#gemrb@irc.freenode.net logs for 11 Jan 2012 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:24:35] --> joneirik has joined #gemrb
[00:31:50] <-- thomas_sch has left IRC (Read error: Connection reset by peer)
[00:56:22] <-- joneirik has left IRC (Ping timeout: 248 seconds)
[01:37:33] <gembot> build #78 of nmake-msvc++10 is complete: Exception [exception git] Build details are at http://buildbot.gemrb.org/builders/nmake-msvc%2B%2B10/builds/78 blamelist: avenger_teambg@sourceforge.net
[01:47:48] --> harijan_ has joined #gemrb
[01:50:38] <-- harijan has left IRC (Ping timeout: 240 seconds)
[01:52:15] <-- harijan_ has left IRC (Ping timeout: 252 seconds)
[01:52:49] --> harijan has joined #gemrb
[01:58:29] <scriptedfate> So if I'm wanting to do something as abjectly foolish as officially supporting a new platform (BlackBerry PlayBook), where do I start?
[02:34:54] <scriptedfate> Well, fwiw, I've got so far as it can now recognize inter-file imports. Still missing something, though, as it doesn't appear to be including libc
[02:52:12] <scriptedfate> Ah-ha. It didn't include cstdlib because it borked on being unable to find config.h
[02:52:18] <scriptedfate> Once again, config.h is evil
[02:52:29] <-- scriptedfate has left IRC (Quit: Leaving)
[05:04:21] <-- Gekz has left IRC (Ping timeout: 240 seconds)
[05:57:41] --> PixelScum has joined #gemrb
[05:58:46] <-- Drakkar has left IRC (Ping timeout: 248 seconds)
[06:08:40] --> Drakkar has joined #gemrb
[06:10:33] <-- PixelScum has left IRC (Ping timeout: 240 seconds)
[06:11:07] --> Gekz has joined #gemrb
[06:11:07] <-- Gekz has left IRC (Changing host)
[06:11:07] --> Gekz has joined #gemrb
[06:11:50] <-- Drakkar has left IRC (Read error: Connection reset by peer)
[06:12:15] --> Drakkar has joined #gemrb
[06:20:55] --> Herrman has joined #gemrb
[06:54:36] <-- Herrman has left IRC (Quit: Mit Einmal Luft holen taucht Chuck Norris 10'000 Meilen unter dem Meer. (2 Mal!))
[07:04:15] --> PixelScum has joined #gemrb
[07:06:33] <-- Drakkar has left IRC (Ping timeout: 244 seconds)
[07:19:10] <-- PixelScum has left IRC (Ping timeout: 276 seconds)
[08:52:21] <-- demitar has left IRC (Quit: Ex-Chat)
[09:02:33] --> lynxlynxlynx has joined #gemrb
[09:02:34] <-- lynxlynxlynx has left IRC (Changing host)
[09:02:34] --> lynxlynxlynx has joined #gemrb
[09:02:34] --- ChanServ gives channel operator status to lynxlynxlynx
[10:31:09] --> SiENcE has joined #gemrb
[11:26:07] <-- SiENcE has left IRC (Quit: @all: cya)
[11:43:21] <-- Gekz has left IRC (*.net *.split)
[11:43:21] <-- Baldurer has left IRC (*.net *.split)
[11:43:24] <-- |Cable| has left IRC (*.net *.split)
[11:43:24] <-- tomprince has left IRC (*.net *.split)
[11:43:24] <-- CIA-36 has left IRC (*.net *.split)
[11:43:25] <-- wjp has left IRC (*.net *.split)
[11:43:25] <-- Calchan has left IRC (*.net *.split)
[11:43:26] <-- Kiranos has left IRC (*.net *.split)
[11:43:27] <-- basse has left IRC (*.net *.split)
[11:43:27] <-- lynxlynxlynx has left IRC (*.net *.split)
[11:43:27] <-- edheldil has left IRC (*.net *.split)
[11:43:28] <-- harijan has left IRC (*.net *.split)
[11:43:28] <-- gembot has left IRC (*.net *.split)
[11:43:28] <-- xrogaan has left IRC (*.net *.split)
[11:43:28] <-- fuzzie has left IRC (*.net *.split)
[11:50:00] --> lynxlynxlynx has joined #gemrb
[11:50:00] --> Gekz has joined #gemrb
[11:50:00] --> harijan has joined #gemrb
[11:50:00] --> gembot has joined #gemrb
