Updated on 2022-09-15 GMT+08:00

To Hive

Sample JSON File

"to-config-values": {
        "configs": [
          {
            "inputs": [
              {
                "name": "toJobConfig.hive",
                "value": "hive"
              },
              {
                "name": "toJobConfig.database",
                "value": "rf_database"
              },
              {
                "name": "toJobConfig.table",
                "value": "rf_to"
              },
              {
                "name": "toJobConfig.tablePreparation",
                "value": "DO_NOTHING"
              },
              {
                "name": "toJobConfig.columnList",
                "value": "aa&bb&cc&dd"
              },
              {
                "name": "toJobConfig.shouldClearTable",
                "value": "true"
              }
            ],
            "name": "toJobConfig"
          }
        ]
      }

Parameter Description

Parameter

Mandatory

Type

Description

toJobConfig.hive

No

String

Data source to which data is written

toJobConfig.database

No

String

Name of the database to which data is written. For example, default.

toJobConfig.table

Yes

String

Name of the table to which data is written

toJobConfig.tablePreparation

Yes

Enumeration

The options for data write to tables are as follows:
  • DO_NOTHING: Do not create the table automatically.
  • CREATE_WHEN_NOT_EXIST: If the destination database does not contain the table specified by tableName, CDM automatically creates the table.
  • DROP_AND_CREATE: Delete the table specified by tableName, and then create the table again.

toJobConfig.columnList

No

String

List of fields to be loaded. Use & to separate field names. For example, id&gid&name.

toJobConfig.shouldClearTable

No

Boolean

Whether to clear the data in the target table before data import. If this parameter is set to true, the data in the target table is cleared before the job is started.