You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As I understand, there is no way to highlight childs in parent/child query (#1764), so this API can also be a workaround to highlight childs after search is done.
The analyze API works because it is a process which happens before data is indexed. Highlighters require indexed terms for query matching and for term statistics. So in the end, you have to index the text.
Its a bit more complex than that. 2/3 of the highlighters require indexed text. The plain highlighter always reanalyzes on the fly, last I checked. It'd technically be possible to implement "here is some text and a query please highlight it" for all highlighters using a MemoryIndex. Its probably overkill though.
Indices already have an _analyze API from which we can see how analyzers work.
We need the "same" API to highlight sample text with a query according to index settings (same thing is already implemented in sphinx ).
This can be useful when we do not need to store text in document, but need to highlight some text.
For now, if we want to highlight sample text, we have to index "tmp" text, highlight it, and then delete it.
The text was updated successfully, but these errors were encountered: