2026 bug fixes

.lintr

@@ -4,6 +4,7 @@ linters: linters_with_defaults(
     brace_linter = NULL,
     object_length_linter = NULL,
     line_length_linter = NULL,
-    cyclocomp_linter = NULL
+    cyclocomp_linter = NULL,
+    pipe_consistency_linter = NULL
   ) # see vignette("lintr")
 encoding: "UTF-8"

R/LoadSyntheaTables.r

@@ -38,6 +38,7 @@ LoadSyntheaTables <-
       csvList <- list.files(syntheaFileLoc, pattern = "*.csv")
 
       conn <- DatabaseConnector::connect(connectionDetails)
+      dateColumns <- c("START", "STOP", "DATE", "START_DATE", "END_DATE", "BIRTHDATE", "DEATHDATE")
 
       for (csv in csvList) {
         syntheaTable <-
@@ -52,33 +53,11 @@ LoadSyntheaTables <-
         writeLines(paste0("Loading: ", csv))
 
         # experiencing type conversion errors and need to explicitly case some columns
-        if ("START"       %in% colnames(syntheaTable)) {
-          syntheaTable$START <-
-            as.Date(syntheaTable$START, format = "%Y-%m-%d")
-        }
-        if ("STOP"        %in% colnames(syntheaTable)) {
-          syntheaTable$STOP         <-
-            as.Date(syntheaTable$STOP, format = "%Y-%m-%d")
-        }
-        if ("DATE"        %in% colnames(syntheaTable)) {
-          syntheaTable$DATE         <-
-            as.Date(syntheaTable$DATE, format = "%Y-%m-%d")
-        }
-        if ("START_DATE"        %in% colnames(syntheaTable)) {
-          syntheaTable$START_DATE         <-
-            as.Date(syntheaTable$START_DATE, format = "%Y-%m-%d")
-        }
-        if ("END_DATE"        %in% colnames(syntheaTable)) {
-          syntheaTable$END_DATE         <-
-            as.Date(syntheaTable$END_DATE, format = "%Y-%m-%d")
-        }
-        if ("BIRTHDATE"   %in% colnames(syntheaTable)) {
-          syntheaTable$BIRTHDATE    <-
-            as.Date(syntheaTable$BIRTHDATE, format = "%Y-%m-%d")
-        }
-        if ("DEATHDATE"   %in% colnames(syntheaTable)) {
-          syntheaTable$DEATHDATE    <-
-            as.Date(syntheaTable$DEATHDATE, format = "%Y-%m-%d")
+        tableDateColumns <- intersect(dateColumns, colnames(syntheaTable))
+        for (dateColumn in tableDateColumns) {
+          syntheaTable[[dateColumn]] <- as.character(syntheaTable[[dateColumn]])
+          syntheaTable[[dateColumn]][nchar(syntheaTable[[dateColumn]]) == 0] <- NA_character_
+          syntheaTable[[dateColumn]] <- substr(syntheaTable[[dateColumn]], 1, 10)
         }
         if ("CODE"        %in% colnames(syntheaTable)) {
           syntheaTable$CODE         <- as.character(syntheaTable$CODE)
@@ -96,11 +75,17 @@ LoadSyntheaTables <-
             as.numeric(syntheaTable$UTILIZATION)
         }
 
+        insertData <- as.data.frame(syntheaTable, stringsAsFactors = FALSE)
+        insertDateColumns <- intersect(dateColumns, colnames(insertData))
+        for (dateColumn in insertDateColumns) {
+          insertData[[dateColumn]] <- as.Date(insertData[[dateColumn]], format = "%Y-%m-%d")
+        }
+
         suppressWarnings({
           DatabaseConnector::insertTable(
             conn,
             tableName = paste0(syntheaSchema, ".", strsplit(csv, "[.]")[[1]][1]),
-            data = as.data.frame(syntheaTable),
+            data = insertData,
             dropTableIfExists = FALSE,
             createTable = FALSE,
             bulkLoad = bulkLoad,

