AWS::KinesisFirehose::DeliveryStream OpenXJsonSerDe

The OpenX SerDe. Used by Firehose for deserializing data, which means converting it from the JSON format in preparation for serializing it to the Parquet or ORC format. This is one of two deserializers you can choose, depending on which one offers the functionality you need. The other option is the native Hive / HCatalog JsonSerDe.

Syntax

To declare this entity in your AWS CloudFormation template, use the following syntax:

JSON

{
  "CaseInsensitive" : Boolean,
  "ColumnToJsonKeyMappings" : {Key: Value, ...},
  "ConvertDotsInJsonKeysToUnderscores" : Boolean
}

YAML

  CaseInsensitive: Boolean
  ColumnToJsonKeyMappings: 
    Key: Value
  ConvertDotsInJsonKeysToUnderscores: Boolean

Properties

CaseInsensitive

When set to true, which is the default, Firehose converts JSON keys to lowercase before deserializing them.

Required: No

Type: Boolean

Update requires: No interruption

ColumnToJsonKeyMappings

Maps column names to JSON keys that aren't identical to the column names. This is useful when the JSON contains keys that are Hive keywords. For example, timestamp is a Hive keyword. If you have a JSON key named timestamp, set this parameter to {"ts": "timestamp"} to map this key to a column named ts.

Required: No

Type: Object of String

Pattern: [a-zA-Z0-9]+

Update requires: No interruption

ConvertDotsInJsonKeysToUnderscores

When set to true, specifies that the names of the keys include dots and that you want Firehose to replace them with underscores. This is useful because Apache Hive does not allow dots in column names. For example, if the JSON contains a key whose name is "a.b", you can define the column name to be "a_b" when using this option.

The default is false.

Required: No

Type: Boolean

Update requires: No interruption