Follow-up Comment #8, bug #15684 (project freeciv):

One cause of such instability can be if a border source has been destroyed
this turn. For instance, say a buoy on a tile that would have otherwise been
within your borders is pillaged. The tile->source (claimer) will have been
the buoy tile itself, and after the buoy is destroyed there's no cheap way to
determine that tile->source should become the city (or whatever) that would
otherwise have claimed the tile. So it becomes NULL and your borders
(temporarily) shrink.

If you save and re-load at this point, this tile will once again be within
your borders, because there is an expensive way to determine correct borders
-- map_calculate_borders() -- which is called when savefiles are loaded.
(It's also called on turn end, so you would have got the tile back soon
enough anyway.)

The same applies but more so when a fortress is pillaged (and presumably when
a city is destroyed, I haven't tested) -- not only does the border source tile
become unowned, so do any other tiles that were under its influence, and this
state is resolved either by turn-end or save/reload.

I assume that it is too expensive to do a full map_calculate_borders()
whenever a border source is destroyed, even though it doesn't happen that
often?

I haven't checked whether any of this is what's going on in your test case,
though.

    _______________________________________________________

Reply to this item at:

  <http://gna.org/bugs/?15684>

_______________________________________________
  Message sent via/by Gna!
  http://gna.org/


_______________________________________________
Freeciv-dev mailing list
Freeciv-dev@gna.org
https://mail.gna.org/listinfo/freeciv-dev

Reply via email to