You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by GitBox <gi...@apache.org> on 2021/09/02 19:37:40 UTC
[GitHub] [spark] sadikovi commented on a change in pull request #33695: [SPARK-34863][SQL] Support complex types for Parquet vectorized reader
sadikovi commented on a change in pull request #33695:
URL: https://github.com/apache/spark/pull/33695#discussion_r701352457
##########
File path: sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala
##########
@@ -886,6 +886,13 @@ object SQLConf {
.booleanConf
.createWithDefault(true)
+ val PARQUET_VECTORIZED_READER_NESTED_COLUMN_ENABLED =
+ buildConf("spark.sql.parquet.enableNestedColumnVectorizedReader")
+ .doc("Enables vectorized Parquet decoding for nested columns (e.g., struct, list, map).")
Review comment:
Should we mention that this is only enabled when vectorised reader is enabled?
##########
File path: sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/ParquetColumn.java
##########
@@ -0,0 +1,316 @@
+/*
+ * 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.spark.sql.execution.datasources.parquet;
+
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Set;
+
+import com.google.common.base.Preconditions;
+import org.apache.spark.memory.MemoryMode;
+import org.apache.spark.sql.execution.vectorized.OffHeapColumnVector;
+import org.apache.spark.sql.execution.vectorized.OnHeapColumnVector;
+import org.apache.spark.sql.execution.vectorized.WritableColumnVector;
+import org.apache.spark.sql.types.ArrayType;
+import org.apache.spark.sql.types.DataType;
+import org.apache.spark.sql.types.DataTypes;
+import org.apache.spark.sql.types.MapType;
+import org.apache.spark.sql.types.StructType;
+
+/**
+ * Contains necessary information representing a Parquet column, either of primitive or nested type.
+ */
+final class ParquetColumn {
+ private final ParquetType type;
+ private final List<ParquetColumn> children;
+ private final WritableColumnVector vector;
+
+ /**
+ * repetition & definition levels
+ * these are allocated only for leaf columns; for non-leaf columns, they simply maintain
+ * references to that of the former.
+ */
+ private WritableColumnVector repetitionLevels;
+ private WritableColumnVector definitionLevels;
+
+ /** whether this column is primitive (i.e., leaf column) */
+ private final boolean isPrimitive;
+
+ /** reader for this column - only set if 'isPrimitive' is true */
+ private VectorizedColumnReader columnReader;
+
+ ParquetColumn(
+ ParquetType type,
+ WritableColumnVector vector,
+ int capacity,
+ MemoryMode memoryMode,
+ Set<ParquetType> missingColumns) {
+ DataType sparkType = type.sparkType();
+ if (!sparkType.sameType(vector.dataType())) {
+ throw new IllegalArgumentException("Spark type: " + type.sparkType() +
+ " doesn't match the type: " + vector.dataType() + " in column vector");
+ }
+
+ this.type = type;
+ this.vector = vector;
+ this.children = new ArrayList<>();
+ this.isPrimitive = type.isPrimitive();
Review comment:
Interesting, `isPrimitive == children.isEmpty()` would this hold?
##########
File path: sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/ParquetColumn.java
##########
@@ -0,0 +1,316 @@
+/*
+ * 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.spark.sql.execution.datasources.parquet;
+
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Set;
+
+import com.google.common.base.Preconditions;
+import org.apache.spark.memory.MemoryMode;
+import org.apache.spark.sql.execution.vectorized.OffHeapColumnVector;
+import org.apache.spark.sql.execution.vectorized.OnHeapColumnVector;
+import org.apache.spark.sql.execution.vectorized.WritableColumnVector;
+import org.apache.spark.sql.types.ArrayType;
+import org.apache.spark.sql.types.DataType;
+import org.apache.spark.sql.types.DataTypes;
+import org.apache.spark.sql.types.MapType;
+import org.apache.spark.sql.types.StructType;
+
+/**
+ * Contains necessary information representing a Parquet column, either of primitive or nested type.
+ */
+final class ParquetColumn {
+ private final ParquetType type;
+ private final List<ParquetColumn> children;
+ private final WritableColumnVector vector;
+
+ /**
+ * repetition & definition levels
Review comment:
nit: Would it be possible to capitalise all of the javadoc comments in this file? I think it would look neater 😄 . Thanks!
