[SPARK-51942] Support selectExpr in DataFrame

### What changes were proposed in this pull request?

This PR aims to support `selectExpr` API in `DataFrame`.

### Why are the changes needed?

For feature parity.

### Does this PR introduce _any_ user-facing change?

No behavior change.

### How was this patch tested?

Pass the CIs.

### Was this patch authored or co-authored using generative AI tooling?

No.

Closes #93 from dongjoon-hyun/SPARK-51942.

Authored-by: Dongjoon Hyun <[email protected]>
Signed-off-by: Dongjoon Hyun <[email protected]>

Author: dongjoon-hyun

Sources/SparkConnect/DataFrame.swift

@@ -269,6 +269,13 @@ public actor DataFrame: Sendable {
     return DataFrame(spark: self.spark, plan: SparkConnectClient.getProject(self.plan.root, cols))
   }
 
+  /// Projects a set of expressions and returns a new ``DataFrame``.
+  /// - Parameter exprs: Expression strings
+  /// - Returns: A ``DataFrame`` with subset of columns.
+  public func selectExpr(_ exprs: String...) -> DataFrame {
+    return DataFrame(spark: self.spark, plan: SparkConnectClient.getProjectExprs(self.plan.root, exprs))
+  }
+
   /// Returns a new Dataset with a column dropped. This is a no-op if schema doesn't contain column name.
   /// - Parameter cols: Column names
   /// - Returns: A ``DataFrame`` with subset of columns.

Sources/SparkConnect/SparkConnectClient.swift

@@ -390,6 +390,22 @@ public actor SparkConnectClient {
     return plan
   }
 
+  static func getProjectExprs(_ child: Relation, _ exprs: [String]) -> Plan {
+    var project = Project()
+    project.input = child
+    let expressions: [Spark_Connect_Expression] = exprs.map {
+      var expression = Spark_Connect_Expression()
+      expression.exprType = .expressionString($0.toExpressionString)
+      return expression
+    }
+    project.expressions = expressions
+    var relation = Relation()
+    relation.project = project
+    var plan = Plan()
+    plan.opType = .root(relation)
+    return plan
+  }
+
   static func getWithColumnRenamed(_ child: Relation, _ colsMap: [String: String]) -> Plan {
     var withColumnsRenamed = WithColumnsRenamed()
     withColumnsRenamed.input = child

Tests/SparkConnectTests/DataFrameTests.swift

@@ -208,6 +208,20 @@ struct DataFrameTests {
     try await #require(throws: Error.self) {
       let _ = try await spark.range(1).select("invalid").schema
     }
+    try await #require(throws: Error.self) {
+      let _ = try await spark.range(1).select("id + 1").schema
+    }
+    await spark.stop()
+  }
+
+  @Test
+  func selectExpr() async throws {
+    let spark = try await SparkSession.builder.getOrCreate()
+    let schema = try await spark.range(1).selectExpr("id + 1 as id2").schema
+    #expect(
+      schema
+        == #"{"struct":{"fields":[{"name":"id2","dataType":{"long":{}}}]}}"#
+    )
     await spark.stop()
   }
 

Tests/SparkConnectTests/SparkConnectClientTests.swift

@@ -84,6 +84,7 @@ struct SparkConnectClientTests {
     await client.stop()
   }
 
+#if !os(Linux)  // TODO: Enable this with the offical Spark 4 docker image
   @Test
   func jsonToDdl() async throws {
     let client = SparkConnectClient(remote: TEST_REMOTE)
@@ -95,4 +96,5 @@ struct SparkConnectClientTests {
     }
     await client.stop()
   }
+#endif
 }