HIVE-29695: ClassCastException in MapJoin when reading Parquet Strings as Date, Timestamp, or TimestampLocalTZ

ql/src/java/org/apache/hadoop/hive/ql/exec/vector/VectorizedBatchUtil.java

@@ -129,7 +129,7 @@
     return createColumnVector(typeInfo, DataTypePhysicalVariation.NONE);
   }
 
   public static ColumnVector createColumnVector(TypeInfo typeInfo,
       DataTypePhysicalVariation dataTypePhysicalVariation) {
     switch(typeInfo.getCategory()) {
     case PRIMITIVE:
@@ -146,6 +146,7 @@
         case DATE:
           return new DateColumnVector(VectorizedRowBatch.DEFAULT_SIZE);
         case TIMESTAMP:
+        case TIMESTAMPLOCALTZ:
           return new TimestampColumnVector(VectorizedRowBatch.DEFAULT_SIZE);
         case INTERVAL_DAY_TIME:
           return new IntervalDayTimeColumnVector(VectorizedRowBatch.DEFAULT_SIZE);
@@ -294,7 +295,7 @@
     }
   }
 
   private static void setVector(Object row,
                                 StructObjectInspector oi,
                                 StructField field,
                                 VectorizedRowBatch batch,
@@ -643,7 +644,7 @@
     copyNonSelectedColumnVector(sourceBatch, sourceColumnNum, targetBatch, targetColumnNum, size, 0);
   }
 
   public static void copyNonSelectedColumnVector(
       VectorizedRowBatch sourceBatch, int sourceColumnNum,
       VectorizedRowBatch targetBatch, int targetColumnNum,
       int size, int startIndex) {

ql/src/java/org/apache/hadoop/hive/ql/io/parquet/ParquetTypeUtils.java

@@ -0,0 +1,55 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied.  See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+package org.apache.hadoop.hive.ql.io.parquet;
+
+import org.apache.hadoop.hive.common.type.Date;
+import org.apache.hadoop.hive.common.type.Timestamp;
+import org.apache.hadoop.hive.common.type.TimestampTZ;
+import org.apache.hadoop.hive.common.type.TimestampTZUtil;
+import java.nio.charset.StandardCharsets;
+import java.time.ZoneId;
+
+import java.util.function.Function;
+
+public class ParquetTypeUtils {
+
+  private static <T> T parseString(byte[] bytes, Function<String, T> parser) {
+    if (bytes == null || bytes.length < 8) {
+      return null;
+    }
+    try {
+      String s = new String(bytes, StandardCharsets.UTF_8);
+      return parser.apply(s);
+    } catch (Exception e) {
+      return null;
+    }
+  }
+
+  public static Date parseDate(byte[] bytes) {
+    return parseString(bytes, Date::valueOf);
+  }
+
+  public static Timestamp parseTimestamp(byte[] bytes) {
+    return parseString(bytes, Timestamp::valueOf);
+  }
+
+  public static TimestampTZ parseTimestampTZ(byte[] bytes, ZoneId defaultTimeZone) {
+    return parseString(bytes, s -> TimestampTZUtil.parse(s, defaultTimeZone));
+  }
+}

ql/src/java/org/apache/hadoop/hive/ql/io/parquet/convert/ETypeConverter.java

@@ -24,8 +24,10 @@
 import java.util.TimeZone;
 
