Use sensible column name for bundles (#20)

* Use bundle_name correctly

* Add styler to project

DESCRIPTION

@@ -19,6 +19,7 @@ Suggests:
   pkgload,
   lintr,
   stringr,
+  styler,
   testthat (>= 3.0.0),
   usethis
 Config/testthat/edition: 3

R/bundles.R

@@ -9,7 +9,7 @@
 #' @description If a bundle has multiple names, then the id will be duplicated across rows
 #'
 #' @param version Requested version, if not defined, the latest will be used
-#' @return dataframe that contains a "concept_name", "version" and a "domain" column for each available concept
+#' @return dataframe that contains a "bundle_name", "version" and a "domain" column for each available bundle
 #' @export
 #' @examples
 #' available_bundles()
@@ -36,7 +36,7 @@ available_bundles <- function(version = "latest") {
 #' @export
 #' @examples
 #' # Usage with available_bundles, from a single row
-#' smoking <- available_bundles() |> dplyr::filter(concept_name == "Smoking")
+#' smoking <- available_bundles() |> dplyr::filter(bundle_name == "Smoking")
 #' concept_by_bundle(domain = smoking$domain, id = smoking$id, version = smoking$version)
 #' # Using if you know the details directly
 #' concept_by_bundle(domain = "observation", id = "smoking")

R/raw-data.R

@@ -12,7 +12,7 @@
 #'
 #' @description If a bundle has multiple names, then the id will be duplicated across rows
 #'
-#' @return dataframe that contains a "concept_name" and a "domain" column for each available concept
+#' @return dataframe that contains a "bundle_name" and a "domain" column for each available bundle
 #' @keywords internal
 raw_bundles <- function() {
   raw_dir <- get_raw_dir()
@@ -38,12 +38,12 @@ get_raw_dir <- function(..., version = "latest") {
 
 
 parse_bundle_names <- function(bundle_name_path) {
-  bundle_name <- bundle_name_path |>
+  domain_name <- bundle_name_path |>
     dirname() |>
     basename()
 
   readr::read_csv(bundle_name_path, col_types = "cc") |>
-    mutate(domain = bundle_name)
+    mutate(domain = domain_name)
 }
 
 

inst/data-raw/measurement/bundle_names.csv

@@ -1,4 +1,4 @@
-id,concept_name
+id,bundle_name
 HBA1c,HBA1c
 LDH,LDH
 antibodies_to_hiv,Antibodies to HIV

inst/data-raw/observation/bundle_names.csv

@@ -1,2 +1,2 @@
-id,concept_name
+id,bundle_name
 smoking,Smoking

inst/data-raw/race/bundle_names.csv

@@ -1,2 +1,2 @@
-id,concept_name
+id,bundle_name
 snomed_race,Race

inst/data-raw/unknown/bundle_names.csv

@@ -1,2 +1,2 @@
-id,concept_name
+id,bundle_name
 indices_of_deprivation,Indices of deprivation

renv.lock

@@ -1233,6 +1233,11 @@
       ],
       "Hash": "960e2ae9e09656611e0b8214ad543207"
     },
+    "styler": {
+      "Package": "styler",
+      "Version": "1.10.3",
+      "Source": "Repository"
+    },
     "sys": {
       "Package": "sys",
       "Version": "3.4.2",

tests/testthat/test-bundles.R

@@ -1,16 +1,15 @@
 test_that("available_bundles isn't empty and have correct columns", {
   result <- omopbundles::available_bundles()
   expect_true(nrow(result) > 0, info = "The dataframe should not be empty")
-  hiv_ab <- dplyr::filter(result, concept_name == "Antibodies to HIV")
+  hiv_ab <- dplyr::filter(result, bundle_name == "Antibodies to HIV")
   expect_equal(hiv_ab$version, "latest")
   expect_equal(hiv_ab$id, "antibodies_to_hiv")
   expect_equal(hiv_ab$domain, "measurement")
-
 })
 
 test_that("Smoking exists as an observation", {
   result <- available_bundles() |>
-    dplyr::filter(concept_name == "Smoking")
+    dplyr::filter(bundle_name == "Smoking")
 
   expect_true(nrow(result) == 1, info = "Smoking should only exist as a single row")
   expect_equal(result$domain, "observation")
@@ -25,7 +24,7 @@ test_that("Concept by bundle works with character values", {
 
 test_that("Available bundles and concept_by_bundle play nicely together", {
   smoking_bundle <- available_bundles() |>
-    dplyr::filter(concept_name == "Smoking")
+    dplyr::filter(bundle_name == "Smoking")
 
   smoking_concepts <- concept_by_bundle(smoking_bundle$domain, smoking_bundle$id)
 

tests/testthat/test-raw-data.R

@@ -12,8 +12,10 @@ assert_bundle_has_name <- function(bundle) {
   bundle_name_file <- get_raw_dir(bundle$domain, "bundle_names.csv")
   bundle_names <- read_csv(bundle_name_file, show_col_types = FALSE)
 
-  expect_true(bundle$id %in% bundle_names$id,
-              glue::glue("{bundle$id} should at least one name in: {bundle$domain}/bundle_names.csv"))
+  expect_true(
+    bundle$id %in% bundle_names$id,
+    glue::glue("{bundle$id} should at least one name in: {bundle$domain}/bundle_names.csv")
+  )
 }
 
 test_that("All raw bundles have at least one name", {
@@ -38,5 +40,4 @@ test_that("All raw bundle names map to a bundle file that has at least one conce
     # Check that at least one concept
     expect_true(nrow(concepts) > 0)
   })
-
 })