Complete.Org: Mailing Lists: Archives: freeciv-dev: February 2003:
[Freeciv-Dev] Re: (PR#3013) gen topol safe goto in server/AI
Home

[Freeciv-Dev] Re: (PR#3013) gen topol safe goto in server/AI

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: per@xxxxxxxxxxx, jdorje@xxxxxxxxxxxxxxxxxxxxx
Cc: jdorje@xxxxxxxxxxxxxxxxxxxxx
Subject: [Freeciv-Dev] Re: (PR#3013) gen topol safe goto in server/AI
From: "Gregory Berkolaiko" <Gregory.Berkolaiko@xxxxxxxxxxxx>
Date: Tue, 18 Feb 2003 04:17:30 -0800
Reply-to: rt@xxxxxxxxxxxxxx

On Mon, 17 Feb 2003, Jason Short wrote:

> 
> Here's an updated version of the patch, including PR#3455, a capstring
> update, and removal of cruft in stdinhand.c.
> 

PR#3454 is still not fixed, so it crashes joyfully on client goto 
command.  To see why, have a look at do_unit_goto in client.

Which raises a big question: for user-specified gotos, where do we set and 
where do we unset "go" pointer?

A natural place would be handle_activity_request and clones.  But these 
functions do not discriminate between requests from human/client and AI 
and if they set it for AI as well, the whole patch collapses.

My proposal:
have server access goto_struct directly and AI (including 
server/settler.c) access it only via "go" pointer.

Also, I just noticed now, why are we removing this nice feature of showing 
unit's goto destination?

G.




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