Recent changes to bugs

Planets don't display in map without a Parent Star System

Ryan Neufeld — Mon, 02 Jun 2008 16:44:07 -0000

At present a Planet object with a parent other than a Star system will not display on the game map (in my case the parent is a Galaxy object.) It should be noted that even though the Planet object is not displayed, the object is still listed properly in the hierarchy of objects.

Zombie Fleets

Sean Hanna — Mon, 07 Apr 2008 01:21:31 -0000

I have noticed that there are fleets which can no longer receive orders. I can't reliably cause a fleet to become a zombie. The version of the client i am using is 0.3.1.1. The server is running the minisec ruleset.

The game is still active on (04/06/2008) (demo1.thousandparsec.net) and the fleet id's 160, 207 both exhibit this behavior.

When trying to give an order to these zombie fleets the client crashes:

Traceback (most recent call last):
File "tp\client\threads.pyo", line 471, in OnCacheDirty
IOError: Unable to insert the order...
============================================

A full copy of the log has been attached.

libtpclient-py 0.3.1.1 on the website extracts to wrong dir

Anonymous — Wed, 02 Apr 2008 02:49:00 -0000

libtpclient-py 0.3.1.1 on the website extracts to wrong dir

It extracts to

libtpclient-py-0.3.1

but it should be

libtpclient-py-0.3.1.1

Sorry to be pedantic. I need it for some packaging scripts.

Eugene
jmingtan@gmail.com

World downloading gets stuck on downloading orders

Hellwing — Fri, 28 Mar 2008 00:02:04 -0000

I created an account on the demo1 server, and right now whenever I connect and it tries to download the world, it gets stuck on downloading the orders for my home planet.

The account name is: test1

Steps to reproduce the bug:

- Create new account
- Play for a few turns, sending out ships to colonize and giving production orders to your home planet
- Then, whenever you are prompted to download data for the new turn, ignore it for the duration of several turns (I ignored it for at least half an hour). Then agree to download the world.
- The download will be stuck when trying to get your home world's orders.

Escape character not removed in from server description

Jure Repinc — Sun, 23 Mar 2008 12:23:36 -0000

Steps to reproduce:
1. open server browser in tpclient-pywx
2. see the description of llnz-dev2

Actual result:
The description contains "Lee\'s"

Expected result:
The description should contain "Lee's", without backslash.

Piechart remains after switching from resources to systems

Jure Repinc — Sun, 23 Mar 2008 11:51:32 -0000

Steps to reproduce
1. connected to demo2 which is running the RFTS ruleset
2. switched from systems to resources view
3. switched back to systems view

Actual result:
The piecharts were not cleared. Only systems were added on top of them.

Expected:
Piechart should be cleared and only systems should be shown.

