Complete.Org: Mailing Lists: Archives: freeciv-ai: March 2004:
[freeciv-ai] Re: simple historian agent
Home

[freeciv-ai] Re: simple historian agent

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: nikodimka <nikodimka@xxxxxxxxx>
Cc: freeciv-ai@xxxxxxxxxxx
Subject: [freeciv-ai] Re: simple historian agent
From: Raimar Falke <i-freeciv-lists@xxxxxxxxxxxxx>
Date: Fri, 26 Mar 2004 13:28:39 +0100

On Thu, Mar 25, 2004 at 04:08:06PM -0800, nikodimka wrote:
> 
> Raimar,
> 
> The majority of your suggestions accepted with thankfull silence.
> Thank you.
> Some require more clarifications.
> 
> 
> > >   strcpy(simple_historian_agent.name, "Simple Historian");
> > 
> > strlcpy
> 
> What is it? I don't think you mean the cool OpenBSDism.

Yes. Or more precisely the freeciv specific variant: mystrlcpy.

> > >   simple_historian_agent.level = 98; /* well which level this should
> > >                                   be to recieve change
> > >                                   notification after all others? */
> > 
> > Good question.
> 
> I can add
> 
> #define DEEPEST_AGENT_LEVEL 99 /* Don't use the last level unless you know 
> what you're doing */
> 
> into agents.h
> 
> and change the line into simple_historian_agent.level = LAST_AGENT_LEVEL;
> 
> if you have no better suggestion.

I would call it LAST_AGENT_LEVEL.

> > > void sha_tile_update(int index) {
> > >
> > >   int x; int y;
> > >   
> > >   freelog(LOG_DEBUG, "sha got tile: %d\n", index);
> > >   
> > >   index_to_map_pos(&x, &y, index); 
> > >   /* Yes. decode and encode again. I could not find a function :( */
> > >   previous_tiles[index] = *map_get_tile(x, y);
> > > }
> > 
> > IMHO the agents.c should do the callback with x and y and not with
> > index.
> 
> This is because struct call's arg is of int type 
> so I am not sure if you plan to extend this to something other type 
> like union of possible arg types. Hence on the basis of Occam's razor 
> I made it int.

There is no problem adding the callbacks with x and y.

> The second issue is: "in which coordinate system these x,y should
> be?"  the "index" allways has well defined meaning whereas x,y is
> something pretty fuzzy as far as I can tell.

Well you can blame Jason ;)

No honestly there is only coordinate system which 95% of the functions
take. I think Jason called it "natural".

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
  "Heuer's Law: Any feature is a bug unless it can be turned off."


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