dplyr)
+library(metatools)
 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) %>%
+  add_labels(
+    VISIT = "Visit Name",
+    VISITNUM = "Visit Number"
+  )
+
+
 # Label dataset ----
 attr(sv, "label") <- "Subject Visits"