[SOLR-5276] highlighter working using stemmed tokens from another field and text from another

• Type: Improvement
• Status: Closed
• Priority: Minor
• Resolution: Duplicate
• Component/s: highlighter
• Assignee: Unassigned
• Reporter: Maciej Lizewski
• Created: 26/Sep/13 12:14
• Updated: 09/Mar/17 15:52
• Resolved: 09/Mar/17 15:52

Description

The case is like this:
I have 'content' field with content extracted with tika and several fields for language versions (like content_pl, content_en, content_es, content_ru, etc).
I also use custom langid component which can copy 'content' to serveral content_* fields if it detects more than one language (so those parts are properly stemmed in every language present in text).

Now to use highlighter in such scenario I need to store all those language fields even if their contents is always same as the one in 'content' field.

Would be nice if I could configure language specific fields to be not stored, and configure highlighter to take tokens positions from those fields and apply them to text in 'content' field...
In other words - to say: take tokens from 'content_pl', and prepare highlight based on text in 'content' field.
It could be administrator responsibility to guarantee that mapped fields have same content.

Issue Links

• duplicates: SOLR-1105 Use a different stored field for highlighting