Release Version: 3.22.1

- getPhenotypeLog() now returns the original user submitted cohort name, instead of the name forced by atlas-phenotype. 
- This is in line with the original requirement. It allows for long names, names with special characters.

- Deprecated listPhenotypes(). This was an older function that did the same as getPhenotypeLog() but returned only two fields.

- Added requirement on R (4.10) because tidyverse seems to cause problems with older versions of R.

DESCRIPTION

@@ -1,11 +1,13 @@
 Package: PhenotypeLibrary
 Type: Package
 Title: The OHDSI Phenotype Library
-Version: 3.22.0
-Date: 2023-09-23
+Version: 3.22.1
+Date: 2023-09-24
 Author: Gowtham Rao [aut, cre]
 Maintainer: Gowtham Rao <[email protected]>
 Description: A repository to store the content of the OHDSI Phenotype library.
+Depends:
+  R (>= 4.1.0)
 Imports:
   checkmate,
 	readr,

NAMESPACE

@@ -4,5 +4,4 @@ export(getPhenotypeLog)
 export(getPlCohortDefinitionSet)
 export(getPlConceptDefinitionSet)
 export(listPhenotypes)
-import(dplyr)
 importFrom(rlang,.data)

NEWS.md

@@ -1,3 +1,13 @@
+PhenotypeLibrary 3.22.1
+======================
+
+- getPhenotypeLog() now returns the original user submitted cohort name, instead of the name forced by atlas-phenotype. 
+- This is in line with the original requirement. It allows for long names, names with special characters.
+
+- Deprecated listPhenotypes(). This was an older function that did the same as getPhenotypeLog() but returned only two fields.
+
+- Added requirement on R (4.10) because tidyverse seems to cause problems with older versions of R.
+
 PhenotypeLibrary 3.22.0
 ======================
 Accepted Cohorts: No cohorts were accepted in this release.

R/PhenotypeLibrary.R

@@ -17,6 +17,5 @@
 #' @keywords internal
 "_PACKAGE"
 
-#' @import dplyr
 #' @importFrom rlang .data
 NULL

R/Phenotypes.R

@@ -14,18 +14,22 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-#' List all phenotypes in the library
+#' Deprecated. List all phenotypes in the library.
 #'
 #' @return
 #' A tibble with the cohort ID and name.
+#' Deprecated. Please use getPhenotypeLog
 #'
 #' @examples
 #' listPhenotypes()
 #'
 #' @export
 listPhenotypes <- function() {
-  cohorts <- readr::read_csv(system.file("Cohorts.csv", package = "PhenotypeLibrary"), col_types = readr::cols())
-  return(cohorts)
+  .Deprecated(
+    new = "getPhenotypeLog",
+    msg = "listPhenotypes is deprecated. use getPhenotypeLog"
+  )
+  getPhenotypeLog()
 }
 
 #' Get a cohort definition set
@@ -37,7 +41,7 @@ listPhenotypes <- function() {
 #' `CohortGenerator` package.
 #'
 #' @examples
-#' cohorts <- listPhenotypes()
+#' cohorts <- getPhenotypeLog()
 #' subsetIds <- cohorts$cohortId[1:3]
 #' getPlCohortDefinitionSet(subsetIds)
 #'
@@ -47,8 +51,8 @@ getPlCohortDefinitionSet <- function(cohortIds) {
   checkmate::assertIntegerish(cohortIds, min.len = 1, add = errorMessages)
   checkmate::reportAssertions(collection = errorMessages)
 