##########
File path: sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/orc/OrcV1SchemaPruningSuite.scala
##########
@@ -25,9 +25,12 @@ class OrcV1SchemaPruningSuite extends SchemaPruningSuite {
override protected val dataSourceName: String = "orc"
override protected val vectorizedReaderEnabledKey: String =
SQLConf.ORC_VECTORIZED_READER_ENABLED.key
+ override protected val vectorizedReaderNestedEnabledKey: String =
+ SQLConf.ORC_VECTORIZED_READER_NESTED_COLUMN_ENABLED.key
override protected def sparkConf: SparkConf =
super
.sparkConf
.set(SQLConf.USE_V1_SOURCE_LIST, "orc")
+
Review comment:
nit: Let's remove new line.
##########
File path: sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/ParquetColumn.java
##########
@@ -0,0 +1,316 @@
+/*
+ * 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.spark.sql.execution.datasources.parquet;
+
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Set;
+
+import com.google.common.base.Preconditions;
+import org.apache.spark.memory.MemoryMode;
+import org.apache.spark.sql.execution.vectorized.OffHeapColumnVector;
+import org.apache.spark.sql.execution.vectorized.OnHeapColumnVector;
+import org.apache.spark.sql.execution.vectorized.WritableColumnVector;
+import org.apache.spark.sql.types.ArrayType;
+import org.apache.spark.sql.types.DataType;
+import org.apache.spark.sql.types.DataTypes;
+import org.apache.spark.sql.types.MapType;
+import org.apache.spark.sql.types.StructType;
+
+/**
+ * Contains necessary information representing a Parquet column, either of primitive or nested type.
+ */
+final class ParquetColumn {
+ private final ParquetType type;
+ private final List<ParquetColumn> children;
+ private final WritableColumnVector vector;
+
+ /**
+ * repetition & definition levels
+ * these are allocated only for leaf columns; for non-leaf columns, they simply maintain
+ * references to that of the former.
+ */
+ private WritableColumnVector repetitionLevels;
+ private WritableColumnVector definitionLevels;
+
+ /** whether this column is primitive (i.e., leaf column) */
+ private final boolean isPrimitive;
+
+ /** reader for this column - only set if 'isPrimitive' is true */
+ private VectorizedColumnReader columnReader;
+
+ ParquetColumn(
+ ParquetType type,
+ WritableColumnVector vector,
+ int capacity,
+ MemoryMode memoryMode,
+ Set<ParquetType> missingColumns) {
Review comment:
Is `missingColumns` a list of columns excluded from projection?
##########
File path: sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/ParquetColumn.java
##########
@@ -0,0 +1,316 @@
+/*
+ * 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.spark.sql.execution.datasources.parquet;
+
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Set;
+
+import com.google.common.base.Preconditions;
+import org.apache.spark.memory.MemoryMode;
+import org.apache.spark.sql.execution.vectorized.OffHeapColumnVector;
+import org.apache.spark.sql.execution.vectorized.OnHeapColumnVector;
+import org.apache.spark.sql.execution.vectorized.WritableColumnVector;
+import org.apache.spark.sql.types.ArrayType;
+import org.apache.spark.sql.types.DataType;
+import org.apache.spark.sql.types.DataTypes;
+import org.apache.spark.sql.types.MapType;
+import org.apache.spark.sql.types.StructType;
+
+/**
+ * Contains necessary information representing a Parquet column, either of primitive or nested type.
+ */
+final class ParquetColumn {
+ private final ParquetType type;
+ private final List<ParquetColumn> children;
+ private final WritableColumnVector vector;
+
+ /**
+ * repetition & definition levels
+ * these are allocated only for leaf columns; for non-leaf columns, they simply maintain
+ * references to that of the former.
+ */
+ private WritableColumnVector repetitionLevels;
+ private WritableColumnVector definitionLevels;
+
+ /** whether this column is primitive (i.e., leaf column) */
+ private final boolean isPrimitive;
+
+ /** reader for this column - only set if 'isPrimitive' is true */
+ private VectorizedColumnReader columnReader;
+
+ ParquetColumn(
+ ParquetType type,
+ WritableColumnVector vector,
+ int capacity,
+ MemoryMode memoryMode,
+ Set<ParquetType> missingColumns) {
+ DataType sparkType = type.sparkType();
+ if (!sparkType.sameType(vector.dataType())) {
+ throw new IllegalArgumentException("Spark type: " + type.sparkType() +
+ " doesn't match the type: " + vector.dataType() + " in column vector");
+ }
+
+ this.type = type;
+ this.vector = vector;
+ this.children = new ArrayList<>();
+ this.isPrimitive = type.isPrimitive();
+
+ if (missingColumns.contains(type)) {
+ vector.setAllNull();
+ return;
+ }
+
+ if (isPrimitive) {
+ repetitionLevels = allocateLevelsVector(capacity, memoryMode);
Review comment:
Does the code handle situations when there are no repetition or definition levels for the type? Would it allocate vectors even when the column is null?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org