Add incremental setting for CDM execution

R/Module-Characterization.R

@@ -39,7 +39,7 @@ CharacterizationModule <- R6::R6Class(
         csvFilePrefix = self$tablePrefix,
         minCellCount =  jobContext$moduleExecutionSettings$minCellCount,
         minCharacterizationMean = jobContext$moduleExecutionSettings$minCharacterizationMean,
-        incremental = T, # any Strartegus param for this?
+        incremental = jobContext$moduleExecutionSettings$incremental,
         threads = as.double(ifelse(Sys.getenv('CharacterizationThreads') == "", 1,Sys.getenv('CharacterizationThreads') ))
       )
 

R/Module-CohortDiagnostics.R

@@ -133,7 +133,6 @@ CohortDiagnosticsModule <- R6::R6Class(
     #'                                    will help reduce the file size of the characterization output, but will remove information
     #'                                    on covariates that have very low values. The default is 0.001 (i.e. 0.1 percent)
     #' @param irWashoutPeriod             Number of days washout to include in calculation of incidence rates - default is 0
-    #' @param incremental                 Create only cohort diagnostics that haven't been created before?
     createModuleSpecifications = function(cohortIds = NULL,
                                           runInclusionStatistics = TRUE,
                                           runIncludedSourceConcepts = TRUE,
@@ -146,8 +145,7 @@ CohortDiagnosticsModule <- R6::R6Class(
                                           runTemporalCohortCharacterization = TRUE,
                                           temporalCovariateSettings = private$.getDefaultCovariateSettings(),
                                           minCharacterizationMean = 0.01,
-                                          irWashoutPeriod = 0,
-                                          incremental = FALSE) {
+                                          irWashoutPeriod = 0) {
       analysis <- list()
       for (name in names(formals(self$createModuleSpecifications))) {
         analysis[[name]] <- get(name)

R/Module-CohortGenerator.R

@@ -46,7 +46,7 @@ CohortGeneratorModule <- R6::R6Class(
         detectOnDescendants = negativeControlOutcomeSettings$detectOnDescendants,
         outputFolder = resultsFolder,
         databaseId = jobContext$moduleExecutionSettings$cdmDatabaseMetaData$databaseId,
-        incremental = jobContext$settings$incremental,
+        incremental = jobContext$moduleExecutionSettings$incremental,
         incrementalFolder = jobContext$moduleExecutionSettings$workSubFolder
       )
 
@@ -79,13 +79,9 @@ CohortGeneratorModule <- R6::R6Class(
       )
     },
     #' @description Creates the CohortGenerator Module Specifications
-    #' @param incremental When TRUE, the module will keep track of the cohorts
-    #' generated so that subsequent runs will skip any previously generated
-    #' cohorts.
     #' @param generateStats When TRUE, the Circe cohort definition SQL will
     #' include steps to compute inclusion rule statistics.
-    createModuleSpecifications = function(incremental = TRUE,
-                                          generateStats = TRUE) {
+    createModuleSpecifications = function(generateStats = TRUE) {
       analysis <- list()
       for (name in names(formals(self$createModuleSpecifications))) {
         analysis[[name]] <- get(name)

R/Settings.R

@@ -253,6 +253,9 @@ createEmptyAnalysisSpecificiations <- function() {
 #' @param logFileName                Logging information from Strategus and all modules will be located in this file. Individual modules will continue to have their own module-specific logs. By default this will be written to the root of the `resultsFolder`
 #' @param minCellCount               The minimum number of subjects contributing to a count before it can be included
 #'                                   in results.
+#' @param incremental                This value will be passed to each module that supports execution in an incremental manner. Modules
+#'                                   and their underlying packages may use the `workFolder` contents to determine their state of execution
+#'                                   and attempt to pick up where they left off when this value is set to TRUE.
 #'
 #' @return
 #' An object of type `ExecutionSettings`.
@@ -265,7 +268,8 @@ createCdmExecutionSettings <- function(workDatabaseSchema,
                                        workFolder,
                                        resultsFolder,
                                        logFileName = file.path(resultsFolder, "strategus-log.txt"),
-                                       minCellCount = 5) {
+                                       minCellCount = 5,
+                                       incremental = TRUE) {
   errorMessages <- checkmate::makeAssertCollection()
   checkmate::assertCharacter(workDatabaseSchema, len = 1, add = errorMessages)
   checkmate::assertCharacter(cdmDatabaseSchema, len = 1, add = errorMessages)
@@ -274,6 +278,7 @@ createCdmExecutionSettings <- function(workDatabaseSchema,
   checkmate::assertCharacter(resultsFolder, len = 1, add = errorMessages)
   checkmate::assertCharacter(logFileName, len = 1, add = errorMessages)
   checkmate::assertInt(minCellCount, add = errorMessages)
+  checkmate::assertLogical(incremental, add = errorMessages)
   checkmate::reportAssertions(collection = errorMessages)
 
   # Normalize paths to convert relative paths to absolute paths