#gemrb@irc.freenode.net logs for 5 Sep 2011 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:10:06] <-- yyz has left IRC (Read error: Connection reset by peer)
[00:10:36] --> yyz has joined #gemrb
[00:11:32] --> Gekz has joined #gemrb
[00:58:37] <-- Demitar has left IRC (Ping timeout: 258 seconds)
[01:05:14] <-- yyz has left IRC (Quit: Lost terminal)
[04:16:00] --> brad_a has joined #gemrb
[04:16:24] <-- brad_a has left IRC (Client Quit)
[05:46:23] --> Drakkar has joined #gemrb
[05:48:26] <-- PixelScum has left IRC (Ping timeout: 260 seconds)
[07:27:17] --> lynxlynxlynx has joined #gemrb
[07:27:17] --- ChanServ gives channel operator status to lynxlynxlynx
[07:32:16] --> lubos has joined #gemrb
[10:11:32] --> Demitar has joined #gemrb
[11:37:28] --> yyz has joined #gemrb
[12:08:16] <-- yyz has left IRC (Read error: Connection reset by peer)
[12:14:24] --> yyz has joined #gemrb
[12:29:30] --> SiENcE has joined #gemrb
[14:01:38] <-- SiENcE has left IRC (Quit: cya)
[14:23:30] --> boriskr has joined #gemrb
[15:04:33] <-- boriskr has left IRC (Read error: Connection reset by peer)
[15:07:49] <-- yyz has left IRC (Read error: Connection reset by peer)
[15:13:49] --> yyz has joined #gemrb
[15:50:52] <-- yyz has left IRC (Ping timeout: 252 seconds)
[15:52:40] <-- lubos has left IRC (Quit: Leaving.)
[15:52:55] --> yyz has joined #gemrb
[16:37:21] --> Yoshimo has joined #gemrb
[16:38:00] --> Maighstir has joined #gemrb
[16:47:03] --> brad_a has joined #gemrb
[16:51:15] <-- brad_a has left IRC (Remote host closed the connection)
[16:51:24] --> brad_a has joined #gemrb
[17:26:27] --> brad_a_ has joined #gemrb
[17:28:44] <brad_a_> does anybody else constantly get this openAL error: "Failed to delete buffers"
[17:28:55] <brad_a_> happens to me on os x and ios
[17:30:14] <-- brad_a has left IRC (Ping timeout: 264 seconds)
[17:30:15] --- brad_a_ is now known as brad_a
[17:36:56] <brad_a> i found a fix/workaround but i still have no explanation why it happens at all. the fix is to loop the call to delete the buffers until no error is returned. no more memory leak, but it doesn't seem like i should have to do that....
[17:42:25] <brad_a> is it a threading thing? I'm not very knowledgeable about such things.
[17:53:58] <lynxlynxlynx> audio is run in a separate thread
[17:54:25] <lynxlynxlynx> i can't say i've noticed any problems, since when i play, i play and not watch the console ;)
[17:54:40] <lynxlynxlynx> is there some audible glitch or just a leak?
[18:01:10] <brad_a> just a leak
[18:06:16] <brad_a> i was able to find this: http://opensource.creative.com/pipermail/openal-devel/2009-March/005405.html
[18:08:11] <Yoshimo> i get "unable to buffer music data: 0xa003 [error
[18:08:14] <brad_a> it doesn't really help at all but it kinda confirms its an apple implementation thing (probably threading) and i can't find any documentation on apples implementation that helps me know what i should do.
[18:08:43] <brad_a> yoshimo: that kind of sounds like the opposite problem :-)
[18:09:14] <lynxlynxlynx> brad_a: just ifdef it
[18:09:25] <brad_a> ok
[18:09:27] <lynxlynxlynx> but please clean up the rotation patch first
[18:09:31] <brad_a> i did
[18:09:33] <brad_a> :)
[18:09:43] <lynxlynxlynx> link or it doesn't exist :)
[18:10:12] <brad_a> http://dl.dropbox.com/u/13866402/megapatch.patch
[18:10:21] <lynxlynxlynx> scary name
[18:10:25] <brad_a> i assume you can just leave out the patch you just told me to ifdef for now
[18:10:27] <brad_a> lol
[18:10:33] <brad_a> it has a lot in it
[18:11:04] <lynxlynxlynx> it will get better
[18:11:35] <Yoshimo> during the first cutscene in bg1, where gorion is attacked by sarevok , wave files collide, and overlap on my computer, in the logfile i see wal_09c,wav wal_09a.wav and sarev10.wav
[18:12:22] <Yoshimo> what does "one instance of ch1cut03 is dropped from 1." mean btw?
[18:13:50] <lynxlynxlynx> would have to check in the code
[18:14:14] <lynxlynxlynx> naively it sounds a bit like a cutscene was triggered multiple times
[18:18:14] <Yoshimo> also when i visit the temple east of beregost, gemrb is llooking for soundfilse with the name "*" "*a" "*b" and "*c", which seriously, looks buggy
[18:19:19] <lynxlynxlynx> that's from the data
[18:19:49] <lynxlynxlynx> you could maybe use dltcep's search to find out which creature or table or x has ti
[18:19:56] <lynxlynxlynx> ti/them
[18:20:59] <Yoshimo> so the data is wrong you mean?
[18:23:19] <lynxlynxlynx> almost certainly
[18:25:38] <Yoshimo> couldnt load animation: mdogg2, cycle 20, how about that?
[18:25:43] <Yoshimo> same`
[18:25:45] <Yoshimo> ?
[18:28:41] <lynxlynxlynx> not really
[18:29:01] <lynxlynxlynx> our problem; due to bad understanding of the animation data
[18:29:27] <lynxlynxlynx> brad_a: whatever you're coding in, please set it to use tabs for indentation
[18:33:33] <brad_a> done. sorry
[18:37:20] <lynxlynxlynx> also, any reason the three vars aren't inited to what is in the config example now?
[18:37:55] <brad_a> no. i just didn't think about that.
[18:37:59] <lynxlynxlynx> ok
[18:46:39] <-- yyz has left IRC (Read error: Connection reset by peer)
[18:47:13] --> yyz has joined #gemrb
[19:28:14] <lynxlynxlynx> brad_a: i wonder how Subject: [PATCH 10/13] implemented formation rotation handling
[19:28:19] <lynxlynxlynx> even compiles for you
[19:29:03] <lynxlynxlynx> ah, nevermind :)
[19:29:04] <brad_a> I've been known to mess up when using git
[19:29:23] <brad_a> i do know my code works but sometimes i leave out lines in git
[19:30:49] <brad_a> which compile flag is cake using that produces this error: "C++ style comments are not allowed in ISO C90"
[19:30:59] <brad_a> cmake. not cake ;-)
[19:33:34] <lynxlynxlynx> -fpedantic ?
[19:33:56] <tomprince> Did you add a new file, which is being compiled as C?
[19:34:49] <brad_a> i added an objective c wrapper
[19:35:10] <brad_a> so that i can have a proper application bundle on os x and remove the sdl dependancy etc
[19:35:23] <brad_a> it works great when i compile with xcode
[19:35:26] <brad_a> of coursse
[19:35:35] <-- Yoshimo has left IRC (Quit: Yoshimo)
[19:35:35] <brad_a> but I'm trying to add it to cmake
[19:48:46] <brad_a> there is very little documentation about compiling obj-c from command line…
[19:51:05] <CIA-26> GemRB: 03brada * r7e97e2688ca5 10gemrb/gemrb/GemRB.cfg.sample.in:
[19:51:05] <CIA-26> GemRB: alter config "Dragons" warning to apply to paths only
[19:51:05] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[19:51:05] <CIA-26> GemRB: 03brada * r0df94c84cd6d 10gemrb/gemrb/plugins/SDLVideo/SDLVideo.cpp:
[19:51:05] <CIA-26> GemRB: clean up touch gesture code. adds a define MIN_GESTURE_DELTA_PIXELS
[19:51:06] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[19:51:17] <brad_a> removing pedantic indeed let it compile fine tho
[20:02:23] <brad_a> except not its erring out when linking saying _main isn't found...