[11:50:00] --> Baldurer has joined #gemrb
[11:50:00] --> |Cable| has joined #gemrb
[11:50:00] --> CIA-36 has joined #gemrb
[11:50:00] --> basse has joined #gemrb
[11:50:00] --> Calchan has joined #gemrb
[11:50:00] --> edheldil has joined #gemrb
[11:50:00] --> xrogaan has joined #gemrb
[11:50:00] --> wjp has joined #gemrb
[11:50:00] --> tomprince has joined #gemrb
[11:50:00] --> Kiranos has joined #gemrb
[11:50:00] --> fuzzie has joined #gemrb
[13:53:36] <-- Baldurer has left IRC (Remote host closed the connection)
[14:23:39] --> Drakkar has joined #gemrb
[14:35:07] --> PixelScum has joined #gemrb
[14:37:51] <-- Drakkar has left IRC (Ping timeout: 252 seconds)
[15:01:24] --> Drakkar has joined #gemrb
[15:05:25] <-- PixelScum has left IRC (Ping timeout: 252 seconds)
[15:06:30] <-- gembot has left IRC (Ping timeout: 248 seconds)
[15:13:22] --> PixelScum has joined #gemrb
[15:15:45] <-- Drakkar has left IRC (Ping timeout: 240 seconds)
[15:52:18] --> Drakkar has joined #gemrb
[15:55:55] <-- PixelScum has left IRC (Ping timeout: 276 seconds)
[16:04:07] --> PixelScum has joined #gemrb
[16:06:57] <-- Drakkar has left IRC (Ping timeout: 252 seconds)
[16:09:11] --> SiENcE has joined #gemrb
[16:42:02] --> Maighstir has joined #gemrb
[16:47:25] <-- SiENcE has left IRC (Quit: @all: cya)
[16:51:03] --> Drakkar has joined #gemrb
[16:53:17] <-- PixelScum has left IRC (Ping timeout: 252 seconds)
[16:59:16] --> gembot has joined #gemrb
[17:14:21] --> brad_a has joined #gemrb
[17:51:18] <brad_a> im having difficulty distinguishing IE_GUI_TEXTAREA_SMOOTHSCROLL and IE_GUI_TEXTAREA_AUTOSCROLL
[18:18:02] --> PixelScum has joined #gemrb
[18:20:52] <-- Drakkar has left IRC (Ping timeout: 276 seconds)
[18:26:12] <lynxlynxlynx> autoscroll is used when feeding text
[18:26:24] <lynxlynxlynx> eg with the message window
[18:26:37] <lynxlynxlynx> smoothscroll only with the chapter text afaik
[18:26:58] <lynxlynxlynx> it does something like perpixel scrolling
[18:27:23] --> Yoshimo has joined #gemrb
[18:29:17] <brad_a> yes you are correct
[18:29:35] <brad_a> well i does auto per-pixel scrolling
[18:32:17] <brad_a> since we do per-pixel scrolling now with scrollbar/scrollwheel that name is a bit misleading, but autoscroll is already taken
[18:37:01] <Yoshimo> hi brad, what non-font related work did you pick as next task ?
[18:38:30] --> demitar has joined #gemrb
[18:40:21] <brad_a> i still need to merge some touchscreen stuff from the ios side
[18:40:48] <brad_a> and i need to get the touch input android compatible
[18:43:06] <brad_a> immediately i am doing some minor things like os alert boxes for ios when a fatal error occurs
[18:43:28] <brad_a> and fixing a bug in text areas that allows me to scroll chapter text
[18:45:13] --> Drakkar has joined #gemrb
[18:46:56] <-- PixelScum has left IRC (Ping timeout: 240 seconds)
[18:54:40] <-- DrMcCoy has left IRC (Remote host closed the connection)
[19:14:38] --> PixelScum has joined #gemrb
[19:17:15] <-- Drakkar has left IRC (Ping timeout: 252 seconds)
[19:19:50] <brad_a> lynx/fuzie/tomprince: if you have a moment id like to see what you have to say about the sanity of these commits
[19:19:51] <brad_a> http://dl.dropbox.com/u/13866402/patches.patch
[19:20:07] <brad_a> primarily the way im doing the ios/mac alerts in error
[19:23:13] <tomprince> You should be careful, printSTatus("ERROR", red), and error(...) behave differently.
