dev_docs/key_concepts/data_views.mdx
Note: Kibana index patterns are currently being renamed to data views. There will be some naming inconsistencies until the transition is complete.
Data views (formerly Kibana index patterns or KIPs) are the central method of describing sets of indices for queries. Usage is strongly recommended as a number of high level <DocLink id="kibBuildingBlocks" text="building blocks"/> rely on them. Further, they provide a consistent view of data across a variety Kibana apps.
Data views are defined by a wildcard string (an index pattern) which matches indices, data streams, and index aliases, optionally specify a timestamp field for time series data, and are stored as a <DocLink id="kibDevDocsSavedObjectsIntro" text="saved object"/>. They have a field list which comprises all the fields in matching indices plus fields defined specifically on the data view via runtime fields. Schema-on-read functionality is provided by data view defined runtime fields.
The data view API is made available via the data plugin (data.indexPatterns, soon to be renamed) and most commonly used with <DocLink id="kibDevTutorialDataSearchAndSessions" section="high-level-search" text="SearchSource" />
(data.search.search.SearchSource) to perform queries. SearchSource will apply existing filters and queries from the search bar UI.
Users can create data views via Data view management. Additionally, they can be created through the data view API.
Data views also allow formatters and custom labels to be defined for fields.
hasData: A standardized way to check the empty state for indices and data views.
hasESData: () => Promise<boolean>; // Check to see if ES data existshasDataView: () => Promise<boolean>; // Check to see if any data view exists (managed or user created)hasUserDataView: () => Promise<boolean>; // Check to see if user created data views exists