Subj : oxp: nodediff fails :(
To   : Martin Foster
From : August Abolins
Date : Sat Feb 08 2020 19:44:49

On 08/02/2020 8:32 a.m., Martin Foster : August Abolins wrote:


 MF>> Finally, finally, finally, I've figured this out .....

 AA> Let me know if I must replace the current Z2 DAILY. 038 that I
 AA> have inorder for the diffs to work.


NEVERMIND.  Looking at the DIFF files and reading about how they work,
the first line of the new diff *has* to match the first line of the
previous full nodelist to which it applies.

The first line of the full .038 from Ambrosia (Z2) does not match the
first line of Tommi's .038.

Apparently, freq'ing for NODELIST from Tommi's sytem produces a DAILY
version (which I only realized a couple weeks ago), which according to
the first line discovery above, will not be a match for next week.

In the past, I had noticed that the order of the zone segments were
different depending on where I grabbed "the" nodelist.  I attributed
that to a "my zone is better than your zone" thing, or to a timing issue
when each zone segment arrived for "global" processing.  I put that
matter behind me.  But it didn't occur to me that they wouldn't work
with "the" scheduled diff later on.

In the past, I thought there was only one "full" global nodelist. Now
there is a version for each zone. I am not sure if that is an
improvement. When did the zone-specific versions start?

--- TB68.4.1/Win7 (the abbr. string!)
 * Origin: nntp://rbb.fidonet.fi - Lake Ylo - Finland (2:221/360.0)

.