[19:25:12] <brad_a> i should have only replaced the printstatus where init was going to return an error
[19:25:51] --> Drakkar has joined #gemrb
[19:26:10] <brad_a> maybe error should output ERROR in red anyway
[19:27:32] <tomprince> Well, but I think the output may be messed up sometimes, since the old code was expecting printStatus to write a \n, which isn't happening now.
[19:27:33] <-- lynxlynxlynx has left IRC (Read error: Connection reset by peer)
[19:28:13] <brad_a> oh i see
[19:28:39] <-- PixelScum has left IRC (Ping timeout: 240 seconds)
[19:29:17] <tomprince> I don't like the define.
[19:29:46] <tomprince> would 'namespace GemRB { class Point ...; } using namespace GemRB;' fix it?
[19:30:16] <brad_a> isnt namespace a c++ thing?
[19:30:38] <tomprince> Which is what GemRB is coded in, yes.
[19:30:49] <Yoshimo> seperating real errors from non.errors would a good idea too
[19:31:22] <brad_a> oh i see lets try the namespace thing
[19:31:23] <tomprince> I have no idea about the textarea stuff, which seems unrelated anyway.
[19:31:32] <brad_a> sure i know those are fine
[19:33:39] <tomprince> Logging is starting to become a mess of #ifdef's, which is anoying.
[19:34:37] <tomprince> Also, I think some people might prefer console errors on mac, at least some of the time.
[19:34:37] <brad_a> yes. i could alternatively externalize it in a separate file i suppose
[19:34:47] <brad_a> that does both
[19:35:04] <brad_a> well currently only console on mac
[19:35:17] <brad_a> i havent gotten around to a gui alert on mac
[19:36:26] <tomprince> I know that I would likely find gui alerts annoying, at least when I am hacking on gemrb. So it would be good to make them optional.
[19:36:41] --> PixelScum has joined #gemrb
[19:37:13] <brad_a> sure easy enough
[19:37:48] <brad_a> the main question is wether i should be mixing that stuff into logging.cpp in the firstplace
[19:38:59] <brad_a> error xits anyway so its not like you would get a barrage of popups
[19:39:34] <-- Drakkar has left IRC (Ping timeout: 252 seconds)
[19:40:03] <tomprince> I think what I would do is define a guialert function + config variable.
[19:40:26] <tomprince> And put the definition in a seperate file for each config, with cmake picking the right one for each system.
[19:40:43] <tomprince> Maybe?
[19:40:49] <tomprince> That is just my first thought.
[19:41:31] <tomprince> Looking at Logging.h, the android code is almost totally disjoint from the rest of it.
[19:43:04] <tomprince> And anyway, I think the log might be slightly messed up, since every individual print is forced to a seperate line there, I think.
[19:44:19] <tomprince> fuzzie: Thoughts? ---^
[19:45:11] <brad_a> about the namespace it will probably work but i need to do something to use the proper namespace do i just replace point with GemRB::Point
[19:45:15] <brad_a> or is there a better way?
[19:46:03] <tomprince> Well, the 'using namespace GemRB;' Should mean you don't have to change any other code.
[19:46:36] <brad_a> where do i put that?
[19:47:03] <tomprince> Try right after closing the namespace.
[19:47:15] <brad_a> just did :-P
[19:47:19] <brad_a> we will see
[19:48:40] <tomprince> I suspect if that works, the proper fix would be to just move everything into the GemRB namespace, so we don't need to worry about name clashes like this.
[19:49:00] <brad_a> sounds good to me
[19:50:27] <tomprince> Which would involve wrapping all the .h files with a 'namespace GemRB', after the includes, and putting a using namespace GemRB; in the .cpp, again after the includes.
[19:50:34] <tomprince> (I think)
[19:51:10] <tomprince> Perhaps something different for the plugins.
