[Freeciv-Dev] Re: Incorrect occupation (PR#628)

[Thue <thue@xxxxxxx>]

First, I realized I needed to check the tiles when a pact changed, since then 
a unit could suddently block a tile it hadn't blocked before.
Then I realized I needed not only to update enemy cities, but also when a 
worker was removed from an enemy city I should show the square as free in 
nearby friendly cities.
I think I need to read this again in the morning, but here it is...Oh, maybe 
I should also test it before commiting it... (hah? ;))
I think it sends city info more than it needs, I may find a way to limit that 
later...

-Thue

On Saturday 16 December 2000 15:53, Thue wrote:
> Here is an implementation of 1)
>
> -Thue
>
> On Fri, 15 Dec 2000 21:50:29 Thue wrote:
> > >The turn after an enemy unit moves off of a city tile, it still is
> >
> > prohibited
> >
> > >from working that tile.
> >
> > Ok, how should this work?
> > 1) Should the unavailable be updated as a unit moves (which would be very
> > easy to do),
> >
> > or
> >
> > 2) should squares that a unit spend the between-turn time in be
> > unavailable
> > until the squares have spend a between-turns update unoccupied?
> >
> > I say we update the squares as units are moved.
> >
> > It is currently handled as 2), but it isn't consistant. If you cause a
> > call
> > to send_adjacent_cities(), fx by moving a worker in a nearby city, any
> > unavailable squares will be updated.
> > I think it is handled as 2) in civ II, but I consider it a bug, and so
> > not
> > worthy of emulation...
> > (Just getting your opinion before making the patch.)
> >
> > -Thue

citizens.diff
Description: Text Data