sphinx preprocssing

Version 3 - Updated on 14 Nov 2017 at 12:15AM by Joachim Hansen

Description

  1. using RT (real time indexing) I dont need to have a mysql database http://sphinxsearch.com/docs/current/rt-overview.html
  2. I need to have unique ID... maybe I can use the line number as ID... will it not auto generate ID?
  3. Default conservative RAM chunk limit (rt_mem_limit) of 32M can lead to poor performance on bigger indexes, you should raise it to 256..1024M if you're planning to index gigabytes.
  4. The above can be changed in sphinx.conf and in the index
  5. searching can be done with select 
    mysql> SELECT * FROM rtindexname WHERE MATCH('@title test');
  6. I could add in bulk like this 
    mysql> INSERT INTO rt ( id, title ) VALUES ( 3, 'third row' ), ( 4, 'fourth entry' );