Complete.Org: Mailing Lists: Archives: freeciv-dev: September 2002:
[Freeciv-Dev] Re: CMA failed assertion caused by Freight (PR#1684)
Home

[Freeciv-Dev] Re: CMA failed assertion caused by Freight (PR#1684)

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: "Per I. Mathisen" <per@xxxxxxxxxxx>
Cc: freeciv-dev@xxxxxxxxxxx, bugs@xxxxxxxxxxxxxxxxxxx
Subject: [Freeciv-Dev] Re: CMA failed assertion caused by Freight (PR#1684)
From: rf13@xxxxxxxxxxxxxxxxx
Date: Fri, 27 Sep 2002 03:14:55 -0700

On Thu, Sep 26, 2002 at 09:54:24AM -0700, Per I. Mathisen wrote:
> On Thu, 26 Sep 2002 rf13@xxxxxxxxxxxxxxxxx wrote:
> > > I don't understand why you cannot leave it at sending the new
> > > informationto the client. The use of auto arrange workers here
> > > should be unnecessary for the CMA.
> >
> > The citymap state and the city status has to be
> > consistent. server_set_tile_city doesn't update the city
> > status. auto_arrange_workers is just a bonus here.
> 
> I would say the opposite of a bonus. It is bad and dangerous.

It should be possible to also work with only a refresh_city and
send_city_info. I can't code/test a patch.

> > > As I said earlier, auto arrange worker code should not be called this deep
> > > in the code. That also applies to current case C_TILE_WORKER. Instead, I
> > > suggest displaced workers are turned into elvises, and that a new city
> > > bitvector is used to tellif a tile has been previously displaced, so that
> > > if the tile becomes available again, we remove the last specialist and put
> > > a worker in the tile instead.
> >
> > Sounds complicated if the server uses CM.
> 
> Why?

You have to decide at which events the CM will be activated. The
client side is activated at every city change. If you do the same for
the server the above is unnecessary.

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx


[Prev in Thread] Current Thread [Next in Thread]