Segfault in isAlive

Tim Ansell — Thu, 13 Mar 2008 10:45:25 -0000

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0xb7abb6c0 (LWP 2859)]
0x080da58c in IGObject::isAlive (this=0x0) at object.cpp:82
82 return alive;
(gdb) bt
#0 0x080da58c in IGObject::isAlive (this=0x0) at object.cpp:82
#1 0x081278ec in ObjectView::packFrame (this=0x81b9898, frame=0x81b2070, playerid=1) at objectview.cpp:56
#2 0x081213f3 in PlayerView::processGetObject (this=0x81b0be0, objid=135, frame=0x81b2070) at playerview.cpp:115
#3 0x08118158 in PlayerAgent::processGetObjectById (this=0x81b7080, frame=0x81aaa80) at playeragent.cpp:243
#4 0x08118237 in PlayerAgent::processIGFrame (this=0x81b7080, frame=0x81aaa80) at playeragent.cpp:91
#5 0x080e7341 in PlayerConnection::inGameFrame (this=0x81b1418) at playerconnection.cpp:301
#6 0x080e8b27 in PlayerConnection::process (this=0x81b1418) at playerconnection.cpp:80
#7 0x080ceaf3 in Network::masterLoop (this=0x8153350) at net.cpp:334
#8 0x080beded in main (argc=3, argv=0xbf908e84) at main.cpp:135
(gdb) bt full
#0 0x080da58c in IGObject::isAlive (this=0x0) at object.cpp:82
No locals.
#1 0x081278ec in ObjectView::packFrame (this=0x81b9898, frame=0x81b2070, playerid=1) at objectview.cpp:56
object = (IGObject *) 0x0
#2 0x081213f3 in PlayerView::processGetObject (this=0x81b0be0, objid=135, frame=0x81b2070) at playerview.cpp:115
object = (class ObjectView *) 0x81b9898
#3 0x08118158 in PlayerAgent::processGetObjectById (this=0x81b7080, frame=0x81aaa80) at playeragent.cpp:243
objectID = 135
i = 57
len = 58
of = (Frame *) 0x81b2070
#4 0x08118237 in PlayerAgent::processIGFrame (this=0x81b7080, frame=0x81aaa80) at playeragent.cpp:91
of = (Frame *) 0x200
#5 0x080e7341 in PlayerConnection::inGameFrame (this=0x81b1418) at playerconnection.cpp:301
frame = (Frame *) 0x81aaa80
#6 0x080e8b27 in PlayerConnection::process (this=0x81b1418) at playerconnection.cpp:80
No locals.
#7 0x080ceaf3 in Network::masterLoop (this=0x8153350) at net.cpp:334
itcurr = {_M_node = 0x81b1760}
netstat = true
write_set = {fds_bits = {0 <repeats 32 times>}}
tv = {tv_sec = 462, tv_usec = 940000}
cur_set = {fds_bits = {512, 0 <repeats 31 times>}}
#8 0x080beded in main (argc=3, argv=0xbf908e84) at main.cpp:135
tpschemename = {static npos = 4294967295, _M_dataplus = {<std::allocator<char>> = {<__gnu_cxx::new_allocator<char>> = {<No data fields>}, <No data fields>},
_M_p = 0xb7dd7624 ""}}
persistencename = {static npos = 4294967295,
_M_dataplus = {<std::allocator<char>> = {<__gnu_cxx::new_allocator<char>> = {<No data fields>}, <No data fields>}, _M_p = 0xb7dd7624 ""}}
rulesetname = {static npos = 4294967295, _M_dataplus = {<std::allocator<char>> = {<__gnu_cxx::new_allocator<char>> = {<No data fields>}, <No data fields>},
_M_p = 0x8154474 "rfts"}}
myNetwork = (Network *) 0x8153350
myConsole = (class Console *) 0x81520f8
myGame = (Game *) 0x815d600
myPlugins = (PluginManager *) 0x815d568
myLogger = (class Logger *) 0x8154828
mySettings = (Settings *) 0x8152028
(gdb)

Universe view improvements

Mon, 03 Mar 2008 21:19:23 -0000

At the moment, the tpclient-pywx feels very little like a game, and more like a graphing and accounting program.

To remedy this, I suggest a number of things:

*When dragging the map/view, have it "float" (continue moving) a short bit after mouse button release. How much should depend on the acceleration of the pointer when the button is released. However, it should not be hard to release the button in such a way that the map does not continue floating for usability reasons.

Note: Inspiration and code can be gathered from OpenMoko, where this has been implemented in menus.

*Zooming is done in smaller steps, and there should be a "zooming" effect consisting of even smaller steps in between these. Ability to use the mouse wheel for this would be great, and even greater if the zooming centres on where the pointer is located rather than the centre of the screen.

Note: The non-OSS game Supreme Commander does this in a most excellent manner. See video here: http://tinyurl.com/yrgwbg

*Zooming should cause the star icons to scale up. At the moment their size is a strain on the eyes for us with large resolutions, and it would create an illusion of dimension even though the amount they scale might be entirely unrealistic.

Note: FreeOrion does this in a nice and gameplay-friendly manner.

*The black background should include several layers of stars in order to look less empty and dull. These stars should be seen moving when the user pans and zooms the view. The different layers would obviously move differently, to create a more authentic illusion of dimension.

Note: FreeOrion does this aswell, and it is a very nice effect.

tpserver-cpp should inbed the git commit version

Tim Ansell — Sun, 02 Mar 2008 12:58:50 -0000

It would be really good if tpserver-cpp included the git version in some form. This would allow you to check what version of tpserver-cpp you are running and hence better report bugs.

Turn info in System tree not correct

Jure Repinc — Sun, 02 Mar 2008 10:18:11 -0000

If you see the first try of the tutorial video and jump to time 7:45 you can see two orders on a planet:
1 turn to do the first fleet and 8 turns for the second fleet. Now if you look at the System tree it shows [1, 7]. It should show [1, 9] next to that planet. Maybe this can be reproduced by issuing the same orders as in the video before 7:45 time mark.