Make columns have true NAs instead of string "NA" (#86)

- Closes #31
traitecoevo · Oct 11, 2023 · 2ac30e0 · 2ac30e0
1 parent e4c01ff
commit 2ac30e0
Show file tree

Hide file tree

Showing 2 changed files with 20 additions and 90 deletions.
diff --git a/R/process.R b/R/process.R
@@ -1601,15 +1601,27 @@ process_format_methods <- function(metadata, dataset_id, sources, contributors)
           dataset_id = dataset_id,
           source_primary_key = source_primary_key,
           source_primary_citation = bib_print(sources[[source_primary_key]]),
-          source_secondary_key = source_secondary_keys %>% paste(collapse = "; "),
-          source_secondary_citation = ifelse(length(source_secondary_keys) == 0, NA_character_,
-            purrr::map_chr(source_secondary_keys, ~sources[[.x]] %>% bib_print) %>% paste(collapse = "; ") %>%
+          source_secondary_key = ifelse(
+            length(source_secondary_keys) > 0,
+            source_secondary_keys %>% paste(collapse = "; "),
+            NA_character_
+          ),
+          source_secondary_citation = ifelse(
+            length(source_secondary_keys) == 0, NA_character_,
+            purrr::map_chr(source_secondary_keys, ~sources[[.x]] %>% bib_print) %>%
+              paste(collapse = "; ") %>%
+              stringr::str_replace_all("\\.;", ";")
+          ),
+          source_original_dataset_key = ifelse(
+            length(source_original_dataset_keys) > 0,
+            source_original_dataset_keys %>% paste(collapse = "; "),
+            NA_character_
+          ),
+          source_original_dataset_citation = ifelse(
+            length(source_original_dataset_keys) == 0, NA_character_,
+            purrr::map_chr(source_original_dataset_keys, ~sources[[.x]] %>% bib_print) %>%
+              paste(collapse = "; ") %>%
               stringr::str_replace_all("\\.;", ";")
-            ),
-          source_original_dataset_key = source_original_dataset_keys %>% paste(collapse = "; "),
-          source_original_dataset_citation = ifelse(length(source_original_dataset_keys) == 0, NA_character_,
-            purrr::map_chr(source_original_dataset_keys, ~sources[[.x]] %>% bib_print) %>% paste(collapse = "; ") %>%
-            stringr::str_replace_all("\\.;", ";")
           )
         )
       ) %>%

diff --git a/tests/testthat/test-xamples.R b/tests/testthat/test-xamples.R
@@ -35,23 +35,6 @@ testthat::test_that("Test Dataset 1 builds correctly", {
   # Todo: also load and test non-csv outputs
   names(expected_output) <- tables
 
-  # Temporary modifications to get these tests to pass
-  Test_2023_1$traits <-
-    Test_2023_1$traits %>%
-    mutate(across(
-      c(temporal_context_id, entity_context_id, plot_context_id, treatment_context_id, method_context_id),
-      ~if_else(.x == "NA", NA_character_, .x)
-    ))
-  Test_2023_1$methods <-
-    Test_2023_1$methods %>%
-    mutate(across(c(source_secondary_key, source_original_dataset_key), ~NA_character_))
-  Test_2023_1$excluded_data <-
-    Test_2023_1$excluded_data %>%
-    mutate(across(
-      c(temporal_context_id, entity_context_id, plot_context_id, treatment_context_id, method_context_id),
-      ~if_else(.x == "NA", NA_character_, .x)
-    ))
-
   # Check all tables are equal to expected output tables
   for (v in tables) {
     expect_equal(Test_2023_1[[v]], expected_output[[v]])
@@ -86,19 +69,6 @@ testthat::test_that("Test Dataset 2 builds correctly", {
   # Todo: also load and test non-csv outputs
   names(expected_output) <- tables
 
