[SPARK-10151] [SQL] Support invocation of hive macro

sql/hive/compatibility/src/test/scala/org/apache/spark/sql/hive/execution/HiveCompatibilitySuite.scala

@@ -684,6 +684,7 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
     "load_file_with_space_in_the_name",
     "loadpart1",
     "louter_join_ppr",
+    "macro",
     "mapjoin_distinct",
     "mapjoin_filter_on_outerjoin",
     "mapjoin_mapjoin",

sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveContext.scala

@@ -21,6 +21,7 @@ import java.io.File
 import java.net.{URL, URLClassLoader}
 import java.sql.Timestamp
 import java.util.concurrent.TimeUnit
+import java.util.regex.Pattern
 
 import scala.collection.JavaConverters._
 import scala.collection.mutable.HashMap
@@ -592,10 +593,14 @@ class HiveContext private[hive](
     )
   }
 
+  private def functionOrMacroDDLPattern(command: String) = Pattern.compile(
+    ".*(create|drop)\\s+(temporary\\s+)?(function|macro).+", Pattern.DOTALL).matcher(command)
+
   protected[hive] def runSqlHive(sql: String): Seq[String] = {
-    if (sql.toLowerCase.contains("create temporary function")) {
+    val command = sql.trim.toLowerCase
+    if (functionOrMacroDDLPattern(command).matches()) {
       executionHive.runSqlHive(sql)
-    } else if (sql.trim.toLowerCase.startsWith("set")) {
+    } else if (command.startsWith("set")) {
       metadataHive.runSqlHive(sql)
       executionHive.runSqlHive(sql)
     } else {

sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveInspectors.scala

@@ -723,6 +723,10 @@ private[hive] trait HiveInspectors {
         inspectorToDataType(m.getMapValueObjectInspector))
     case _: WritableStringObjectInspector => StringType
     case _: JavaStringObjectInspector => StringType
+    case _: WritableHiveVarcharObjectInspector => StringType
+    case _: JavaHiveVarcharObjectInspector => StringType
+    case _: WritableHiveCharObjectInspector => StringType
+    case _: JavaHiveCharObjectInspector => StringType
     case _: WritableIntObjectInspector => IntegerType
     case _: JavaIntObjectInspector => IntegerType
     case _: WritableDoubleObjectInspector => DoubleType

sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveQl.scala

@@ -117,6 +117,7 @@ private[hive] object HiveQl extends Logging {
     "TOK_CREATEDATABASE",
     "TOK_CREATEFUNCTION",
     "TOK_CREATEINDEX",
+    "TOK_CREATEMACRO",
     "TOK_CREATEROLE",
 
     "TOK_DESCDATABASE",
@@ -125,6 +126,7 @@ private[hive] object HiveQl extends Logging {
     "TOK_DROPDATABASE",
     "TOK_DROPFUNCTION",
     "TOK_DROPINDEX",
+    "TOK_DROPMACRO",
     "TOK_DROPROLE",
     "TOK_DROPTABLE_PROPERTIES",
     "TOK_DROPVIEW",

sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveShim.scala

@@ -117,9 +117,10 @@ private[hive] object HiveShim {
    * Detail discussion can be found at https://github.com/apache/spark/pull/3640
    *
    * @param functionClassName UDF class name
+   * @param instance optional UDF instance which contains additional information (for macro)
    */
-  private[hive] case class HiveFunctionWrapper(var functionClassName: String)
-    extends java.io.Externalizable {
+  private[hive] case class HiveFunctionWrapper(var functionClassName: String,
+    private var instance: AnyRef = null) extends java.io.Externalizable {
 
     // for Serialization
     def this() = this(null)
@@ -154,8 +155,6 @@ private[hive] object HiveShim {
       serializeObjectByKryo(Utilities.runtimeSerializationKryo.get(), function, out)
     }
 
-    private var instance: AnyRef = null
-
     def writeExternal(out: java.io.ObjectOutput) {
       // output the function name
       out.writeUTF(functionClassName)
@@ -184,7 +183,7 @@ private[hive] object HiveShim {
         // read the function in bytes
         val functionInBytesLength = in.readInt()
         val functionInBytes = new Array[Byte](functionInBytesLength)
-        in.read(functionInBytes, 0, functionInBytesLength)
+        in.readFully(functionInBytes)
 
         // deserialize the function object via Hive Utilities
         instance = deserializePlan[AnyRef](new java.io.ByteArrayInputStream(functionInBytes),

sql/hive/src/main/scala/org/apache/spark/sql/hive/hiveUDFs.scala

@@ -64,7 +64,10 @@ private[hive] class HiveFunctionRegistry(underlying: analysis.FunctionRegistry)
       // don't satisfy the hive UDF, such as type mismatch, input number mismatch, etc. Here we
       // catch the exception and throw AnalysisException instead.
       try {
-        if (classOf[UDF].isAssignableFrom(functionInfo.getFunctionClass)) {
+        if (classOf[GenericUDFMacro].isAssignableFrom(functionInfo.getFunctionClass)) {
+          HiveGenericUDF(
+            new HiveFunctionWrapper(functionClassName, functionInfo.getGenericUDF), children)
+        } else if (classOf[UDF].isAssignableFrom(functionInfo.getFunctionClass)) {
           HiveSimpleUDF(new HiveFunctionWrapper(functionClassName), children)
         } else if (classOf[GenericUDF].isAssignableFrom(functionInfo.getFunctionClass)) {
           HiveGenericUDF(new HiveFunctionWrapper(functionClassName), children)

sql/hive/src/test/resources/golden/macro-0-50131c0ba7b7a6b65c789a5a8497bada

@@ -0,0 +1 @@
+0

sql/hive/src/test/resources/golden/macro-11-f55b8684c77f6eefc2618ba79e5e0587

@@ -0,0 +1 @@
+0

sql/hive/src/test/resources/golden/macro-13-87e53d2b4c84098e662779e8f0a59084

@@ -0,0 +1 @@
+1

sql/hive/src/test/resources/golden/macro-17-b3864f1d19fdb88b3b74f6d74a0ba548

@@ -0,0 +1 @@
+10

sql/hive/src/test/resources/golden/macro-2-fde44c7854a9897acb4c2f78f24c8eec

@@ -0,0 +1 @@
+0.8807970779778823

sql/hive/src/test/resources/golden/macro-7-decde0a59183a393e580941c633d3c5c

@@ -0,0 +1 @@
+2