README.md

@@ -47,8 +47,9 @@ ETLSyntheaBuilder::CreateCDMTables(connectionDetails = cd, cdmSchema = cdmSchema
 ETLSyntheaBuilder::CreateSyntheaTables(connectionDetails = cd, syntheaSchema = syntheaSchema, syntheaVersion = syntheaVersion)
 
 ETLSyntheaBuilder::LoadSyntheaTables(connectionDetails = cd, syntheaSchema = syntheaSchema, syntheaFileLoc = syntheaFileLoc)
-
-ETLSyntheaBuilder::LoadVocabFromCsv(connectionDetails = cd, cdmSchema = cdmSchema, vocabFileLoc = vocabFileLoc)
+
+## Note this step assumes you have built the CDM using the ETLSyntheaBuilder::CreateCDMTables command. If you have executed the DDL using the OHDSI/CommonDataModel package you will encounter constraint errors.
+ETLSyntheaBuilder::LoadVocabFromCsv(connectionDetails = cd, cdmSchema = cdmSchema, vocabFileLoc = vocabFileLoc,delimiter='\t')
 
 ETLSyntheaBuilder::CreateMapAndRollupTables(connectionDetails = cd, cdmSchema = cdmSchema, syntheaSchema = syntheaSchema, cdmVersion = cdmVersion, syntheaVersion = syntheaVersion)
 

extras/codeToRun.R

@@ -32,7 +32,8 @@ ETLSyntheaBuilder::CreateSyntheaTables(connectionDetails,syntheaSchema, syntheaV
 # Populate synthea tables
 ETLSyntheaBuilder::LoadSyntheaTables(connectionDetails,syntheaSchema,syntheaFileLoc)
 # Populate vocabulary tables
-ETLSyntheaBuilder::LoadVocabFromCsv(connectionDetails,cdmDatabaseSchema,vocabFileLoc)
+## Note this step assumes you have built the CDM using the ETLSyntheaBuilder::CreateCDMTables command. If you have executed the DDL using the OHDSI/CommonDataModel package you will encounter constraint errors.
+ETLSyntheaBuilder::LoadVocabFromCsv(connectionDetails,cdmDatabaseSchema,vocabFileLoc,delimiter='\t')
 # Create intermediate vocabulary mapping and visit rollup tables
 ETLSyntheaBuilder::CreateMapAndRollupTables(connectionDetails, cdmDatabaseSchema,syntheaSchema,cdmVersion,syntheaVersion)
 ## Optional Step to create extra indices

inst/sql/sql_server/synthea_version/v310/create_synthea_tables.sql

@@ -291,8 +291,8 @@ create table @synthea_schema.payer_transitions (
 --HINT DISTRIBUTE_ON_RANDOM
 create table @synthea_schema.payers (
   id                       varchar(1000),
-  "name"                     varchar(1000),
-  ownership                varchar NULL,
+  "name"                   varchar(1000),
+  ownership                varchar(1000),
   address                  varchar(1000),
   city                     varchar(1000),
   state_headquartered      varchar(1000),

inst/sql/sql_server/synthea_version/v320/create_synthea_tables.sql

@@ -291,8 +291,8 @@ create table @synthea_schema.payer_transitions (
 --HINT DISTRIBUTE_ON_RANDOM
 create table @synthea_schema.payers (
   id                       varchar(1000),
-  "name"                     varchar(1000),
-  ownership                varchar NULL,
+  "name"                   varchar(1000),
+  ownership                varchar(1000),
   address                  varchar(1000),
   city                     varchar(1000),
   state_headquartered      varchar(1000),

inst/sql/sql_server/synthea_version/v330/create_synthea_tables.sql

@@ -294,8 +294,8 @@ create table @synthea_schema.payer_transitions (
 --HINT DISTRIBUTE_ON_RANDOM
 create table @synthea_schema.payers (
   id                       varchar(1000),
-  "name"                     varchar(1000),
-  ownership                varchar NULL,
+  "name"                   varchar(1000),
+  ownership                varchar(1000),
   address                  varchar(1000),
   city                     varchar(1000),
   state_headquartered      varchar(1000),