[Freeciv-Dev] Re: [Patch] [RFC] Path finding version 14

[Raimar Falke <rf13@xxxxxxxxxxxxxxxxx>]

On Mon, Sep 16, 2002 at 11:53:49AM +0100, Gregory Berkolaiko wrote:
> > Problems:
> > 
> > 2: PF: path (at 0xbfffd440) is valid
> > 2: PF:  consists of 4 positions:
> > 2: PF:    0/ 4: (15,30) t/m= 0/ 3 dir=NW/ 6 BMC= 0 EC=0 COP=0
> >                              ^^^^ this is known but it isn't required to 
> > fill this out in TM_NONE mode
> > 2: PF:    1/ 4: (14,29) t/m=-1/-1 dir=W / 9 BMC= 6 EC=291400 COP=291406
> > 2: PF:    2/ 4: (13,29) t/m=-1/-1 dir=SW/ 3 BMC=15 EC=582700 COP=582715
> > 2: PF:    3/ 4: (12,30) t/m=-1/-1 dir=--/-1 BMC=18 EC=883900 COP=883918
> > 2: PF: path (at 0xbfffb7f0) is valid
> > 2: PF:  consists of 4 positions:
> > 2: PF:    0/ 4: (15,30) t/m=-1/-1 dir=NW/ 0 BMC= 0 EC=0 COP=0
> >                                           ^ missing
> > 2: PF:    1/ 4: (14,29) t/m=-1/-1 dir=W / 0 BMC= 6 EC=291400 COP=0
> >                                                        same here ^
> > 2: PF:    2/ 4: (13,29) t/m=-1/-1 dir=SW/ 0 BMC=15 EC=582700 COP=0
> > 2: PF:    3/ 4: (12,30) t/m=-1/-1 dir=--/ 0 BMC=18 EC=883900 COP=0
> 
> I didn't do the things which I don't think should be there.  All 3 above 
> discrepancies are of this sort.  IMO only BMC of the next step might ever 
> be useful.

The first one isn't a problem. And I agree that we may remove the COP
field.

> > 2: PF:  consists of 2 positions:
> > 2: PF:    0/ 2: (62,24) t/m= 0/ 3 dir=NW/ 1 BMC= 0 EC=0 COP=0
> > 2: PF:    1/ 2: (61,23) t/m=-1/-1 dir=--/-1 BMC= 1 EC=236100 COP=236101
> >                                                  ^
> > 2: PF: path (at 0xbfffb7f0) is valid
> > 2: PF:  consists of 2 positions:
> > 2: PF:    0/ 2: (62,24) t/m=-1/-1 dir=NW/ 0 BMC= 0 EC=0 COP=0
> > 2: PF:    1/ 2: (61,23) t/m=-1/-1 dir=--/ 0 BMC= 3 EC=236100 COP=0
> >                                                  ^
> > 
> > Your implementation also doesn't handle sea units.
> 
> It did, but then it was deleted by you :)
> I'll put it in.  It never had support for F_IGTER though (see above).  But 
> adding it shouldn't be a problem.

Good.

> > You should use a larger test case. A random savegame which had CMA
> > problems ;) attached. This should provide a good mix of unit types.
> 
> Savegame is missing :(

Uhh.

> login as who?

Nobody. I set timeout to -1.

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
 "SIGDANGER - The System is likely to crash soon"

testcase.gz
Description: application/gunzip