Kafka Connect Filter (Confluent) SMT Usage Reference for Confluent Platform¶

The following provides usage information for the Confluent SMT io.confluent.connect.transforms.Filter. Include or drop records that match the filter.condition Predicate.

Description¶

Include or drop records that match the filter.condition predicate.

The filter.condition is a predicate specifying JSON Path that is applied to each record processed, and when this predicate successfully matches the record is either included (when filter.type=include) or excluded (when filter.type=exclude).

The missing.or.null.behavior property defines how the transform behaves when a record does not have the field(s) used in the filter condition predicate. By default the behavior is to fail. This property can also be set to include or exclude the record that is missing the predicate’s field(s).

Use the transformation type designed for the record key (io.confluent.connect.transforms.Filter$Key) or value (io.confluent.connect.transforms.Filter$Value).

Installation¶

This transformation is developed by Confluent and does not ship by default with Kafka or Confluent Platform. You can install this transformation using the confluent connect plugin install command:

confluent connect plugin install confluentinc/connect-transforms:latest

Examples¶

The configuration snippet below shows how to use and configure the Confluent Filter SMT.

The filter.condition is a predicate specifying JSON path. For more information on JSON path format, see https://github.com/json-path/JsonPath.

filter.condition specified in the following examples is satisfied when the nested field 'nestedKey' in field 'key' is equal to either 'value1' or 'value2'.

Defines the configuration for including records satisfying the filter condition. The filter condition is applied on record value. The fail behavior in the configuration specifies to throw an exception and fail the connector task when the record does not have the field(s) used in the filter condition.

"transforms": "filterExample1",
"transforms.filterExample1.type": "io.confluent.connect.transforms.Filter$Value",
"transforms.filterExample1.filter.condition": "$.key[?(@.nestedKey == "value1" || @.nestedKey == "value2")]",
"transforms.filterExample1.filter.type": "include", "transforms.filterExample1.missing.or.null.behavior": "fail"

Defines the configuration for excluding records that satisfy the filter condition. The filter condition is applied on record key. The include behavior in the configuration specifies to pass the record through when the record does not have the field(s) used in the filter condition.

"transforms": "filterExample2",
"transforms.filterExample2.type": "io.confluent.connect.transforms.Filter$Key", "transforms.filterExample2.filter.condition": "$.key[?(@.nestedKey == "value1" || @.nestedKey == "value2")]", "transforms.filterExample2.filter.type": "exclude", "transforms.filterExample2.missing.or.null.behavior": "include"

This example includes syntax to filter on a simple schema:

"transforms":"filterExample3",
"transforms.filterExample3.type":"io.confluent.connect.transforms.Filter$Value",
"transforms.filterExample3.filter.condition": "$[?(@.fieldToFilter == 'valueToMatch')]",
"transforms.filterExample3.filter.type":"exclude",
"transforms.filterExample3.missing.or.null.behavior":"include",

For additional examples, see Filter (Confluent) for fully-managed connectors.

Properties¶

Name	Description	Type	Default	Valid Values	Importance
`filter.condition`	Specifies the criteria used to match records to be included or excluded by this transformation. Use JSON Path predicate notation defined in: https://github.com/json-path/JsonPath.	string			high
`filter.type`	Specifies the action to perform with records that match the `filter.condition` predicate. Use `include` to pass through all records that match the predicate and drop all records that do not satisfy the predicate, or use `exclude` to drop all records that match the predicate.	string		[include, exclude]	high
`missing.or.null.behavior`	Specifies the behavior when the record does not have the field(s) used in the `filter.condition`. Use `fail` to throw an exception and fail the connector task, `include` to pass the record through, or `exclude` to drop the record.	string	`fail`	[fail, include, exclude]	medium