Merge pull request #302 from intersystems-community/main

Introducing support for InterSystems IRIS

DESCRIPTION

@@ -13,14 +13,15 @@ Authors@R: c(
     person("Amazon Inc.", role = c("cph"), comment = "RedShift  JDBC driver")
   )
 Description: An R 'DataBase Interface' ('DBI') compatible interface to various database platforms ('PostgreSQL', 'Oracle', 'Microsoft SQL Server', 
-    'Amazon Redshift', 'Microsoft Parallel Database Warehouse', 'IBM Netezza', 'Apache Impala', 'Google BigQuery', 'Snowflake', 'Spark', and 'SQLite'). Also includes support for
-    fetching data as 'Andromeda' objects. Uses either 'Java Database Connectivity' ('JDBC') or other 'DBI' drivers to connect to databases.
+    'Amazon Redshift', 'Microsoft Parallel Database Warehouse', 'IBM Netezza', 'Apache Impala', 'Google BigQuery', 'Snowflake', 'Spark', 'SQLite', 
+    and 'InterSystems IRIS'). Also includes support for fetching data as 'Andromeda' objects. Uses either 'Java Database Connectivity' ('JDBC') or 
+    other 'DBI' drivers to connect to databases.
 SystemRequirements: Java (>= 8)
 Depends:
     R (>= 4.0.0)
 Imports:
     rJava,
-    SqlRender (>= 1.16.0),
+    SqlRender (>= 1.19.1),
     methods,
     stringr,
     readr,

R/Connect.R

@@ -32,7 +32,8 @@ checkIfDbmsIsSupported <- function(dbms) {
     "spark",
     "snowflake",
     "synapse",
-    "duckdb"
+    "duckdb",
+    "iris"
   )
   deprecated <- c(
     "hive",
@@ -332,6 +333,8 @@ connectUsingJdbc <- function(connectionDetails) {
     return(connectSpark(connectionDetails))
   } else if (dbms == "snowflake") {
     return(connectSnowflake(connectionDetails))
+  } else if (dbms == "iris") {
+    return(connectIris(connectionDetails))
   } else {
     abort("Something went wrong when trying to connect to ", dbms)
   }
@@ -747,6 +750,36 @@ connectSqlite <- function(connectionDetails) {
   return(connection)
 }
 
+connectIris <- function(connectionDetails) {
+  inform("Connecting using InterSystems IRIS driver")
+  jarPath <- findPathToJar("^intersystems-jdbc-.*\\.jar$", connectionDetails$pathToDriver)
+  driver <- getJbcDriverSingleton("com.intersystems.jdbc.IRISDriver", jarPath)
+  if (is.null(connectionDetails$connectionString()) || connectionDetails$connectionString() == "") {
+    if (is.null(connectionDetails$port())) {
+      port <- "1972"
+    } else {
+      port <- connectionDetails$port()
+    }
+    connectionString <- paste0("jdbc:IRIS://", connectionDetails$server(), ":", port, "/USER")  # use a full connection string for nondefault database
+    if (!is.null(connectionDetails$extraSettings)) {
+      connectionString <- paste(connectionString, connectionDetails$extraSettings, sep = ";")
+    }
+  } else {
+    connectionString <- connectionDetails$connectionString()
+  }
+  if (is.null(connectionDetails$user())) {
+    connection <- connectUsingJdbcDriver(driver, connectionString, dbms = connectionDetails$dbms)
+  } else {
+    connection <- connectUsingJdbcDriver(driver,
+      connectionString,
+      user = connectionDetails$user(),
+      password = connectionDetails$password(),
+      dbms = connectionDetails$dbms
+    )
+  }
+  return(connection)
+}
+
 connectUsingJdbcDriver <- function(jdbcDriver,
                                    url,
                                    identifierQuote = "'",

R/DatabaseConnector.R

@@ -56,7 +56,7 @@ NULL
 #' functions to point to the driver. Alternatively, you can set the 'DATABASECONNECTOR_JAR_FOLDER' environmental
 #' variable, for example in your .Renviron file (recommended).
 #'
-#' # SQL Server, Oracle, PostgreSQL, PDW, Snowflake, Spark, RedShift, Azure Synapse, BigQuery
+#' # SQL Server, Oracle, PostgreSQL, PDW, Snowflake, Spark, RedShift, Azure Synapse, BigQuery, InterSystems IRIS
 #' 
 #' Use the [downloadJdbcDrivers()] function to download these drivers from the OHDSI GitHub pages.
 #'
@@ -75,7 +75,7 @@ NULL
 #' 
 #' For SQLite we actually don't use a JDBC driver. Instead, we use the RSQLite package, which can be installed
 #' using `install.packages("RSQLite")`.
-#'
+#' 
 NULL
 
 globalVars <- new.env()

R/Drivers.R

@@ -33,6 +33,7 @@ jdbcDrivers <- new.env()
 #' - "spark" for Spark
 #' - "snowflake" for Snowflake
 #' - "bigquery" for Google BigQuery
+#' - "iris" for InterSystems IRIS
 #' - "all" for all aforementioned platforms
 #'  
 #' @param method The method used for downloading files. See `?download.file` for details and options.
@@ -48,6 +49,7 @@ jdbcDrivers <- new.env()
 #' - Spark (Databricks): V2.6.36
 #' - Snowflake: V3.16.01
 #' - BigQuery: v1.3.2.1003
+#' - InterSystems IRIS: v3.10.2
 #' 
 #' @return Invisibly returns the destination if the download was successful.
 #' @export
@@ -81,9 +83,9 @@ downloadJdbcDrivers <- function(dbms, pathToDriver = Sys.getenv("DATABASECONNECT
     warn(paste0("The folder location '", pathToDriver, "' does not exist. Attempting to create."))
     dir.create(pathToDriver, recursive = TRUE)
   }
-  
-  stopifnot(is.character(dbms), length(dbms) == 1, dbms %in% c("all", "postgresql", "redshift", "sql server", "oracle", "pdw", "snowflake", "spark", "bigquery"))
-  
+
+  stopifnot(is.character(dbms), length(dbms) == 1, dbms %in% c("all", "postgresql", "redshift", "sql server", "oracle", "pdw", "snowflake", "spark", "bigquery", "iris"))
+
   if (dbms == "pdw" || dbms == "synapse") {
     dbms <- "sql server"
   }
@@ -96,7 +98,8 @@ downloadJdbcDrivers <- function(dbms, pathToDriver = Sys.getenv("DATABASECONNECT
     4,oracle,oracleV19.8.zip,https://ohdsi.github.io/DatabaseConnectorJars/
     5,spark,DatabricksJDBC42-2.6.36.1062.zip,https://databricks-bi-artifacts.s3.us-east-2.amazonaws.com/simbaspark-drivers/jdbc/2.6.36/
     6,snowflake,snowflake-jdbc-3.16.1.jar,https://repo1.maven.org/maven2/net/snowflake/snowflake-jdbc/3.16.1/
-    7,bigquery,SimbaJDBCDriverforGoogleBigQuery42_1.6.2.1003.zip,https://storage.googleapis.com/simba-bq-release/jdbc/"
+    7,bigquery,SimbaJDBCDriverforGoogleBigQuery42_1.6.2.1003.zip,https://storage.googleapis.com/simba-bq-release/jdbc/
+    8,iris,intersystems-jdbc-3.10.2.jar,https://repo1.maven.org/maven2/com/intersystems/intersystems-jdbc/3.10.2/"
   )
 
   if (dbms == "all") {

R/RStudio.R

@@ -76,7 +76,7 @@ unregisterWithRStudio <- function(connection) {
 }
 
 hasCatalogs <- function(connection) {
-  return(dbms(connection) %in% c("pdw", "postgresql", "sql server", "synapse", "redshift", "snowflake", "spark", "bigquery", "duckdb"))
+  return(dbms(connection) %in% c("pdw", "postgresql", "sql server", "synapse", "redshift", "snowflake", "spark", "bigquery", "duckdb", "iris"))
 }
 
 listDatabaseConnectorColumns <- function(connection,

R/Sql.R

@@ -291,9 +291,10 @@ lowLevelExecuteSql.default <- function(connection, sql) {
 
   statement <- rJava::.jcall(connection@jConnection, "Ljava/sql/Statement;", "createStatement")
   on.exit(rJava::.jcall(statement, "V", "close"))
-  if (dbms(connection) == "spark") {
+  if ((dbms(connection) == "spark") || (dbms(connection) == "iris")) {
     # For some queries the DataBricks JDBC driver will throw an error saying no ROWCOUNT is returned
-    # when using executeLargeUpdate, so using execute instead. 
+    # when using executeLargeUpdate, so using execute instead.
+    # Also use this approach for IRIS JDBC driver, which does not support executeLargeUpdate() directly.
     rJava::.jcall(statement, "Z", "execute", as.character(sql), check = FALSE)
     rowsAffected <- rJava::.jcall(statement, "I", "getUpdateCount", check = FALSE)
     if (rowsAffected == -1) {
@@ -435,7 +436,13 @@ executeSql <- function(connection,
       tryCatch(
         {
           startQuery <- Sys.time()
-          rowsAffected <- c(rowsAffected, rJava::.jcall(statement, "[J", "executeLargeBatch"))
+          # InterSystems IRIS JDBC supports batch updates but does not have a separate
+          # executeLargeBatch() method
+          if (con@dbms == "iris") {
+            rowsAffected <- c(rowsAffected, rJava::.jcall(statement, "[I", "executeBatch"))
+          } else {
+            rowsAffected <- c(rowsAffected, rJava::.jcall(statement, "[J", "executeLargeBatch"))
+          }
           delta <- Sys.time() - startQuery
           if (profile) {
             inform(paste("Statements", start, "through", end, "took", delta, attr(delta, "units")))

README.md

@@ -27,6 +27,7 @@ Features
   - IBM Netezza
   - SQLite
   - Spark
+  - InterSystems IRIS
 - Statements for executing queries with 
   - Error reporting to file
   - Progress reporting

man-roxygen/Dbms.R

@@ -11,4 +11,5 @@
 #' - "sqlite extended" for SQLite with extended types (DATE and DATETIME)
 #' - "spark" for Spark
 #' - "snowflake" for Snowflake
+#' - "iris" for InterSystems IRIS
 #'                           

man-roxygen/DefaultConnectionDetails.R

@@ -131,6 +131,14 @@
 #'  - `user`. The user name used to access the server.
 #'  - `password`. The password for that user.
 #' 
+#' InterSystems IRIS:
+#'  - `connectionString`. The connection string (e.g. starting with
+#'         'jdbc:IRIS://host:port/namespace'). Alternatively, you can provide
+#'         values for `server` and `port`, in which case the default `USER` namespace
+#'         is used to connect.
+#'  - `user`. The user name used to access the server.
+#'  - `password`. The password for that user.
+#'  - `pathToDriver`. The path to the folder containing the InterSystems IRIS JDBC driver JAR file.
 #'
 #' ## Windows authentication for SQL Server:
 #'