Merge pull request #588 from yjunechoe/row_distinct-extracts

Bugfix `rows_distinct()` dropping non-tested columns

NEWS.md

@@ -1,8 +1,11 @@
 # pointblank (development version)
 
-* The `brief` argument of validation functions now also supports `{glue}` syntax (#587)
 
-* Validation step `brief`s correctly recycle to match expanded steps (#564)
+- Data extracts for `rows_distinct()` preserves columns other than the ones tested (#588)
+
+- The `brief` argument of validation functions now also supports `{glue}` syntax (#587)
+
+- Validation step `brief`s correctly recycle to match expanded steps (#564)
 
 # pointblank 0.12.2
 

R/interrogate.R

@@ -2392,9 +2392,8 @@ interrogate_distinct <- function(
     column_validity_has_columns(columns = column_names)
 
     table %>%
-      dplyr::select({{ column_names }}) %>%
       dplyr::group_by(!!!col_syms) %>%
-      dplyr::mutate(`pb_is_good_` = ifelse(dplyr::n() == 1, TRUE, FALSE)) %>%
+      dplyr::mutate(`pb_is_good_` = dplyr::n() == 1L) %>%
       dplyr::ungroup()
   }
 
@@ -2413,15 +2412,13 @@ interrogate_distinct <- function(
 
     unduplicated <-
       table %>%
-      dplyr::select({{ column_names }}) %>%
       dplyr::count(!!!col_syms, name = "pb_is_good_") %>%
-      dplyr::mutate(`pb_is_good_` = ifelse(`pb_is_good_` == 1, TRUE, FALSE)) %>%
-      dplyr::filter(`pb_is_good_` == TRUE)
+      dplyr::mutate(`pb_is_good_` = `pb_is_good_` == 1L) %>%
+      dplyr::filter(`pb_is_good_`)
 
     table %>%
-      dplyr::select({{ column_names }}) %>%
       dplyr::left_join(unduplicated, by = column_names) %>%
-      dplyr::mutate(`pb_is_good_` = ifelse(is.na(`pb_is_good_`), FALSE, TRUE))
+      dplyr::mutate(`pb_is_good_` = is.na(`pb_is_good_`))
   }
 
   # Perform the validation of the table

tests/testthat/test-get_data_extracts.R

@@ -268,3 +268,16 @@ test_that("Data extracts of different sizes are possible to create", {
   get_data_extracts(agent) %>% names() %>% expect_null()
   get_data_extracts(agent) %>% length() %>% expect_equal(0)
 })
+
+test_that("`rows_distinct()` extracts all columns", {
+
+  agent <- create_agent(small_table) %>%
+    rows_distinct(c(date, date_time)) %>%
+    interrogate()
+
+  expect_identical(
+    colnames(get_data_extracts(agent)[[1]]),
+    colnames(small_table)
+  )
+
+})