Add percentage and default sorting to miss_scan_count (Fix #341)

R/miss-scan-count.R

@@ -39,27 +39,20 @@
 #'
 #'
 miss_scan_count <- function(data,search){
-  # if there is only one value to search
-  if (length(search) == 1) {
-    res <- purrr::map_dfc(data,
-                         ~length(grep(search,
-                                      x = .))) %>%
-      # return the dataframe with the columns "
-      tidyr::pivot_longer(cols = dplyr::everything(),
-                          names_to = "Variable",
-                          values_to = "n")
-    # but if there are more than one, we need to combine the search terms
-  }
 
-  if (length(search) > 1) {
-    res <- purrr::map_dfc(data,
-                         ~length(grep(paste0(search,
-                                             collapse ="|"),
-                                      x = .))) %>%
-      tidyr::pivot_longer(cols = dplyr::everything(),
-                          names_to = "Variable",
-                          values_to = "n")
-  }
+  pattern <- paste0(search, collapse = "|")
+
+  res <- data %>%
+    # if the value is in the search terms, return TRUE
+    dplyr::mutate(dplyr::across(dplyr::everything(), ~grepl(pattern, .x, perl = TRUE))) %>%
+    # sum the number of times the value is found
+    dplyr::summarise(dplyr::across(dplyr::everything(), sum)) %>%
+    # present the data in a long format
+    tidyr::pivot_longer(cols = dplyr::everything(), names_to = "Variable", values_to = "n") %>%
+    # calculate the percentage
+    dplyr::mutate(pct = (n/nrow(data))*100) %>%
+    # order the data in descending order
+    dplyr::arrange(-n)
 
   return(res)
 }

tests/testthat/test-miss-scan-count.R

@@ -25,37 +25,30 @@ test_that("miss_scan_count returns an error when no data is provided", {
 test_that("miss_scan_count returns a data.frame of the right size", {
   expect_equal(
     dim(miss_scan_count(dat_ms, -99)),
-    c(3, 2)
+    c(3, 3)
   )
 })
 
+test_that("miss_scan_count returns results sorted in descending order by 'n'", {
+  result <- miss_scan_count(dat_ms, c(-99, -98))
+  n_values <- result$n
+  expect_true(all(n_values == sort(n_values, decreasing = TRUE)))
+})
+
 correct_answer_1 <- tibble::tribble(
-  ~Variable, ~n,
-  "x", 1L,
-  "y", 0L,
-  "z", 1L
+  ~Variable, ~n, ~pct,
+  "x", 1L, 20,
+  "z", 1L, 20,
+  "y", 0L, 0
 )
 
 correct_answer_2 <- tibble::tribble(
-  ~Variable, ~n,
-  "x", 2L,
-  "y", 0L,
-  "z", 2L
+  ~Variable, ~n, ~pct,
+  "x", 2L, 40,
+  "z", 2L, 40,
+  "y", 0L, 0
 )
 
-correct_answer_3 <- tibble::tribble(
-  ~Variable, ~n,
-  "x", 2L,
-  "y", 1L,
-  "z", 2L
-)
-
-correct_answer_4 <- tibble::tribble(
-  ~Variable, ~n,
-  "x", 2L,
-  "y", 1L,
-  "z", 2L
-)
 
 test_that("miss_scan_count returns the right answer", {
   expect_equal(miss_scan_count(dat_ms,-99),