fix: add input check to merge_start_stop() to ensure unique rows in all supplied datasets

Author: RobinDenz1

NAMESPACE

@@ -43,4 +43,5 @@ importFrom(data.table,setkey)
 importFrom(data.table,setkeyv)
 importFrom(data.table,setnames)
 importFrom(data.table,shift)
+importFrom(data.table,uniqueN)
 importFrom(fastmatch,"%fin%")

R/merge_start_stop.r

@@ -11,6 +11,7 @@
 #' @importFrom data.table :=
 #' @importFrom data.table rbindlist
 #' @importFrom data.table shift
+#' @importFrom data.table uniqueN
 #' @export
 merge_start_stop <- function(x, y, ..., dlist, by, start="start",
                              stop="stop", all=FALSE, all.x=all, all.y=all,
@@ -58,7 +59,7 @@ merge_start_stop <- function(x, y, ..., dlist, by, start="start",
     if (!all(unlist(type_value_i) %in% c("logical", "numeric", "character",
                                          "integer"))) {
       stop("All columns containing variables (columns except 'by', 'start' ",
-           "and 'stop') must be of type:\n 'logical', 'numeric' or ",
+           "and 'stop') must be of type:\n 'logical', 'numeric', 'integer' or ",
            "'character'.", call.=FALSE)
     }
 
@@ -74,6 +75,12 @@ merge_start_stop <- function(x, y, ..., dlist, by, start="start",
                         variable.factor=FALSE)
       )
     }
+
+    # check if all values in a dataset are unique
+    if (uniqueN(dlist[[i]]) != nrow(dlist[[i]])) {
+      stop("Duplicate rows found in dataset ", i, ". Remove duplicates",
+           " and re-run the function.", call.=FALSE)
+    }
   }
 
   # put together all datasets in one

tests/testthat/test_merge_start_stop.r

@@ -601,3 +601,21 @@ test_that("general test cases, 2 datasets, integers", {
   output <- merge_start_stop(dlist=dlist, by="ID", center_on_first=TRUE)
   expect_true(all(output[, .(start = min(start)), by=ID]$start==0))
 })
+
+test_that("error with duplicate rows", {
+
+  d1 <- data.table(ID=c(1, 1, 1, 1, 2, 2, 3, 5),
+                   start=c(20, 210, 370, 370, 55, 98, 1, 9),
+                   stop=c(189, 301, 375, 375, 90, 190, 900, 10),
+                   d1=c(TRUE, TRUE, FALSE, FALSE, TRUE, FALSE, FALSE, TRUE))
+
+  d2 <- data.table(ID=c(1, 1, 1, 2, 2, 3, 5),
+                   start=c(17, 211, 370, 58, 98, 1, 9),
+                   stop=c(189, 321, 375, 90, 191, 94, 11),
+                   d2=c(TRUE, TRUE, FALSE, TRUE, FALSE, FALSE, TRUE))
+  dlist <- list(d1, d2)
+
+  expect_error({merge_start_stop(d1, d2, by="ID")},
+               paste0("Duplicate rows found in dataset 1. Remove ",
+                      "duplicates and re-run the function."))
+})