Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-12162

Update fts_matcher to use sub-document language-aware tokenization

    XMLWordPrintableJSON

Details

    • Icon: Bug Bug
    • Resolution: Duplicate
    • Icon: Major - P3 Major - P3
    • None
    • 2.5.5
    • Internal Code
    • None
    • ALL

    Description

      Subdocument language specs are used at index time, but not yet applied in a parallel manner in the fts_matcher post-processing. In order to make this work smoothly, we refactor the indexer mod to use an fts_iterator class which can also be used by phrase and negation post-processing.

      Attachments

        Activity

          People

            Unassigned Unassigned
            paul.pedersen Paul Pedersen
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: