Complete.Org: Mailing Lists: Archives: freeciv-dev: April 2000:
[Freeciv-Dev] Re: Pacts patch.

[Freeciv-Dev] Re: Pacts patch.

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: freeciv-dev@xxxxxxxxxxx
Cc: Paul.Thomas@xxxxxxxxxxxxx
Subject: [Freeciv-Dev] Re: Pacts patch.
From: Thue Janus Kristensen <thue@xxxxxxx>
Date: Wed, 12 Apr 2000 20:08:40 +0200

> This is a patch to add various states of "alliance" to FreeCiv.  It's
> against the current CVS (late pm 21 March here in .au).
> It's somewhat tested, but for a change this big of course I'd like
> lots more help playtesting, both for code errors and game balance /
> compatability issues.  All feedback is most welcome.
> Some notes on the implementation:
>  * there are six states of relation, corresponding to (my memory of)
>    CivII rules:
>    "Neutral" is the default state.  The normal ZOC and other rules
>    apply in this state.
>    "Alliance" can be entered from the diplomacy dialog.  If two
>    players have an alliance, they can share territory; ZOC is ignored
>    between the two players; they can occupy each others cities and be
>    transported on each others boats; and they don't interfere with
>    each others production.
>    "Peace" can be entered from the diplomacy dialog or by breaking an
>    alliance.  ZOC is ignored between the players, and you can't
>    attack.
>    "War" is entered when a peace treaty is broken.  Normal ZOC rules
>    and other rules apply.
>    "Ceasefire" can be entered from the diplomacy dialog.  It's like
>    peace, but only lasts a set number of turns before expiring.
>    "No capability" is a special state for unpatched clients and
>    servers.

You have defined the capability as just "pacts" and not "+pacts", butt
when I tried to connect with an unpatched gtk client it crashed.

>  * I've introduced a new capability, "pacts"; the player information
>    packet is modified, and there are a bunch of utility functions in
>    common/player.c for dealing with pacts.
>  * a lot of places in the code that had checks like this:
>      if (player_number_one != player_number_two)
>    or
>      if (player_number_one == player_number_two)
>    have been rewritten to use the new players_at_war() and
>    players_allied() functions, as appropriate.
>  * players have a reputation, which is known to all.  Your reputation
>    is damaged by breaking pacts and diplomatic incidents.
>  * governments with a senate (republic, democracy) may overrule your
>    decision to break a pact, and may even dissolve and plunge you into
>    revolution.  This follows (my memory of) CivII rules.
> There are some known issues and unresolved bugs.  Help with these is
> very much appreciated:
>  * this patch is big and affects both the server and client.
>  * only the Xaw client is modified.  If someone who uses Gtk could
>    provide a matching set of patches for the Gtk client, that would be
>    great.
>  * the "Cancel Pact" button is disabled in too many cases; it should
>    be enabled whenever a ceasefire, peace, or alliance is in effect.
>    At the moment it's enabled whenever the above holds and an embassy
>    has been established.

I simply allowed it to be enabled in all cases where there is a pact, and
not as before where you also had to have an embassy.
Note that the text on the button should change with it's consequences,
and not allways read "cancel pact". At least it should read "cancel

>  * the button in the diplomacy reads '_("Pacts")' instead of 'Pacts'
>    (or the localised equivalent).  This one has me stumped!

It surely looks all right in the code...

>  * when an alliance is broken, resolve_unit_stack needs to be called
>    to resolve stacks of previously-allied units that can no longer
>    share a square.  This isn't yet done.

Well, quick and untested implementation done. Note that we still need to
modify resolve_unit_stack to take into account if a unit is in a
foreign city and if the units it shares stack with are allies, but I am
reluctant to do that right now where there is allready 3 (!) patches
modifying resolve_unit_stack pending. :)

>  * the AI knows nothing about pacts or reputation.
> Falk may recognise some of his handiwork from a year or so ago, when
> we joined forces and I incorporated his patches.  He deserves a lot of
> the credit.  Thanks again, Falk!
> -- paul

Otherwise the patch looks very reasonable.

The patch is updated against latest CVS. (apr 12)
And unlike the previous patch it should actually be possible to apply it
without manually typing in all the filenames :)


Attachment: pacts.diff
Description: Text document

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