-  cohorts <- listPhenotypes() |>
-    filter(.data$cohortId %in% cohortIds) |>
+  cohorts <- getPhenotypeLog() |>
+    dplyr::filter(.data$cohortId %in% cohortIds) |>
     dplyr::select(
       "cohortId",
       "cohortName"
@@ -68,15 +72,15 @@ getPlCohortDefinitionSet <- function(cohortIds) {
     sql <-
       readFile(file.path(sqlFolder, paste0(cohorts$cohortId[i], ".sql")))
     output <- cohorts[i, ] |>
-      mutate(
+      dplyr::mutate(
         json = !!json,
         sql = !!sql
       )
     return(output)
   }
 
   result <- lapply(seq_len(nrow(cohorts)), getJsonAndSql) |>
-    bind_rows()
+    dplyr::bind_rows()
 
   return(result)
 }
@@ -98,19 +102,48 @@ getPlCohortDefinitionSet <- function(cohortIds) {
 #' getPhenotypeLog(cohortIds = c(1, 2))
 #'
 #' @export
-getPhenotypeLog <- function(cohortIds = listPhenotypes()$cohortId) {
-  log <-
-    readr::read_csv(
-      system.file("Cohorts.csv", package = "PhenotypeLibrary"),
+getPhenotypeLog <- function(cohortIds = NULL) {
+  checkmate::assertIntegerish(cohortIds,
+    min.len = 0,
+    null.ok = TRUE
+  )
+
+  cohorts <-
+    readr::read_csv(system.file("Cohorts.csv", package = "PhenotypeLibrary"),
       col_types = readr::cols()
     ) |>
-    dplyr::filter(.data$cohortId %in% c(cohortIds)) |>
+    dplyr::mutate(
+      newCohortName = dplyr::case_when(
+        !is.na(.data$cohortNameLong) &
+          cohortNameLong != "" ~ .data$cohortNameLong, !is.na(.data$cohortNameFormatted) &
+          cohortNameFormatted != "" ~ .data$cohortNameFormatted,
+        TRUE ~ .data$cohortName
+      )
+    ) |>
+    dplyr::rename(
+      cohortNameAtlas = .data$cohortName,
+      cohortName = .data$newCohortName
+    ) |>
+    dplyr::relocate(
+      .data$cohortId,
+      .data$cohortName
+    ) |>
+    dplyr::arrange(.data$cohortId)
+
+  cohorts <- transformColumns(cohorts)
+
+  cohorts <- cohorts |>
     dplyr::mutate(
       addedDate = as.Date(.data$createdDate),
       updatedDate = as.Date(.data$modifiedDate)
     ) |>
     dplyr::arrange(.data$cohortId)
-  return(log)
+
+  if (!is.null(cohortIds)) {
+    cohorts <- cohorts |>
+      dplyr::filter(.data$cohortId %in% c(cohortIds))
+  }
+  return(cohorts)
 }
 
 
@@ -130,9 +163,40 @@ getPhenotypeLog <- function(cohortIds = listPhenotypes()$cohortId) {
 #'
 #' @export
 getPlConceptDefinitionSet <-
-  function(cohortIds = listPhenotypes()$cohortId) {
+  function(cohortIds = getPhenotypeLog()$cohortId) {
     conceptSets <-
       system.file("ConceptSetsInCohortDefinition.RDS", package = "PhenotypeLibrary") |>
       readRDS()
+
+    if (!is.null(cohortIds)) {
+      conceptSets <- conceptSets |>
+        dplyr::filter(.data$cohortId %in% c(cohortIds))
+    }
+
+    conceptSets <- transformColumns(df = conceptSets)
+
     return(conceptSets)
   }
+
+
+# Function to transform columns based on conditions
+transformColumns <- function(df) {
+  # Iterate through each column
+  for (col_name in names(df)) {
+    col_data <- df[[col_name]]
+
+    # If column contains alphabets
+    if (any(grepl("[a-zA-Z]", col_data, ignore.case = TRUE))) {
+      # Replace NA with empty string
+      df[[col_name]][is.na(df[[col_name]])] <- ""
+    }
+
+    # If column contains only 0, 1, and NA
+    else if (all(col_data %in% c(0, 1, NA))) {
+      # Replace NA with 0
+      df[[col_name]][is.na(df[[col_name]])] <- 0
+    }
+  }
+  return(df |>
+    dplyr::tibble())
+}