[19:53:44] <brad_a> well i get a bunch of "reference to 'Point' is ambiguous" errors now
[19:54:00] --> Avenger has joined #gemrb
[19:54:17] <brad_a> this is puzzleing
[19:54:22] <Avenger> this is why i hate vectors... valgrind still complains about npclevels
[19:54:53] <Avenger> the best would be to rewrite it with simpler data type
[20:01:23] <brad_a> I have to run to work now, but im prolly gonna just externalize this to avoid name clashes etc.
[20:01:44] <brad_a> unless somebody knows another way
[20:01:46] <brad_a> later
[20:01:47] <-- brad_a has left #gemrb
[20:02:26] <tomprince> Avenger: what are you seeing?
[20:04:28] <-- edheldil has left IRC (Remote host closed the connection)
[20:06:04] <Avenger> without my change: npclevels leaks
[20:06:32] <Avenger> with change: ==4187== Mismatched free() / delete / delete []
[20:06:41] <Avenger> by 0x4EABEA1: Game::~Game() (Game.cpp:199)
[20:08:00] <wjp> that means you call free() on something allocated with new[]/new, or delete[] on something allocated with malloc/new, or delete on something allocated with new[]/malloc
[20:08:48] <Avenger> i don't call free, i call delete
[20:08:56] <tomprince> And it should be delete[]
[20:09:00] <Avenger> delete npclevels[i][j]
[20:09:27] <Avenger> hmm, delete [] npclevels[i][j] ?
[20:09:32] <tomprince> Yes.
[20:09:35] <tomprince> That should do it.
[20:10:13] <tomprince> I you should need any of the .clear()s there either.
[20:10:18] <tomprince> s/I/And/
[20:11:13] --> Baldurer has joined #gemrb
[20:11:33] <Avenger> ok, i guess you wanted to say I should NOT need the clear
[20:12:01] <tomprince> That is what I meant to say, yes. :)
[20:14:43] <CIA-36> GemRB: 03avenger_teambg * r8126b86a6755 10gemrb/gemrb/core/Game.cpp: proper delete of npclevels
[20:14:48] <Avenger> thanks
[20:15:20] <wjp> hm, if all we store in there is char[9]'s, shouldn't need the manual dynamic allocation at all
[20:15:50] <Avenger> it is actually an ieResRef type
[20:15:58] <Avenger> at least, it should be
[20:16:19] <tomprince> I think the problem is that we can't have a vector of char[9], at least until C++11, since char[9] isn't copyable. :(
[20:18:23] <Avenger> well, i'm happy with this solution, no leaks, no mismatch
[20:18:28] <Avenger> now only fonts leak
[20:20:07] <Avenger> any idea what causes this leak
[20:20:11] <Avenger> ==7036== 4,209 (168 direct, 4,041 indirect) bytes in 3 blocks are definitely lost in loss record 429 of 454
[20:20:13] <Avenger> ==7036== at 0x4C27CC1: operator new(unsigned long) (vg_replace_malloc.c:261)
[20:20:15] <Avenger> ==7036== by 0x65141CA: SDLVideoDriver::CreateSprite8(int, int, int, void*, void*, bool, int) (SDLVideo.cpp:807)
[20:20:16] <Avenger> ==7036== by 0xB7E8E77: BAMImporter::GetFont(unsigned short, unsigned short) (BAMImporter.cpp:334)
[20:20:18] <Avenger> ==7036== by 0xB7E99DF: BAMFontManager::GetFont(unsigned short, unsigned short, unsigned short, FontStyle, Palette*) (BAMFontManager.cpp:43)
[20:20:20] <Avenger> ==7036== by 0x4EBDFA8: Interface::LoadSprites() (Interface.cpp:1402)
[20:20:21] <Avenger> ==7036== by 0x4EC50DB: Interface::Init() (Interface.cpp:1720)
[20:20:22] <Avenger> ==7036== by 0x400D91: main (GemRB.cpp:84)
[20:20:46] <Avenger> the weird thing is, it is only 3 blocks, we have more fonts or sprites in fonts
[20:20:58] <Avenger> so it is only one type of fonts?
