ticket summary component version milestone type owner status created _changetime _description _reporter 32 Network game crashes abuse 0.7.1 1.0 defect Sam Hocevar new 2011-04-09T01:39:36+02:00 2011-04-09T01:39:36+02:00 "Network game crashes when client chooses ""network"" from the menu. The process acting as server crashes when exiting. Traces of both follows: {{{ Creating responder on port 37008 net driver : could not bind socket to port 37008 zsh: segmentation fault (core dumped) /usr/ports/pobj/abuse-0.7.1/abuse-0.7.1/src/abuse (gdb) bt #0 0x00000000004b145c in tcpip_protocol::find_address (this=0x8e1ae0, port=37008, name=0x7f7ffffbf9e0 """") at tcpip.cpp:435 #1 0x0000000000406565 in net_configuration::input (this=0x203900a00) at netcfg.cpp:516 #2 0x000000000045cc47 in menu_handler (ev=@0x7f7ffffbfcc0, inm=0x2062fb400) at menu.cpp:551 #3 0x000000000045d023 in main_menu () at menu.cpp:735 #4 0x000000000046c499 in game::step (this=0x20397a400) at game.cpp:2134 #5 0x000000000046d6dd in main (argc=1, argv=0x7f7ffffbff28) at game.cpp:2727 }}} {{{ Thank you for playing Abuse! zsh: segmentation fault (core dumped) ./abuse (gdb) bt #0 0x000000000047530d in ldelete_syms (root=0x200190020) at lisp.cpp:889 #1 0x0000000000475316 in ldelete_syms (root=0x204edaf00) at lisp.cpp:889 #2 0x0000000000475323 in ldelete_syms (root=0x20bf42d40) at lisp.cpp:890 #3 0x0000000000475316 in ldelete_syms (root=0x20bf42900) at lisp.cpp:889 #4 0x0000000000475323 in ldelete_syms (root=0x20bf42000) at lisp.cpp:890 #5 0x0000000000475323 in ldelete_syms (root=0x20344ed00) at lisp.cpp:890 #6 0x0000000000475323 in ldelete_syms (root=0x20344ea00) at lisp.cpp:890 #7 0x0000000000475323 in ldelete_syms (root=0x20344e300) at lisp.cpp:890 #8 0x0000000000475316 in ldelete_syms (root=0x20851fcc0) at lisp.cpp:889 #9 0x0000000000475316 in ldelete_syms (root=0x20851ffc0) at lisp.cpp:889 #10 0x0000000000475316 in ldelete_syms (root=0x20851f840) at lisp.cpp:889 #11 0x0000000000475316 in ldelete_syms (root=0x203022e00) at lisp.cpp:889 #12 0x0000000000475323 in ldelete_syms (root=0x203022940) at lisp.cpp:890 #13 0x0000000000475316 in ldelete_syms (root=0x203022dc0) at lisp.cpp:889 #14 0x0000000000475316 in ldelete_syms (root=0x202c524c0) at lisp.cpp:889 #15 0x0000000000475323 in ldelete_syms (root=0x204ed60c0) at lisp.cpp:890 #16 0x0000000000475323 in ldelete_syms (root=0x20acadcc0) at lisp.cpp:890 #17 0x0000000000475323 in ldelete_syms (root=0x20acad340) at lisp.cpp:890 #18 0x0000000000475323 in ldelete_syms (root=0x20acad040) at lisp.cpp:890 #19 0x0000000000475316 in ldelete_syms (root=0x20acad2c0) at lisp.cpp:889 #20 0x0000000000475323 in ldelete_syms (root=0x20acad300) at lisp.cpp:890 #21 0x000000000047535a in lisp_uninit () at lisp.cpp:3237 #22 0x000000000046da79 in main (argc=1, argv=0x7f7ffffcfe98) at game.cpp:2796 }}} " guest 12 free sounds abuse SVN 1.0 defect Sam Hocevar new 2008-03-04T18:17:11+01:00 2008-03-04T18:17:11+01:00 "Almost everything in Abuse was freed by Crack dot Com, except the sound and music files. They are still copyrighted by Bobby Prince (who also did the Doom sfx). I approached him in May 2004 and his answer was: While the sfx are not in the public domain, you have my permission to use them for your purposes below. I'll check your operating system out -- it sounds interesting :-) Unfortunately he did not clarify what license he wanted to use. Two solutions: * ask Bobby Prince again to release the SFX under a free license * create our own SFX or find free ones (hint: Golgotha Forever has a lot of free sounds) " Sam Hocevar 19 invalid casts in input.cpp abuse SVN 1.0 defect Sam Hocevar new 2008-03-15T10:59:02+01:00 2008-03-15T10:59:02+01:00 "{{{ % rgrep '\*((int *\*' src src/imlib/input.cpp: if (*((int *)b->read())==0) src/imlib/input.cpp: if (*((int *)b2->read())==0) src/imlib/input.cpp: if (*((int *)b->read())==0) // did the user press or release the button src/imlib/input.cpp: if ((b!