Add modulesToExecute in execution settings. (#170)

DESCRIPTION

@@ -60,7 +60,7 @@ Remotes:
   ohdsi/SelfControlledCaseSeries
 VignetteBuilder: knitr
 NeedsCompilation: no
-RoxygenNote: 7.3.1
+RoxygenNote: 7.3.2
 Roxygen: list(markdown = TRUE)
 Encoding: UTF-8
 Config/testthat/edition: 3

R/Execution.R

@@ -80,6 +80,61 @@ execute <- function(analysisSpecifications,
     }
   }
 
+  # Determine if the user has opted to subset to specific modules
+  # in the analysis specification. If so, validate that the
+  # modulesToExecute are present in the analysis specification
+  # before attempting to subset the analyses to run.
+  if (length(executionSettings$modulesToExecute) > 0) {
+    # Get the modules in the analysis specification with their
+    # index in the array
+    modulesWithIndex <- lapply(
+      X = seq_along(analysisSpecifications$moduleSpecifications),
+      FUN = function(i) {
+        list(
+          idx = i,
+          module = analysisSpecifications$moduleSpecifications[[i]]$module
+        )
+      }
+    )
+    modulesInAnalysisSpecification <- sapply(
+      X = modulesWithIndex,
+      FUN = function(x) {
+        x$module
+      }
+    )
+
+    modulesToExecuteString <- paste(executionSettings$modulesToExecute, collapse = ", ")
+    modulesInAnalysisSpecificationString <- paste(modulesInAnalysisSpecification, collapse = ", ")
+
+    # Stop if we cannot find all of the requested modules
+    # to execute in the overall analysis specification
+    if(!all(tolower(executionSettings$modulesToExecute) %in% tolower(modulesInAnalysisSpecification))) {
+      errorMsg <- paste0(
+        "The executionSettings specified to run only the modules: ",
+        modulesToExecuteString,
+        ".\n However the analysis specification includes the following modules: ",
+        modulesInAnalysisSpecificationString
+      )
+      stop(errorMsg)
+    }
+
+    # Subset the analysis specifications to those modules
+    # specified by the user
+    cli::cli_alert_info(paste0("Runnning a subset of modules: ", modulesToExecuteString))
+    moduleSubset <- unlist(
+      lapply(
+        X = modulesWithIndex,
+        FUN = function(x) {
+          if (tolower(x$module) %in% tolower(executionSettings$modulesToExecute)) {
+            return(x$idx)
+          }
+        }
+      )
+    )
+    analysisSpecifications$moduleSpecifications <- analysisSpecifications$moduleSpecifications[moduleSubset]
+  }
+
+
   # Set up logging
   if (!dir.exists(dirname(executionSettings$logFileName))) {
     dir.create(dirname(executionSettings$logFileName), recursive = T)

R/Module-CohortIncidence.R

@@ -198,9 +198,9 @@ CohortIncidenceModule <- R6::R6Class(
       return(data)
     },
     .getResultsDataModelSpecification = function() {
-      rdms <- readr::read_csv(
+      rdms <- CohortGenerator::readCsv(
         file = private$.getResultsDataModelSpecificationFileLocation(),
-        show_col_types = FALSE
+        warnOnCaseMismatch = FALSE
       )
       rdms$tableName <-paste0(self$tablePrefix, rdms$tableName)
       return(rdms)

R/Settings.R

@@ -258,6 +258,8 @@ createEmptyAnalysisSpecificiations <- function() {
 #'                                   and attempt to pick up where they left off when this value is set to TRUE.
 #' @param maxCores                   The maximum number of processing cores to use for execution. The default is to
 #'                                   use all available cores on the machine.
+#' @param modulesToExecute           (Optional) A vector with the list of modules to execute. When an empty vector/NULL is supplied (default),
+#'                                   all modules in the analysis specification are executed.
 #'
 #' @return
 #' An object of type `ExecutionSettings`.
@@ -272,7 +274,8 @@ createCdmExecutionSettings <- function(workDatabaseSchema,
                                        logFileName = file.path(resultsFolder, "strategus-log.txt"),
                                        minCellCount = 5,
                                        incremental = TRUE,
-                                       maxCores = parallel::detectCores()) {
+                                       maxCores = parallel::detectCores(),
+                                       modulesToExecute = c()) {
   errorMessages <- checkmate::makeAssertCollection()
   checkmate::assertCharacter(workDatabaseSchema, len = 1, add = errorMessages)
   checkmate::assertCharacter(cdmDatabaseSchema, len = 1, add = errorMessages)
@@ -283,6 +286,7 @@ createCdmExecutionSettings <- function(workDatabaseSchema,
   checkmate::assertInt(minCellCount, add = errorMessages)
   checkmate::assertLogical(incremental, add = errorMessages)
   checkmate::assertInt(maxCores, add = errorMessages)
+  checkmate::assertVector(modulesToExecute, null.ok = TRUE, add = errorMessages)
   checkmate::reportAssertions(collection = errorMessages)
 
   # Normalize paths to convert relative paths to absolute paths
@@ -308,6 +312,9 @@ createCdmExecutionSettings <- function(workDatabaseSchema,
 #'                                   in results.
 #' @param maxCores                   The maximum number of processing cores to use for execution. The default is to
 #'                                   use all available cores on the machine.
+#' @param modulesToExecute           (Optional) A vector with the list of modules to execute. When an empty vector/NULL is supplied (default),
+#'                                   all modules in the analysis specification are executed.
+#'
 #' @return
 #' An object of type `ExecutionSettings`.
 #'
@@ -317,14 +324,16 @@ createResultsExecutionSettings <- function(resultsDatabaseSchema,
                                            resultsFolder,
                                            logFileName = file.path(resultsFolder, "strategus-log.txt"),
                                            minCellCount = 5,
-                                           maxCores = parallel::detectCores()) {
+                                           maxCores = parallel::detectCores(),
+                                           modulesToExecute = c()) {
   errorMessages <- checkmate::makeAssertCollection()
   checkmate::assertCharacter(resultsDatabaseSchema, len = 1, add = errorMessages)
   checkmate::assertCharacter(workFolder, len = 1, add = errorMessages)
   checkmate::assertCharacter(resultsFolder, len = 1, add = errorMessages)
   checkmate::assertCharacter(logFileName, len = 1, add = errorMessages)
   checkmate::assertInt(minCellCount, add = errorMessages)
   checkmate::assertInt(maxCores, add = errorMessages)
+  checkmate::assertVector(modulesToExecute, null.ok = TRUE, add = errorMessages)
   checkmate::reportAssertions(collection = errorMessages)
 
   # Normalize paths to convert relative paths to absolute paths

extras/PackageMaintenance.R

@@ -115,8 +115,7 @@ cdModuleSpecifications <- cdModuleSettingsCreator$createModuleSpecifications(
   runBreakdownIndexEvents = TRUE,
   runIncidenceRate = TRUE,
   runCohortRelationship = TRUE,
-  runTemporalCohortCharacterization = TRUE,
-  incremental = FALSE
+  runTemporalCohortCharacterization = TRUE
 )
 
 # Cohort Generator -----------------
@@ -447,9 +446,8 @@ cdmModulesAnalysisSpecifications <- createEmptyAnalysisSpecificiations() |>
   addCharacterizationModuleSpecifications(cModuleSpecifications) |>
   addCohortDiagnosticsModuleSpecifications(cdModuleSpecifications) |>
   addCohortGeneratorModuleSpecifications(cgModuleSpecifications) |>
-  #addCohortIncidenceModuleSpecifications(ciModuleSpecifications) |>
+  addCohortIncidenceModuleSpecifications(ciModuleSpecifications) |>
   addCohortMethodeModuleSpecifications(cmModuleSpecifications) |>
-  #addEvidenceSynthesisModuleSpecifications(evidenceSynthesisAnalysisSpecifications) |>
   addSelfControlledCaseSeriesModuleSpecifications(sccsModuleSpecifications) |>
   addPatientLevelPredictionModuleSpecifications(plpModuleSpecifications)