[20:25:44] <CIA-36> GemRB: 03tom.prince * r61be930a7142 10gemrb/gemrb/core/Game.cpp: Game: Allocate the right sized vectors for npclevels.
[20:26:29] <Avenger> hehe right, i was sure cols*rows is not correct
[20:27:16] <Avenger> can you take a peek at font creation too?
[20:27:23] <wjp> numeric fonts look fishy
[20:28:02] <wjp> see BAMImporter.cpp line 303
[20:28:06] <-- PixelScum has left IRC (Ping timeout: 248 seconds)
[20:28:20] <Avenger> the problem is somewhere around BAMImporter.cpp line 315
[20:28:22] <wjp> it allocates 11 glyphs, but then tells the Font constructor it has chars '0' up to '9'
[20:29:09] <Avenger> the problem is ==7036== by 0xB7E8E77: BAMImporter::GetFont(unsigned short, unsigned short) (BAMImporter.cpp:334)
[20:29:17] <wjp> (so the 11th glyph is leaked)
[20:29:26] <Avenger> ahh you are right
[20:29:39] <Avenger> yes that would explain why i have only a few leaks
[20:31:20] <Avenger> what is the next char after 9...
[20:31:29] --> Drakkar has joined #gemrb
[20:31:30] <Avenger> the numeric fonts contain 'lbs'
[20:31:36] <tomprince> '9'+1
[20:31:42] <Avenger> hehe
[20:31:46] <wjp> it's a colon
[20:32:17] <Avenger> yes :
[20:33:28] <wjp> but '9'+1 is clearer :-)
[20:33:40] <Avenger> Wouldn't '0'+Count-1 the best?
[20:33:58] <Avenger> or maybe '0'+10 :)
[20:34:10] <wjp> up to you :-)
[20:34:29] <wjp> (as long as you paint the bikeshed grey, of course)
[20:42:53] <Avenger> i think i will just remove that stuff
[20:44:35] <wjp> it's not the most readable function, no
[20:48:01] <Avenger> well, i don't like it restricting the numeric fonts to 11
[20:48:09] <Avenger> so using if (FirstChar+Count != (unsigned int) LastChar+1) { error
[20:48:26] <Avenger> and fixing the font.2da's to properly use 58 as last char for numeric fonts
[20:50:10] <CIA-36> GemRB: 03avenger_teambg * r3aca0ab3267d 10gemrb/gemrb/core/Game.cpp: Merge branch 'master' of ssh://gemrb.git.sourceforge.net/gitroot/gemrb/gemrb
[20:50:23] <CIA-36> GemRB: 03avenger_teambg * r05e6586b9867 10gemrb/gemrb/ (7 files in 7 dirs): fixed a leak and removed some glyph count restriction of numeric fonts
[20:51:03] <tomprince> That seems sensible.
[20:52:20] <-- Baldurer has left IRC (Ping timeout: 276 seconds)
[21:07:17] <gembot> build #387 of cmake g++-4.2.4 is complete: Failure [failed minimal test] Build details are at http://buildbot.gemrb.org/builders/cmake%20g%2B%2B-4.2.4/builds/387 blamelist: avenger_teambg@sourceforge.net
[21:08:35] <wjp> numtest.bam?
[21:09:40] --> SiENcE has joined #gemrb
[21:10:49] <wjp> I suppose fonts.2da for tests/minimal still has the wrong firstchar/lastchar
[21:11:27] --> Baldurer has joined #gemrb
[21:11:41] <CIA-36> GemRB: 03tom.prince * r34287ac56e99 10gemrb/gemrb/tests/minimal/data/ (fonts.2da numtest.bam): Fix minimal tests for saner font handling in 05e6586b9867b28b1e2b5e91798321ddf9d4ff7d.