=b2 || maxdown==0) && *((int *)b2->read())==0) }}} There may be more." Sam Hocevar 6 get rid of ridiculously small files abuse SVN 1.0 defect Sam Hocevar new 2008-03-02T14:35:49+01:00 2008-04-12T21:16:07+02:00 Some files such as src/imlib/globals.cpp are really small and don't serve any reasonable purpose. They should disappear. Sam Hocevar 7 unneeded symbols exported abuse SVN 1.0 defect Sam Hocevar new 2008-03-02T16:10:17+01:00 2008-03-02T16:12:43+01:00 "There are many global symbols that should not be exported because they aren't used outside their .cpp file. To generate a list I do: {{{ % for x in **/*.o ; do nm $x >| $x.sym ; done % grep ' T ' **/*sym | cut -f3 -d' ' | while read i ; do N=$(grep $i **/*sym | wc -l); echo ""$N $i ($(c++filt $i))""; done | grep '^1 ' }}}" Sam Hocevar 10 better graphics options abuse SVN 1.0 defect Sam Hocevar new 2008-03-04T11:01:00+01:00 2008-03-04T11:01:00+01:00 Make OpenGL, screen resolution etc. available through a graphics menu. Sam Hocevar 16 joystick support missing abuse SVN 1.0 defect Sam Hocevar new 2008-03-06T22:03:52+01:00 2008-03-06T22:03:52+01:00 `sdlport/joystick.cpp` is empty. The API is very simple, joystick support should not be hard to implement. Sam Hocevar 17 network play is broken abuse SVN 1.0 defect Sam Hocevar new 2008-03-07T14:45:01+01:00 2008-03-07T14:45:01+01:00 Network play just segfaults when launched (server mode). Sam Hocevar 22 No sound with fRaBs on OS X abuse SVN 1.0 defect Sam Hocevar new 2009-04-12T16:10:33+02:00 2011-05-09T03:25:27+02:00 doesnt have sound guest 23 Update bootstrap script to find SDL in /usr/local abuse SVN 1.0 defect Sam Hocevar new 2009-10-03T07:30:48+02:00 2009-10-03T07:30:48+02:00 "I'm on OS X (either 10.5 or 10.6) doing a command-line build after installing SDL in /usr/local. (This is the result of using 'brew' to install SDL, see: http://github.com/mxcl/homebrew ) The bootstrap script doesn't look in /usr/local/... to find M4 macros, so it fails to detect SDL unless I edit the end of the script as seen in the attached patch. " guest 28 #ctxtnav www SVN 1.0 defect carmie new 2010-08-20T15:55:32+02:00 2010-08-20T16:26:45+02:00 #ctxtnav bouge quand on resize la fenetre, N'apparait que quand on est pas loggué. carmie 29 check ie6/7/8, safari & opera www SVN 1.0 defect carmie new 2010-08-20T15:56:21+02:00 2010-08-20T16:27:15+02:00 carmie 30 styles : closed tickets www SVN 1.0 defect carmie new 2010-08-20T16:25:24+02:00 2010-08-20T16:25:24+02:00 carmie 31 fix layout www SVN 1.0 defect carmie new 2010-08-20T16:46:53+02:00 2010-08-20T16:46:53+02:00 "Plein de bug placements menus (exemple page milestone), et les tableaux ""depassent"" du cadre." carmie 8 improve main menu abuse SVN 1.0 enhancement Sam Hocevar new 2008-03-04T10:59:28+01:00 2008-03-04T10:59:28+01:00 Currently abuse relies on the commandline to select the game/flavour to play. It would be better to have a startup menu listing all available games. Then maybe another menu listing separate levels. Sam Hocevar 41 Design away kludge in file opening code abuse SVN 1.0 enhancement Sam Hocevar new 2011-09-20T21:55:40+02:00 2011-09-20T21:55:40+02:00 "In order to determine whether get_filename_prefix()/spec_prefix should be prepended to the requested filename, both open_FILE() in game.cpp and jFILE::open_external() in imlib/specs.cpp currently use a kludge that bases the decision on whether or not the requested filename begins with a forward slash. This is problematic for porting because some platforms require absolute paths that begin with