just an FYI - don't panic

cengiz at ISI.EDU (Cengiz Alaettinoglu) writes
 * 
 * There is a way to do this with ripe-181 line continuations (at least
 * according to the documentation, the tools (i.e. dbupdate) may not
 * implement this correctly):

The tools understand the documented continuation lines (at least the
tools I have been involved with ;-)

The idea is that you can break up the policy expression anywhere, just
repeat the neighbors and weights and continue on the next line.

 * as-in: ... accept ASxxx AND NOT { 1, 2, 3
 * as-in: ... accept , 4, 5}
 * 
 * Note that ... must be identical on both lines.

Indeed.

-Marten

 * 
 * urtis Villamizar (curtis at ans.net) on May 11:
 * > 
 * > In message <199505111942.PAA25372 at lovefm.reston.mci.net>, Tony Bates write
 * s:
 * > > Curtis,
 * > > 	whay cant these lines be broken up with the current RIPE-181
 * > > line wrap rules ?
 * > > 
 * > > 		--Tony.
 * > 
 * > Because
 * > 
 * >    ASxxx AND NOT { 1 2 3 4 5 }
 * > 
 * > does not equal
 * > 
 * >    ASxxx AND NOT { 1 2 3 }
 * >    ASxxx AND NOT { 4 5 }
 * > 
 * > In other words:
 * > 
 * >    A AND NOT ( B OR C )
 * > 
 * > 	!=
 * > 
 * >    ( A AND NOT B ) OR ( A AND NOT C )
 * > 
 * >    for example:	A = 1 2 3 4 5
 * > 		B = 1 2
 * > 		C = 3
 * > 
 * >    A AND NOT ( B OR C )			-> 4 5
 * > 
 * >    ( A AND NOT B )			-> 3 4 5
 * >    ( A AND NOT C )			-> 1 2 4 5
 * >    ( A AND NOT B ) OR ( A AND NOT C )	-> 1 2 3 4 5
 * > 
 * > Is that clear?
 * > 
 * > The problem is in lists where ASxxx expands to a few thousand networks
 * > and there are perhaps a dozen legitimate exceptions (like DMZs) that
 * > get slightly different treatment.
 * > 
 * > Curtis
 * 
 * 
 * Cengiz
 * 
 * -- 
 * Cengiz Alaettinoglu       Information Sciences Institute
 * (310) 822-1511            University of Southern California
 * http://www.isi.edu/div7/people/cengiz.home



-------- Logged at  Sat May 13 02:37:03 MET DST 1995  ---------