Complete.Org: Mailing Lists: Archives: freeciv-dev: July 2005:
[Freeciv-Dev] Re: (PR#13262) RFC: pubserver-in-a-diff
Home

[Freeciv-Dev] Re: (PR#13262) RFC: pubserver-in-a-diff

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: per@xxxxxxxxxxx
Subject: [Freeciv-Dev] Re: (PR#13262) RFC: pubserver-in-a-diff
From: "Mike Kaufman" <kaufman@xxxxxxxxxxxxxxxxxxxxxx>
Date: Sun, 24 Jul 2005 17:34:41 -0700
Reply-to: bugs@xxxxxxxxxxx

<URL: http://bugs.freeciv.org/Ticket/Display.html?id=13262 >

On Sun, Jul 24, 2005 at 07:31:26AM -0700, Per I. Mathisen wrote:
> 
> <URL: http://bugs.freeciv.org/Ticket/Display.html?id=13262 >
> 
> On Wed, 22 Jun 2005, Mike Kaufman wrote:
> 
> How is this code used now? Just copied over server/userdb in the cvs
> checkout on pubserver, or what?

no. the userdb library is a compile time option:

"--disable-client --disable-nls --disable-cvs-deps --without-readline
--with-auth=/home/freeciv/lib/db/libuserdb.a"

that defaults to <freeciv>/server/userdb/libuserdb.a

so you can drop in any module you want. There is a pitfall though, if say a
header [connection.h] changes between the time you compile the library and
compile the server. This is not remotely a perfect solution. In fact, it's
a real PITA.

> Since the server anyway has a direct link to the mysql db here, why not
> just insert the game results into the db directly, instead of going the
> way of parsing the gamelog?

that's possible, but that means putting ranking policy into the server
of which I'm not that big a fan... of course, we have a bit in gamelog.c
already.

> I have no problem with making both authentication and pubserver mode
> require mysql, BTW. In terms of third-party usability, it will be
> light-years ahead of what we (do not) have now anyway.

As I might have already made clear, I'm not all that hot on a "pubserver"
mode in the server. I think we ought to be very careful how much we change
the default codebase so we can run pubserver.

-mike





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