JDBC ドライバー (OSS) を使用したクエリの実行

この記事には、クエリを同期および非同期的に実行する方法を示す例が含まれています。

例: クエリを実行する

次の例は、Databricks JDBCドライバー(OSS)を使用して、Databricks コンピュート リソースを使用してDatabricks SQLのクエリを実行する方法を示しています。

Java

import java.sql.Connection;
import java.sql.DriverManager;
import java.sql.ResultSet;
import java.sql.ResultSetMetaData;
import java.sql.Statement;
import java.util.Properties;

public class DatabricksJDBCExample {

    public static void main(String[] args) {

        Class.forName("com.databricks.client.jdbc.Driver");

        // Set JDBC URL properties
        String jdbcUrl = "jdbc:databricks://dbc-a1b2345c-d6e7.cloud.databricks.com:443";
        Properties connectionProperties = new Properties();
        connectionProperties.put("httpPath", "sql/protocolv1/o/123456780012345/0123-123450-z000pi22");
        connectionProperties.put("ssl", "1");

        // Set authentication properties (personal access token)
        connectionProperties.put("AuthMech", "3");
        connectionProperties.put("user", "token");
        connectionProperties.put("password", "12345678901234667890abcdabcd");

        // Set logging properties
        connectionProperties.put("logPath", "logs/myapplication.log");

        // Establish connection and execute query
        try (Connection connection = DriverManager.getConnection(jdbcUrl, connectionProperties);
             Statement statement = connection.createStatement();
             ResultSet resultSet = statement.executeQuery("SELECT * FROM samples.nyctaxi.trips")) {

            // Get metadata and column names
            ResultSetMetaData metaData = resultSet.getMetaData();
            String[] columns = new String[metaData.getColumnCount()];
            for (int i = 0; i < columns.length; i++) {
                columns[i] = metaData.getColumnName(i + 1);
            }

            // Process and print the result set
            while (resultSet.next()) {
                System.out.print("Row " + resultSet.getRow() + "=[");
                for (int i = 0; i < columns.length; i++) {
                    if (i != 0) {
                        System.out.print(", ");
                    }
                    System.out.print(columns[i] + "='" + resultSet.getObject(i + 1) + "'");
                }
                System.out.println("]");
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
}

例: クエリを非同期的に実行する

次の例は、 Databricks JDBC ドライバー (OSS) を使用して非同期 Databricks SQL クエリを実行および処理する方法を示しています。

API リファレンスについては、「 Databricks JDBC ドライバー (OSS) の Java API リファレンス」を参照してください。

ステートメントの非同期実行を開始します。

Java

Statement statement = conn.createStatement();
IDatabricksStatement dbStatement = statement.unwrap(IDatabricksStatement.class);

ResultSet result = dbStatement.executeAsync(sql);
IDatabricksResultSet asyncResult = result.unwrap(IDatabricksResultSet.class);
IExecutionStatus asyncStatus = asyncResult.getExecutionStatus();
long startTime = System.currentTimeMillis();
while ((asyncStatus.getExecutionState() == ExecutionState.RUNNING |  asyncStatus.getExecutionState() == ExecutionState.PENDING) || (startTime + timeout < System.currentTimeMillis())) {
  Thread.sleep(1000); // Sleep for 1000 ms
  asyncResult = dbStatement.getExecutionResult().unwrap(IDatabricksResultSet.class);
  asyncStatus = asyncResult.getExecutionStatus();
}

if (asyncStatus.getExecutionStatus() == ExecutionState.RUNNING | ExecutionState.PENDING) {
  dbStatement.cancel();
}

if (asyncStatus.getExecutionStatus() == ExecutionState.SUCCEEDED) {
  // process result set
}

if (asyncStatus.getExecutionStatus() == ExecutionState.FAILED) {
  String sqlState = asyncStatus.getSqlState();
  String errorMessage = asyncStatus.getErrorMessage();
  // log error code and message
}

ステートメントを別のスレッドで処理します。

Java

Statement statement = conn1.createStatement();
IDatabricksStatement dbStatement = statement.unwrap(IDatabricksStatement.class);

ResultSet asyncResult = dbStatement.executeAsync(sql);
IDatabricksResultSet drs = asyncResult.unwrap(IDatabricksResultSet.class);
String statementId = drs.getStatementId();

ExecutionState state = drs.getExecutionStatus().getExecutionState();

while (state != ExecutionState.SUCCEEDED) {
  Thread.sleep(sleepInterval);
  asyncResult = dbStatement.getExecutionResult();
  state = asyncResult.unwrap(IDatabricksResultSet.class).getExecutionStatus().getExecutionState();
}


// In another thread
IDatabricksConnection dbConn2 = conn2.unwrap(IDatabricksConnection.class);
IDatabricksStatement asyncStatementHandle = dbConn2.getStatement(statementId).unwrap(IDatabricksStatement.class);
IDatabricksResultSet asyncResultHandle = asyncStatementHandle.getExecutionResult().unwrap(IDatabricksResultSet.class);

// Cancel if needed
if (asyncResultHandle.getExecutionStatus().getExecutionState() == ExecutionState.PENDING | asyncResultHandle.getExecutionStatus().getExecutionState() == ExecutionState.RUNNING) {
  asyncStatementHandle.cancel();
}

接続 ID を使用して接続を閉じます。

Java

// Get connection-Id from existing connection
String connectionId = conn.unwrap(IDatabricksConnection.class).getConnectionId();

// Close the connection from other thread using same JDBC Url and connection properties and connection-Id retrieved from above
com.databricks.client.jdbc.Driver.getInstance().closeConnection(jdbcUrl, properties, connectionId);