[
https://issues.apache.org/jira/browse/ARROW-1780?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16431015#comment-16431015
]
ASF GitHub Bot commented on ARROW-1780:
---------------------------------------
atuldambalkar commented on a change in pull request #1759: ARROW-1780 - [WIP]
JDBC Adapter to convert Relational Data objects to Arrow Data Format Vector
Objects
URL: https://github.com/apache/arrow/pull/1759#discussion_r180185358
##########
File path:
java/adapter/jdbc/src/main/java/org/apache/arrow/adapter/jdbc/JdbcToArrowUtils.java
##########
@@ -0,0 +1,343 @@
+/**
+ * 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.arrow.adapter.jdbc;
+
+import org.apache.arrow.vector.*;
+import org.apache.arrow.vector.types.DateUnit;
+import org.apache.arrow.vector.types.TimeUnit;
+import org.apache.arrow.vector.types.pojo.ArrowType;
+import org.apache.arrow.vector.types.pojo.Field;
+import org.apache.arrow.vector.types.pojo.FieldType;
+import org.apache.arrow.vector.types.pojo.Schema;
+
+import java.nio.charset.Charset;
+import java.sql.*;
+import java.util.ArrayList;
+import java.util.List;
+
+import static org.apache.arrow.vector.types.FloatingPointPrecision.DOUBLE;
+import static org.apache.arrow.vector.types.FloatingPointPrecision.SINGLE;
+
+
+/**
+ * Class that does most of the work to convert JDBC ResultSet data into Arrow
columnar format Vector objects.
+ *
+ * @since 0.10.0
+ */
+public class JdbcToArrowUtils {
+
+ private static final int DEFAULT_BUFFER_SIZE = 256;
+
+ /**
+ * Create Arrow {@link Schema} object for the given JDBC {@link
ResultSetMetaData}.
+ *
+ * This method currently performs following type mapping for JDBC SQL data
types to corresponding Arrow data types.
+ *
+ * CHAR --> ArrowType.Utf8
+ * NCHAR --> ArrowType.Utf8
+ * VARCHAR --> ArrowType.Utf8
+ * NVARCHAR --> ArrowType.Utf8
+ * LONGVARCHAR --> ArrowType.Utf8
+ * LONGNVARCHAR --> ArrowType.Utf8
+ * NUMERIC --> ArrowType.Decimal(precision, scale)
+ * DECIMAL --> ArrowType.Decimal(precision, scale)
+ * BIT --> ArrowType.Bool
+ * TINYINT --> ArrowType.Int(8, signed)
+ * SMALLINT --> ArrowType.Int(16, signed)
+ * INTEGER --> ArrowType.Int(32, signed)
+ * BIGINT --> ArrowType.Int(64, signed)
+ * REAL --> ArrowType.FloatingPoint(FloatingPointPrecision.SINGLE)
+ * FLOAT --> ArrowType.FloatingPoint(FloatingPointPrecision.SINGLE)
+ * DOUBLE --> ArrowType.FloatingPoint(FloatingPointPrecision.DOUBLE)
+ * BINARY --> ArrowType.Binary
+ * VARBINARY --> ArrowType.Binary
+ * LONGVARBINARY --> ArrowType.Binary
+ * DATE --> ArrowType.Date(DateUnit.MILLISECOND)
+ * TIME --> ArrowType.Time(TimeUnit.MILLISECOND, 32)
+ * TIMESTAMP --> ArrowType.Timestamp(TimeUnit.MILLISECOND, timezone=null)
+ * CLOB --> ArrowType.Utf8
+ * BLOB --> ArrowType.Binary
+ *
+ * @param rsmd
+ * @return {@link Schema}
+ * @throws SQLException
+ */
+ public static Schema jdbcToArrowSchema(ResultSetMetaData rsmd) throws
SQLException {
+
+ assert rsmd != null;
+
+// ImmutableList.Builder<Field> fields = ImmutableList.builder();
+ List<Field> fields = new ArrayList<>();
+ int columnCount = rsmd.getColumnCount();
+ for (int i = 1; i <= columnCount; i++) {
+ String columnName = rsmd.getColumnName(i);
+ switch (rsmd.getColumnType(i)) {
+ case Types.BOOLEAN:
+ case Types.BIT:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Bool()), null));
+ break;
+ case Types.TINYINT:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Int(8, true)), null));
+ break;
+ case Types.SMALLINT:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Int(16, true)), null));
+ break;
