Exclusion Filter Index Stage

The Exclusion Filter index stage is used to remove fields or documents that match items in a pre-defined exclusion list.

There are two ways to supply an exclusion list:

  • Upload a file containing a newline-separated list, using the Blob Store. When configuring the index stage, reference the list by its blob name in the location property (Exclusion List URI in the Fusion UI).

  • When configuring the index stage, enter an array of values for exclusion in the excludeValues property (Exclusion List in the Fusion UI).

The Exclusion Filter stage can be configured using one or both of these methods; Fusion combines them into one list. If regexPattern is configured, the pattern is applied to the field before the result is compared to the combined list.

By default, any matching field is excluded from indexing. To exclude the whole document, set skipDocument to "true" (Skip Document in the Fusion UI).

Uploading an exclusion list

Before you can configure the location property, you must upload one or more exclusion lists to Fusion using the Blob Store API.

Fusion comes with an example exclusion list at fusion/4.2.x/data/nlp/excludes/excludes.txt. Here is an example of how to upload this file using curl, where admin:pass are the credentials for an admin-level user:

curl -u admin:pass -X PUT --data-binary @data/nlp/excludes/excludes.txt -H 'Content-type: text/plain' http://localhost:8764/api/blobs/excludes.txt


Use an exclusion list for entities found in the author field:

    "type" : "exclusion-filter",
    "id" : "iw",
    "filters" : [ {
      "sourceField" : "author_s",
      "location" : "excludes.txt",
      "caseSensitive" : false
    } ],
    "skip" : false
  } ]


When entering configuration values in the UI, use unescaped characters, such as \t for the tab character. When entering configuration values in the API, use escaped characters, such as \\t for the tab character.