[21:11:51] <gembot> build #388 of autotools g++-4.4.5 is complete: Failure [failed minimal test] Build details are at http://buildbot.gemrb.org/builders/autotools%20g%2B%2B-4.4.5/builds/388 blamelist: avenger_teambg@sourceforge.net
[21:11:53] <gembot> build #384 of autotools g++-4.2.4 is complete: Failure [failed minimal test] Build details are at http://buildbot.gemrb.org/builders/autotools%20g%2B%2B-4.2.4/builds/384 blamelist: avenger_teambg@sourceforge.net
[21:14:33] <gembot> build #382 of cmake clang++ is complete: Failure [failed minimal test] Build details are at http://buildbot.gemrb.org/builders/cmake%20clang%2B%2B/builds/382 blamelist: avenger_teambg@sourceforge.net
[21:17:16] <gembot> build #379 of mingw32 is complete: Failure [failed minimal test] Build details are at http://buildbot.gemrb.org/builders/mingw32/builds/379 blamelist: avenger_teambg@sourceforge.net
[21:18:49] <gembot> build #375 of autotools clang++ is complete: Failure [failed minimal test] Build details are at http://buildbot.gemrb.org/builders/autotools%20clang%2B%2B/builds/375 blamelist: avenger_teambg@sourceforge.net
[21:26:52] <gembot> build #388 of cmake g++-4.2.4 is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/cmake%20g%2B%2B-4.2.4/builds/388
[21:31:38] <gembot> build #385 of autotools g++-4.2.4 is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/autotools%20g%2B%2B-4.2.4/builds/385
[21:31:44] <-- SiENcE has left IRC (Quit: cya)
[21:32:54] <gembot> build #383 of cmake clang++ is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/cmake%20clang%2B%2B/builds/383
[21:38:51] <gembot> build #376 of autotools clang++ is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/autotools%20clang%2B%2B/builds/376
[21:43:16] <gembot> build #390 of autotools g++-4.4.5 is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/autotools%20g%2B%2B-4.4.5/builds/390
[21:43:51] --> DrMcCoy has joined #gemrb
[21:46:09] <-- Baldurer has left IRC (Read error: Connection reset by peer)
[21:50:00] <gembot> build #82 of nmake-msvc++10 is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/nmake-msvc%2B%2B10/builds/82
[22:01:12] --> PixelScum has joined #gemrb
[22:01:22] <gembot> build #380 of mingw32 is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/mingw32/builds/380
[22:03:29] <-- Drakkar has left IRC (Ping timeout: 252 seconds)
[22:13:47] --> Drakkar has joined #gemrb
[22:15:10] <Avenger> heh thanks for fixing the minimal test
[22:16:24] <-- PixelScum has left IRC (Ping timeout: 244 seconds)
[22:16:27] <Yoshimo> usually buildbot is a bit more quiet
[22:20:25] <-- Avenger has left IRC (Quit: bye!)
[22:24:29] <-- Yoshimo has left IRC (Quit: Yoshimo)
[22:24:43] --> PixelScum has joined #gemrb
[22:27:21] <-- Drakkar has left IRC (Ping timeout: 240 seconds)
[22:32:04] <tomprince> Although it seems that there are more valgrind errors, when I run the minimal test on my desktop.
[22:44:21] --> brad_a has joined #gemrb
[22:45:25] <brad_a> heh. i actually knew about the font leak. I had been meaning to find out what the extraneous characters were in numeric fonts and fix it apropriately. obviously that slipped my mind :-P thaks for fixing
[22:46:02] <brad_a> but since it got brought up id like to make sure TTF fonts have appropriate replacement glyphs for whatever those extra characters are
[22:46:22] <brad_a> i saw one was 'lbs' *shudders*
[22:46:52] <brad_a> what are the others so i can have the ttf importer create custom replacements?
[22:48:44] <brad_a> tomprince: any further thoughts on the error alerts? have we come to the conclusion we should just use separate source files?
[22:49:44] <tomprince> I haven't any thoughts. I'd like to hear fuzzies thoughts first.
[22:50:29] <fuzzie> am confused as to question
[22:50:41] <fuzzie> can confirm that android log is split into a lot of lines and it is annoying, if that helps
[22:51:38] <fuzzie> If it's about guialert - seperate files makes sense to me.
[22:51:38] <tomprince> See the discussion before that, on adding gui errors on osx/ios, and how to structure the code.
[22:52:32] <tomprince> I tried https://gist.github.com/1597262 but apparently a bunch of files are technically corrupt, or something, and we depend on the silent failure. :(
[22:52:36] <fuzzie> I'm not so sure about logging in general. Really it would be nice to have platform subclasses there and etc and sigh..