+ case Types.INTEGER:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Int(32, true)), null));
+ break;
+ case Types.BIGINT:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Int(64, true)), null));
+ break;
+ case Types.NUMERIC:
+ case Types.DECIMAL:
+ int precision = rsmd.getPrecision(i);
+ int scale = rsmd.getScale(i);
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Decimal(precision, scale)), null));
+ break;
+ case Types.REAL:
+ case Types.FLOAT:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.FloatingPoint(SINGLE)), null));
+ break;
+ case Types.DOUBLE:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.FloatingPoint(DOUBLE)), null));
+ break;
+ case Types.CHAR:
+ case Types.NCHAR:
+ case Types.VARCHAR:
+ case Types.NVARCHAR:
+ case Types.LONGVARCHAR:
+ case Types.LONGNVARCHAR:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Utf8()), null));
+ break;
+ case Types.DATE:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Date(DateUnit.MILLISECOND)), null));
+ break;
+ case Types.TIME:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Time(TimeUnit.MILLISECOND, 32)), null));
+ break;
+ case Types.TIMESTAMP:
+ // timezone is null
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Timestamp(TimeUnit.MILLISECOND, null)), null));
+ break;
+ case Types.BINARY:
+ case Types.VARBINARY:
+ case Types.LONGVARBINARY:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Binary()), null));
+ break;
+ case Types.ARRAY:
+ // not handled
+// fields.add(new Field("list", FieldType.nullable(new
ArrowType.List()), null));
+ break;
+ case Types.CLOB:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Utf8()), null));
+ break;
+ case Types.BLOB:
+ fields.add(new Field(columnName, FieldType.nullable(new
ArrowType.Binary()), null));
+ break;
+
+ default:
+ // no-op
+ break;
+ }
+ }
+
+ return new Schema(fields, null);
+ }
+
+ private static void allocateVectors(VectorSchemaRoot root, int size) {
+ List<FieldVector> vectors = root.getFieldVectors();
+ for (FieldVector fieldVector: vectors) {
+ if (fieldVector instanceof BaseFixedWidthVector) {
+ ((BaseFixedWidthVector) fieldVector).allocateNew(size);
+ } else {
+ fieldVector.allocateNew();
+ }
+ fieldVector.setInitialCapacity(size);
+ }
+ }
+
+ /**
+ * Iterate the given JDBC {@link ResultSet} object to fetch the data and
transpose it to populate
+ * the given Arrow Vector objects.
+ *
+ * @param rs ResultSet to use to fetch the data from underlying database
+ * @param root Arrow {@link VectorSchemaRoot} object to populate
+ * @throws Exception
+ */
+ public static void jdbcToArrowVectors(ResultSet rs, VectorSchemaRoot root)
throws Exception {
+
+ assert rs != null;
+ assert root != null;
+
+ ResultSetMetaData rsmd = rs.getMetaData();
+ int columnCount = rsmd.getColumnCount();
+
+ allocateVectors(root, DEFAULT_BUFFER_SIZE);
+
+ int rowCount = 0;
+ while (rs.next()) {
+ // for each column get the value based on the type
+
+ // need to change this to build Java lists and then build Arrow
vectors
+ for (int i = 1; i <= columnCount; i++) {
+ String columnName = rsmd.getColumnName(i);
+ switch (rsmd.getColumnType(i)) {
+ case Types.BOOLEAN:
+ case Types.BIT:
+ BitVector bitVector = (BitVector)
root.getVector(columnName);
+ bitVector.setSafe(rowCount, rs.getBoolean(i)? 1: 0);
+ bitVector.setValueCount(rowCount + 1);
+ break;
+ case Types.TINYINT:
+ TinyIntVector tinyIntVector =
(TinyIntVector)root.getVector(columnName);
+ tinyIntVector.setSafe(rowCount, rs.getInt(i));
+ tinyIntVector.setValueCount(rowCount + 1);
+ break;
+ case Types.SMALLINT:
+ SmallIntVector smallIntVector =
(SmallIntVector)root.getVector(columnName);
+ smallIntVector.setSafe(rowCount, rs.getInt(i));
+ smallIntVector.setValueCount(rowCount + 1);
+ break;
+ case Types.INTEGER:
+ IntVector intVector =
(IntVector)root.getVector(columnName);
+ intVector.setSafe(rowCount, rs.getInt(i));
+ intVector.setValueCount(rowCount + 1);
+ break;
+ case Types.BIGINT:
+ BigIntVector bigIntVector =
(BigIntVector)root.getVector(columnName);
+ bigIntVector.setSafe(rowCount, rs.getInt(i));
+ bigIntVector.setValueCount(rowCount + 1);
+ break;
+ case Types.NUMERIC:
+ case Types.DECIMAL:
+ DecimalVector decimalVector =
(DecimalVector)root.getVector(columnName);
+ decimalVector.setSafe(rowCount, rs.getBigDecimal(i));
+ decimalVector.setValueCount(rowCount + 1);
+ break;
+ case Types.REAL:
+ case Types.FLOAT:
+ Float4Vector float4Vector =
(Float4Vector)root.getVector(columnName);
+ float4Vector.setSafe(rowCount, rs.getFloat(i));
+ float4Vector.setValueCount(rowCount + 1);
+ break;
+ case Types.DOUBLE:
+ Float8Vector float8Vector =
(Float8Vector)root.getVector(columnName);
+ float8Vector.setSafe(rowCount, rs.getDouble(i));
+ float8Vector.setValueCount(rowCount + 1);
+ break;
+ case Types.CHAR:
+ case Types.NCHAR:
+ case Types.VARCHAR:
+ case Types.NVARCHAR:
+ case Types.LONGVARCHAR:
+ case Types.LONGNVARCHAR:
+ VarCharVector varcharVector =
(VarCharVector)root.getVector(columnName);
+ String value = rs.getString(i) != null ?
rs.getString(i) : "";
+ varcharVector.setIndexDefined(rowCount);
+ varcharVector.setValueLengthSafe(rowCount,
value.length());
+ varcharVector.setSafe(rowCount,
value.getBytes(Charset.forName("UTF-8")), 0, value.length());
+ varcharVector.setValueCount(rowCount + 1);
+ break;
+ case Types.DATE:
+ Date date = rs.getDate(i);
+ DateMilliVector dateMilliVector = (DateMilliVector)
root.getVector(columnName);
+ dateMilliVector.setValueCount(rowCount + 1);
+ if (date != null) {
+ dateMilliVector.setSafe(rowCount,
rs.getDate(i).getTime());
Review comment:
Hi @laurentgo Would you have some time to discuss the TimeZone issue. I gave
some more thought on this and I am not entirely sure how to really handle
issues surrounding this. As we have no control over the TimeZone setting for
the running JVM and also the underlying Database. Is there a way I can reach
you over the phone? I am based out of Cupertino, Bay Area, CA - 650 797 8289.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> JDBC Adapter for Apache Arrow
> -----------------------------
>
> Key: ARROW-1780
> URL: https://issues.apache.org/jira/browse/ARROW-1780
> Project: Apache Arrow
> Issue Type: New Feature
> Reporter: Atul Dambalkar
> Assignee: Atul Dambalkar
> Priority: Major
> Labels: pull-request-available
> Fix For: 0.10.0
>
>
> At a high level the JDBC Adapter will allow upstream apps to query RDBMS data
> over JDBC and get the JDBC objects converted to Arrow objects/structures. The
> upstream utility can then work with Arrow objects/structures with usual
> performance benefits. The utility will be very much similar to C++
> implementation of "Convert a vector of row-wise data into an Arrow table" as
> described here -
> https://arrow.apache.org/docs/cpp/md_tutorials_row_wise_conversion.html
> The utility will read data from RDBMS and covert the data into Arrow
> objects/structures. So from that perspective this will Read data from RDBMS,
> If the utility can push Arrow objects to RDBMS is something need to be
> discussed and will be out of scope for this utility for now.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
