TheNeuralBit commented on a change in pull request #12498: URL: https://github.com/apache/beam/pull/12498#discussion_r468722620
########## File path: sdks/java/io/jdbc/src/main/java/org/apache/beam/sdk/io/jdbc/JdbcSchemaIOProvider.java ########## @@ -0,0 +1,172 @@ +/* + * 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.beam.sdk.io.jdbc; + +import com.google.auto.service.AutoService; +import java.io.Serializable; +import java.util.List; +import java.util.stream.Collectors; +import java.util.stream.StreamSupport; +import javax.annotation.Nullable; +import org.apache.beam.sdk.annotations.Internal; +import org.apache.beam.sdk.schemas.Schema; +import org.apache.beam.sdk.schemas.Schema.FieldType; +import org.apache.beam.sdk.schemas.io.SchemaIO; +import org.apache.beam.sdk.schemas.io.SchemaIOProvider; +import org.apache.beam.sdk.transforms.PTransform; +import org.apache.beam.sdk.values.PBegin; +import org.apache.beam.sdk.values.PCollection; +import org.apache.beam.sdk.values.PDone; +import org.apache.beam.sdk.values.Row; + +/** + * An implementation of {@link SchemaIOProvider} for reading and writing JSON payloads with {@link + * JdbcIO}. + */ +@Internal +@AutoService(SchemaIOProvider.class) +public class JdbcSchemaIOProvider implements SchemaIOProvider { + + /** Returns an id that uniquely represents this IO. */ + @Override + public String identifier() { + return "jdbc"; + } + + /** + * Returns the expected schema of the configuration object. Note this is distinct from the schema + * of the data source itself. + */ + @Override + public Schema configurationSchema() { + return Schema.builder() + .addStringField("driverClassName") + .addStringField("jdbcUrl") + .addStringField("username") + .addStringField("password") + .addNullableField("connectionProperties", FieldType.STRING) + .addNullableField("connectionInitSqls", FieldType.iterable(FieldType.STRING)) + .addNullableField("readQuery", FieldType.STRING) + .addNullableField("writeStatement", FieldType.STRING) + .addNullableField("fetchSize", FieldType.INT16) + .addNullableField("outputParallelization", FieldType.BOOLEAN) + .build(); + } + + /** + * Produce a SchemaIO given a String representing the data's location, the schema of the data that + * resides there, and some IO-specific configuration object. + */ + @Override + public JdbcSchemaIO from(String location, Row configuration, Schema dataSchema) { + return new JdbcSchemaIO(location, configuration); + } + + @Override + public boolean requiresDataSchema() { + return false; + } + + @Override + public PCollection.IsBounded isBounded() { + return PCollection.IsBounded.BOUNDED; + } + + /** An abstraction to create schema aware IOs. */ + static class JdbcSchemaIO implements SchemaIO, Serializable { + protected final Row config; + protected final String location; + + JdbcSchemaIO(String location, Row config) { + this.config = config; + this.location = location; + } + + @Override + public Schema schema() { + return null; + } + + @Override + public PTransform<PBegin, PCollection<Row>> buildReader() { + String readQuery; + if (config.getString("readQuery") != null) { + readQuery = config.getString("readQuery"); + } else { + readQuery = String.format("SELECT f_int FROM %s", location); Review comment: Similarly this shouldn't mention specific field names. In this case I think we could just do `SELECT * FROM %s` ########## File path: sdks/java/io/jdbc/src/main/java/org/apache/beam/sdk/io/jdbc/JdbcSchemaIOProvider.java ########## @@ -0,0 +1,172 @@ +/* + * 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.beam.sdk.io.jdbc; + +import com.google.auto.service.AutoService; +import java.io.Serializable; +import java.util.List; +import java.util.stream.Collectors; +import java.util.stream.StreamSupport; +import javax.annotation.Nullable; +import org.apache.beam.sdk.annotations.Internal; +import org.apache.beam.sdk.schemas.Schema; +import org.apache.beam.sdk.schemas.Schema.FieldType; +import