[Freeciv-Dev] Re: (PR#7282) connect as orders

["Gregory Berkolaiko" <Gregory.Berkolaiko@xxxxxxxxxxxxx>]

<URL: http://rt.freeciv.org/Ticket/Display.html?id=7282 >


On Mon, 17 May 2004, Jason Short wrote:

>
> <URL: http://rt.freeciv.org/Ticket/Display.html?id=7282 >
>
> Gregory Berkolaiko wrote:
>
> > But adding a get_moveEC to PF is easier and more benefitial.  Can you
> > make a patch?  Raimar, do you object?
>
> I don't think I'm qualified to make a patch.

Just imagine it's mid-term and I am not around ;)
But you are right, one should stick to things one knows to increase
productivity, unless there is a good reason to diversify.

On Mon, 17 May 2004, Jason Short wrote:

> Jason Short wrote:
> > <URL: http://rt.freeciv.org/Ticket/Display.html?id=7282 >
> >
> > Gregory Berkolaiko wrote:
> >
> >
> >>But adding a get_moveEC to PF is easier and more benefitial.  Can you
> >>make a patch?  Raimar, do you object?
> >
> > I don't think I'm qualified to make a patch.
>
> What about the possibility of doing it "right"?
>
> - The EC is 1 for roads, 0 (or NULL callback) for everything else.  PF
> is given an option to consider EC _before_ MC.

This option existed in early Raimars version of PF under the guise of
user-define-COP (cost of path).  It was removed because noone could think
of a situation when it might become necessary.  We can resurrect it.

> - The move_cost is a normal move cost.
>
> - A new callback is added that tells how many extra turns must be spent
> at each tile (or for each move, maybe).  PF integrates this value in
> with the move_cost.
>
> We can then not only calculate the exact "best" path but we will know
> how long it will take (which can be shown to the user).

I still think that introducing a move_EC callback is a better solution
because I remember thinking about ferries and banging my head in
frustration because I badly needed the freedom of such a callback.
However now I cannot remember the exact situation when I needed it.  When
I remember I will explain the reasons and also make the patch.

G.