#gemrb@irc.freenode.net logs for 14 Mar 2013 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:02:17] <-- i30817 has left IRC (Ping timeout: 256 seconds)
[00:02:17] --- i30817_ is now known as i30817
[00:17:47] <-- edheldil has left IRC (Remote host closed the connection)
[02:33:11] <-- i30817 has left #gemrb
[03:10:03] <brada> ha ha ha GUI config is awesome :D
[03:28:40] <-- brada has left IRC (Quit: brada)
[03:29:11] --> brada has joined #gemrb
[03:49:26] <-- brada has left IRC (Quit: brada)
[04:21:56] --> brada has joined #gemrb
[04:39:34] <-- brada has left IRC (Quit: brada)
[08:16:32] --> edheldil has joined #gemrb
[08:16:32] --- ChanServ gives channel operator status to edheldil
[08:56:56] --> lynxlynxlynx has joined #gemrb
[08:56:56] --- ChanServ gives channel operator status to lynxlynxlynx
[09:13:06] --> WingedHussar has joined #gemrb
[12:26:30] <-- WingedHussar has left IRC (Ping timeout: 256 seconds)
[12:54:14] --> kida has joined #gemrb
[13:06:28] --> WingedHussar has joined #gemrb
[13:26:39] <-- WingedHussar has left IRC (Ping timeout: 276 seconds)
[13:50:03] --> WingedHussar has joined #gemrb
[14:11:19] --> brada has joined #gemrb
[14:37:24] <-- WingedHussar has left IRC (Ping timeout: 256 seconds)
[14:49:49] --> WingedHussar has joined #gemrb
[14:53:05] --> mikol79 has joined #gemrb
[14:56:12] <mikol79> I seem to be getting errors when compiling from git on Xubuntu 12.04, 32 bit. Seems to have started happening after yesterday's commits. Here's the paste: http://paste.ubuntu.com/5613792/
[14:56:13] <Seniorita> Ubuntu Pastebin
[14:58:16] <mikol79> Tried several times with a normal build and a debug build: it is always to do with VFS.cpp.o
[15:00:33] <fuzzie> hi
[15:01:36] <fuzzie> it is indeed broken
[15:01:40] <fuzzie> let me fix that
[15:02:30] <mikol79> Thanks
[15:02:41] <fuzzie> thanks for the report :)
[15:03:03] <fuzzie> of course it doesn't work for me now
[15:03:17] <fuzzie> brada: phffffft, I say
[15:03:24] <fuzzie> but that is a different problem.
[15:04:05] <brada> my bad :D
[15:04:23] <fuzzie> brada: gemrb no longer reads from '.'
[15:04:28] <brada> really?
[15:04:33] <fuzzie> so it says, I didn't find a gemrb.cfg in these weird places, so I will quit.
[15:04:52] <brada> let me look
[15:04:57] <fuzzie> mikol79: should work now btw.
[15:04:58] <brada> give me a min to finish what im doing
[15:05:02] <fuzzie> brada: no rush.
[15:05:16] <fuzzie> just probably not a good idea for me to go messing with it without knowing what I'm doing.
[15:12:00] <brada> fuzzie: is DATADIR defined by default when you build?
[15:12:03] <mikol79> fuzzie: thanks, it did compile this time without errors
[15:12:06] <brada> that could be the problem
[15:12:59] <brada> oh i see
[15:13:44] <fuzzie> brada: I think the problem here is the bit where you removed the "FIXME: temporary hack, to be deleted??" bit without actually adding a non-hack. :P
[15:13:55] <brada> sort of
[15:14:24] <brada> its that the data path only gets set to ./ if there is no home path
[15:14:52] <brada> i should check if home + gemrb exists
[15:16:04] <brada> do we have any path exists functions?
[15:16:35] <fuzzie> can't you just try all sensible paths?
[15:16:59] <fuzzie> and, i have no idea
[15:17:18] <fuzzie> dir_exists/file_exists?
[15:20:34] <mikol79> Anyway, must go now, will report any further issues. Thanks again.
[15:20:37] <-- mikol79 has left #gemrb
[15:34:18] <brada> fuzzie: the question is i guess what is a non-hack for that
[15:35:43] <brada> is that even a hack?