org.apache.beam.sdk.schemas.io.SchemaIO; +import org.apache.beam.sdk.schemas.io.SchemaIOProvider; +import org.apache.beam.sdk.transforms.PTransform; +import org.apache.beam.sdk.values.PBegin; +import org.apache.beam.sdk.values.PCollection; +import org.apache.beam.sdk.values.PDone; +import org.apache.beam.sdk.values.Row; + +/** + * An implementation of {@link SchemaIOProvider} for reading and writing JSON payloads with {@link + * JdbcIO}. + */ +@Internal +@AutoService(SchemaIOProvider.class) +public class JdbcSchemaIOProvider implements SchemaIOProvider { + + /** Returns an id that uniquely represents this IO. */ + @Override + public String identifier() { + return "jdbc"; + } + + /** + * Returns the expected schema of the configuration object. Note this is distinct from the schema + * of the data source itself. + */ + @Override + public Schema configurationSchema() { + return Schema.builder() + .addStringField("driverClassName") + .addStringField("jdbcUrl") + .addStringField("username") + .addStringField("password") + .addNullableField("connectionProperties", FieldType.STRING) + .addNullableField("connectionInitSqls", FieldType.iterable(FieldType.STRING)) + .addNullableField("readQuery", FieldType.STRING) + .addNullableField("writeStatement", FieldType.STRING) + .addNullableField("fetchSize", FieldType.INT16) + .addNullableField("outputParallelization", FieldType.BOOLEAN) + .build(); + } + + /** + * Produce a SchemaIO given a String representing the data's location, the schema of the data that + * resides there, and some IO-specific configuration object. + */ + @Override + public JdbcSchemaIO from(String location, Row configuration, Schema dataSchema) { + return new JdbcSchemaIO(location, configuration); + } + + @Override + public boolean requiresDataSchema() { + return false; + } + + @Override + public PCollection.IsBounded isBounded() { + return PCollection.IsBounded.BOUNDED; + } + + /** An abstraction to create schema aware IOs. */ + static class JdbcSchemaIO implements SchemaIO, Serializable { + protected final Row config; + protected final String location; + + JdbcSchemaIO(String location, Row config) { + this.config = config; + this.location = location; + } + + @Override + public Schema schema() { + return null; + } + + @Override + public PTransform<PBegin, PCollection<Row>> buildReader() { + String readQuery; + if (config.getString("readQuery") != null) { + readQuery = config.getString("readQuery"); + } else { + readQuery = String.format("SELECT f_int FROM %s", location); + } + + JdbcIO.DataSourceConfiguration dataSourceConfiguration = getDataSourceConfiguration(); + + JdbcIO.ReadRows readRows = + JdbcIO.readRows() + .withDataSourceConfiguration(dataSourceConfiguration) + .withQuery(readQuery); + + if (config.getInt16("fetchSize") != null) { + readRows = readRows.withFetchSize(config.getInt16("fetchSize")); + } + if (config.getBoolean("outputParallelization") != null) { + readRows = readRows.withOutputParallelization(config.getBoolean("outputParallelization")); + } + return readRows; + } + + @Override + public PTransform<PCollection<Row>, PDone> buildWriter() { + String writeStatement; + if (config.getString("writeStatement") != null) { + writeStatement = config.getString("writeStatement"); + } else { + writeStatement = String.format("INSERT INTO %s VALUES(?, ?, ?)", location); Review comment: The number of ?s in this query should depend on the number of fields in the schema of the input PCollection. `(?, ?, ?)` will only work when writing exactly three fields. I'd suggest returning a PTransform that looks at the input PCollection in its expand method. Something like this: ```java return new PTransform<PCollection<Row>, PDone>() { public PDone expand(PCollection<Row> input) { Schema schema = input.getSchema() // Generate writeStatement based on input schema return input.apply(JdbcIO.<Row>write()...); } } ``` ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected]
