Modifying the maprdb Format Settings within the dfs Storage Plugin Configuration

The following table lists the maprdb format options that you can set within the dfs storage plugin configuration:

Option	Description	Value
allTextMode	When enabled, Drill reads all values as type varchar. Useful when the underlying data set has type values of mixed scalar types, such as integers, floating point, varchars, date, time, and timestamp. Disabled by default.	true|false
disableCountOptimization	When enabled, this option disables optimization for queries with the COUNT (*) aggregate function. Disabled by default.	true|false
enablePushdown	When enabled, Drill pushes down filters to MapR Database. Disabling this option is not recommended unless you intend to use it for troubleshooting purposes. Enabled by default.	true|false
ignoreSchemaChange	When enabled, Drill ignores schema changes. Disabled by default.	true|false
nonExistentColumnsProjection	When enabled, Drill can distinguish between null and non-existent fields. Disabled by default.	true|false
readNumbersAsDouble	When enabled, Drill reads all numeric values as type double. Useful when the underlying data set has type values of mixed numeric types, such as integers and floating point. Disabled by default.	true|false

The following example configuration shows you how to include the options in the maprdb format configuration within the dfs storage plugin configuration:

{
  "type": "file",
  "enabled": true,
  "connection": "maprfs:///",
  "config": null,
  "workspaces": {
    "root": {
      "location": "/",
      "writable": false,
      "defaultInputFormat": "maprdb",
      "allowAccessOutsideWorkspace": false
    },
  "formats": {
    "maprdb": {
      "type": "maprdb",
      "allTextMode": true,
      "disableCountOptimization": true,
      "enablePushdown": false
    },
    "parquet": {
      "type": "parquet"
    },
    "json": {
      "type": "json",
      "extensions": [
        "json"
      ]
    }
  }
}