something other than a slash (e.g., a device name), but it's also an odd design in general that requires the caller to know about and prepend the save path prefix but not the data path prefix. I'm pretty sure that there are only two classes of files that the game should be attempting to open via these functions: 1. Game data files, which should have get_filename_prefix() prepended, and 2. Game save files, which should have get_save_filename_prefix() prepended. As a result, it seems to me that a better design would be to either force the caller to prepend the prefixes ahead of time, or else parameterize(*) the functions to force the caller to specify which prefix should be prepended. ''* - the parameter could be optional, defaulting to whichever prefix makes more sense.'' ---- Additionally, open_FILE() prepends a space between get_filename_prefix() and the filename, which might be a bug." guest 4 Reorganise source abuse SVN 1.0 task Sam Hocevar new 2008-03-02T01:59:51+01:00 2008-03-02T01:59:51+01:00 The source is still a huge mess. Splitting it into more modules might be worthwile. Sam Hocevar 20 i18n support abuse SVN 1.0 defect Sam Hocevar new 2008-03-16T12:48:26+01:00 2008-03-16T12:48:26+01:00 There's not much text in the game. It should be possible to translate it. Sam Hocevar 33 Music volume not restored abuse SVN 1.0 defect Sam Hocevar new 2011-05-04T14:54:17+02:00 2011-05-04T14:54:17+02:00 "Music volume setting works in game, but does not get restored on start up." guest 36 No GL or antialiasing disable switch exists abuse SVN 1.0 defect Sam Hocevar new 2011-05-04T15:01:16+02:00 2011-05-04T15:01:16+02:00 Once those are enabled, they can be removed only from abuserc. guest 39 Regain health bug abuse SVN 1.0 defect Sam Hocevar new 2011-05-16T09:56:54+02:00 2011-05-16T09:56:54+02:00 How to reproduce: lose HP, save and load. HP has been restored to 100. guest 11 external LISP engine? abuse SVN 1.0 enhancement Sam Hocevar new 2008-03-04T11:14:37+01:00 2011-04-28T01:10:26+02:00 The LISP engine is nice and interesting, and I think I fixed most of its most important bugs. But it's very difficult to understand. I therefore would like to evaluate the use of an external LISP engine. Sam Hocevar 38 Realtime debug logging to the screen abuse SVN 1.0 enhancement Sam Hocevar new 2011-05-04T15:08:12+02:00 2011-05-04T15:08:12+02:00 "I am requesting a realtime debug logging to the screen, to see which sfx is currently playing to ease audio sfx development." guest 40 Allow multiple game engine events to be queued by a single SDL event abuse SVN 1.0 enhancement Sam Hocevar new 2011-09-20T21:34:20+02:00 2011-09-20T21:34:20+02:00 "Currently, the SDL event processing code only allows a single game engine event to be generated for each SDL event that is processed. Some reasons why this is problematic: * Mouse status is polled first thing in the SDL event handler, but a mouse event is only passed on to the game engine if no other events (e.g. keyboard key presses/releases) have occurred. This could potentially cause a loss of mouse responsiveness if/when the player is frantically mashing the keyboard keys. * Implementing SDL joystick or other controller support by simulating keyboard and/or mouse events is problematic. For example, it's impossible to generate both ""left arrow key released"" and ""right arrow key pressed"" events in the case that the player moves the joystick from left to right so fast that SDL doesn't generate a joystick move event with a position in the center dead-zone. A more ideal solution might be to set up a mechanism where the SDL event processing code creates and queues as many game engine events as needed, and then bursts the queue contents to the game engine." guest