[leafnode-list] adding IHAVE support to leafnode

["Cory C. Albrecht" <cory+spam@xxxxxxxxxxxxxx>]

Hello,

I was just looking at the message.id tree, trying to figure out how it 
is used. What confuses me is why the individual files in that tree seem 
to be near-identical copies of the files in the normal articles tree. 
The only difference appears to be that the file in the message.id tree 
lacks a couple of the header lines.

Why is this? Wouldn't it be more efficitent to use a database hash 
where the message-id is the key and the  value is the newsgroup article 
file? IIRC, most free un*xes come with some form of dbm/ndbm/gdbm, or 
even the old db.h from original BSD would work.

With a hash like that would mean there's no reason to waste diskspace 
by storing articles twice. It would also make adding IHAVE support 
easier - query the hash and you know whether or not to accept it.

I'm interested in learning what advantages the current scheme offers 
compared to using DB functions.

BTW, 1.9.52.rc8 works beautifully for me on OpenBSD 3.4-current (built 
late December 2003).
--
Cory C. Albrecht
http://s94306345.onlinehome.us/
Only a fool tells the bald truth on social occasions.
 -- Robert A. Heinlein


-- 
_______________________________________________
leafnode-list mailing list
leafnode-list@xxxxxxxxxxxxxxxxxxxxxxxxxxxx
http://www.dt.e-technik.uni-dortmund.de/mailman/listinfo/leafnode-list
http://leafnode.sourceforge.net/