Atomic counters allow you to implement an atomic counter on a column. This feature provides statistic data for online applications such as the number of page views (PVs) on various topics.

Prerequisites

  • The OTSClient instance is initialized. For more information, see Initialization.
  • A data table is created. Data is written to the table.

Limits

  • You can implement atomic counters only on INTEGER columns.
  • By default, if a column that is specified as an atomic counter does not exist, the value of the column is 0 before you write data. If a column that is specified as an atomic counter is not an INTEGER column, an OTSParameterInvalid error occurs.
  • You can update an atomic counter by using a positive or negative number, but you must avoid an integer overflow. If an integer overflow occurs, an OTSParameterInvalid error is returned.
  • By default, the value of an atomic counter is not returned in the response to an update row request. You can specify that the increased value of an atomic counter is returned.
  • You cannot specify a column as an atomic counter and update the column in a single request. If Attribute Column A is set to an atomic counter, you cannot perform other operations such as overwrite and delete operations on the attribute column A.
  • You can perform multiple update operations on the same row by using a BatchWriteRow request. However, if you perform an atomic counter operation on a row, you can perform only one update operation on the row in a BatchWriteRow request.
  • Only the value of the latest version of an atomic counter is increased. You cannot increase the value of a specified version of an atomic counter. After you update a row, a new version of data is inserted to the atomic counter in the row.

API operations

The API operations to implement the atomic counter feature are added to the updateRowChange class. The following table lists these operations.
Operation Description
IncrementColumn(columnName string, value int64) Increases or decreases a value by a number.
AppendIncrementColumnToReturn(name string) Specifies the name of the column on which operations are performed to implement atomic counter. This API operation is configured to return the value of the column.
SetReturnIncrementValue() Specifies the return type. The updated value in the column on which operations are performed to implement atomic counter must be returned.

Parameters

Parameter Description
TableName The name of the table.
ColumnName The name of the column on which operations are performed to implement atomic counter. Columns whose valid values are integers are supported to implement the atomic counter feature.
Value The value to update the columns on which operations are performed to implement atomic counter.

Examples

The following code provides an example on how to call the updateRowChange operation to update the columns whose valid values are integers and return the updated values when you write a row of data:
func UpdateRowWithIncrement(client *tablestore.TableStoreClient, tableName string) {
    fmt.Println("begin to update row")
    updateRowRequest = new(tablestore.UpdateRowRequest)
    updateRowChange = new(tablestore.UpdateRowChange)
    // Specify the name of the table.
    updateRowChange.TableName = tableName
    updatePk = new(tablestore.PrimaryKey)
    updatePk.AddPrimaryKeyColumn("pk1", "pk1increment")
    updatePk.AddPrimaryKeyColumn("pk2", int64(2))
    updatePk.AddPrimaryKeyColumn("pk3", []byte("pk3"))
    updateRowChange.PrimaryKey = updatePk
    // Specify that the value the col2 column on which operations are performed to implement atomic counter is increased by 30. Do not set timestamps.
    updateRowChange.IncrementColumn("col2", int64(30))
    // Return the values of the columns on which operations are performed to implement atomic counter.
    updateRowChange.SetReturnIncrementValue()
    updateRowChange.SetCondition(tablestore.RowExistenceExpectation_IGNORE)
    updateRowChange.AppendIncrementColumnToReturn("col2")
    updateRowRequest.UpdateRowChange = updateRowChange

    resp, err := client.UpdateRow(updateRowRequest)
    if err != nil {
        fmt.Println("update failed with error:", err)
        return
    } else {
        fmt.Println("update row finished")
        fmt.Println(resp)
        fmt.Println(len(resp.Columns))
        fmt.Println(resp.Columns[0].ColumnName)
        fmt.Println(resp.Columns[0].Value)
        fmt.Println(resp.Columns[0].Timestamp)
    }
}