Base Filter

A version of this document for releases prior to release-2023-07 is available here.

Base filters are used to filter the result set on a dashboard or a search modal. They can be visible (displayFilter) or invisible (all other types). Base filters are used in different places and have a common format to construct the filter query.

Filter Query Format

There are different places, where one can define base filter:

At all these places, one can use the same query format, e.g.

{key: 'contentType', term: 'regular'}

Learn more about the filter queries format here

Filter Query Examples

This are all available queryTypes which can be used to form a filter query.

// contentType {key: 'string', term: string || array}
{key: 'contentType', term: 'regular'}

// not contentType {key: 'string', term: string || array}
{not: {key: 'contentType', term: 'regular'}}

// documentType {key: 'string', term: string || array}
{key: 'documentType', term: 'article'}
{key: 'documentType', term: ['article', 'page']}

// language
{key: 'language', term: 'de'}

// ownerId
{key: 'ownerId', term: 1}

// dateRange
{key: 'updatedAt', range: {gte: '2023-06-21T07:55:00.000Z', lte: '2023-06-23T07:55:00.000Z'}}

// metadata
{key: 'metadata.foo', term: 'bar'}
{key: 'metadata.foo', exists: true}
{key: 'metadata.teaserImage.mediaId', exists: true}
{
  key: 'metadata.publicationDate',
  range: {gte: '2023-06-21T07:55:00.000Z', lte: '2023-06-23T07:55:00.000Z'}
}

// tasks
{key: 'metadata.proofreading.state', taskValue: 'pending'}
{key: 'metadata.review.state', taskValue: 'done'}

// reference
{key: 'references', term: ['document:123']}
{key: 'references', term: ['image:W8GRbmju4grG']}
{key: 'references', term: ['file:A7JRa4mS3xBQ']}
{key: 'references', term: ['video:P2rSblad3aUS']}

// state for the mediaLibrary
{key: 'state', term: 'active'}
{key: 'state', term: 'revoked'}

// userInTeam (Added in: release-2023-03)
{key: 'metadata.myTeamHandle.activeUsers.id', termPattern: '{{ userId }}'}

// term variables (Added in: release-2025-01)
// only supported by li-teaser and li-document-search
{key: 'metadata.category.id', termVariable: 'metadata.category.id'}

// embargo (Added in: release-2025-07)
{key: 'publishControl.embargo.enforced', term: true},
{or: [
    {key: 'publishControl.embargo.until', exists: false},
    {key: 'publishControl.embargo.until', range: {gt: 'now'}}
  ]
}

Example - Filter by metadata with key/value

baseFilters: [{key: 'metadata.foo', term: 'bar'}]

This would filter for only documents that have the value bar in the metadata field foo. You have to make sure that foo is an indexed metadata field.

Example - Filter by metadata with objects

defaultQueries: [{key: 'metadata.author.reference.id', term: 42}]

More complex metadata fields are indexed as an object (instead of key/value). In this case one can filter based on subproperties.

// This is how the metadata field has been indexed into the search index
{
  metadata: {
    author: {
      reference: {
        id: 42
      }
    }
  }
}

This example would filter documents that have the value 42 in the metadata field author with properties reference.id. You have to make sure that author is an indexed metadata field.

Example - Filter by task

defaultQueries: [{key: 'metadata.proofreading.state', term: 'requested'}]

This would filter for only documents that have had a successful proofreading. The core only exposes the proofreading task, but you can define your own custom tasks. The values are todo, doing, done for the 3 states that a task can have.

Example - Filter by metadata with dataType keyword for mediaIndex

baseFilters: [{key: 'metadata.transformed', term: true}]

This would filter for only transformed assets.

Example - Filter by document statistics

baseFilters: [
  {
    and: [
      {key: 'statistics.componentCount.image', exists: false},
      {
        or: [
          {key: 'statistics.characterCount', range: {gte: 1000}},
          {key: 'statistics.componentCount.paragraph', range: {gte: 3}}
        ]
      }
    ]
  }
]

This would filter for documents without an image, but with more than 2 components of name paragraph or +1000 characters. To query for documents without a component, use the exists: false notation instead of a range, since components with zero-counts are not indexed.