Skip to content
ExamineX
These docs are for ExamineX 11 · Umbraco 19 (pre-release). For the current release see ExamineX 6–10 · Umbraco 13–18.

Known limitations

This page lists what doesn’t work the same as Umbraco’s default Lucene (Examine) provider, and what isn’t supported yet.

Faceting through the Umbraco Search APIs isn’t fully supported yet:

  • Integer facet fields (FacetInteger) aren’t mapped as facet fields, so they don’t return facet values.
  • Integer range facets are sent with long.MinValue and long.MaxValue as their open bounds.

Faceting with the Examine APIs is fully supported, see Faceting for Azure AI Search or Elasticsearch.

ExamineX’s license and health status isn’t shown in the backoffice yet.

Pick your provider; the choice is remembered on other pages.

Lucene fields can always hold multiple values, but Azure AI Search fields can’t unless they are a collection, and Azure AI Search can’t sort on collections. ExamineX makes Umbraco Search’s keyword and text fields collections, but numeric and date fields stay single value fields so they can be sorted.

Examine has no DateTimeOffset range facet, so use the ExamineX FacetDateTimeRange extension method (ExamineX.Shared namespace). It’s ignored on an index that isn’t an ExamineX index (for example a disabled index). Ranges can have gaps but can’t overlap, see Range boundaries.

OData $filter queries can’t be passed to NativeQuery, use the Azure Search APIs directly. Range queries can’t be boosted, use scoring profiles instead.

The Free tier’s 3 index limit is less than the 5 indexes Umbraco 19 creates, see Azure Search Limits. To stay on the Free tier, disable some indexes or use Elasticsearch for some of them.

The Umbraco 19 version of ExamineX.AzureSearch.Umbraco.Media isn’t published yet. It depends on Umbraco.StorageProviders.AzureBlob for Umbraco 19, and will be published once that ships. When it is:

  • Culture-variant media is skipped.
  • Image analysis runs again every time a media item is saved.
  • The AI generated fields (OCR text, descriptions and tags) aren’t searchable from the backoffice search yet.