[SPARK-22771][SQL] Concatenate binary inputs into a binary output

## What changes were proposed in this pull request?
This pr modified `concat` to concat binary inputs into a single binary output.
`concat` in the current master always output data as a string. But, in some databases (e.g., PostgreSQL), if all inputs are binary, `concat` also outputs binary.

## How was this patch tested?
Added tests in `SQLQueryTestSuite` and `TypeCoercionSuite`.

Author: Takeshi Yamamuro <yamamuro@apache.org>

Closes #19977 from maropu/SPARK-22771.

Author: maropu

R/pkg/R/functions.R

@@ -2133,7 +2133,8 @@ setMethod("countDistinct",
           })
 
 #' @details
-#' \code{concat}: Concatenates multiple input string columns together into a single string column.
+#' \code{concat}: Concatenates multiple input columns together into a single column.
+#' If all inputs are binary, concat returns an output as binary. Otherwise, it returns as string.
 #'
 #' @rdname column_string_functions
 #' @aliases concat concat,Column-method

common/unsafe/src/main/java/org/apache/spark/unsafe/types/ByteArray.java

@@ -74,4 +74,29 @@ public static byte[] subStringSQL(byte[] bytes, int pos, int len) {
     }
     return Arrays.copyOfRange(bytes, start, end);
   }
+
+  public static byte[] concat(byte[]... inputs) {
+    // Compute the total length of the result
+    int totalLength = 0;
+    for (int i = 0; i < inputs.length; i++) {
+      if (inputs[i] != null) {
+        totalLength += inputs[i].length;
+      } else {
+        return null;
+      }
+    }
+
+    // Allocate a new byte array, and copy the inputs one by one into it
+    final byte[] result = new byte[totalLength];
+    int offset = 0;
+    for (int i = 0; i < inputs.length; i++) {
+      int len = inputs[i].length;
+      Platform.copyMemory(
+        inputs[i], Platform.BYTE_ARRAY_OFFSET,
+        result, Platform.BYTE_ARRAY_OFFSET + offset,
+        len);
+      offset += len;
+    }
+    return result;
+  }
 }

docs/sql-programming-guide.md

@@ -1780,6 +1780,8 @@ options.
 
  - Since Spark 2.3, when either broadcast hash join or broadcast nested loop join is applicable, we prefer to broadcasting the table that is explicitly specified in a broadcast hint. For details, see the section [Broadcast Hint](#broadcast-hint-for-sql-queries) and [SPARK-22489](https://issues.apache.org/jira/browse/SPARK-22489).
 
+ - Since Spark 2.3, when all inputs are binary, `functions.concat()` returns an output as binary. Otherwise, it returns as a string. Until Spark 2.3, it always returns as a string despite of input types. To keep the old behavior, set `spark.sql.function.concatBinaryAsString` to `true`.
+
 ## Upgrading From Spark SQL 2.1 to 2.2
 
   - Spark 2.1.1 introduced a new configuration key: `spark.sql.hive.caseSensitiveInferenceMode`. It had a default setting of `NEVER_INFER`, which kept behavior identical to 2.1.0. However, Spark 2.2.0 changes this setting's default value to `INFER_AND_SAVE` to restore compatibility with reading Hive metastore tables whose underlying file schema have mixed-case column names. With the `INFER_AND_SAVE` configuration value, on first access Spark will perform schema inference on any Hive metastore table for which it has not already saved an inferred schema. Note that schema inference can be a very time consuming operation for tables with thousands of partitions. If compatibility with mixed-case column names is not a concern, you can safely set `spark.sql.hive.caseSensitiveInferenceMode` to `NEVER_INFER` to avoid the initial overhead of schema inference. Note that with the new default `INFER_AND_SAVE` setting, the results of the schema inference are saved as a metastore key for future use. Therefore, the initial schema inference occurs only at a table's first access.

python/pyspark/sql/functions.py

@@ -1374,7 +1374,8 @@ def hash(*cols):
 @ignore_unicode_prefix
 def concat(*cols):
     """
-    Concatenates multiple input string columns together into a single string column.
+    Concatenates multiple input columns together into a single column.
+    If all inputs are binary, concat returns an output as binary. Otherwise, it returns as string.
 
