Indexing object text

To make text searchable, identify text to be indexed.

Overview

Indexing makes content searchable by extracting and storing text from documents and properties in a search index. When you enable content-based retrieval (CBR) for classes and properties, the search service indexes the content and makes it available for full-text searches.

To configure the core indexing elements, see Setting up indexing.

Indexable base classes

You can index objects that belong to a CBR-enabled class. You can index the following base classes and their subclasses:

  • Document
  • Annotation
  • Folder
  • Custom Object

String properties of these classes are also indexable. For a string property to be indexed, both the class and the property must be CBR-enabled. For example, to index the FolderName property, both the Folder class and the FolderName property must be CBR-enabled.

Configuration elements

The following elements determine what content is indexed.

  • CBR-enabled status of classes
  • CBR-enabled status of properties
  • MIME type eligibility for indexing
  • Surplus XML element definitions (XML filtering)
  • Text language identification and analyzer selection for the object store

Indexing options

You can configure object text indexing by using the following search services.

  • Indexing object text with Content Search Services
  • Indexing object text with Elasticsearch or OpenSearch
  • Indexing object text in dual mode (indexing to both Content Search Services and Elasticsearch simultaneously during migration)
Tip: If you change the indexing configuration after content is already indexed, reindex the appropriate classes or indexes to apply the changes to existing content.