Quoting Scott Little on Tue 7 Jan 2003 19:53 to Jan Vermeulen:

 JV>>     You get a new diff each week to update last week's nodelist. As
 JV>> long as you have kept your records in order (and why shouldn't you?),
 JV>> you move them to an auxiliary base record by record, inserting the new
 JV>> nodes as per the diff telling you Axx, copying where it says Cxx and
 JV>> skipping where it says Dxx. Then kill the old file and rename the new
 JV>> one to the old one's name.

 sl> Thats extra complication that wouldn't be necessary if updates
 sl> specified the node.  Line based updates also mean taking the table
 sl> offline (totally unacceptable) to do a complete copy/rename, using
 sl> transactions (may not be available), or a post-diff comparison of
 sl> the old and new tables to determine whats changed and update the
 sl> live table accordingly (*shudder*).

    I've never said you should do that on line. You can prepare all your
files in a separate operation and than swap them. Should take not more than
a few seconds on the kind of system you apparently use...


    -==-