[20:14:57] <brad_a> well i guess its not compiling correctly because the o file is essentially empty…
[20:19:07] <tomprince> Is it combiling as objc?
[20:24:56] <lynxlynxlynx> doesn't look like you submitted everything
[20:25:03] <lynxlynxlynx> found out what's wrong with patch 3
[20:26:30] <lynxlynxlynx> one space -.-
[20:29:20] <CIA-26> GemRB: 03brada * r09488be92a2d 10gemrb/gemrb/plugins/OpenALAudio/OpenALAudio.cpp:
[20:29:20] <CIA-26> GemRB: clear OpenAL errors in OpenALAudioDriver::SetupNewStream() and OpenALAudioDriver::CreateStream(). Stagnant OpenAL errors were preventing music and video audio from playing on Apple (OS X, iOS) devices.
[20:29:20] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[20:29:20] <CIA-26> GemRB: 03brada * r4e7b96eaeb55 10gemrb/gemrb/ (5 files in 3 dirs):
[20:29:21] <CIA-26> GemRB: implement config settings for touch gestures. NumFingScroll, NumFingKboard, and NumFingInfo.
[20:29:21] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[20:29:22] <CIA-26> GemRB: 03brada * rf0859106ccbc 10gemrb/gemrb/ (8 files in 8 dirs):
[20:29:22] <CIA-26> GemRB: repurposed a reserved auto pause flag for generic use.
[20:29:23] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[20:29:23] <CIA-26> GemRB: 03brada * r50d9474eb9b3 10gemrb/gemrb/plugins/SDLVideo/SDLVideo.cpp:
[20:29:24] <CIA-26> GemRB: implemented an auto pause for when the game window is minimized. Requires SDL 1.3. Useful for Android/iPhone interruptions.
[20:29:24] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[20:31:49] <lynxlynxlynx> brad_a: what's up with this:
[20:31:52] <lynxlynxlynx> - drawtarget = Selected && markerfeedback >= 4;
[20:31:52] <lynxlynxlynx> + drawtarget = (Selected && (InternalFlags|IF_NORECTICLE) && markerfeedback >= 4 && GetPathLength());
[20:32:07] <lynxlynxlynx> (InternalFlags|IF_NORECTICLE) is always true
[20:32:33] <brad_a> i don't think that is needed then
[20:32:47] <brad_a> i was just copying from the old code in Movable IIRC
[20:33:02] <lynxlynxlynx> nope, you can see the previous line
[20:33:16] <brad_a> is (Selected && markerfeedback >= 4 && GetPathLength()); sufficient then?
[20:33:37] <lynxlynxlynx> ohhh
[20:33:39] <brad_a> i mean the old drawTargetPoint method that had that
[20:33:41] <lynxlynxlynx> you did something very bad
[20:33:44] <brad_a> oh no!
[20:34:01] <lynxlynxlynx> when you claim moving a function, it should stay untouched
[20:34:08] <brad_a> oh
[20:34:17] <brad_a> well it got touched
[20:34:19] <lynxlynxlynx> one commit for the move and the obvious rename, others for the rest
[20:34:26] <brad_a> oh ok i get it
[20:35:00] <lynxlynxlynx> it can be misleading when debugging, plus the diff is unneccessarily big
[20:36:01] <brad_a> noted
[20:49:49] <lynxlynxlynx> brad_a: two things need more attention
[20:50:19] <lynxlynxlynx> if mouse up happens on an actor, the rotation sticks up
[20:51:35] <lynxlynxlynx> secondly, it tries to also work for a single character, but fails (no that there is any sense in trying): wherever you rotate, the reticle will be moved under the cursor on mouse up
[20:53:38] <lynxlynxlynx> it appears the handle length from the pivot is constant in that case too
[20:54:48] <brad_a> ah yes. don't know how that escaped me
[20:57:42] <CIA-26> GemRB: 03brada * r100f66d16030 10gemrb/gemrb/core/GUI/GameControl.cpp:
[20:57:42] <CIA-26> GemRB: change mouse cursor when rotating formation.