[15:37:07] <brada> and should i check it first or last or what?
[15:38:50] <fuzzie> not sure the order matters
[15:39:15] <fuzzie> it was a hack because it hard-coded 'GemRB' in the fielname if nothing else
[15:39:18] <fuzzie> i guess
[15:48:49] <brada> hopefully that didnt make things worse ;)
[15:52:13] <fuzzie> $ gemrb/gemrb
[15:52:13] <fuzzie> Segmentation fault (core dumped)
[15:52:24] <fuzzie> that seems fairly worse :P
[15:52:48] <edheldil> hehe
[15:53:25] <fuzzie> brada: you construct CFGConfig before you construct Interface
[15:53:27] <fuzzie> --> crash
[15:54:22] <edheldil> I deemed it a hack because it loaded from the current dir
[15:54:37] <fuzzie> edheldil: we all use that though :p
[15:54:50] <brada> im on spring holiday from school so ive nothing better to do then work on gemrb
[15:55:01] <edheldil> I know, I have never deleted it, did I? :-D
[15:55:01] <fuzzie> brada: this isn't platform-specific code so presumably you can also reproduce it, as long as you don't pass '-c' to gemrb?
[15:55:08] <fuzzie> edheldil: :)
[15:55:20] <brada> fuzzie: mac uses a diffrent init code
[15:55:34] <brada> i must have messed something up on the shared side
[15:55:36] <fuzzie> ah.
[15:55:40] <fuzzie> but still
[15:55:43] <brada> yes
[15:55:54] <fuzzie> should break on iOS at least, and anything using GemRB.cpp too
[15:56:05] <brada> ios doesnt use gemrb.cpp either
[15:56:05] <edheldil> why not check %UserSettings% or whatever on windows, btw?
[15:56:10] <fuzzie> brada: no, but iOS has the same flaw
[15:56:13] <brada> ed: i dont know
[15:56:16] <brada> dont use windows :p
[15:56:21] <fuzzie> well, assuming ios/GEM_AppDelegate.mm is iOS
[15:56:27] <fuzzie> if it isn't then I have to ask, what is it? :P
[15:56:29] <brada> yes
[15:56:36] <brada> it worked last night
[15:56:41] <fuzzie> but obviously if you pass '-c' then it works.
[15:56:42] <brada> well after i made that change....
[15:56:46] <brada> ah
[15:56:47] <fuzzie> it's just that you have cleverly broken the other case.
[15:56:55] <brada> gotcha
[15:56:56] <fuzzie> because you call functions which need core, and you didn't construct core yet.
[15:57:00] <fuzzie> make sense?
[15:57:03] <brada> yes
[15:57:09] <brada> ios always passes -c (for now)
[15:58:34] <edheldil> btw, since you mess with it, is not it the right time to pull LoadConfig outside of Interface? :)
[15:58:38] <gembot> build #304 of nmake-msvc++6 is complete: Failure [4failed compile] Build details are at http://buildbot.gemrb.org/builders/nmake-msvc%2B%2B6/builds/304 blamelist: Alyssa Milburn <fuzzie@fuzzie.org>, Brad Allred <bradallred@me.com>
[16:00:44] <-- WingedHussar has left IRC (Quit: WingedHussar)
[16:02:57] <lynxlynxlynx> EnableCheatKeys is on by default btw
[16:03:40] <edheldil> it should not
[16:03:56] <fuzzie> brada: the 'int i' you added in the for loop on Interface.cpp:1579 isn't allowed
[16:04:40] <brada> by gcc?
[16:04:51] <fuzzie> by msvc6
[16:05:00] <brada> ah
[16:06:05] <brada> what do you have to declare it outside the loop?