     >>> df = spark.createDataFrame([('abcd','123')], ['s', 'd'])
     >>> df.select(concat(df.s, df.d).alias('s')).collect()

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala

@@ -150,7 +150,7 @@ class Analyzer(
       TimeWindowing ::
       ResolveInlineTables(conf) ::
       ResolveTimeZone(conf) ::
-      TypeCoercion.typeCoercionRules ++
+      TypeCoercion.typeCoercionRules(conf) ++
       extendedResolutionRules : _*),
     Batch("Post-Hoc Resolution", Once, postHocResolutionRules: _*),
     Batch("View", Once,

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/TypeCoercion.scala

@@ -27,6 +27,7 @@ import org.apache.spark.sql.catalyst.expressions._
 import org.apache.spark.sql.catalyst.expressions.aggregate._
 import org.apache.spark.sql.catalyst.plans.logical._
 import org.apache.spark.sql.catalyst.rules.Rule
+import org.apache.spark.sql.internal.SQLConf
 import org.apache.spark.sql.types._
 
 
@@ -45,13 +46,14 @@ import org.apache.spark.sql.types._
  */
 object TypeCoercion {
 
-  val typeCoercionRules =
+  def typeCoercionRules(conf: SQLConf): List[Rule[LogicalPlan]] =
     InConversion ::
       WidenSetOperationTypes ::
       PromoteStrings ::
       DecimalPrecision ::
       BooleanEquality ::
       FunctionArgumentConversion ::
+      ConcatCoercion(conf) ::
       CaseWhenCoercion ::
       IfCoercion ::
       StackCoercion ::
@@ -660,6 +662,28 @@ object TypeCoercion {
     }
   }
 
+  /**
+   * Coerces the types of [[Concat]] children to expected ones.
+   *
+   * If `spark.sql.function.concatBinaryAsString` is false and all children types are binary,
+   * the expected types are binary. Otherwise, the expected ones are strings.
+   */
+  case class ConcatCoercion(conf: SQLConf) extends TypeCoercionRule {
+
+    override protected def coerceTypes(plan: LogicalPlan): LogicalPlan = plan transform { case p =>
+      p transformExpressionsUp {
+        // Skip nodes if unresolved or empty children
+        case c @ Concat(children) if !c.childrenResolved || children.isEmpty => c
+        case c @ Concat(children) if conf.concatBinaryAsString ||
+            !children.map(_.dataType).forall(_ == BinaryType) =>
+          val newChildren = c.children.map { e =>
+            ImplicitTypeCasts.implicitCast(e, StringType).getOrElse(e)
+          }
+          c.copy(children = newChildren)
+      }
+    }
+  }
+
   /**
    * Turns Add/Subtract of DateType/TimestampType/StringType and CalendarIntervalType
    * to TimeAdd/TimeSub

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringExpressions.scala

@@ -24,11 +24,10 @@ import java.util.regex.Pattern
 
 import scala.collection.mutable.ArrayBuffer
 
-import org.apache.spark.sql.AnalysisException
 import org.apache.spark.sql.catalyst.InternalRow
 import org.apache.spark.sql.catalyst.analysis.TypeCheckResult
 import org.apache.spark.sql.catalyst.expressions.codegen._
-import org.apache.spark.sql.catalyst.util.{ArrayData, GenericArrayData}
+import org.apache.spark.sql.catalyst.util.{ArrayData, GenericArrayData, TypeUtils}
 import org.apache.spark.sql.types._
 import org.apache.spark.unsafe.types.{ByteArray, UTF8String}
 
@@ -38,7 +37,8 @@ import org.apache.spark.unsafe.types.{ByteArray, UTF8String}
 
 
 /**
- * An expression that concatenates multiple input strings into a single string.
+ * An expression that concatenates multiple inputs into a single output.
+ * If all inputs are binary, concat returns an output as binary. Otherwise, it returns as string.
  * If any input is null, concat returns null.
  */
 @ExpressionDescription(
@@ -48,17 +48,37 @@ import org.apache.spark.unsafe.types.{ByteArray, UTF8String}
       > SELECT _FUNC_('Spark', 'SQL');
        SparkSQL
   """)
-case class Concat(children: Seq[Expression]) extends Expression with ImplicitCastInputTypes {
+case class Concat(children: Seq[Expression]) extends Expression {
 
-  override def inputTypes: Seq[AbstractDataType] = Seq.fill(children.size)(StringType)
-  override def dataType: DataType = StringType
+  private lazy val isBinaryMode: Boolean = dataType == BinaryType
+
+  override def checkInputDataTypes(): TypeCheckResult = {
+    if (children.isEmpty) {
+      TypeCheckResult.TypeCheckSuccess
+    } else {
+      val childTypes = children.map(_.dataType)
+      if (childTypes.exists(tpe => !Seq(StringType, BinaryType).contains(tpe))) {
+        TypeCheckResult.TypeCheckFailure(
+          s"input to function $prettyName should have StringType or BinaryType, but it's " +
+            childTypes.map(_.simpleString).mkString("[", ", ", "]"))
+      }
+      TypeUtils.checkForSameTypeInputExpr(childTypes, s"function $prettyName")
+    }
+  }
+
+  override def dataType: DataType = children.map(_.dataType).headOption.getOrElse(StringType)
 
   override def nullable: Boolean = children.exists(_.nullable)
   override def foldable: Boolean = children.forall(_.foldable)
 
   override def eval(input: InternalRow): Any = {
-    val inputs = children.map(_.eval(input).asInstanceOf[UTF8String])
-    UTF8String.concat(inputs : _*)
+    if (isBinaryMode) {
+      val inputs = children.map(_.eval(input).asInstanceOf[Array[Byte]])
+      ByteArray.concat(inputs: _*)
+    } else {
+      val inputs = children.map(_.eval(input).asInstanceOf[UTF8String])
+      UTF8String.concat(inputs : _*)
+    }
   }
 
   override protected def doGenCode(ctx: CodegenContext, ev: ExprCode): ExprCode = {
@@ -73,17 +93,27 @@ case class Concat(children: Seq[Expression]) extends Expression with ImplicitCas
         }
       """
     }
+
+    val (concatenator, initCode) = if (isBinaryMode) {
+      (classOf[ByteArray].getName, s"byte[][] $args = new byte[${evals.length}][];")
+    } else {
+      ("UTF8String", s"UTF8String[] $args = new UTF8String[${evals.length}];")
+    }
     val codes = ctx.splitExpressionsWithCurrentInputs(
       expressions = inputs,
       funcName = "valueConcat",
-      extraArguments = ("UTF8String[]", args) :: Nil)
+      extraArguments = (s"${ctx.javaType(dataType)}[]", args) :: Nil)
     ev.copy(s"""
-      UTF8String[] $args = new UTF8String[${evals.length}];
+      $initCode
       $codes
-      UTF8String ${ev.value} = UTF8String.concat($args);
+      ${ctx.javaType(dataType)} ${ev.value} = $concatenator.concat($args);
       boolean ${ev.isNull} = ${ev.value} == null;
     """)
   }
+
+  override def toString: String = s"concat(${children.mkString(", ")})"
+
+  override def sql: String = s"concat(${children.map(_.sql).mkString(", ")})"
 }
 
 

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/expressions.scala

@@ -21,6 +21,7 @@ import scala.collection.immutable.HashSet
 import scala.collection.mutable.{ArrayBuffer, Stack}
 
 import org.apache.spark.sql.catalyst.analysis._
+import org.apache.spark.sql.catalyst.analysis.TypeCoercion.ImplicitTypeCasts
 import org.apache.spark.sql.catalyst.expressions._
 import org.apache.spark.sql.catalyst.expressions.Literal.{FalseLiteral, TrueLiteral}
 import org.apache.spark.sql.catalyst.expressions.aggregate._
@@ -645,15 +646,27 @@ object CombineConcats extends Rule[LogicalPlan] {
       stack.pop() match {
         case Concat(children) =>
           stack.pushAll(children.reverse)
+        // If `spark.sql.function.concatBinaryAsString` is false, nested `Concat` exprs possibly
+        // have `Concat`s with binary output. Since `TypeCoercion` casts them into strings,
+        // we need to handle the case to combine all nested `Concat`s.
+        case c @ Cast(Concat(children), StringType, _) =>
+          val newChildren = children.map { e => c.copy(child = e) }
+          stack.pushAll(newChildren.reverse)
         case child =>
           flattened += child
       }
     }
     Concat(flattened)
   }
 
+  private def hasNestedConcats(concat: Concat): Boolean = concat.children.exists {
+    case c: Concat => true
+    case c @ Cast(Concat(children), StringType, _) => true
+    case _ => false
+  }
+
   def apply(plan: LogicalPlan): LogicalPlan = plan.transformExpressionsDown {
-    case concat: Concat if concat.children.exists(_.isInstanceOf[Concat]) =>
+    case concat: Concat if hasNestedConcats(concat) =>
       flattenConcats(concat)
   }
 }

sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala

@@ -1044,6 +1044,12 @@ object SQLConf {
         "When this conf is not set, the value from `spark.redaction.string.regex` is used.")
       .fallbackConf(org.apache.spark.internal.config.STRING_REDACTION_PATTERN)
 
