So I'm working on making the amavisnewsql plugin work with the latest  
tables.  (patch coming)

One of the changes that he made in the tables was to unify mailaddr  
and wblist.  So instead of

# ' FROM wblist JOIN mailaddr ON wblist.sid=mailaddr.id'.
#   ' WHERE wblist.rid=? AND mailaddr.email IN (%k)'.
#   ' ORDER BY mailaddr.priority DESC';

It's just

# ' FROM wblist ' WHERE wblist.rid=? AND wblist.email IN (%k)'.
#   ' ORDER BY wblist.priority DESC';

After looking at the the tables, and testing repeatedly with some  
sample datasets, it strikes me that amavisd should probably be doing  
this.  The size of repeated data isn't large enough to cause a  
problem, and dropping the join improves performance on both mysql and  
postgres even with the larger tables.  I used 100k and 1mil table  
sizes for my example with 100 entries per address, and it still came  
out faster.

Is there any reason that this change shouldn't be adopted by amavisd  
in 2.6 ?

-- 
Jo Rhett
Net Consonance : consonant endings by net philanthropy, open source  
and other randomness



-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
AMaViS-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/amavis-user
AMaViS-FAQ:http://www.amavis.org/amavis-faq.php3
AMaViS-HowTos:http://www.amavis.org/howto/

Reply via email to