Beispiel: Flexible Java App Engine-Umgebung

Dieses Beispiel besteht aus einer App Engine-Anwendung, geschrieben in Java, die einige "Hello World"-Grüße in eine Tabelle schreibt und dann aus ihr liest. Die Anwendung verwendet die App Engine-Laufzeit und hat uneingeschränkten Zugriff auf App Engine APIs. Sie wird in der Cloud in der flexiblen App Engine-Umgebung ausgeführt. Der Code für die Anwendung ist im GitHub-Repository GoogleCloudPlatform/cloud-bigtable-examples im Verzeichnis java/gae-flexible-helloworld enthalten.

Codebeispiel – Übersicht

Das Codebeispiel beinhaltet die folgenden Klassen:

BigtableHelper

BigtableHelper stellt eine Methode zum Erstellen einer Verbindung zu Cloud Bigtable bereit. Sie speichert auch die Verbindung im Cache und stellt eine Methode zur Verfügung, die zwischengespeicherte Verbindung aufzurufen, sofern vorhanden. Dies liegt daran, dass das Erstellen der Verbindung ein relativ teurer Vorgang ist.

public static void connect() throws IOException {

  if (PROJECT_ID == null || INSTANCE_ID == null ) {
    if (sc != null) {
      sc.log("environment variables BIGTABLE_PROJECT, and BIGTABLE_INSTANCE need to be defined.");
    }
    return;
  }

  connection = BigtableConfiguration.connect(PROJECT_ID, INSTANCE_ID);
}

public static Connection getConnection() {
  if (connection == null) {
    try {
      connect();
    } catch (IOException e) {
      if (sc != null) {
        sc.log("connect ", e);
      }
    }
  }
  if (connection == null) {
    if (sc != null) {
      sc.log("BigtableHelper-No Connection");
    }
  }
  return connection;
}

BigtableHelloWorld

BigtableHelloWorld wird verwendet, um eine Reihe von Grüßen in Cloud Bigtable zu schreiben, diese dann zu lesen und anzuzeigen. Die Klasse erhält eine Cloud Bigtable-Verbindung von BigtableHelper und ruft darüber ein Table-Objekt ab, mit dem Sie Werte lesen und schreiben können. Schließlich schreibt sie mithilfe des Objekts Table in die Tabelle und liest daraus:

/**
 * A minimal application that connects to Cloud Bigtable using the native HBase API
 * and performs some basic operations.
 */
public class BigtableHelloWorld {

  // Refer to table metadata names by byte array in the HBase API
  private static final byte[] TABLE_NAME = Bytes.toBytes("Hello-Bigtable");
  private static final byte[] COLUMN_FAMILY_NAME = Bytes.toBytes("cf1");
  private static final byte[] COLUMN_NAME = Bytes.toBytes("greeting");

  // Write some friendly greetings to Cloud Bigtable
  private static final String[] GREETINGS =
      { "Hello World!", "Hello Cloud Bigtable!", "Hello HBase!" };

  public static String create(Connection connection) {
    try {
        // The admin API lets us create, manage and delete tables
      Admin admin = connection.getAdmin();

      // Create a table with a single column family
      HTableDescriptor descriptor = new HTableDescriptor(TableName.valueOf(TABLE_NAME));
      descriptor.addFamily(new HColumnDescriptor(COLUMN_FAMILY_NAME));

      admin.createTable(descriptor);
    } catch (IOException e) {
      return "Table exists.";
    }
    return "Create table " + Bytes.toString(TABLE_NAME);
  }

  /**
   * Connects to Cloud Bigtable, runs some basic operations and prints the results.
   */
  public static String doHelloWorld() {

    StringBuilder result = new StringBuilder();

    // Create the Bigtable connection, use try-with-resources to make sure it gets closed
    Connection connection = BigtableHelper.getConnection();
    result.append(create(connection));
    result.append("<br><br>");
    try (Table table = connection.getTable(TableName.valueOf(TABLE_NAME))) {

      // Retrieve the table we just created so we can do some reads and writes

      // Write some rows to the table
      result.append("Write some greetings to the table<br>");
      for (int i = 0; i < GREETINGS.length; i++) {
        // Each row has a unique row key.
        //
        // Note: This example uses sequential numeric IDs for simplicity, but
        // this can result in poor performance in a production application.
        // Since rows are stored in sorted order by key, sequential keys can
        // result in poor distribution of operations across nodes.
        //
        // For more information about how to design a Bigtable schema for the
        // best performance, see the documentation:
        //
        //     https://cloud.google.com/bigtable/docs/schema-design
        String rowKey = "greeting" + i;

        // Put a single row into the table. We could also pass a list of Puts to write a batch.
        Put put = new Put(Bytes.toBytes(rowKey));
        put.addColumn(COLUMN_FAMILY_NAME, COLUMN_NAME, Bytes.toBytes(GREETINGS[i]));
        table.put(put);
      }

      // Get the first greeting by row key
      String rowKey = "greeting0";
      Result getResult = table.get(new Get(Bytes.toBytes(rowKey)));
      String greeting = Bytes.toString(getResult.getValue(COLUMN_FAMILY_NAME, COLUMN_NAME));
      result.append("Get a single greeting by row key<br>");
      result.append("     ");
      result.append(rowKey);
      result.append("= ");
      result.append(greeting);
      result.append("<br>");

      // Now scan across all rows.
      Scan scan = new Scan();

      result.append("Scan for all greetings:");
      ResultScanner scanner = table.getScanner(scan);
      for (Result row : scanner) {
        byte[] valueBytes = row.getValue(COLUMN_FAMILY_NAME, COLUMN_NAME);
        result.append("    ");
        result.append(Bytes.toString(valueBytes));
        result.append("<br>");
      }

    } catch (IOException e) {
      result.append("Exception while running HelloWorld: " + e.getMessage() + "<br>");
      result.append(e.toString());
      return result.toString();
    }

    return result.toString();
  }

}