[20:57:42] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[20:57:43] <CIA-26> GemRB: 03brada * r807a306f79c8 10gemrb/gemrb/core/Interface.cpp:
[20:57:43] <CIA-26> GemRB: properly initialize UseSoftKeyboard variable.
[20:57:43] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[20:57:46] <CIA-26> GemRB: 03brada * r8915130d0fdc 10gemrb/gemrb/core/GUI/ (GameControl.cpp GameControl.h):
[20:57:46] <CIA-26> GemRB: implemented formation rotation handling.
[20:57:46] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[20:57:46] <CIA-26> GemRB: 03brada * refcf11ac61fa 10gemrb/gemrb/core/ (5 files in 2 dirs):
[20:57:46] <CIA-26> GemRB: moved and adapted target reticle draw code from Movable to GameControl in
[20:57:46] <CIA-26> GemRB: preparation for formation rotation code.
[20:57:47] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[20:57:47] <CIA-26> GemRB: 03brada * r60ca070217cc 10gemrb/gemrb/core/Scriptable/Scriptable.cpp:
[20:57:48] <CIA-26> GemRB: fixed crash in Movable::GetPathLength() when there is no path.
[20:57:48] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[21:04:03] <lynxlynxlynx> just one left
[21:17:29] <CIA-26> GemRB: 03brada * r7fdc589d721f 10gemrb/gemrb/plugins/SDLVideo/SDLVideo.cpp:
[21:17:29] <CIA-26> GemRB: made formation rotation possible with touch input.
[21:17:29] <CIA-26> GemRB: Signed-off-by: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[21:21:50] <CIA-26> GemRB: 03lynxlupodian * r2a8af85e8ffb 10gemrb/gemrb/plugins/SDLVideo/SDLVideo.cpp:
[21:21:50] <CIA-26> GemRB: sdlvideo: fixed spaces vs tabs mess
[21:21:50] <CIA-26> GemRB: -w suggested
[21:24:45] <brad_a> sorry bout the spaces junk :|
[21:29:44] <lynxlynxlynx> actually this was for past sins, i fixed the patches manually
[21:33:09] <lynxlynxlynx> hmpf, the last touch commit doesn't compile
[21:33:25] <lynxlynxlynx> murphy's law
[21:34:11] <brad_a> lousy murphy
[21:36:02] <lynxlynxlynx> you didn't have to touch any cmake stuff?
[21:36:23] <brad_a> no but I'm currently trying to
[21:36:37] <brad_a> I'm failing at making it play nice with objective-c
[21:36:38] <lynxlynxlynx> gemrb/gemrb/plugins/SDLVideo/SDLVideo.cpp:28:85: fatal error: GameControl.h: No such file or directory <-- it's one level deeper
[21:37:35] <lynxlynxlynx> i guess you have something extra for apple
[21:37:39] <brad_a> yes i do
[21:37:53] <brad_a> I'm trying to get cake to work tho then i can do better testing
[21:40:45] <CIA-26> GemRB: 03lynxlupodian * r9bc8eb7adbb9 10gemrb/gemrb/plugins/SDLVideo/SDLVideo.cpp: fixed include path from 7fdc58
[21:40:46] <CIA-26> GemRB: 03lynxlupodian * r06e7b0fa2cdd 10gemrb/gemrb/core/ (5 files in 3 dirs):
[21:40:46] <CIA-26> GemRB: recticle is just a common misspelling of reticle
[21:40:46] <CIA-26> GemRB: (the function names were left intact to match script use)
[21:42:34] <tomprince> speaking of failed compiles, my buildbot is back up.
[21:45:29] <brad_a> are there other files cmake uses other than the cmakelists.txt files?
[21:46:11] <-- yyz has left IRC (Read error: Connection reset by peer)
[21:47:18] --> yyz has joined #gemrb
[21:48:39] <tomprince> Nothing it doesn't generate itself.
[21:48:45] <tomprince> At least, I don't think so.