[22:53:14] <tomprince> Platform subclasses?
[22:53:53] <fuzzie> e.g. AndroidSystem which re-implemented the logging functionality and the GUI alerts from a superclass with default functionality, or somesuch
[22:54:34] <fuzzie> and us depending on silent failure in some places seems like a very good thing to know about and look into further. but I really don't have time right now.
[22:54:47] <brad_a> yes i think a logging class would be swell
[22:55:03] <brad_a> then i only have to subclass and override
[22:55:42] <fuzzie> Well it is my first thought as to a nice way to build a nice solution, but obviously you'd end up splitting into files anyway, so splitting into files seems a good first step even if it's a silly idea.
[22:55:58] <tomprince> fuzzie: Why the sigh?
[22:56:13] <gembot> build #50 of osx-test is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/osx-test/builds/50
[22:56:19] <fuzzie> tomprince: I am so terribly spoiled by projects which, well, aren't GemRB. :-)
[22:56:27] <tomprince> brad_a: I'd say go ahead and turn stuff into a class, if you want.
[22:56:40] <brad_a> i can give it a shot
[22:56:57] <brad_a> if somebody has a basic idea of methods/structure that would help
[22:58:20] <tomprince> Well, we essentialy have a bunch of methods. They just aren't wrapped in a class.
[22:58:23] <brad_a> jsut do like core and have a global singleton?
[22:59:00] <brad_a> or have member for core?
[22:59:07] <brad_a> a member of the logging class
[22:59:19] <tomprince> Except that they are currently insufficiently stateful (see the discussion re android logs)
[22:59:23] <brad_a> probably the first one so we can log without core
[22:59:24] <fuzzie> Well, either way works.
[22:59:37] <fuzzie> Right.
[22:59:52] <fuzzie> I would guess you probably want some wrapper functions so you can still just call error(), but not sure I have any input other than that.
[23:00:15] <brad_a> right
[23:00:18] <tomprince> I think I would be inclined to just have free functions, to hide the fact that you have a singleton. (i.e. what fuzzie said)
[23:00:33] <brad_a> dont feel like rewriting all the calls to error() atm :-P
[23:13:07] <tomprince> brad_a: Did you say you had valgrind on your machine?
[23:13:14] <brad_a> yes
[23:15:00] <-- gembot has left IRC (Quit: buildmaster reconfigured: bot disconnecting)
[23:15:16] --> gembot has joined #gemrb
[23:19:31] <gembot> build #51 of osx-test is complete: Failure [failed minimal test] Build details are at http://buildbot.gemrb.org/builders/osx-test/builds/51 blamelist: Avenger <avenger_teambg@sourceforge.net>, Brad Allred <bradallred@me.com>, Jaka Kranjc <lynxlupodian@users.sourceforge.net>, pintin neil <npintin@gmail.com>, Tom Prince <tom.prince@ualberta.net>
[23:19:32] <gembot> build #389 of cmake g++-4.2.4 is complete: Failure [failed minimal test] Build details are at http://buildbot.gemrb.org/builders/cmake%20g%2B%2B-4.2.4/builds/389 blamelist: Avenger <avenger_teambg@sourceforge.net>, Brad Allred <bradallred@me.com>, Jaka Kranjc <lynxlupodian@users.sourceforge.net>, pintin neil <npintin@gmail.com>, Tom Prince <tom.prince@ualberta.net>
[23:23:08] <gembot> build #52 of osx-test is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/osx-test/builds/52
[23:27:18] <brad_a> out of curiosity sake i really do want to know how to resolve this ambiguous Point symbol business without using that define
[23:27:51] <tomprince> Did you try the namespace hack I suggested?
