update (ddsjoberg#2039)

NEWS.md

@@ -2,6 +2,8 @@
 
 * Bug fix in `add_ci.tbl_svysummary()` for factor variables where order was alphabetical instead of the factor levels. (#2036)
 
+* Addressing encoding issue where `sort()` and `dplyr::arrange()` sorted differently, and the order of the `by` levels was inconsistent in the resulting table. (#2038)
+
 # gtsummary 2.0.3
 
 ### New Features and Functions

R/tbl_summary.R

@@ -452,8 +452,8 @@ tbl_summary <- function(data,
           !cards$context %in% "attributes",
         ) |>
           dplyr::select(cards::all_ard_groups(), "variable", "context") |>
-          dplyr::distinct() |>
-          dplyr::arrange(unlist(!!sym("group1_level"))) |>
+          dplyr::distinct() %>%
+          {.[order(unlist(.$group1_level)), ]} |> # styler: off
           dplyr::mutate(
             .by = cards::all_ard_groups(),
             gts_column = paste0("stat_", dplyr::cur_group_id())

tests/testthat/test-tbl_summary.R

@@ -666,3 +666,18 @@ test_that("tbl_summary() data frame column labels are not dropped", {
     "Age"
   )
 })
+
+# addressing issue #2038
+test_that("tbl_summary() test encoding/sorting difference between sort() and dplyr::arrange()", {
+  # before this fix, the columns would print in reverse order, e.g. stat_3, stat_2, stat_1
+  expect_equal(
+    data.frame(
+      groupe_etude = c(rep("Tem", 13), rep("TTA-", 7), rep("TTA+", 18)),
+      tympan_g_inc = rep(1,38)
+    ) |>
+      tbl_summary(by = groupe_etude, include = tympan_g_inc) |>
+      as.data.frame(col_labels = FALSE) |>
+      names(),
+    c("label", "stat_1", "stat_2", "stat_3")
+  )
+})