From d9d51b978b6ccbe92380d8bf2614e8faa1c592dc Mon Sep 17 00:00:00 2001
From: Trevor Riley <89118428+TNRiley@users.noreply.github.com>
Date: Thu, 23 May 2024 23:12:23 -0400
Subject: [PATCH] Update compare.R

Replaced base::unique() with dplyr::distinct() for improved performance in count_unique function.
---
 R/compare.R | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/R/compare.R b/R/compare.R
index 87b26869..c1a6dd20 100644
--- a/R/compare.R
+++ b/R/compare.R
@@ -27,7 +27,14 @@ count_unique <- function(unique_data, include_references = FALSE) {
       type = ifelse(.data$unique, "unique", "duplicated") %>% factor(levels = c("unique", "duplicated"))
     ) %>%
     dplyr::ungroup() %>%
-    unique()
+    dplyr::distinct()
+
+  if (include_references == TRUE) {
+    out %>% dplyr::left_join(unique_data %>% dplyr::select(-dplyr::all_of(setdiff(intersect(names(.), names(out)), "duplicate_id"))), by = "duplicate_id")
+  } else {
+    out
+  }
+}
 
   if (include_references == TRUE) {
     out %>% dplyr::left_join(unique_data %>% dplyr::select(-dplyr::all_of(setdiff(intersect(names(.), names(out)), "duplicate_id"))), by = "duplicate_id")