[21:51:48] <CIA-26> GemRB: 03lynxlupodian * r1e790f69c17c 10gemrb/gemrb.6.in: added new options to the manpage
[21:51:59] <brad_a> i know I'm missing something obvious but my googling is failing me
[21:59:57] <tomprince> What is the issue?
[22:02:09] <brad_a> the issue is when i compile the objective c file it produces an o file that is essentially empty vs when i compile with Xcode i get a correct o
[22:02:31] <brad_a> it says it compiles and goes to the linking step but fails to link because the symbol isn't in the o
[22:03:19] <brad_a> I'm watching the Xcode build log now to see if i can ssee what options it is passing to gcc
[22:05:23] <tomprince> And what command is cmake calling?
[22:06:39] <brad_a> it that in the makefile generated by cake?
[22:06:43] <brad_a> I'm a make/cmake noob
[22:06:56] <tomprince> Try passing VERBOSE=1 to make
[22:07:05] <brad_a> ok
[22:08:35] <lynxlynxlynx> cake is a much better name :)
[22:10:30] <brad_a> lol that the stupid autocorrect
[22:10:43] <brad_a> cd /Users/brada/Developer/gemrb/build/gemrb && /usr/bin/c++ -DHAVE_CONFIG_H -Werror -Wall -W -Wpointer-arith -Wcast-align -Wno-format-y2k -Wno-long-long -fno-strict-aliasing -fvisibility=hidden -O2 -g -I/Users/brada/Developer/gemrb/build -I/Users/brada/Developer/gemrb/gemrb/includes -I/Users/brada/Developer/gemrb/gemrb/core -I/Library/Frameworks/SDL.framework/Headers -o CMakeFiles/gemrb.dir/CocoaWrapper.o -c
[22:10:44] <brad_a> /Users/brada/Developer/gemrb/gemrb/CocoaWrapper.m
[22:10:44] <brad_a> "/Applications/CMake 2.8-5.app/Contents/bin/cmake" -E cmake_progress_report /Users/brada/Developer/gemrb/build/CMakeFiles 33
[22:12:12] <brad_a> hmmm i don't think it should be calling /usr/bin/c++ for this but i would also thing that would cause a compile error
[22:12:18] <tomprince> Now, I don't know objc or osx, but does that look sane? And does it generate something sensible if you run it by hand?
[22:12:53] <brad_a> Xcode is calling some binary or alias called CompileC
[22:12:58] <tomprince> I think you may need to add some incatations to cmake to make it know about objc.
[22:13:29] <brad_a> i have been googling that but I'm failing
[22:14:12] <brad_a> the only thing i see in samples is changing the c compiler and I'm doing that. or at least think i am
[22:14:31] <brad_a> I'm using this SET (CMAKE_C_COMPILER "/Developer/usr/llvm-gcc-4.2/bin/llvm-gcc-4.2")
[22:14:39] <brad_a> and this: SET (CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -x objective-c")
[22:14:46] <tomprince> http://www.cmake.org/pipermail/cmake/2009-July/031089.html
[22:15:33] <brad_a> shouldn't my objective c cause a compile error with the g++ compiler tho?
[22:15:54] <brad_a> ill certainly try this tho
[22:16:49] <tomprince> c++ is probably jsut a symlink to gcc, and it adds -lstdc++ to the args, and a couple of other things, but is mostly just gcc.
[22:22:58] <tomprince> http://vtk.org/gitweb?p=VTK.git;a=tree is by the authors of cmake, and it looks like it handles some objc stuff. I don't have time to look through it now, though.
[22:27:06] <brad_a> i don't see anything in their cake file that stands out ATM
[22:27:13] <brad_a> but I'm getting closer i think
[22:27:28] <brad_a> it is finally using the right compiler but I'm still getting the same o
[22:27:47] <brad_a> so it should be a simple matter of finding the correct flags
[22:27:54] <brad_a> but i must go now
[22:27:59] <-- brad_a has left IRC (Quit: brad_a)
[22:44:57] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[23:59:35] --> PixelScum has joined #gemrb