BigQuery export to Parquet (via Storage API) template

The BigQuery export to Parquet template is a batch pipeline that reads data from a BigQuery table and writes it to a Cloud Storage bucket in Parquet format. This template utilizes the BigQuery Storage API to export the data.

Pipeline requirements

• The input BigQuery table must exist before running the pipeline.
• The output Cloud Storage bucket must exist before running the pipeline.

Template parameters

Parameter	Description
tableRef	The BigQuery input table location. For example, <my-project>:<my-dataset>.<my-table>.
bucket	The Cloud Storage folder in which to write the Parquet files. For example, gs://mybucket/exports.
numShards	(Optional) The number of output file shards. The default value is 1.
fields	(Optional) A comma-separated list of fields to select from the input BigQuery table.

Run the template

Console

1. Go to the Dataflow Create job from template page.
Go to Create job from template
2. In the Job name field, enter a unique job name.
3. Optional: For Regional endpoint, select a value from the drop-down menu. The default region is us-central1.

   For a list of regions where you can run a Dataflow job, see Dataflow locations.

4. From the Dataflow template drop-down menu, select the BigQuery export to Parquet (via Storage API) template.
5. In the provided parameter fields, enter your parameter values.
6. Click Run job.

gcloud

Note: To use the Google Cloud CLI to run flex templates, you must have Google Cloud CLI version 284.0.0 or later.

In your shell or terminal, run the template:

gcloud dataflow flex-template run JOB_NAME \
    --project=PROJECT_ID \
    --template-file-gcs-location=gs://dataflow-templates-REGION_NAME/VERSION/flex/BigQuery_to_Parquet \
    --region=REGION_NAME \
    --parameters \
tableRef=BIGQUERY_TABLE,\
bucket=OUTPUT_DIRECTORY,\
numShards=NUM_SHARDS,\
fields=FIELDS

Replace the following:

• PROJECT_ID: the Google Cloud project ID where you want to run the Dataflow job
• JOB_NAME: a unique job name of your choice
• VERSION: the version of the template that you want to use

  You can use the following values:

  • latest to use the latest version of the template, which is available in the non-dated parent folder in the bucket— gs://dataflow-templates-REGION_NAME/latest/
  • the version name, like 2023-09-12-00_RC00, to use a specific version of the template, which can be found nested in the respective dated parent folder in the bucket— gs://dataflow-templates-REGION_NAME/
  Caution: The latest version of templates might update with breaking changes. Your production environments should use templates kept in the most recent dated parent folder to prevent these breaking changes from affecting your production workflows.
• REGION_NAME: the region where you want to deploy your Dataflow job—for example, us-central1
• BIGQUERY_TABLE: your BigQuery table name
• OUTPUT_DIRECTORY: your Cloud Storage folder for output files
• NUM_SHARDS: the desired number of output file shards
• FIELDS: the comma-separated list of fields to select from the input BigQuery table

API

To run the template using the REST API, send an HTTP POST request. For more information on the API and its authorization scopes, see projects.templates.launch.

POST https://dataflow.googleapis.com/v1b3/projects/PROJECT_ID/locations/LOCATION/flexTemplates:launch
{
   "launch_parameter": {
      "jobName": "JOB_NAME",
      "parameters": {
          "tableRef": "BIGQUERY_TABLE",
          "bucket": "OUTPUT_DIRECTORY",
          "numShards": "NUM_SHARDS",
          "fields": "FIELDS"
      },
      "containerSpecGcsPath": "gs://dataflow-templates-LOCATION/VERSION/flex/BigQuery_to_Parquet",
   }
}

Replace the following:

• PROJECT_ID: the Google Cloud project ID where you want to run the Dataflow job
• JOB_NAME: a unique job name of your choice
• VERSION: the version of the template that you want to use

  You can use the following values:

  • latest to use the latest version of the template, which is available in the non-dated parent folder in the bucket— gs://dataflow-templates-REGION_NAME/latest/
  • the version name, like 2023-09-12-00_RC00, to use a specific version of the template, which can be found nested in the respective dated parent folder in the bucket— gs://dataflow-templates-REGION_NAME/
  Caution: The latest version of templates might update with breaking changes. Your production environments should use templates kept in the most recent dated parent folder to prevent these breaking changes from affecting your production workflows.
• LOCATION: the region where you want to deploy your Dataflow job—for example, us-central1
• BIGQUERY_TABLE: your BigQuery table name
• OUTPUT_DIRECTORY: your Cloud Storage folder for output files
• NUM_SHARDS: the desired number of output file shards
• FIELDS: the comma-separated list of fields to select from the input BigQuery table

Template source code

Java

/*
 * Copyright (C) 2019 Google LLC
 *
 * Licensed 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 com.google.cloud.teleport.v2.templates;

import com.google.api.gax.rpc.InvalidArgumentException;
import com.google.api.services.bigquery.model.TableReference;
import com.google.cloud.bigquery.storage.v1beta1.BigQueryStorageClient;
import com.google.cloud.bigquery.storage.v1beta1.ReadOptions.TableReadOptions;
import com.google.cloud.bigquery.storage.v1beta1.Storage.CreateReadSessionRequest;
import com.google.cloud.bigquery.storage.v1beta1.Storage.ReadSession;
import com.google.cloud.bigquery.storage.v1beta1.TableReferenceProto;
import com.google.cloud.teleport.metadata.Template;
import com.google.cloud.teleport.metadata.TemplateCategory;
import com.google.cloud.teleport.metadata.TemplateParameter;
import com.google.cloud.teleport.v2.common.UncaughtExceptionLogger;
import com.google.cloud.teleport.v2.templates.BigQueryToParquet.BigQueryToParquetOptions;
import com.google.common.base.Splitter;
import com.google.common.base.Strings;
import java.io.IOException;
import java.util.Arrays;
import java.util.List;
import org.apache.avro.Schema;
import org.apache.avro.generic.GenericRecord;
import org.apache.beam.sdk.Pipeline;
import org.apache.beam.sdk.PipelineResult;
import org.apache.beam.sdk.extensions.avro.coders.AvroCoder;
import org.apache.beam.sdk.io.FileIO;
import org.apache.beam.sdk.io.gcp.bigquery.BigQueryHelpers;
import org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO;
import org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO.TypedRead;
import org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO.TypedRead.Method;
import org.apache.beam.sdk.io.gcp.bigquery.SchemaAndRecord;
import org.apache.beam.sdk.io.parquet.ParquetIO;
import org.apache.beam.sdk.options.Default;
import org.apache.beam.sdk.options.PipelineOptions;
import org.apache.beam.sdk.options.PipelineOptionsFactory;
import org.apache.beam.sdk.options.Validation.Required;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

/**
 * The {@link BigQueryToParquet} pipeline exports data from a BigQuery table to Parquet file(s) in a
 * Google Cloud Storage bucket.
 *
 * <p>Check out <a
 * href="https://github.com/GoogleCloudPlatform/DataflowTemplates/blob/main/v2/bigquery-to-parquet/README_BigQuery_to_Parquet.md">README</a>
 * for instructions on how to use or modify this template.
 */
@Template(
    name = "BigQuery_to_Parquet",
    category = TemplateCategory.BATCH,
    displayName = "BigQuery export to Parquet (via Storage API)",
    description =
        "The BigQuery export to Parquet template is a batch pipeline that reads data from a BigQuery table and writes it to a Cloud Storage bucket in Parquet format. "
            + "This template utilizes the <a href=\"https://cloud.google.com/bigquery/docs/reference/storage\">BigQuery Storage API</a> to export the data.",
    optionsClass = BigQueryToParquetOptions.class,
    flexContainerName = "bigquery-to-parquet",
    documentation =
        "https://cloud.google.com/dataflow/docs/guides/templates/provided/bigquery-to-parquet",
    contactInformation = "https://cloud.google.com/support",
    requirements = {
      "The input BigQuery table must exist before running the pipeline.",
      "The output Cloud Storage bucket must exist before running the pipeline."
    })
public class BigQueryToParquet {