[16:06:13] <fuzzie> yes
[16:06:18] <fuzzie> msvc6 defines it *above* the loop
[16:06:28] <fuzzie> and so it breaks because there's another 'int i' defined later in the function
[16:06:52] <brada> oh i see
[16:07:36] <fuzzie> mushing this all into one even huger function isn't very nice maybe
[16:07:46] <fuzzie> it was crazy enough as it was :P
[16:07:49] <brada> yes
[16:07:55] <brada> let me fix it first
[16:08:08] <brada> then we can talk about what best to do
[16:13:14] <-- kida has left IRC (Ping timeout: 255 seconds)
[16:23:36] <brada> i wonder how long NullSource plugin has been missing from mac/ios builds
[16:42:06] <gembot> build #309 of msvc++6 is complete: Failure [4failed compile] Build details are at http://buildbot.gemrb.org/builders/msvc%2B%2B6/builds/309 blamelist: Brad Allred <bradallred@me.com>
[16:44:58] <brada> 1860 is a blank line...
[16:49:49] <fuzzie> that's still an old revision
[16:54:10] <brada> ah
[16:56:25] <brada> fuzzie: is that stillbroken btw?
[16:56:46] <brada> i am able to have CFGConfig locate and init gemrb without -c
[16:57:18] <brada> i am having a problem with NullSource tho
[16:57:30] <brada> oh
[16:57:31] <brada> god
[16:57:54] <brada> i think i have an ancient nullsource lingering around in the app support folder from days of yore
[16:58:51] <brada> at least im gettin an already loaded error when it tries to load the one in the bundle
[17:00:30] <brada> bbl
[17:00:33] <-- brada has left IRC (Quit: brada)
[17:01:54] --> brada has joined #gemrb
[17:02:56] <-- brada has left IRC (Client Quit)
[17:23:30] <gembot> build #310 of msvc++6 is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/msvc%2B%2B6/builds/310
[17:27:43] <fuzzie> \o/
[17:32:23] <gembot> build #305 of nmake-msvc++6 is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/nmake-msvc%2B%2B6/builds/305
[18:19:38] --> brada has joined #gemrb
[19:02:13] --> rocket_hamster has joined #gemrb
[19:03:53] <brada> fuzzie: if you have any suggestions for improving the core config/init process i have plenty of free time till next week.
[19:04:30] <brada> i imagine there are a number of one off variables in core that we can get rid of
[19:04:41] <brada> i dont see why core needs width and height for example
[19:06:57] <fuzzie> well, because the config code is all mixed up with it.
[19:07:41] <fuzzie> but certainly splitting up Init would be a good idea..
[19:08:57] <brada> because it is massive?
[19:08:59] <fuzzie> yes
[19:09:28] <fuzzie> at some point I think wjp was trying to move more of the game data stuff into GameData but that stalled due to lack of time etc
[19:12:35] <fuzzie> but in any case..
[19:12:43] <fuzzie> you don't have things you'd rather be doing? :)
[19:12:50] <lynxlynxlynx> you can also look at adding fontconfig paths for font lookup
[19:13:20] <brada> not sure how to do that xplatform
[19:18:04] <brada> i an make a stub function with mac functionality
[19:18:13] <brada> somebody else would have to add other platforms
[19:19:07] <lynxlynxlynx> i don't think this would work on windows anyway
[19:19:25] <lynxlynxlynx> not even sure if macs use fontconfig
[19:20:27] <brada> kinda hard for me to implement it then :)
[19:25:40] --> Yoshimo has joined #gemrb
[19:26:03] <lynxlynxlynx> needs exploring
[19:35:15] <-- brada has left IRC (Quit: brada)
[19:36:45] --> brada has joined #gemrb
[22:02:56] <-- Yoshimo has left IRC (Quit: Yoshimo)
[22:10:47] <-- rocket_hamster has left IRC (Remote host closed the connection)
[22:33:20] <-- brada has left IRC (Quit: brada)
[22:36:49] --> brada has joined #gemrb
[22:38:21] --- ermo^ is now known as ermo
[22:40:34] <-- brada has left IRC (Client Quit)
[22:55:27] --- ermo is now known as ermo^
[23:00:54] --> brada has joined #gemrb
[23:24:21] --> rocket_hamster has joined #gemrb
[23:34:16] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[23:49:54] <rocket_hamster> http://devopsreactions.tumblr.com/post/42345786031/senior-developer-explaining-how-to-use-his-library
[23:49:57] <Seniorita> Senior Developer explaining how to use his library - DevOps Reactions