+  val CONCAT_BINARY_AS_STRING = buildConf("spark.sql.function.concatBinaryAsString")
+    .doc("When this option is set to false and all inputs are binary, `functions.concat` returns " +
+      "an output as binary. Otherwise, it returns as a string. ")
+    .booleanConf
+    .createWithDefault(false)
+
   val CONTINUOUS_STREAMING_EXECUTOR_QUEUE_SIZE =
     buildConf("spark.sql.streaming.continuous.executorQueueSize")
     .internal()
@@ -1378,6 +1384,8 @@ class SQLConf extends Serializable with Logging {
   def continuousStreamingExecutorPollIntervalMs: Long =
     getConf(CONTINUOUS_STREAMING_EXECUTOR_POLL_INTERVAL_MS)
 
+  def concatBinaryAsString: Boolean = getConf(CONCAT_BINARY_AS_STRING)
+
   /** ********************** SQLConf functionality methods ************ */
 
   /** Set Spark SQL configuration properties. */

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/analysis/TypeCoercionSuite.scala

@@ -869,6 +869,60 @@ class TypeCoercionSuite extends AnalysisTest {
         Literal.create(null, IntegerType), Literal.create(null, StringType))))
   }
 
+  test("type coercion for Concat") {
+    val rule = TypeCoercion.ConcatCoercion(conf)
+
+    ruleTest(rule,
+      Concat(Seq(Literal("ab"), Literal("cde"))),
+      Concat(Seq(Literal("ab"), Literal("cde"))))
+    ruleTest(rule,
+      Concat(Seq(Literal(null), Literal("abc"))),
+      Concat(Seq(Cast(Literal(null), StringType), Literal("abc"))))
+    ruleTest(rule,
+      Concat(Seq(Literal(1), Literal("234"))),
+      Concat(Seq(Cast(Literal(1), StringType), Literal("234"))))
+    ruleTest(rule,
+      Concat(Seq(Literal("1"), Literal("234".getBytes()))),
+      Concat(Seq(Literal("1"), Cast(Literal("234".getBytes()), StringType))))
+    ruleTest(rule,
+      Concat(Seq(Literal(1L), Literal(2.toByte), Literal(0.1))),
+      Concat(Seq(Cast(Literal(1L), StringType), Cast(Literal(2.toByte), StringType),
+        Cast(Literal(0.1), StringType))))
+    ruleTest(rule,
+      Concat(Seq(Literal(true), Literal(0.1f), Literal(3.toShort))),
+      Concat(Seq(Cast(Literal(true), StringType), Cast(Literal(0.1f), StringType),
+        Cast(Literal(3.toShort), StringType))))
+    ruleTest(rule,
+      Concat(Seq(Literal(1L), Literal(0.1))),
+      Concat(Seq(Cast(Literal(1L), StringType), Cast(Literal(0.1), StringType))))
+    ruleTest(rule,
+      Concat(Seq(Literal(Decimal(10)))),
+      Concat(Seq(Cast(Literal(Decimal(10)), StringType))))
+    ruleTest(rule,
+      Concat(Seq(Literal(BigDecimal.valueOf(10)))),
+      Concat(Seq(Cast(Literal(BigDecimal.valueOf(10)), StringType))))
+    ruleTest(rule,
+      Concat(Seq(Literal(java.math.BigDecimal.valueOf(10)))),
+      Concat(Seq(Cast(Literal(java.math.BigDecimal.valueOf(10)), StringType))))
+    ruleTest(rule,
+      Concat(Seq(Literal(new java.sql.Date(0)), Literal(new Timestamp(0)))),
+      Concat(Seq(Cast(Literal(new java.sql.Date(0)), StringType),
+        Cast(Literal(new Timestamp(0)), StringType))))
+
+    withSQLConf("spark.sql.function.concatBinaryAsString" -> "true") {
+      ruleTest(rule,
+        Concat(Seq(Literal("123".getBytes), Literal("456".getBytes))),
+        Concat(Seq(Cast(Literal("123".getBytes), StringType),
+          Cast(Literal("456".getBytes), StringType))))
+    }
+
+    withSQLConf("spark.sql.function.concatBinaryAsString" -> "false") {
+      ruleTest(rule,
+        Concat(Seq(Literal("123".getBytes), Literal("456".getBytes))),
+        Concat(Seq(Literal("123".getBytes), Literal("456".getBytes))))
+    }
+  }
+
   test("BooleanEquality type cast") {
     val be = TypeCoercion.BooleanEquality
     // Use something more than a literal to avoid triggering the simplification rules.