Complete.Org: Mailing Lists: Archives: freeciv-dev: February 2002:
[Freeciv-Dev] Re: agent assertion failed (PR#1259)
Home

[Freeciv-Dev] Re: agent assertion failed (PR#1259)

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: freeciv-dev@xxxxxxxxxxx
Cc: bugs@xxxxxxxxxxxxxxxxxxx
Subject: [Freeciv-Dev] Re: agent assertion failed (PR#1259)
From: rf13@xxxxxxxxxxxxxxxxxxxxxx
Date: Mon, 11 Feb 2002 06:22:38 -0800 (PST)

On Mon, Feb 11, 2002 at 01:27:59PM +0000, Ben Webb wrote:
> On Mon, 11 Feb 2002, Raimar Falke wrote:
> 
> > On Mon, Feb 11, 2002 at 03:31:05AM -0800, Christian Knoke wrote:
> > > CVS 11 FEB 2002 12:00 CET GTK+ Linux i386
> > > 
> > > chris@max:~> LANG= civclient -t trident
> > > civclient: agents.c:340: agents_game_joined: Assertion
> > > Has_capability("turn", aconnection.capability)' failed.
> > > Abgebrochen (core dumped)
> > > chris@max:~> 
> > 
> > Are you sure that you used a recent server? The agents won't work with
> > stock 1.12.0.
> 
>       This code shouldn't be triggered with a stock 1.12.0 server 
> anyway, since it's called from client/packhand.c as:-
> 
>   if (strcmp(s_capability, our_capability) == 0) {
>     agents_game_joined();
>     return;
>   }
> 
> i.e. only if the capabilities match. Incidentally, this causes major 
> problems with FreecivAC, or in fact anything that adds extra capabilities. 
> Since the capabilities will no longer match when using (e.g.) a FreecivAC 
> client with a Freeciv server, agents_game_joined() will never be called. 
> Since this would have called freeze(), an assert failure occurs pretty 
> quickly when Freeciv tries to trigger the corresponding thaw(). Oh dear...

Mhhh ... I looks like the test has be to moved:

  if (packet->you_can_join) {
    freelog(LOG_VERBOSE, "join game accept:%s", packet->message);
    aconnection.established = 1;
    game.conn_id = packet->conn_id;
+   agents_game_joined();
  } else {

> (I was going to suggest playing with FREECIV_CAPS to reproduce this 
> behaviour, but unfortunately I can't start a game at all with current CVS. 
> Looks like the same problem that Christian has observed - the reason why 
> it occurs is that at that point in the code both s_capability and 
> our_capability are blank strings - but why, I don't know...)

        Raimar
-- 
 email: rf13@xxxxxxxxxxxxxxxxx
  One nuclear bomb can ruin your whole day.



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