colossus-tracker Mailing List for Colossus (Page 3)

Support Requests item #3558706, was opened at 2012-08-16 20:26
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3558706&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: https://www.google.com/accounts ()
Assigned to: Nobody/Anonymous (nobody)
Summary: can't start local hotseat game

Initial Comment:
Trying to run web start, select network client.  Just running single human with AI opponents.  When I try to start the game with network client, it tries to open a port at 127.0.0.1:26567, but fails  "Connection refused".  Running Win 7 ultimate.

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2012-08-17 01:08

Message:
When you want to play just local human(s) plus AIs, start the game with
"New Game".

The Network client is only needed for the case that somebody else is
hosting the game on his computer and you want your Colossus to be just a
client. 

Giving 127.0.0.1 and the port would indicate, that on your own computer one
(some other than the one you are in) game instance is expecting a network
(=remote) client to connect. 


Admittedly the arrangement of the buttons is rather uninituitive :-(

-Clemens



----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3558706&group_id=1939

Support Requests item #3558706, was opened at 2012-08-16 20:26
Message generated for change (Tracker Item Submitted) made by 
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3558706&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 5
Private: No
Submitted By: https://www.google.com/accounts ()
Assigned to: Nobody/Anonymous (nobody)
Summary: can't start local hotseat game

Initial Comment:
Trying to run web start, select network client.  Just running single human with AI opponents.  When I try to start the game with network client, it tries to open a port at 127.0.0.1:26567, but fails  "Connection refused".  Running Win 7 ultimate.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=201939&aid=3558706&group_id=1939

Bugs item #3541505, was opened at 2012-07-08 20:57
Message generated for change (Tracker Item Submitted) made by nsillito
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3541505&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Babbling Balrog (nsillito)
Assigned to: Nobody/Anonymous (nobody)
Summary: Error attempting to view game in progress

Initial Comment:
> 04:40:03 BitBard: @admin: connected to game 23259, then checked out 
> two other games, went back to 23259, got the game status window up, 
> then an error message
> 04:40:58 BitBard: @admin: logged out, downloaded Public Test build, 
> logged in, tried 23259 again, got the same error message - "problem 
> connecting to game?"
> Thanks, yeah. Does not surprise me. Code was written with the use 
> cases "connect first time as spectator" or "reconnect as alive player 
> within 180 seconds" in mind.
>
> For anything else (e.g. reconnect as spectator, where you had been 
> connected few mins before, or as dead player), might not work. This is 
> because data from a previously-being-connected still lurks around in 
> some variables and so the server code chooses the wrong thing to do 
> when a client connects with same identity as before.
>
> I mean, there were good reasons why I didn't make a official release 
> from the new source code yet - I know this was just a first shot, 
> trying things out and fixing the edges is clearly still needed  :>)
>


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3541505&group_id=1939

Bugs item #3531582, was opened at 2012-06-02 16:24
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3531582&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
>Category: Game Server
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Babbling Balrog (nsillito)
Assigned to: Nobody/Anonymous (nobody)
Summary: Reconnect Functionality not working

Initial Comment:
Game 22371 , Turn 12 , Bitbard lost connection.
Used reconnect. Log window opened and said reconnected.
 
Locked up after end of 1st battle and after 
clicking on second battle. No response from Tawn recieved - fight or flee.
 
Clicked on test connection. Recieved 
confirmation from server after 216 ms - connection to server is ok!
 
Clicking OK would not close the message box, 
Clicking the X in the corner would not close the box.
 
Tried the web client to chat with Tawn - got 
a message box (empty) which I then could not close either., Locked out 
of the game.
 
Killed the game
 
Tawn, did you get any messages? I believe he did and said fight.

I used windows Task Manager to end the hung 
web client. It also ended the game. at least on my end.


----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2012-06-03 03:24

Message:
I'd say a more correct description is: Reconnect occasionally (rarely) gets
stuck, and then hangs up the EDT (GUI) thread.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3531582&group_id=1939

Bugs item #3531582, was opened at 2012-06-02 16:24
Message generated for change (Tracker Item Submitted) made by nsillito
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3531582&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Babbling Balrog (nsillito)
Assigned to: Nobody/Anonymous (nobody)
Summary: Reconnect Functionality not working

Initial Comment:
Game 22371 , Turn 12 , Bitbard lost connection.
Used reconnect. Log window opened and said reconnected.
 
Locked up after end of 1st battle and after 
clicking on second battle. No response from Tawn recieved - fight or flee.
 
Clicked on test connection. Recieved 
confirmation from server after 216 ms - connection to server is ok!
 
Clicking OK would not close the message box, 
Clicking the X in the corner would not close the box.
 
Tried the web client to chat with Tawn - got 
a message box (empty) which I then could not close either., Locked out 
of the game.
 
Killed the game
 
Tawn, did you get any messages? I believe he did and said fight.

I used windows Task Manager to end the hung 
web client. It also ended the game. at least on my end.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3531582&group_id=1939

Feature Requests item #3529880, was opened at 2012-05-25 18:41
Message generated for change (Comment added) made by jeffmatthews
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 3
Private: No
Submitted By: Jeff (jeffmatthews)
Assigned to: Clemens Katzer (cleka)
Summary: Moving creatures in battle

Initial Comment:
Just an idea.   It would be cool to be able to drag and drop a creature to any slot where it is allowed.   Pointing and clicking is okay, too.   One issue that could be made more convenient is where you block the paths of creatures and have to undo moves.   For example, if you have some Cyclopses and Gargoyles, the Cyclopses could be placed on the 2nd hex during battle phase 1, but not if you place the Gargoyles on the first hex.   This forces undo.   There really isn't a need to force an undo to allow this kind of movement.  Requiring undo doesn't promote anything that I can figure.

----------------------------------------------------------------------

>Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 06:16

Message:
"It would lead to confusion about the rules if people were sometimes (but
not always) allowed
to move non-flying creatures over their allies."

That part wouldn't be so bad.   But it would definitely cause confusion
when you move a creature that blocks the path of another ("Blocking
Creature").   You move the other creature now that it is not blocked, and
then, you can't move the Blocking Creature back to what appears to be an
obviously open hex.

Yes, I concede.   You guys were right.

The only place where it would make any sense is in the Phase 1, where you
initially place creatures on the board - which happens to be the reason why
I made this post in the first place.   

Thanks again.

----------------------------------------------------------------------

Comment By: David Ripton (dripton)
Date: 2012-05-31 05:36

Message:
We actually have that kind of speculative movement code in the AI.  But I
think it's a bad idea to use it for regular movement.  It would lead to
confusion about the rules if people were sometimes (but not always) allowed
to move non-flying creatures over their allies.  I think it's better to
have the player move his creatures in the correct order manually.

I think allowing drag and drop for moves (in addition to the current
two-click method) would be nice, but I've never done drag and drop in Swing
so I don't know how easy it is.

----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:30

Message:
OK, sorry for posting more, but my logic was flawed.   It would take more
work as suggested by you guys.   You'd have to also store at the beginning
of a phase to say, for example, "Hex13 is available if creature currently
occupying Hex 13 or in the path of movement to Hex13 is moved out of the
way by the user," and similarly, maybe an additional consideration or 2.  
Can it be done?   Yes.   Is it worth it?   Probably not, unless someone is
bored and wants to take on a logic challenge for fun.

Thanks for your comments.   

----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:17

Message:
BTW, I realize that what I suggested does not eliminate the need for undo
totally.   It does, as was said previously, prevent the need for undos in
"trivial example" situations.  But, you can also see how the logic I
suggested can still apply to creatures even after they have been moved.  
In other words, you can use the available hexes to allow a re-move of a
creature, rather than an undo.   User could re-move a creature all he
wanted until he presses "Done" for battle.

----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:09

Message:

Some keyboard button made it post before I was finished describing what I
thought is the proper and easy algorithm.


1. At the beginning of the turn, before any creature has been moved, what
are all the available hexes for EACH creature? These would INCLUDE even
those hexes that are currently occupied by ANY creature, including the
opponent's creatures.  Basically, for each of MY creatures, I want to know
every hex within creature's moving distance - just like it's an empty
board, but we still respect terrain issues, like bogs.

2. At the moment of moving Creature X, show all hexes that were available
at beginning of turn, except those hexes that are now occupied (not
blocked
- there is a difference).  In other words, just treat it like the creature
can fly to whatever of the available hexes remains available.


----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:03

Message:
"Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only
complicated
to program, it would make it possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to
consider)."


This is the second comment that suggests it is not as easily done as said.

I don't think it is hard to do at all.   What you need to know is this:   

1.   At the beginning of the turn, before any creature has been moved, what
are all the available hexes for EACH creature?   These would INCLUDE

2.  At the moment of moving Creature X, show all hexes that were available
at beginning of turn, except those hexes that are now occupied (not blocked
- there is a difference).

Honestly, guys, I used to be a programmer, but am not much of one any more.
  I am 44, and my programming days ended back in the days of DOS and dBase
III, plus.   Things have changed a bit with event based and object based
programming.   We didn't have that technology back then.

If you guys think it's a good idea and not all too hard, I think it would
be nice, too.   It just avoids having to undo a bunch of moves and redo
them in order to set up your creatures the way you want.  Honesty, I am
happy this game even exists.   I used to play it on the board 25 years ago.
  



----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2012-05-30 23:24

Message:
The current implementation is the easiest way to ensure rules are
fulfilled. Therefore the thing that "requiring undo" promotes is: easy
implementation.

Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only complicated
to program, it would make it  possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to consider).
 

[[ Even worse, not only would the client side, i.e the user's GUI have to
validate the move, the server side, e.g. the public server, needs to
re-validate whether the client did sent a valid move once again [[could
have been a hacked client that allows "illegal" moves]]. CPU time on the
server is extremely precious - it's a single-core virtual server. That's
actually the reason why no AIs can run on the public server ]].

This feature here has a really bad "value vs. cost" ratio. I can ad hoc
figure out 25 things which would me more useful and less work ;-)
IMHO, resources to implement a "nice to have" thing is typically the most
precious resource in open source projects, in Colossus even more than
others.... - except if the requester is a possible resource itself.
So, if you are willing to roll up the sleeves to implement it yourself
you're welcome :)

Besides, I'll move this to feature requests (it's not a bug).


I personally also am not very keen on drag'n'drop for Colossus. 
1) I don't like it much myself in general (getting a cramp in the fingers
easily), 
2) Colossus isn't using drag'n'drop anywhere, so it's 
  a) more work to get it implemented [can't just copy-paste from somewhere
else inside Colossus] and 
  b) it would mean inconsistency in the way-of-usage.
Again - I don't mind if you do it, as long as it stays an optional feature
and the 2-clicks-mode stays the default behavior. :>)


----------------------------------------------------------------------

Comment By: Tim Hoban (tjhoban)
Date: 2012-05-26 17:06

Message:
If you think about this beyond the trivial example you have provided, I
think you will agree that any code to validate this behaviour rapidly gets
fairly involved.




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Feature Requests item #3529880, was opened at 2012-05-25 18:41
Message generated for change (Comment added) made by dripton
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 3
Private: No
Submitted By: Jeff (jeffmatthews)
Assigned to: Clemens Katzer (cleka)
Summary: Moving creatures in battle

Initial Comment:
Just an idea.   It would be cool to be able to drag and drop a creature to any slot where it is allowed.   Pointing and clicking is okay, too.   One issue that could be made more convenient is where you block the paths of creatures and have to undo moves.   For example, if you have some Cyclopses and Gargoyles, the Cyclopses could be placed on the 2nd hex during battle phase 1, but not if you place the Gargoyles on the first hex.   This forces undo.   There really isn't a need to force an undo to allow this kind of movement.  Requiring undo doesn't promote anything that I can figure.

----------------------------------------------------------------------

>Comment By: David Ripton (dripton)
Date: 2012-05-31 05:36

Message:
We actually have that kind of speculative movement code in the AI.  But I
think it's a bad idea to use it for regular movement.  It would lead to
confusion about the rules if people were sometimes (but not always) allowed
to move non-flying creatures over their allies.  I think it's better to
have the player move his creatures in the correct order manually.

I think allowing drag and drop for moves (in addition to the current
two-click method) would be nice, but I've never done drag and drop in Swing
so I don't know how easy it is.

----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:30

Message:
OK, sorry for posting more, but my logic was flawed.   It would take more
work as suggested by you guys.   You'd have to also store at the beginning
of a phase to say, for example, "Hex13 is available if creature currently
occupying Hex 13 or in the path of movement to Hex13 is moved out of the
way by the user," and similarly, maybe an additional consideration or 2.  
Can it be done?   Yes.   Is it worth it?   Probably not, unless someone is
bored and wants to take on a logic challenge for fun.

Thanks for your comments.   

----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:17

Message:
BTW, I realize that what I suggested does not eliminate the need for undo
totally.   It does, as was said previously, prevent the need for undos in
"trivial example" situations.  But, you can also see how the logic I
suggested can still apply to creatures even after they have been moved.  
In other words, you can use the available hexes to allow a re-move of a
creature, rather than an undo.   User could re-move a creature all he
wanted until he presses "Done" for battle.

----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:09

Message:

Some keyboard button made it post before I was finished describing what I
thought is the proper and easy algorithm.


1. At the beginning of the turn, before any creature has been moved, what
are all the available hexes for EACH creature? These would INCLUDE even
those hexes that are currently occupied by ANY creature, including the
opponent's creatures.  Basically, for each of MY creatures, I want to know
every hex within creature's moving distance - just like it's an empty
board, but we still respect terrain issues, like bogs.

2. At the moment of moving Creature X, show all hexes that were available
at beginning of turn, except those hexes that are now occupied (not
blocked
- there is a difference).  In other words, just treat it like the creature
can fly to whatever of the available hexes remains available.


----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:03

Message:
"Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only
complicated
to program, it would make it possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to
consider)."


This is the second comment that suggests it is not as easily done as said.

I don't think it is hard to do at all.   What you need to know is this:   

1.   At the beginning of the turn, before any creature has been moved, what
are all the available hexes for EACH creature?   These would INCLUDE

2.  At the moment of moving Creature X, show all hexes that were available
at beginning of turn, except those hexes that are now occupied (not blocked
- there is a difference).

Honestly, guys, I used to be a programmer, but am not much of one any more.
  I am 44, and my programming days ended back in the days of DOS and dBase
III, plus.   Things have changed a bit with event based and object based
programming.   We didn't have that technology back then.

If you guys think it's a good idea and not all too hard, I think it would
be nice, too.   It just avoids having to undo a bunch of moves and redo
them in order to set up your creatures the way you want.  Honesty, I am
happy this game even exists.   I used to play it on the board 25 years ago.
  



----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2012-05-30 23:24

Message:
The current implementation is the easiest way to ensure rules are
fulfilled. Therefore the thing that "requiring undo" promotes is: easy
implementation.

Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only complicated
to program, it would make it  possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to consider).
 

[[ Even worse, not only would the client side, i.e the user's GUI have to
validate the move, the server side, e.g. the public server, needs to
re-validate whether the client did sent a valid move once again [[could
have been a hacked client that allows "illegal" moves]]. CPU time on the
server is extremely precious - it's a single-core virtual server. That's
actually the reason why no AIs can run on the public server ]].

This feature here has a really bad "value vs. cost" ratio. I can ad hoc
figure out 25 things which would me more useful and less work ;-)
IMHO, resources to implement a "nice to have" thing is typically the most
precious resource in open source projects, in Colossus even more than
others.... - except if the requester is a possible resource itself.
So, if you are willing to roll up the sleeves to implement it yourself
you're welcome :)

Besides, I'll move this to feature requests (it's not a bug).


I personally also am not very keen on drag'n'drop for Colossus. 
1) I don't like it much myself in general (getting a cramp in the fingers
easily), 
2) Colossus isn't using drag'n'drop anywhere, so it's 
  a) more work to get it implemented [can't just copy-paste from somewhere
else inside Colossus] and 
  b) it would mean inconsistency in the way-of-usage.
Again - I don't mind if you do it, as long as it stays an optional feature
and the 2-clicks-mode stays the default behavior. :>)


----------------------------------------------------------------------

Comment By: Tim Hoban (tjhoban)
Date: 2012-05-26 17:06

Message:
If you think about this beyond the trivial example you have provided, I
think you will agree that any code to validate this behaviour rapidly gets
fairly involved.




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Feature Requests item #3529880, was opened at 2012-05-25 18:41
Message generated for change (Comment added) made by jeffmatthews
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 3
Private: No
Submitted By: Jeff (jeffmatthews)
Assigned to: Clemens Katzer (cleka)
Summary: Moving creatures in battle

Initial Comment:
Just an idea.   It would be cool to be able to drag and drop a creature to any slot where it is allowed.   Pointing and clicking is okay, too.   One issue that could be made more convenient is where you block the paths of creatures and have to undo moves.   For example, if you have some Cyclopses and Gargoyles, the Cyclopses could be placed on the 2nd hex during battle phase 1, but not if you place the Gargoyles on the first hex.   This forces undo.   There really isn't a need to force an undo to allow this kind of movement.  Requiring undo doesn't promote anything that I can figure.

----------------------------------------------------------------------

>Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:30

Message:
OK, sorry for posting more, but my logic was flawed.   It would take more
work as suggested by you guys.   You'd have to also store at the beginning
of a phase to say, for example, "Hex13 is available if creature currently
occupying Hex 13 or in the path of movement to Hex13 is moved out of the
way by the user," and similarly, maybe an additional consideration or 2.  
Can it be done?   Yes.   Is it worth it?   Probably not, unless someone is
bored and wants to take on a logic challenge for fun.

Thanks for your comments.   

----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:17

Message:
BTW, I realize that what I suggested does not eliminate the need for undo
totally.   It does, as was said previously, prevent the need for undos in
"trivial example" situations.  But, you can also see how the logic I
suggested can still apply to creatures even after they have been moved.  
In other words, you can use the available hexes to allow a re-move of a
creature, rather than an undo.   User could re-move a creature all he
wanted until he presses "Done" for battle.

----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:09

Message:

Some keyboard button made it post before I was finished describing what I
thought is the proper and easy algorithm.


1. At the beginning of the turn, before any creature has been moved, what
are all the available hexes for EACH creature? These would INCLUDE even
those hexes that are currently occupied by ANY creature, including the
opponent's creatures.  Basically, for each of MY creatures, I want to know
every hex within creature's moving distance - just like it's an empty
board, but we still respect terrain issues, like bogs.

2. At the moment of moving Creature X, show all hexes that were available
at beginning of turn, except those hexes that are now occupied (not
blocked
- there is a difference).  In other words, just treat it like the creature
can fly to whatever of the available hexes remains available.


----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:03

Message:
"Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only
complicated
to program, it would make it possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to
consider)."


This is the second comment that suggests it is not as easily done as said.

I don't think it is hard to do at all.   What you need to know is this:   

1.   At the beginning of the turn, before any creature has been moved, what
are all the available hexes for EACH creature?   These would INCLUDE

2.  At the moment of moving Creature X, show all hexes that were available
at beginning of turn, except those hexes that are now occupied (not blocked
- there is a difference).

Honestly, guys, I used to be a programmer, but am not much of one any more.
  I am 44, and my programming days ended back in the days of DOS and dBase
III, plus.   Things have changed a bit with event based and object based
programming.   We didn't have that technology back then.

If you guys think it's a good idea and not all too hard, I think it would
be nice, too.   It just avoids having to undo a bunch of moves and redo
them in order to set up your creatures the way you want.  Honesty, I am
happy this game even exists.   I used to play it on the board 25 years ago.
  



----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2012-05-30 23:24

Message:
The current implementation is the easiest way to ensure rules are
fulfilled. Therefore the thing that "requiring undo" promotes is: easy
implementation.

Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only complicated
to program, it would make it  possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to consider).
 

[[ Even worse, not only would the client side, i.e the user's GUI have to
validate the move, the server side, e.g. the public server, needs to
re-validate whether the client did sent a valid move once again [[could
have been a hacked client that allows "illegal" moves]]. CPU time on the
server is extremely precious - it's a single-core virtual server. That's
actually the reason why no AIs can run on the public server ]].

This feature here has a really bad "value vs. cost" ratio. I can ad hoc
figure out 25 things which would me more useful and less work ;-)
IMHO, resources to implement a "nice to have" thing is typically the most
precious resource in open source projects, in Colossus even more than
others.... - except if the requester is a possible resource itself.
So, if you are willing to roll up the sleeves to implement it yourself
you're welcome :)

Besides, I'll move this to feature requests (it's not a bug).


I personally also am not very keen on drag'n'drop for Colossus. 
1) I don't like it much myself in general (getting a cramp in the fingers
easily), 
2) Colossus isn't using drag'n'drop anywhere, so it's 
  a) more work to get it implemented [can't just copy-paste from somewhere
else inside Colossus] and 
  b) it would mean inconsistency in the way-of-usage.
Again - I don't mind if you do it, as long as it stays an optional feature
and the 2-clicks-mode stays the default behavior. :>)


----------------------------------------------------------------------

Comment By: Tim Hoban (tjhoban)
Date: 2012-05-26 17:06

Message:
If you think about this beyond the trivial example you have provided, I
think you will agree that any code to validate this behaviour rapidly gets
fairly involved.




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Feature Requests item #3529880, was opened at 2012-05-25 18:41
Message generated for change (Comment added) made by jeffmatthews
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 3
Private: No
Submitted By: Jeff (jeffmatthews)
Assigned to: Clemens Katzer (cleka)
Summary: Moving creatures in battle

Initial Comment:
Just an idea.   It would be cool to be able to drag and drop a creature to any slot where it is allowed.   Pointing and clicking is okay, too.   One issue that could be made more convenient is where you block the paths of creatures and have to undo moves.   For example, if you have some Cyclopses and Gargoyles, the Cyclopses could be placed on the 2nd hex during battle phase 1, but not if you place the Gargoyles on the first hex.   This forces undo.   There really isn't a need to force an undo to allow this kind of movement.  Requiring undo doesn't promote anything that I can figure.

----------------------------------------------------------------------

>Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:17

Message:
BTW, I realize that what I suggested does not eliminate the need for undo
totally.   It does, as was said previously, prevent the need for undos in
"trivial example" situations.  But, you can also see how the logic I
suggested can still apply to creatures even after they have been moved.  
In other words, you can use the available hexes to allow a re-move of a
creature, rather than an undo.   User could re-move a creature all he
wanted until he presses "Done" for battle.

----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:09

Message:

Some keyboard button made it post before I was finished describing what I
thought is the proper and easy algorithm.


1. At the beginning of the turn, before any creature has been moved, what
are all the available hexes for EACH creature? These would INCLUDE even
those hexes that are currently occupied by ANY creature, including the
opponent's creatures.  Basically, for each of MY creatures, I want to know
every hex within creature's moving distance - just like it's an empty
board, but we still respect terrain issues, like bogs.

2. At the moment of moving Creature X, show all hexes that were available
at beginning of turn, except those hexes that are now occupied (not
blocked
- there is a difference).  In other words, just treat it like the creature
can fly to whatever of the available hexes remains available.


----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:03

Message:
"Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only
complicated
to program, it would make it possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to
consider)."


This is the second comment that suggests it is not as easily done as said.

I don't think it is hard to do at all.   What you need to know is this:   

1.   At the beginning of the turn, before any creature has been moved, what
are all the available hexes for EACH creature?   These would INCLUDE

2.  At the moment of moving Creature X, show all hexes that were available
at beginning of turn, except those hexes that are now occupied (not blocked
- there is a difference).

Honestly, guys, I used to be a programmer, but am not much of one any more.
  I am 44, and my programming days ended back in the days of DOS and dBase
III, plus.   Things have changed a bit with event based and object based
programming.   We didn't have that technology back then.

If you guys think it's a good idea and not all too hard, I think it would
be nice, too.   It just avoids having to undo a bunch of moves and redo
them in order to set up your creatures the way you want.  Honesty, I am
happy this game even exists.   I used to play it on the board 25 years ago.
  



----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2012-05-30 23:24

Message:
The current implementation is the easiest way to ensure rules are
fulfilled. Therefore the thing that "requiring undo" promotes is: easy
implementation.

Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only complicated
to program, it would make it  possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to consider).
 

[[ Even worse, not only would the client side, i.e the user's GUI have to
validate the move, the server side, e.g. the public server, needs to
re-validate whether the client did sent a valid move once again [[could
have been a hacked client that allows "illegal" moves]]. CPU time on the
server is extremely precious - it's a single-core virtual server. That's
actually the reason why no AIs can run on the public server ]].

This feature here has a really bad "value vs. cost" ratio. I can ad hoc
figure out 25 things which would me more useful and less work ;-)
IMHO, resources to implement a "nice to have" thing is typically the most
precious resource in open source projects, in Colossus even more than
others.... - except if the requester is a possible resource itself.
So, if you are willing to roll up the sleeves to implement it yourself
you're welcome :)

Besides, I'll move this to feature requests (it's not a bug).


I personally also am not very keen on drag'n'drop for Colossus. 
1) I don't like it much myself in general (getting a cramp in the fingers
easily), 
2) Colossus isn't using drag'n'drop anywhere, so it's 
  a) more work to get it implemented [can't just copy-paste from somewhere
else inside Colossus] and 
  b) it would mean inconsistency in the way-of-usage.
Again - I don't mind if you do it, as long as it stays an optional feature
and the 2-clicks-mode stays the default behavior. :>)


----------------------------------------------------------------------

Comment By: Tim Hoban (tjhoban)
Date: 2012-05-26 17:06

Message:
If you think about this beyond the trivial example you have provided, I
think you will agree that any code to validate this behaviour rapidly gets
fairly involved.




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Feature Requests item #3529880, was opened at 2012-05-25 18:41
Message generated for change (Comment added) made by jeffmatthews
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 3
Private: No
Submitted By: Jeff (jeffmatthews)
Assigned to: Clemens Katzer (cleka)
Summary: Moving creatures in battle

Initial Comment:
Just an idea.   It would be cool to be able to drag and drop a creature to any slot where it is allowed.   Pointing and clicking is okay, too.   One issue that could be made more convenient is where you block the paths of creatures and have to undo moves.   For example, if you have some Cyclopses and Gargoyles, the Cyclopses could be placed on the 2nd hex during battle phase 1, but not if you place the Gargoyles on the first hex.   This forces undo.   There really isn't a need to force an undo to allow this kind of movement.  Requiring undo doesn't promote anything that I can figure.

----------------------------------------------------------------------

>Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:09

Message:

Some keyboard button made it post before I was finished describing what I
thought is the proper and easy algorithm.


1. At the beginning of the turn, before any creature has been moved, what
are all the available hexes for EACH creature? These would INCLUDE even
those hexes that are currently occupied by ANY creature, including the
opponent's creatures.  Basically, for each of MY creatures, I want to know
every hex within creature's moving distance - just like it's an empty
board, but we still respect terrain issues, like bogs.

2. At the moment of moving Creature X, show all hexes that were available
at beginning of turn, except those hexes that are now occupied (not
blocked
- there is a difference).  In other words, just treat it like the creature
can fly to whatever of the available hexes remains available.


----------------------------------------------------------------------

Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:03

Message:
"Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only
complicated
to program, it would make it possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to
consider)."


This is the second comment that suggests it is not as easily done as said.

I don't think it is hard to do at all.   What you need to know is this:   

1.   At the beginning of the turn, before any creature has been moved, what
are all the available hexes for EACH creature?   These would INCLUDE

2.  At the moment of moving Creature X, show all hexes that were available
at beginning of turn, except those hexes that are now occupied (not blocked
- there is a difference).

Honestly, guys, I used to be a programmer, but am not much of one any more.
  I am 44, and my programming days ended back in the days of DOS and dBase
III, plus.   Things have changed a bit with event based and object based
programming.   We didn't have that technology back then.

If you guys think it's a good idea and not all too hard, I think it would
be nice, too.   It just avoids having to undo a bunch of moves and redo
them in order to set up your creatures the way you want.  Honesty, I am
happy this game even exists.   I used to play it on the board 25 years ago.
  



----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2012-05-30 23:24

Message:
The current implementation is the easiest way to ensure rules are
fulfilled. Therefore the thing that "requiring undo" promotes is: easy
implementation.

Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only complicated
to program, it would make it  possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to consider).
 

[[ Even worse, not only would the client side, i.e the user's GUI have to
validate the move, the server side, e.g. the public server, needs to
re-validate whether the client did sent a valid move once again [[could
have been a hacked client that allows "illegal" moves]]. CPU time on the
server is extremely precious - it's a single-core virtual server. That's
actually the reason why no AIs can run on the public server ]].

This feature here has a really bad "value vs. cost" ratio. I can ad hoc
figure out 25 things which would me more useful and less work ;-)
IMHO, resources to implement a "nice to have" thing is typically the most
precious resource in open source projects, in Colossus even more than
others.... - except if the requester is a possible resource itself.
So, if you are willing to roll up the sleeves to implement it yourself
you're welcome :)

Besides, I'll move this to feature requests (it's not a bug).


I personally also am not very keen on drag'n'drop for Colossus. 
1) I don't like it much myself in general (getting a cramp in the fingers
easily), 
2) Colossus isn't using drag'n'drop anywhere, so it's 
  a) more work to get it implemented [can't just copy-paste from somewhere
else inside Colossus] and 
  b) it would mean inconsistency in the way-of-usage.
Again - I don't mind if you do it, as long as it stays an optional feature
and the 2-clicks-mode stays the default behavior. :>)


----------------------------------------------------------------------

Comment By: Tim Hoban (tjhoban)
Date: 2012-05-26 17:06

Message:
If you think about this beyond the trivial example you have provided, I
think you will agree that any code to validate this behaviour rapidly gets
fairly involved.




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Feature Requests item #3529880, was opened at 2012-05-25 18:41
Message generated for change (Comment added) made by jeffmatthews
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Priority: 3
Private: No
Submitted By: Jeff (jeffmatthews)
Assigned to: Clemens Katzer (cleka)
Summary: Moving creatures in battle

Initial Comment:
Just an idea.   It would be cool to be able to drag and drop a creature to any slot where it is allowed.   Pointing and clicking is okay, too.   One issue that could be made more convenient is where you block the paths of creatures and have to undo moves.   For example, if you have some Cyclopses and Gargoyles, the Cyclopses could be placed on the 2nd hex during battle phase 1, but not if you place the Gargoyles on the first hex.   This forces undo.   There really isn't a need to force an undo to allow this kind of movement.  Requiring undo doesn't promote anything that I can figure.

----------------------------------------------------------------------

>Comment By: Jeff (jeffmatthews)
Date: 2012-05-31 05:03

Message:
"Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only
complicated
to program, it would make it possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to
consider)."


This is the second comment that suggests it is not as easily done as said.

I don't think it is hard to do at all.   What you need to know is this:   

1.   At the beginning of the turn, before any creature has been moved, what
are all the available hexes for EACH creature?   These would INCLUDE

2.  At the moment of moving Creature X, show all hexes that were available
at beginning of turn, except those hexes that are now occupied (not blocked
- there is a difference).

Honestly, guys, I used to be a programmer, but am not much of one any more.
  I am 44, and my programming days ended back in the days of DOS and dBase
III, plus.   Things have changed a bit with event based and object based
programming.   We didn't have that technology back then.

If you guys think it's a good idea and not all too hard, I think it would
be nice, too.   It just avoids having to undo a bunch of moves and redo
them in order to set up your creatures the way you want.  Honesty, I am
happy this game even exists.   I used to play it on the board 25 years ago.
  



----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2012-05-30 23:24

Message:
The current implementation is the easiest way to ensure rules are
fulfilled. Therefore the thing that "requiring undo" promotes is: easy
implementation.

Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only complicated
to program, it would make it  possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to consider).
 

[[ Even worse, not only would the client side, i.e the user's GUI have to
validate the move, the server side, e.g. the public server, needs to
re-validate whether the client did sent a valid move once again [[could
have been a hacked client that allows "illegal" moves]]. CPU time on the
server is extremely precious - it's a single-core virtual server. That's
actually the reason why no AIs can run on the public server ]].

This feature here has a really bad "value vs. cost" ratio. I can ad hoc
figure out 25 things which would me more useful and less work ;-)
IMHO, resources to implement a "nice to have" thing is typically the most
precious resource in open source projects, in Colossus even more than
others.... - except if the requester is a possible resource itself.
So, if you are willing to roll up the sleeves to implement it yourself
you're welcome :)

Besides, I'll move this to feature requests (it's not a bug).


I personally also am not very keen on drag'n'drop for Colossus. 
1) I don't like it much myself in general (getting a cramp in the fingers
easily), 
2) Colossus isn't using drag'n'drop anywhere, so it's 
  a) more work to get it implemented [can't just copy-paste from somewhere
else inside Colossus] and 
  b) it would mean inconsistency in the way-of-usage.
Again - I don't mind if you do it, as long as it stays an optional feature
and the 2-clicks-mode stays the default behavior. :>)


----------------------------------------------------------------------

Comment By: Tim Hoban (tjhoban)
Date: 2012-05-26 17:06

Message:
If you think about this beyond the trivial example you have provided, I
think you will agree that any code to validate this behaviour rapidly gets
fairly involved.




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Feature Requests item #3529880, was opened at 2012-05-25 18:41
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
>Category: None
Group: None
Status: Open
>Priority: 3
Private: No
Submitted By: Jeff (jeffmatthews)
Assigned to: Clemens Katzer (cleka)
Summary: Moving creatures in battle

Initial Comment:
Just an idea.   It would be cool to be able to drag and drop a creature to any slot where it is allowed.   Pointing and clicking is okay, too.   One issue that could be made more convenient is where you block the paths of creatures and have to undo moves.   For example, if you have some Cyclopses and Gargoyles, the Cyclopses could be placed on the 2nd hex during battle phase 1, but not if you place the Gargoyles on the first hex.   This forces undo.   There really isn't a need to force an undo to allow this kind of movement.  Requiring undo doesn't promote anything that I can figure.

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2012-05-30 23:24

Message:
The current implementation is the easiest way to ensure rules are
fulfilled. Therefore the thing that "requiring undo" promotes is: easy
implementation.

Any kind of auto-logic like "ok, that cyc could go there IF user wouldn't
have moved gargoyle yet but would do it afterwards" is not only complicated
to program, it would make it  possibly unbearably slow (possibly
exponentially-behaving algorithm growing with nr of creatures to consider).
 

[[ Even worse, not only would the client side, i.e the user's GUI have to
validate the move, the server side, e.g. the public server, needs to
re-validate whether the client did sent a valid move once again [[could
have been a hacked client that allows "illegal" moves]]. CPU time on the
server is extremely precious - it's a single-core virtual server. That's
actually the reason why no AIs can run on the public server ]].

This feature here has a really bad "value vs. cost" ratio. I can ad hoc
figure out 25 things which would me more useful and less work ;-)
IMHO, resources to implement a "nice to have" thing is typically the most
precious resource in open source projects, in Colossus even more than
others.... - except if the requester is a possible resource itself.
So, if you are willing to roll up the sleeves to implement it yourself
you're welcome :)

Besides, I'll move this to feature requests (it's not a bug).


I personally also am not very keen on drag'n'drop for Colossus. 
1) I don't like it much myself in general (getting a cramp in the fingers
easily), 
2) Colossus isn't using drag'n'drop anywhere, so it's 
  a) more work to get it implemented [can't just copy-paste from somewhere
else inside Colossus] and 
  b) it would mean inconsistency in the way-of-usage.
Again - I don't mind if you do it, as long as it stays an optional feature
and the 2-clicks-mode stays the default behavior. :>)


----------------------------------------------------------------------

Comment By: Tim Hoban (tjhoban)
Date: 2012-05-26 17:06

Message:
If you think about this beyond the trivial example you have provided, I
think you will agree that any code to validate this behaviour rapidly gets
fairly involved.




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=351939&aid=3529880&group_id=1939

Bugs item #3529880, was opened at 2012-05-25 18:41
Message generated for change (Comment added) made by tjhoban
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3529880&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: GUI
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Jeff (jeffmatthews)
Assigned to: Clemens Katzer (cleka)
Summary: Moving creatures in battle

Initial Comment:
Just an idea.   It would be cool to be able to drag and drop a creature to any slot where it is allowed.   Pointing and clicking is okay, too.   One issue that could be made more convenient is where you block the paths of creatures and have to undo moves.   For example, if you have some Cyclopses and Gargoyles, the Cyclopses could be placed on the 2nd hex during battle phase 1, but not if you place the Gargoyles on the first hex.   This forces undo.   There really isn't a need to force an undo to allow this kind of movement.  Requiring undo doesn't promote anything that I can figure.

----------------------------------------------------------------------

Comment By: Tim Hoban (tjhoban)
Date: 2012-05-26 17:06

Message:
If you think about this beyond the trivial example you have provided, I
think you will agree that any code to validate this behaviour rapidly gets
fairly involved.




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3529880&group_id=1939

Bugs item #3529880, was opened at 2012-05-25 18:41
Message generated for change (Tracker Item Submitted) made by jeffmatthews
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3529880&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: GUI
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Jeff (jeffmatthews)
Assigned to: Clemens Katzer (cleka)
Summary: Moving creatures in battle

Initial Comment:
Just an idea.   It would be cool to be able to drag and drop a creature to any slot where it is allowed.   Pointing and clicking is okay, too.   One issue that could be made more convenient is where you block the paths of creatures and have to undo moves.   For example, if you have some Cyclopses and Gargoyles, the Cyclopses could be placed on the 2nd hex during battle phase 1, but not if you place the Gargoyles on the first hex.   This forces undo.   There really isn't a need to force an undo to allow this kind of movement.  Requiring undo doesn't promote anything that I can figure.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3529880&group_id=1939

Bugs item #3520872, was opened at 2012-04-23 23:45
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3520872&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: GUI
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: John Bauer (angledluffa)
Assigned to: Clemens Katzer (cleka)
Summary: Connection log window spam

Initial Comment:
There's an apparent bug which can get triggered when someone who is
eliminated then has connection problems.  It is now constantly trying
to reconnect to him, which adds a line to the Connection Log Window,
which takes the focus from wherever else you are in the game.  Takes
away hot keys, the right click menus and basically any ability to type
in the chat window.  Perhaps the connection log window shouldn't grab
focus automatically.

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2012-05-02 02:37

Message:
I was not able to reproduce that effect. It let's me, as a dead player,
reconnect; works like a charm.

Also there's no "every second" telling something for "tries to reconnect
case" by itself. There's a "every second telling something" in the case if
writing data to that client has issues. That might be looked into.

I'd not like to remove the "take focus" fully - at least first time there's
something to report it would be good to come up.

But perhaps if there's many things to be printed, upon 3rd time or so, or
"more than 5 within last 3 seconds", stop grabbing focus then.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3520872&group_id=1939

Bugs item #3522403, was opened at 2012-04-29 08:07
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3522403&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
>Category: GUI
Group: None
Status: Open
Resolution: None
>Priority: 3
Private: No
Submitted By: https://me.yahoo.com/a/maeGL.Zv ()
Assigned to: Clemens Katzer (cleka)
Summary: Watch muster creatures 

Initial Comment:
I was trying the public test build, watching a game in progress. Sometimes after the legion musters the newly mustered creature escapes and is left behind when the legion moves. See attachment.
Greenbeyer.

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2012-04-30 11:02

Message:
Well the recruit chit remains displayed on the hex, but it's not "escaped"
from the legion. It's just a cosmetic problem that the chit remains visible
on the hex where the legion was at moment of recruiting.

I am aware of this "cosmetic" problem (noticed it shortly after implemented
the watching), but I don't know an easy/good way to fix it.

Technically, the chits remain visible, because "normally" they are cleaned
up for each player on his board when his turn starts, but a spectator
(watcher) does not have an own turn, so this step does not happen. BTW, in
rare cases the recruit chits stay visible also for normal players

You can press "C" (clear recruit chits, or select that item from phase
menu), if they disturb too much. 

As said, I couldn't come up with an easy/nice fix. Either would "clear" all
chits at begin of turn (first player), but then it's "unbalanced" -- for
first player one sees them all the "player turns of that turn number thru",
for last one only that one player's turn. Or clean them whenever one
player's turn start? Rather short time visible then. 

Conceptually, the reason why this all is a problem, is caused by the fact
how the recruit chits are handled/displayed: they are, in fact, not linked
in any form to the legion they belong to (were recruited into), instead
they are just "thrown" onto the hex of that legion, and painted separately
(that's also the reason why the chit hides the size, and why it's not
"easily possible" to fix that; and also, why in reinforcement in battle
case, the recruited chit is displayed on the center of the hex, instead of
centered on the reinforcing legion).

It's afterwards not "easily" trackable to which legion (and thus, to which
player) an existing recruit chit belongs, therefore selective cleanup for
those of some player is not cleanly implementable.

As proper correction (so I think), one would have to fix the recruit chit
handling so that they are attached to the mustering legion and painted
together with it; I remember once I even started some work into that
direction, but it's not a simple task, and since it's more cosmetic  than
really impacting anything, it never came to completion.... if that were
done, one could e.g. for the watcher clear chits "selectively" from the
legions of a player whose turn is now starting.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3522403&group_id=1939

Bugs item #3522403, was opened at 2012-04-29 08:07
Message generated for change (Tracker Item Submitted) made by 
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3522403&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: https://me.yahoo.com/a/maeGL.Zv ()
Assigned to: Clemens Katzer (cleka)
Summary: Watch muster creatures 

Initial Comment:
I was trying the public test build, watching a game in progress. Sometimes after the legion musters the newly mustered creature escapes and is left behind when the legion moves. See attachment.
Greenbeyer.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3522403&group_id=1939

Bugs item #3520872, was opened at 2012-04-23 23:45
Message generated for change (Tracker Item Submitted) made by angledluffa
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3520872&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: GUI
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: John Bauer (angledluffa)
Assigned to: Clemens Katzer (cleka)
Summary: Connection log window spam

Initial Comment:
There's an apparent bug which can get triggered when someone who is
eliminated then has connection problems.  It is now constantly trying
to reconnect to him, which adds a line to the Connection Log Window,
which takes the focus from wherever else you are in the game.  Takes
away hot keys, the right click menus and basically any ability to type
in the chat window.  Perhaps the connection log window shouldn't grab
focus automatically.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3520872&group_id=1939

Bugs item #3516697, was opened at 2012-04-10 18:28
Message generated for change (Comment added) made by sewerstarfish
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3516697&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: None
Status: Open
Resolution: None
>Priority: 1
Private: No
Submitted By: Patrick Bauer (sewerstarfish)
Assigned to: Clemens Katzer (cleka)
Summary: Coward2 Exception

Initial Comment:
The following was recorded while playing Colossus 0.12.0 Build 2011-02-15 19:52:59 by katzer from revision 4995 java version 1.6.0_24 in Windows 7

Woooah! An exception was caught while processing from client Coward2 the input line:
    === doBattleMove ~ 313 ~ B3 ===

Stack trace:
java.util.ConcurrentModificationException
	at java.util.AbstractList$Itr.checkForComodification(Unknown Source)
	at java.util.AbstractList$Itr.next(Unknown Source)
	at net.sf.colossus.server.Server.allTellBattleMove(Server.java:2102)
	at net.sf.colossus.server.BattleServerSide.moveCritterToHexAndInformClients(BattleServerSide.java:1102)
	at net.sf.colossus.server.BattleServerSide.doMove(BattleServerSide.java:1080)
	at net.sf.colossus.server.Server.doBattleMove(Server.java:2088)
	at net.sf.colossus.server.ClientHandler.callMethod(ClientHandler.java:513)
	at net.sf.colossus.server.ClientHandler.doCallMethodInTryBlock(ClientHandler.java:356)
	at net.sf.colossus.server.ClientHandler.processInput(ClientHandler.java:159)
	at net.sf.colossus.server.Server.handleSelectedKeys(Server.java:528)
	at net.sf.colossus.server.Server.waitOnSelector(Server.java:399)
	at net.sf.colossus.server.Server.run(Server.java:227)


Game might be unstable or hang from now on...

----------------------------------------------------------------------

>Comment By: Patrick Bauer (sewerstarfish)
Date: 2012-04-11 13:13

Message:
1) I'm lazy
2) I believe it was while quitting .
3) Sorry not on Linux

You should probably just ignore it.



----------------------------------------------------------------------

Comment By: Patrick Bauer (sewerstarfish)
Date: 2012-04-11 13:13

Message:

This bug tracker item has been closed, since it is very old and probably
not relevant any more.

Colossus has been changed so much internally, that the problem situation
most likely does not exist any more; but even if it would still exist, it
would be hard to fix with the given info, because attached saved games
don't work nowadays any more.

Should someone again come over this bug situation, now a new bug tracker
item should be created with a recent Colossus version.


----------------------------------------------------------------------

Comment By: Clemens Katzer (cleka)
Date: 2012-04-10 21:12

Message:
Hi!

1) You seem to use a rather old release of Colossus. May I ask why?

2) Can you withcertaintly say whether this was "somewhere during the game"
or "possibly when ending the game" ? I've see similar error messages
happening when one does quit or start new game from File menu, while game
is still ongoing full speed (e.g. only AIs left). 
[technically, the Event Dispatcher Thread (executing the "File-Quit" menu
item) starts the cleanup, which removes clients from a list; but in same
time the "main game thread" is still processing input from a client, in
this case involving iterating over the list of all clients to tell them the
battle results. So while it's iterating, it notices that suddenly "someone
else" has removed an item from the list.]

3) if you're on Linux, could you check whether there's a log file like so:

[katzer@dhcppc2 ws]$ ll /tmp | grep Colossus
-rw-r--r--. 1 katzer  everybody   50237 Apr 10 21:41 Colossus0.log
-rw-r--r--. 1 katzer  everybody    1445 Apr 10 20:18 Colossus1.log
-rw-r--r--. 1 katzer  everybody    1445 Apr 10 20:17 Colossus2.log
[katzer@dhcppc2 ws]$ 

The one with number 0 is always newest. If you haven't played "much" since
then, the log file from that game might still exist.


BR,
Clemens


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3516697&group_id=1939

Bugs item #3516697, was opened at 2012-04-10 18:28
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3516697&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Patrick Bauer (sewerstarfish)
Assigned to: Clemens Katzer (cleka)
Summary: Coward2 Exception

Initial Comment:
The following was recorded while playing Colossus 0.12.0 Build 2011-02-15 19:52:59 by katzer from revision 4995 java version 1.6.0_24 in Windows 7

Woooah! An exception was caught while processing from client Coward2 the input line:
    === doBattleMove ~ 313 ~ B3 ===

Stack trace:
java.util.ConcurrentModificationException
	at java.util.AbstractList$Itr.checkForComodification(Unknown Source)
	at java.util.AbstractList$Itr.next(Unknown Source)
	at net.sf.colossus.server.Server.allTellBattleMove(Server.java:2102)
	at net.sf.colossus.server.BattleServerSide.moveCritterToHexAndInformClients(BattleServerSide.java:1102)
	at net.sf.colossus.server.BattleServerSide.doMove(BattleServerSide.java:1080)
	at net.sf.colossus.server.Server.doBattleMove(Server.java:2088)
	at net.sf.colossus.server.ClientHandler.callMethod(ClientHandler.java:513)
	at net.sf.colossus.server.ClientHandler.doCallMethodInTryBlock(ClientHandler.java:356)
	at net.sf.colossus.server.ClientHandler.processInput(ClientHandler.java:159)
	at net.sf.colossus.server.Server.handleSelectedKeys(Server.java:528)
	at net.sf.colossus.server.Server.waitOnSelector(Server.java:399)
	at net.sf.colossus.server.Server.run(Server.java:227)


Game might be unstable or hang from now on...

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2012-04-10 21:12

Message:
Hi!

1) You seem to use a rather old release of Colossus. May I ask why?

2) Can you withcertaintly say whether this was "somewhere during the game"
or "possibly when ending the game" ? I've see similar error messages
happening when one does quit or start new game from File menu, while game
is still ongoing full speed (e.g. only AIs left). 
[technically, the Event Dispatcher Thread (executing the "File-Quit" menu
item) starts the cleanup, which removes clients from a list; but in same
time the "main game thread" is still processing input from a client, in
this case involving iterating over the list of all clients to tell them the
battle results. So while it's iterating, it notices that suddenly "someone
else" has removed an item from the list.]

3) if you're on Linux, could you check whether there's a log file like so:

[katzer@dhcppc2 ws]$ ll /tmp | grep Colossus
-rw-r--r--. 1 katzer  everybody   50237 Apr 10 21:41 Colossus0.log
-rw-r--r--. 1 katzer  everybody    1445 Apr 10 20:18 Colossus1.log
-rw-r--r--. 1 katzer  everybody    1445 Apr 10 20:17 Colossus2.log
[katzer@dhcppc2 ws]$ 

The one with number 0 is always newest. If you haven't played "much" since
then, the log file from that game might still exist.


BR,
Clemens


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3516697&group_id=1939

Bugs item #3516697, was opened at 2012-04-10 18:28
Message generated for change (Tracker Item Submitted) made by sewerstarfish
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3516697&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Patrick Bauer (sewerstarfish)
Assigned to: Clemens Katzer (cleka)
Summary: Coward2 Exception

Initial Comment:
The following was recorded while playing Colossus 0.12.0 Build 2011-02-15 19:52:59 by katzer from revision 4995 java version 1.6.0_24 in Windows 7

Woooah! An exception was caught while processing from client Coward2 the input line:
    === doBattleMove ~ 313 ~ B3 ===

Stack trace:
java.util.ConcurrentModificationException
	at java.util.AbstractList$Itr.checkForComodification(Unknown Source)
	at java.util.AbstractList$Itr.next(Unknown Source)
	at net.sf.colossus.server.Server.allTellBattleMove(Server.java:2102)
	at net.sf.colossus.server.BattleServerSide.moveCritterToHexAndInformClients(BattleServerSide.java:1102)
	at net.sf.colossus.server.BattleServerSide.doMove(BattleServerSide.java:1080)
	at net.sf.colossus.server.Server.doBattleMove(Server.java:2088)
	at net.sf.colossus.server.ClientHandler.callMethod(ClientHandler.java:513)
	at net.sf.colossus.server.ClientHandler.doCallMethodInTryBlock(ClientHandler.java:356)
	at net.sf.colossus.server.ClientHandler.processInput(ClientHandler.java:159)
	at net.sf.colossus.server.Server.handleSelectedKeys(Server.java:528)
	at net.sf.colossus.server.Server.waitOnSelector(Server.java:399)
	at net.sf.colossus.server.Server.run(Server.java:227)


Game might be unstable or hang from now on...

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3516697&group_id=1939

Bugs item #3509762, was opened at 2012-03-21 09:20
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3509762&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
>Resolution: Invalid
Priority: 5
Private: No
Submitted By: Andrew Hartung (eddie1914)
Assigned to: Nobody/Anonymous (nobody)
Summary: Guardian limited to one movement in jungle

Initial Comment:
While playing the undead variant which has every non tree square in the jungle brambles, the guardian can only move one square. Since it is a flying creature, shouldn't it have its full movement?

----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2012-03-21 10:01

Message:
> Since it is a flying creature, shouldn't it have its full movement?

No. The slowing applies when it lands.

From the rules, 11.4: 

... A flying character ignores all hazards (except Volcanos, which only
Dragons
may enter or cross) until it lands. If a flying character is slowed by a
hazard to a hex in which it chooses to land, landing in that hex counts as
moving two hexes. Hazards have no effect on the movement of
characters leaving a hex (exception: see Cliffs, 11.7).


Seems that's a detail of the rules which is not so widely known - you're
not the first one asking this ;-)


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3509762&group_id=1939

Bugs item #3509762, was opened at 2012-03-21 09:20
Message generated for change (Tracker Item Submitted) made by eddie1914
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3509762&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Andrew Hartung (eddie1914)
Assigned to: Nobody/Anonymous (nobody)
Summary: Guardian limited to one movement in jungle

Initial Comment:
While playing the undead variant which has every non tree square in the jungle brambles, the guardian can only move one square. Since it is a flying creature, shouldn't it have its full movement?

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3509762&group_id=1939

Bugs item #3497085, was opened at 2012-03-05 09:41
Message generated for change (Comment added) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3497085&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: GUI
Group: None
Status: Open
>Resolution: Fixed
>Priority: 3
Private: No
Submitted By: Clemens Katzer (cleka)
Assigned to: Clemens Katzer (cleka)
Summary: Undo battle move errors / undo stack not cleared

Initial Comment:
In newer game logs on game server one can find "Undo battle move" errrors.

[katzer@dhcppc2 games]$ grep "Undo move error" 20*/*/Colossus*.log
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in F3
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in E2
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in E3
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in D3
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in D2
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in D1
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in C1
[katzer@dhcppc2 games]$ 
[katzer@dhcppc2 games]$ grep "Undo move error" 19*/*/Colossus*.log
19100-19199/19169/Colossus0.log:SEVERE: Undo move error: no critter in C1
[katzer@dhcppc2 games]$ 
[katzer@dhcppc2 games]$ grep "Undo move error" 18*/*/Colossus*.log
18000-18099/18097/Colossus0.log:SEVERE: Undo move error: no critter in D5

From 20286:

INFO: xxx's battle turn, number 1
Mar 5, 2012 6:51:54 AM net.sf.colossus.server.BattleServerSide$BattlePhaseAdvancer advancePhaseInternal
INFO: Battle phase advances to Move
Mar 5, 2012 6:52:06 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Angel moves from X1 to D6
Mar 5, 2012 6:52:08 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Titan moves from X1 to E5
Mar 5, 2012 6:52:10 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Ranger moves from X1 to F4
Mar 5, 2012 6:52:11 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Angel moves from X1 to F3
Mar 5, 2012 6:52:19 AM net.sf.colossus.server.CreatureServerSide undoMove
INFO: Angel undoes move and returns to Plains hex X1
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.CreatureServerSide undoMove
INFO: Ranger undoes move and returns to Plains hex X1
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.CreatureServerSide undoMove
INFO: Titan undoes move and returns to Plains hex X1
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.CreatureServerSide undoMove
INFO: Angel undoes move and returns to Plains hex X1
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in F3
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in E2
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in E3
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in D3
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in D2
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in D1
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in C1
Mar 5, 2012 6:52:23 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Angel moves from X1 to D6
Mar 5, 2012 6:52:24 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Ranger moves from X1 to E5

It's clearly that the undo actions are from previous battle.

In 19169, looks like a combination of double/wrong move due to delay and undo.

IN 1807, a combination of "null move" (which is not accepted any more by server) and undo from a move from previous battle/turn of same battle.

Nov 13, 2011 11:44:57 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Gargoyle moves from C3 to D5
Nov 13, 2011 11:45:03 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Gargoyle in Tower (1) hex D5 does not move
Nov 13, 2011 11:45:08 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Gargoyle moves from E3 to D3
Nov 13, 2011 11:45:12 PM net.sf.colossus.server.CreatureServerSide undoMove
INFO: Gargoyle undoes move and returns to Tower (1) hex C3
Nov 13, 2011 11:45:12 PM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in D5
Nov 13, 2011 11:45:24 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Gargoyle moves from C3 to D5
Nov 13, 2011 11:45:26 PM net.sf.colossus.server.BattleServerSide$BattlePhaseAdvancer advancePhaseInternal


----------------------------------------------------------------------

>Comment By: Clemens Katzer (cleka)
Date: 2012-03-05 11:32

Message:
Fixed (most likely/hopefully) in r5117.

Error might still occur/visible in server logs for a long while, until all
users will have taken coming release (probably) 0.13.1 into use.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3497085&group_id=1939

Bugs item #3497085, was opened at 2012-03-05 09:41
Message generated for change (Tracker Item Submitted) made by cleka
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3497085&group_id=1939

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: GUI
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Clemens Katzer (cleka)
Assigned to: Clemens Katzer (cleka)
Summary: Undo battle move errors / undo stack not cleared

Initial Comment:
In newer game logs on game server one can find "Undo battle move" errrors.

[katzer@dhcppc2 games]$ grep "Undo move error" 20*/*/Colossus*.log
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in F3
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in E2
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in E3
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in D3
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in D2
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in D1
20200-20299/20286/Colossus0.log:SEVERE: Undo move error: no critter in C1
[katzer@dhcppc2 games]$ 
[katzer@dhcppc2 games]$ grep "Undo move error" 19*/*/Colossus*.log
19100-19199/19169/Colossus0.log:SEVERE: Undo move error: no critter in C1
[katzer@dhcppc2 games]$ 
[katzer@dhcppc2 games]$ grep "Undo move error" 18*/*/Colossus*.log
18000-18099/18097/Colossus0.log:SEVERE: Undo move error: no critter in D5

From 20286:

INFO: xxx's battle turn, number 1
Mar 5, 2012 6:51:54 AM net.sf.colossus.server.BattleServerSide$BattlePhaseAdvancer advancePhaseInternal
INFO: Battle phase advances to Move
Mar 5, 2012 6:52:06 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Angel moves from X1 to D6
Mar 5, 2012 6:52:08 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Titan moves from X1 to E5
Mar 5, 2012 6:52:10 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Ranger moves from X1 to F4
Mar 5, 2012 6:52:11 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Angel moves from X1 to F3
Mar 5, 2012 6:52:19 AM net.sf.colossus.server.CreatureServerSide undoMove
INFO: Angel undoes move and returns to Plains hex X1
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.CreatureServerSide undoMove
INFO: Ranger undoes move and returns to Plains hex X1
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.CreatureServerSide undoMove
INFO: Titan undoes move and returns to Plains hex X1
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.CreatureServerSide undoMove
INFO: Angel undoes move and returns to Plains hex X1
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in F3
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in E2
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in E3
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in D3
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in D2
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in D1
Mar 5, 2012 6:52:20 AM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in C1
Mar 5, 2012 6:52:23 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Angel moves from X1 to D6
Mar 5, 2012 6:52:24 AM net.sf.colossus.server.BattleServerSide doMove
INFO: Ranger moves from X1 to E5

It's clearly that the undo actions are from previous battle.

In 19169, looks like a combination of double/wrong move due to delay and undo.

IN 1807, a combination of "null move" (which is not accepted any more by server) and undo from a move from previous battle/turn of same battle.

Nov 13, 2011 11:44:57 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Gargoyle moves from C3 to D5
Nov 13, 2011 11:45:03 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Gargoyle in Tower (1) hex D5 does not move
Nov 13, 2011 11:45:08 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Gargoyle moves from E3 to D3
Nov 13, 2011 11:45:12 PM net.sf.colossus.server.CreatureServerSide undoMove
INFO: Gargoyle undoes move and returns to Tower (1) hex C3
Nov 13, 2011 11:45:12 PM net.sf.colossus.server.BattleServerSide undoMove
SEVERE: Undo move error: no critter in D5
Nov 13, 2011 11:45:24 PM net.sf.colossus.server.BattleServerSide doMove
INFO: Gargoyle moves from C3 to D5
Nov 13, 2011 11:45:26 PM net.sf.colossus.server.BattleServerSide$BattlePhaseAdvancer advancePhaseInternal


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101939&aid=3497085&group_id=1939