[23:27:57] <brad_a> yes
[23:27:57] <gembot> build #391 of autotools g++-4.4.5 is complete: Failure [failed minimal test] Build details are at http://buildbot.gemrb.org/builders/autotools%20g%2B%2B-4.4.5/builds/391 blamelist: Avenger <avenger_teambg@sourceforge.net>, Brad Allred <bradallred@me.com>, Jaka Kranjc <lynxlupodian@users.sourceforge.net>, pintin neil <npintin@gmail.com>, Tom Prince <tom.prince@ualberta.net>
[23:27:57] <gembot> build #386 of autotools g++-4.2.4 is complete: Failure [failed minimal test] Build details are at http://buildbot.gemrb.org/builders/autotools%20g%2B%2B-4.2.4/builds/386 blamelist: Avenger <avenger_teambg@sourceforge.net>, Brad Allred <bradallred@me.com>, Jaka Kranjc <lynxlupodian@users.sourceforge.net>, pintin neil <npintin@gmail.com>, Tom Prince <tom.prince@ualberta.net>
[23:28:02] <tomprince> gembot: mute
[23:28:02] <gembot> Shutting up for now.
[23:28:21] <brad_a> i tried it and thats what gave me the ambiguous symbol
[23:28:32] <brad_a> before i was getting a redefinition of point error
[23:28:39] <tomprince> build log
[23:30:14] <brad_a> sure. the errors are all when compiling the logging code so i will just paste that log
[23:30:23] <tomprince> Also, does 'valgrind --suppresions=<gemrb_root>/testing/python.supp gemrb work for you?
[23:31:09] <tomprince> that should be 'valgrind --suppressions=<gemrb_root>/testing/python.supp gemrb -c <config>'
[23:31:12] <brad_a> http://dl.dropbox.com/u/13866402/buildlog.txt
[23:32:14] <tomprince> What happens if you put a using decl after the includes in Logging.cpp
[23:32:49] <brad_a> im sorry what?
[23:33:01] <tomprince> Ignore me ...
[23:35:40] <brad_a> trying the valgrind thing now
[23:37:59] <tomprince> What happens if you move those includes to the top of Logging.cpp (the apple ones)
[23:38:32] <brad_a> i shall try momentaryinly. im building a good build to try the valgrind business
[23:41:37] <brad_a> what is the expected output?
[23:42:21] <tomprince> I just expect it to run succesfully. I am trying to figure out why http://buildbot.gemrb.org/builders/osx-test/builds/52/steps/minimal%20test/logs/stdio failed to run properly.
[23:43:20] <brad_a> well that command fails. it says -c no such file or directory
[23:44:06] <tomprince> !?!?!
[23:44:28] <brad_a> lol yes i dont understand. ill paste exactly what i used
[23:44:40] --> joneirik has joined #gemrb
[23:44:47] <brad_a> valgrind GemRB.app/Contents/MacOS/GemRB -suppressions=~/Developer/gemrb/testing/python.supp -c ~/Library/Application\ Support/GemRB/GemRB.cfg
[23:45:01] <brad_a> gives error "-c: No such file or directory"
[23:45:34] <brad_a> hm
[23:45:39] <brad_a> i see i left out the second -
[23:45:52] <tomprince> The --suppressions need to be before the gemrb exe
[23:46:19] <brad_a> well i just tried it the way i had it except as --supressions and it seems to like it
[23:46:27] <brad_a> oh maybenot
[23:48:10] <brad_a> valgrind --suppressions=~/Developer/gemrb/testing/python.supp GemRB.app/Contents/MacOS/GemRB -c ~/Library/Application\ Support/GemRB/GemRB.cfg
[23:48:25] <brad_a> that says --suppressions=~/Developer/gemrb/testing/python.supp: No such file or directory
[23:48:37] <brad_a> time to man
[23:48:55] <tomprince> Try without the =
[23:49:33] <tomprince> Except that doesn't work here ...
[23:49:44] <brad_a> that jsut says supressions no such file blah blah
[23:50:07] <tomprince> What version of valgrind do you have? And what happens if you replace ~ with ${HOME}.
[23:50:19] <tomprince> Not every shell expands ~ after and =
[23:50:24] <tomprince> s/and/an/
[23:50:55] <brad_a> ive never had issues with ~ and valgrind before
[23:51:16] <brad_a> and just using the standard mac shell (bash)
[23:54:29] <brad_a> absolute path same effect
[23:56:04] <brad_a> something doesn seem to be wrong with the environment/shell tho
[23:57:52] <-- Maighstir has left IRC (Quit: .)
[23:59:29] <brad_a> it seems to want a file first...
[23:59:36] <brad_a> no idea why