 import com.google.common.base.MoreObjects;
+import org.apache.hadoop.hive.common.type.Date;
 import org.apache.hadoop.hive.common.type.HiveDecimal;
 import org.apache.hadoop.hive.common.type.Timestamp;
+import org.apache.hadoop.hive.common.type.TimestampTZ;
 import org.apache.hadoop.hive.conf.HiveConf;
 import org.apache.hadoop.hive.ql.io.parquet.read.DataWritableReadSupport;
 import org.apache.hadoop.hive.ql.io.parquet.timestamp.NanoTime;
@@ -40,6 +42,8 @@
 import org.apache.hadoop.hive.serde2.io.HiveDecimalWritable;
 import org.apache.hadoop.hive.serde2.io.HiveVarcharWritable;
 import org.apache.hadoop.hive.serde2.io.TimestampWritableV2;
+import org.apache.hadoop.hive.serde2.io.TimestampLocalTZWritable;
+import org.apache.hadoop.hive.ql.io.parquet.ParquetTypeUtils;
 import org.apache.hadoop.hive.serde2.typeinfo.CharTypeInfo;
 import org.apache.hadoop.hive.serde2.typeinfo.DecimalTypeInfo;
 import org.apache.hadoop.hive.serde2.typeinfo.HiveDecimalUtils;
@@ -508,8 +512,7 @@
       // of code paths that do not provide the typeInfo in those cases we default to Text. This idiom is also
       // followed by for example the BigDecimal converter in which if there is no type information,
       // it defaults to the widest representation
-      if (hiveTypeInfo instanceof PrimitiveTypeInfo) {
-        PrimitiveTypeInfo t = (PrimitiveTypeInfo) hiveTypeInfo;
+      if (hiveTypeInfo instanceof PrimitiveTypeInfo t) {
         switch (t.getPrimitiveCategory()) {
           case CHAR:
             return new BinaryConverter<HiveCharWritable>(type, parent, index) {
@@ -525,9 +528,33 @@
                 return new HiveVarcharWritable(binary.getBytes(), ((VarcharTypeInfo) hiveTypeInfo).getLength());
               }
             };
+          case DATE:
+            return new BinaryConverter<DateWritableV2>(type, parent, index) {
+              @Override
+              protected DateWritableV2 convert(Binary binary) {
+                Date date = ParquetTypeUtils.parseDate(binary.getBytes());
+                return date != null ? new DateWritableV2(date) : null;
+              }
+            };
+          case TIMESTAMP:
+            return new BinaryConverter<TimestampWritableV2>(type, parent, index) {
+              @Override
+              protected TimestampWritableV2 convert(Binary binary) {
+                Timestamp ts = ParquetTypeUtils.parseTimestamp(binary.getBytes());
+                return ts != null ? new TimestampWritableV2(ts) : null;
+              }
+            };
+          case TIMESTAMPLOCALTZ:
+            return new BinaryConverter<TimestampLocalTZWritable>(type, parent, index) {
+              @Override
+              protected TimestampLocalTZWritable convert(Binary binary) {
+                TimestampTZ tstz = ParquetTypeUtils.parseTimestampTZ(binary.getBytes(), ZoneId.systemDefault());
+                return tstz != null ? new TimestampLocalTZWritable(tstz) : null;
+              }
+            };
         }
       }
-      // STRING type
+      // Default to STRING type (Text)
       return new BinaryConverter<Text>(type, parent, index) {
         @Override
         protected Text convert(Binary binary) {

ql/src/java/org/apache/hadoop/hive/ql/io/parquet/serde/ArrayWritableObjectInspector.java

@@ -29,6 +29,7 @@
 import org.apache.hadoop.hive.serde2.typeinfo.ListTypeInfo;
 import org.apache.hadoop.hive.serde2.typeinfo.MapTypeInfo;
 import org.apache.hadoop.hive.serde2.typeinfo.StructTypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TimestampLocalTZTypeInfo;
 import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
 import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
 import org.apache.hadoop.hive.serde2.typeinfo.VarcharTypeInfo;
@@ -137,6 +138,8 @@
       return PrimitiveObjectInspectorFactory.getPrimitiveWritableObjectInspector((CharTypeInfo) typeInfo);
     } else if (typeInfo.getTypeName().toLowerCase().startsWith(serdeConstants.VARCHAR_TYPE_NAME)) {
       return PrimitiveObjectInspectorFactory.getPrimitiveWritableObjectInspector((VarcharTypeInfo) typeInfo);
+    } else if (typeInfo instanceof TimestampLocalTZTypeInfo) {
+      return PrimitiveObjectInspectorFactory.getPrimitiveWritableObjectInspector((TimestampLocalTZTypeInfo) typeInfo);
     } else {
       throw new UnsupportedOperationException("Unknown field type: " + typeInfo);
     }