  /* Logger for class. */
  private static final Logger LOG = LoggerFactory.getLogger(BigQueryToParquet.class);

  /** File suffix for file to be written. */
  private static final String FILE_SUFFIX = ".parquet";

  /** Factory to create BigQueryStorageClients. */
  static class BigQueryStorageClientFactory {

    /**
     * Creates BigQueryStorage client for use in extracting table schema.
     *
     * @return BigQueryStorageClient
     */
    static BigQueryStorageClient create() {
      try {
        return BigQueryStorageClient.create();
      } catch (IOException e) {
        LOG.error("Error connecting to BigQueryStorage API: " + e.getMessage());
        throw new RuntimeException(e);
      }
    }
  }

  /** Factory to create ReadSessions. */
  static class ReadSessionFactory {

    /**
     * Creates ReadSession for schema extraction.
     *
     * @param client BigQueryStorage client used to create ReadSession.
     * @param tableString String that represents table to export from.
     * @param tableReadOptions TableReadOptions that specify any fields in the table to filter on.
     * @return session ReadSession object that contains the schema for the export.
     */
    static ReadSession create(
        BigQueryStorageClient client, String tableString, TableReadOptions tableReadOptions) {
      TableReference tableReference = BigQueryHelpers.parseTableSpec(tableString);
      String parentProjectId = "projects/" + tableReference.getProjectId();

      TableReferenceProto.TableReference storageTableRef =
          TableReferenceProto.TableReference.newBuilder()
              .setProjectId(tableReference.getProjectId())
              .setDatasetId(tableReference.getDatasetId())
              .setTableId(tableReference.getTableId())
              .build();

      CreateReadSessionRequest.Builder builder =
          CreateReadSessionRequest.newBuilder()
              .setParent(parentProjectId)
              .setReadOptions(tableReadOptions)
              .setTableReference(storageTableRef);
      try {
        return client.createReadSession(builder.build());
      } catch (InvalidArgumentException iae) {
        LOG.error("Error creating ReadSession: " + iae.getMessage());
        throw new RuntimeException(iae);
      }
    }
  }

  /**
   * The {@link BigQueryToParquetOptions} class provides the custom execution options passed by the
   * executor at the command-line.
   */
  public interface BigQueryToParquetOptions extends PipelineOptions {
    @TemplateParameter.BigQueryTable(
        order = 1,
        description = "BigQuery table to export",
        helpText = "BigQuery table location to export in the format <project>:<dataset>.<table>.",
        example = "your-project:your-dataset.your-table-name")
    @Required
    String getTableRef();

    void setTableRef(String tableRef);

    @TemplateParameter.GcsWriteFile(
        order = 2,
        description = "Output Cloud Storage file(s)",
        helpText = "Path and filename prefix for writing output files.",
        example = "gs://your-bucket/export/")
    @Required
    String getBucket();

    void setBucket(String bucket);

    @TemplateParameter.Integer(
        order = 3,
        optional = true,
        description = "Maximum output shards",
        helpText =
            "The maximum number of output shards produced when writing. A higher number of shards"
                + " means higher throughput for writing to Cloud Storage, but potentially higher"
                + " data aggregation cost across shards when processing output Cloud Storage"
                + " files.")
    @Default.Integer(0)
    Integer getNumShards();

    void setNumShards(Integer numShards);

    @TemplateParameter.Text(
        order = 4,
        optional = true,
        description = "List of field names",
        helpText = "Comma separated list of fields to select from the table.")
    String getFields();

    void setFields(String fields);

    @TemplateParameter.Text(
        order = 5,
        optional = true,
        description = "Row restrictions/filter.",
        helpText =
            "Read only rows which match the specified filter, which must be a SQL expression"
                + " compatible with Google standard SQL"
                + " (https://cloud.google.com/bigquery/docs/reference/standard-sql). If no value is"
                + " specified, then all rows are returned.")
    String getRowRestriction();

    void setRowRestriction(String restriction);
  }

  /**
   * The {@link BigQueryToParquet#getTableSchema(ReadSession)} method gets Avro schema for table
   * using from the {@link ReadSession} object.
   *
   * @param session ReadSession that contains schema for table, filtered by fields if any.
   * @return avroSchema Avro schema for table. If fields are provided then schema will only contain
   *     those fields.
   */
  private static Schema getTableSchema(ReadSession session) {
    Schema avroSchema;

    avroSchema = new Schema.Parser().parse(session.getAvroSchema().getSchema());
    LOG.info("Schema for export is: " + avroSchema.toString());

    return avroSchema;
  }

  /**
   * Main entry point for pipeline execution.
   *
   * @param args Command line arguments to the pipeline.
   */
  public static void main(String[] args) {
    UncaughtExceptionLogger.register();

    BigQueryToParquetOptions options =
        PipelineOptionsFactory.fromArgs(args).withValidation().as(BigQueryToParquetOptions.class);

    run(options);
  }

  /**
   * Runs the pipeline with the supplied options.
   *
   * @param options The execution parameters to the pipeline.
   * @return The result of the pipeline execution.
   */
  private static PipelineResult run(BigQueryToParquetOptions options) {

    // Create the pipeline.
    Pipeline pipeline = Pipeline.create(options);

    TableReadOptions.Builder builder = TableReadOptions.newBuilder();

    /* Add fields to filter export on, if any. */
    if (options.getFields() != null) {
      builder.addAllSelectedFields(Arrays.asList(options.getFields().split(",\\s*")));
    }

    TableReadOptions tableReadOptions = builder.build();
    BigQueryStorageClient client = BigQueryStorageClientFactory.create();
    ReadSession session =
        ReadSessionFactory.create(client, options.getTableRef(), tableReadOptions);

    // Extract schema from ReadSession
    Schema schema = getTableSchema(session);
    client.close();

    TypedRead<GenericRecord> readFromBQ =
        BigQueryIO.read(SchemaAndRecord::getRecord)
            .from(options.getTableRef())
            .withTemplateCompatibility()
            .withMethod(Method.DIRECT_READ)
            .withCoder(AvroCoder.of(schema));

    if (options.getFields() != null) {
      List<String> selectedFields = Splitter.on(",").splitToList(options.getFields());
      readFromBQ =
          selectedFields.isEmpty() ? readFromBQ : readFromBQ.withSelectedFields(selectedFields);
    }

    // Add row restrictions/filter if any.
    if (!Strings.isNullOrEmpty(options.getRowRestriction())) {
      readFromBQ = readFromBQ.withRowRestriction(options.getRowRestriction());
    }

    /*
     * Steps: 1) Read records from BigQuery via BigQueryIO.
     *        2) Write records to Google Cloud Storage in Parquet format.
     */
    pipeline
        /*
         * Step 1: Read records via BigQueryIO using supplied schema as a PCollection of
         *         {@link GenericRecord}.
         */
        .apply("ReadFromBigQuery", readFromBQ)
        /*
         * Step 2: Write records to Google Cloud Storage as one or more Parquet files
         *         via {@link ParquetIO}.
         */
        .apply(
            "WriteToParquet",
            FileIO.<GenericRecord>write()
                .via(ParquetIO.sink(schema))
                .to(options.getBucket())
                .withNumShards(options.getNumShards())
                .withSuffix(FILE_SUFFIX));

    // Execute the pipeline and return the result.
    return pipeline.run();
  }
}

What's next

• Learn about Dataflow templates.
• See the list of Google-provided templates.