Update of patch #3346 (project freeciv):
Status: Ready For Test = Done
Open/Closed:Open = Closed
___
Reply to this item at:
Follow-up Comment #16, patch #3346 (project freeciv):
I'm so used to these server notifications that I haven't been able to pay much
attention to them and notice the incoherence.
The problem is solely the message text itself stating the game results in a
draw, as the ranklog is supposed to
Follow-up Comment #15, patch #3346 (project freeciv):
When I /endgame, client popup still says that game ended in draw. Ranklog file
has been written with me as winner.
___
Reply to this item at:
http://gna.org/patch/?3346
Follow-up Comment #13, patch #3346 (project freeciv):
- Removed math.h include from srv_main.c. Changes made to that file could no
way require it. Maybe it was leftover from some earlier experiment you did?
(file #16668)
___
Additional
Update of patch #3346 (project freeciv):
Planned Release: 2.5.0 = 2.4.0, 2.5.0
___
Follow-up Comment #12:
Well, maybe being in right team is enough even if you die. After all, players
are not in teams
Additional Item Attachment, patch #3346 (project freeciv):
File name: RankingFixes.patch Size:7 KB
___
Reply to this item at:
http://gna.org/patch/?3346
___
Message sent via/by
Follow-up Comment #11, patch #3346 (project freeciv):
Oops, i hadn't considered that. I guess things should stay as in comment #8.
___
Reply to this item at:
http://gna.org/patch/?3346
___
Follow-up Comment #8, patch #3346 (project freeciv):
+ if (game.info.turn game.server.end_turn)
+ /* endturn was reached - rank users based on team scores */
+ rank_users(TRUE);
+ else
+ /* game ended for victory conditions - rank users based on survival */
+
Follow-up Comment #9, patch #3346 (project freeciv):
It's not new with behavior with this patch, but I wonder if dead players
should be losers in all cases. Now they are marked as losers in first pass,
but later passes have no aliveness check before the turn players to winners.
Right now, i
Follow-up Comment #10, patch #3346 (project freeciv):
Right now, i can't think of an example of a dead player who
might reasonably have a higher score than the players who
defeated it.
It's team score that's used here, not that of individual player - all members
of the winning team are
Update of patch #3346 (project freeciv):
Category:None = general
Priority: 3 - Low = 5 - Normal
Status:None = Ready For Test
Assigned to:
Follow-up Comment #7, patch #3346 (project freeciv):
I revised the patch and made some improvements;
With the old scorelog file format, it was impossible to parse the scores by
simple means without a small probability to fail, since the log grouped
elements inside parenthesis, but parenthesis
URL:
http://gna.org/patch/?3346
Summary: Ranklog: add player's score to log -- rank even if
game ends for /endturn
Project: Freeciv
Submitted by: davide_at_debian
Submitted on: Wed 27 Jun 2012 11:56:45 AM GMT
Category:
Follow-up Comment #1, patch #3346 (project freeciv):
Also, with this patch, the ranklog file is created even if the
game ends because of endturn is reached
According to rules, player with highest score is the winner in that case. Your
patch does not seem to address that at all - it doesn't
Follow-up Comment #2, patch #3346 (project freeciv):
When the game reaches the endturn limit, no ranklog file would be produced at
all. This patch is meant to always provide a score log when the game ends if
the --ranklog option is passed by the command-line. And, to ease post-game
parsing by
Follow-up Comment #3, patch #3346 (project freeciv):
Yes, but I think current version of your patch does not to produce *correct*
(in respect to winners/losers) ranklog in case of endturn.
___
Reply to this item at:
Follow-up Comment #4, patch #3346 (project freeciv):
Thank you for pointing it out. I was solely concerned about logging the
scores, not the winners/losers classification. I just realized that in case of
/endturn all players would be considered winners, except for space victory.
Follow-up Comment #5, patch #3346 (project freeciv):
I redesigned the rank_users() function. Any ranking is now correct, after
_endturn_, too.
The attached patch to consider is now score_system.diff.
In the function, now I distinguish two conditions:
1) rank_users() is called the traditional
Follow-up Comment #6, patch #3346 (project freeciv):
According to rules, player with highest score is the winner in
that case. Your patch does not seem to address that at all - it
doesn't touch the code that considers all alive players winners
except in case of spacerace victory.
Bug
19 matches
Mail list logo
