Elasticsearch is adding a new feature called "runtime fields" in 7.10. This meta issue tracks the changes that are required to the ML and Transforms Kibana plugins to support the use of runtime fields.
ML
Transforms

We should switch from obtaining field values from _source and switch to using the new Elasticsearch fields API.
Elasticsearch is adding a new feature called "runtime fields" in 7.10. This meta issue tracks the changes that are required to the ML and Transforms Kibana plugins to support the use of runtime fields.
ML
_source(7.10 - [ML] Add runtime fields support #78700)_source(advanced wizard, jobs list) (7.10 - [ML] Add runtime fields support #78700)_source, failing at theNo examples for this field could be foundstep (7.10 - [ML] Add runtime fields support #78700)_source(7.10 - [ML] Add runtime fields support #78700)models/data_visualizer/data_visualizer.tsgets field examples for non aggregatable fields from_source. Probably worth changing to usefieldsAPI in casetextandgeo_pointare added to runtime types. (7.12 - [ML] Add embedded map to geo_point fields for Data Visualizer #88880)airline_rt: a*) removes the runtime field from the runtime_mappings when it isn't used as an influencer or partitioning field so job fails (7.13 - [ML] Improve check for runtime fields in datafeed query #98289)Transforms
_source(example belowresponsetimeplusis a runtime field) (7.10 - [ML] Add runtime fields support #78700)- Cloning transform job that includes runtime_mappings
- Support creating and cloning transform job where index pattern has runtime fields
- Add advanced runtime mapping editor
We should switch from obtaining field values from
_sourceand switch to using the new Elasticsearch fields API.