-  # Temporary modifications to get these tests to pass
-  columns <- c("basis_of_value", "replicates", "life_stage", "collection_date", "measurement_remarks")
-
-  Test_2023_2$traits <-
-    Test_2023_2$traits %>%
-    mutate(across(dplyr::all_of(columns), as.character))
-  Test_2023_2$methods <-
-    Test_2023_2$methods %>%
-    mutate(across(c(source_secondary_key, source_original_dataset_key), ~NA_character_))
-  Test_2023_2$excluded_data <-
-    Test_2023_2$excluded_data %>%
-    mutate(across(dplyr::all_of(columns), as.character))
-
   # Check all tables are equal to expected output tables
   for (v in tables) {
     expect_equal(Test_2023_2[[v]], expected_output[[v]])
@@ -133,19 +103,6 @@ testthat::test_that("Test Dataset 3 builds correctly", {
   # Todo: also load and test non-csv outputs
   names(expected_output) <- tables
 
-  # Temporary modifications to get these tests to pass
-  columns <- c("basis_of_value", "replicates", "life_stage", "collection_date", "measurement_remarks")
-
-  Test_2023_3$traits <-
-    Test_2023_3$traits %>%
-    mutate(across(dplyr::all_of(columns), as.character))
-  Test_2023_3$methods <-
-    Test_2023_3$methods %>%
-    mutate(across(c(source_secondary_key, source_original_dataset_key), ~NA_character_))
-  Test_2023_3$excluded_data <-
-    Test_2023_3$excluded_data %>%
-    mutate(across(dplyr::all_of(columns), as.character))
-
   # Check all tables are equal to expected output tables
   for (v in tables) {
     expect_equal(Test_2023_3[[v]], expected_output[[v]])
@@ -180,19 +137,6 @@ testthat::test_that("Test Dataset 4 builds correctly", {
   # Todo: also load and test non-csv outputs
   names(expected_output) <- tables
 
-  # Temporary modifications to get these tests to pass
-  columns <- c("basis_of_value", "replicates", "life_stage", "collection_date", "measurement_remarks")
-
-  Test_2023_4$traits <-
-    Test_2023_4$traits %>%
-    mutate(across(dplyr::all_of(columns), as.character))
-  Test_2023_4$methods <-
-    Test_2023_4$methods %>%
-    mutate(across(c(source_secondary_key, source_original_dataset_key), ~NA_character_))
-  Test_2023_4$excluded_data <-
-    Test_2023_4$excluded_data %>%
-    mutate(across(dplyr::all_of(columns), as.character))
-
   # Check all tables are equal to expected output tables
   for (v in tables) {
     expect_equal(Test_2023_4[[v]], expected_output[[v]])
@@ -255,19 +199,6 @@ testthat::test_that("Test Dataset 7 builds correctly", {
   # Todo: also load and test non-csv outputs
   names(expected_output) <- tables
 
-  # Temporary modifications to get these tests to pass
-  columns <- c("basis_of_value", "replicates", "life_stage", "collection_date", "measurement_remarks")
-
-  Test_2023_7$traits <-
-    Test_2023_7$traits %>%
-    mutate(across(dplyr::all_of(columns), as.character))
-  Test_2023_7$methods <-
-    Test_2023_7$methods %>%
-    mutate(across(c(source_secondary_key, source_original_dataset_key), ~NA_character_))
-  Test_2023_7$excluded_data <-
-    Test_2023_7$excluded_data %>%
-    mutate(across(dplyr::all_of(columns), as.character))
-
   # Check all tables are equal to expected output tables
   for (v in tables) {
     expect_equal(Test_2023_7[[v]], expected_output[[v]])
@@ -302,19 +233,6 @@ testthat::test_that("Test Dataset 8 builds correctly", {
   # Todo: also load and test non-csv outputs
   names(expected_output) <- tables
 
-  # Temporary modifications to get these tests to pass
-  columns <- c("basis_of_value", "replicates", "life_stage", "collection_date", "measurement_remarks")
-
-  Test_2023_8$traits <-
-    Test_2023_8$traits %>%
-    mutate(across(dplyr::all_of(columns), as.character))
-  Test_2023_8$methods <-
-    Test_2023_8$methods %>%
-    mutate(across(c(source_secondary_key, source_original_dataset_key), ~NA_character_))
-  Test_2023_8$excluded_data <-
-    Test_2023_8$excluded_data %>%
-    mutate(across(dplyr::all_of(columns), as.character))
-
   # Check all tables are equal to expected output tables
   for (v in tables) {
     expect_equal(Test_2023_8[[v]], expected_output[[v]])