[Freeciv-Dev] Re: Goto avoids empty enemy city (PR#1239)

[Gregory Berkolaiko <gberkolaiko@xxxxxxxxxxx>]

 --- Christian Knoke <ChrisK@xxxxxxxx> wrote: 
> On Wed, Jan 23, 2002 at 03:59:41PM -0500, Daniel L Speyer wrote:
> > On Wed, 23 Jan 2002, Christian Knoke wrote:
> > 
> > > CVS 23 JAN 2002 with CMA 3.2-pre3 Gtk+ Linux i386
> > > 
> > > You will say that I'm nitpicking but:
> > > 
> > > Goto avoids an empty enemy city (no ZOC) and takes a
> > > slightly longer way to its goal. If you enter a waypoint
> > > for that city it works.
> > > 
> > > Reproducible. Savegame is here:
> > > 
> > > http://www.enter.de/~c.knoke/bugs/goto-avoids.sav.gz
> > > 
> > > Login as chris (Viking) and send Mech.Inf. to the empty
> > > finn city Aposta. It doesn't go through the empty catalan
> > > city of Berga though that'd save 1/3 of a move point.
> > > 
> > 
> > I think this behaviour is a good thing.  Conquering cities shouldn't
> > be a
> > byproduct of movement.  The alternate route you described would
> > result in
> > an empty Viking city -- possibly a way for the Catalons to gain Labor
> > Union -- even if they've only just researched horseback riding.
> > 
> > If you want to conquer the city, you can decide that easily enough.
> 
> Maybe, but I don't think so. Goto is a command to find the shortest
> way, not to protect the user from him/herself. When I have two empty
> enemy cities in reach I know for sure what to do.

If the majority agrees that it is a bug, it's trivial to fix:
change 0 to 1 in
http://www.freeciv.org/lxr/source/client/goto.c?v=cvs#L365

But I agree with Daniel here, conquering cities on the way is wrong.
Those who wrote goto.c (Thue adapted it from gotohand, I believe) were of
the same opinion if they added this special clause.

And goto code does try to propose a more sensible route, not just a
shortest route.

G.

__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com